Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Sustainability evaluation of transportation infrastructure under uncertainty : a fuzzy-based approach Umer, Adil 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2015_may_umer_adil.pdf [ 2.96MB ]
Metadata
JSON: 24-1.0074439.json
JSON-LD: 24-1.0074439-ld.json
RDF/XML (Pretty): 24-1.0074439-rdf.xml
RDF/JSON: 24-1.0074439-rdf.json
Turtle: 24-1.0074439-turtle.txt
N-Triples: 24-1.0074439-rdf-ntriples.txt
Original Record: 24-1.0074439-source.json
Full Text
24-1.0074439-fulltext.txt
Citation
24-1.0074439.ris

Full Text

SUSTAINABILITY EVALUATION OF TRANSPORTATION INFRASTRUCTURE UNDER UNCERTAINTY: A FUZZY-BASED APPROACH  by  Adil Umer  BSc. Civil Engineering, University of Engineering & Technology Lahore, 2011    A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF   MASTER OF APPLIED SCIENCE  in  THE COLLEGE OF GRADUATE STUDIES (Civil Engineering)   THE UNIVERSITY OF BRITISH COLUMBIA (Okanagan)    April 2015   © Adil Umer, 2015 ii  Abstract The construction and maintenance of transportation infrastructure consume significant natural resources, produces considerable waste and uses extensive human capital. Sustainability evaluation of alternative initiatives and policies for developing transportation infrastructure enables decision makers to make informed choices. Despite the availability of numerous sustainability rating tools for roadway infrastructure, there is a need to develop customizable sustainability evaluation tools for informed decision-making. Such tools, unlike the rating systems, ideally need to handle uncertain data, incorporate expert opinion and adapt to project and geographic specific constraints. Deterministic approaches for life cycle cost analysis (LCCA) and life cycle assessment (LCA) have been extensively applied to select sustainable pavement alternatives. However, the information used to conduct LCCA and LCA is often imprecise and vague in early project phases. Therefore, certain technique is required to incorporate and propagate such uncertainties so that the reliability of final results is transparent. Unlike probabilistic methods, fuzzy based techniques are more appropriate to handle uncertainties due to vagueness and imprecision in a computationally efficient manner.   This study aimed to investigate the use of fuzzy logic to evaluate sustainability under uncertainty at two levels of infrastructures - Roadways as systems and pavements as components. A novel roadway sustainability evaluation framework was developed using indicators from existing green rating system. A customizable excel-based tool was programmed based on the framework to estimate the sustainability index (SI) of roadways under uncertainty using fuzzy synthetic evaluation (FSE) technique. The FSE technique enables the tool to evaluate reliable and informative SI by incorporating expert opinion. Moreover, fuzzy composite programming (FCP) technique was used to estimate the life cycle environmental and economic sustainability indices (SIs) from LCA and LCCA of pavement alternatives under uncertainty. The FCP technique improved the reliability of final results by propagating input uncertainties to the outputs. Scenario analysis was performed using FSE and FCP techniques to demonstrate the influence of uncertainties and decision maker’s preferences on the overall SI of roadways and pavements respectively. This study demonstrated a compelling utility of fuzzy-based techniques to evaluate sustainability under uncertainty in the early project phases for informed decision-making.   iii  Preface This research has been conducted and prepared in the form of thesis and papers by the author under the supervision of Drs. Rehan Sadiq and Kasun Hewage.   First journal article is being prepared based on Chapter 2 and 3 for publication in the Journal of Cleaner Production titled “Green Proforma for Roadway Infrastructures: Evaluating Sustainability using Fuzzy Synthetic Evaluation”.  Second journal article is being prepared based on Chapter 2 and 4 for publication in the Journal of Infrastructure Systems titled “Evaluating Economic and Environmental Sustainability of Pavements under Uncertainty using Fuzzy Composite Programming”.    iv  Table of Contents Abstract ................................................................................................................................................. ii Preface .................................................................................................................................................. iii List of Tables ........................................................................................................................................ vi List of Figures .................................................................................................................................... viii List of Illustrations .............................................................................................................................. ix List of Abbreviations and Acronyms ................................................................................................. xi Acknowledgements ............................................................................................................................ xiii Dedication........................................................................................................................................... xiv Chapter 1 Introduction .................................................................................................................... 1 1.1 Background ....................................................................................................................... 1 1.2 Research Objectives.......................................................................................................... 3 1.3 Research Outline and Methodology ................................................................................. 4 Chapter 2 Sustainability Evaluation Techniques .......................................................................... 6 2.1 Sustainability in Civil Engineering Projects ..................................................................... 6 2.2 Evaluating Sustainability of Roadway Infrastructure ....................................................... 7 2.2.1 Rating Systems and their Deficiencies ......................................................................... 8 2.3 Evaluating Sustainability of Pavements ......................................................................... 10 2.3.1 Pavements Types ........................................................................................................ 11 2.3.2 Characterization of Pavement Alternatives ................................................................ 13 2.3.3 Maintenance and Rehabilitation of Pavements .......................................................... 15 2.3.4 Life Cycle Thinking for Pavements ........................................................................... 18 2.4 Uncertainty in Evaluating Sustainability ........................................................................ 23 2.4.1 Types of Uncertainties involved in Sustainability Evaluation ................................... 24 2.4.2 Fuzzy Sets and Fuzzy Numbers ................................................................................. 26 2.4.3 Fuzzy Operations ........................................................................................................ 28 2.4.4 Fuzzy Synthetic Evaluation ........................................................................................ 30 2.4.5 Fuzzy Composite Programming ................................................................................. 33 Chapter 3 Sustainability Evaluation of Roadway Infrastructure.............................................. 35 3.1 Multicriteria Analysis: Sustainability Evaluation of Roadways ..................................... 35 3.2 Green Proforma: Sustainability Evaluation Framework ................................................. 36 3.3 Green Proforma: Sustainability Evaluation Process ....................................................... 39 3.4 Fuzzification of Indicators .............................................................................................. 41 3.4.1 Defining and Fuzzifying Benchmarks ........................................................................ 41 v  3.4.2 Fuzzifying Inputs and Generating Fuzzy Sets ............................................................ 43 3.5 Prioritization and Aggregation ....................................................................................... 45 3.6 Defuzzification and Result Generation ........................................................................... 45 3.7 Scenario Analysis using Green Proforma ....................................................................... 46 3.8 Summary and Discussion ............................................................................................... 53 Chapter 4 Sustainability Evaluation of Pavements ..................................................................... 55 4.1 Life Cycle Thinking Approach for Pavements ............................................................... 55 4.2 Development of Scenarios .............................................................................................. 56 4.2.1 Input Parameters for Characterizing Alternatives ...................................................... 58 4.2.2 Pavement Section Designs ......................................................................................... 61 4.3 Life Cycle Thinking Approach ....................................................................................... 64 4.3.1 Uncertainty Analysis .................................................................................................. 64 4.3.2 Maintenance and Rehabilitation Schedules ................................................................ 65 4.3.3 Implementation of LCA ............................................................................................. 67 4.3.4 Implementation of LCCA ........................................................................................... 70 4.4 Normalization and Aggregation of Indicators ................................................................ 73 4.5 Summary and Discussion ............................................................................................... 80 Chapter 5 Conclusion and Recommendations............................................................................. 82 References ........................................................................................................................................... 85 Appendices .......................................................................................................................................... 96 Appendix A1: Sustainability Criteria, Indicator and Benchmark List based on Greenroads TM      v 1.5 Manual (Muench et al. 2011) .............................................................................................. 96 Appendix A2: Excel Interface for the Green Proforma .............................................................. 100 Appendix B1: Life Cycle Assessment Results showing fuzzy numbers for Environmental   Impact Indicators for each Scenario ........................................................................................... 105 Appendix B2: Comparison of Environmental and Economic Sustainability Indices of Alternatives for CBR 10% .......................................................................................................... 112 Appendix B3: Aggregated Sustainability Indices of Alternatives for CBR 10% ....................... 114     vi  List of Tables Table 2.1 M&R strategies..................................................................................................................... 16 Table 2.2 TFN for the linguist scale ..................................................................................................... 28 Table 2.3 Arithmetic operations on fuzzy numbers with only low and high values ............................ 29 Table 3.1 Input for the benchmark of indicators (data from Muench et al. 2011) ............................... 42 Table 3.2 Example indicator input value (indicator from Muench et al. 2011) ................................... 44 Table 3.3 Fuzzy Set .............................................................................................................................. 44 Table 3.4 Sensitivity of overall sustainability level based on decision maker’s preference ................ 49 Table 4.1  Common input parameters for pavement alternatives ......................................................... 58 Table 4.2  Alternative specific pavement design inputs and assumptions ........................................... 59 Table 4.3  Unreinforced flexible pavement design ............................................................................... 61 Table 4.4  Rigid pavement design ........................................................................................................ 61 Table 4.5  Geosynthetically reinforced flexible pavement design based on TBR values from            Al-Qadi et. al (1997) .......................................................................................................... 62 Table 4.6  Geosynthetically reinforced flexible pavement design based on LCR values from Maccaferri (2001) ............................................................................................................... 62 Table 4.7  Uncertainty values used for characterizing fuzzy inputs ..................................................... 65 Table 4.8  Pavement maintenance schedule for flexible pavements (adapted from Holt et al. 2011,    © Transportation Association of Canada, with permission) .............................................. 66 Table 4.9  Pavement maintenance schedule for rigid pavements (adapted from Holt et al. 2011, © Transportation Association of Canada, with permission) .................................................. 66 Table 4.10  Pavement maintenance schedule for geosynthetically reinforced flexible pavements  (adapted from Holt et al. 2011, © Transportation Association of Canada, with   permission) ......................................................................................................................... 66 Table 4.11 Transportation distances ..................................................................................................... 67 Table 4.12 Fuzzy numbers for % increase in environmental indicators............................................... 68 Table 4.13  Construction costs ............................................................................................................. 70 Table 4.14  Flexible pavement maintenance and rehabilitation unit costs ........................................... 70 Table 4.15  Rigid pavement maintenance and rehabilitation unit costs ............................................... 70 Table 4.16 Asphalt pavement construction costs ................................................................................. 71 Table 4.17 Asphalt pavement maintenance and rehabilitation costs .................................................... 71 Table 4.18 Hierarchist weighting of environmental impact indicators (Goedkoop and Spriensma 2001) ................................................................................................................................... 74 vii  Table 4.19 Individual weights distribution of environmental indicators based on Goedkoop & Spriensma (2001) ............................................................................................................... 74 Table 4.20 Normalized values for economic and environmental sustainability index values .............. 74    viii  List of Figures Figure 1.1 Outline of research methodology .......................................................................................... 5 Figure 2.1 Cross-sections of flexible pavement (a) and rigid pavement (b) ........................................ 11 Figure 2.2 LCA steps based on ISO 14040 (2006)............................................................................... 23 Figure 2.3 Fuzzy synthetic evaluation process ..................................................................................... 32 Figure 2.4 Example of hierarchical sustainability evaluation framework for a system ....................... 32 Figure 2.5 Fuzzy composite programming process .............................................................................. 34 Figure 3.1 Methodology for evaluating roadway infrastructure sustainability .................................... 36 Figure 3.2 Sustainability evaluation framework (criteria from Muench et al. 2011) ........................... 38 Figure 3.3 Iterative sustainability evaluation process .......................................................................... 40 Figure 3.4 Potential scenarios............................................................................................................... 48 Figure 4.1 Life cycle thinking methodology under uncertainty ........................................................... 56 Figure 4.2 Scenarios for pavement life-cycle thinking approach ......................................................... 57 Figure 4.3 Framework for implementing FCP ..................................................................................... 73    ix  List of Illustrations Illustration 1.1 Stakeholders’ ability to make changes in a project over time ........................................ 2 Illustration 1.2 Roadway infrastructure (image by AkosSzoboszlay 2005 (Own work) [Public domain],  via Wikimedia Commons) ........................................................................... 3 Illustration 2.1 State of Canadian municipal infrastructure .................................................................. 10 Illustration 2.2 Non-woven (left) and woven (right) geotextiles (by Marilyn475 2008a                  (Own work) [Public domain], via Wikimedia Commons) ......................................... 12 Illustration 2.3 Geogrid made of perpendicular arrangement of linear polymeric elements (by   Marilyn475 2008b (Own work) [Public domain], via Wikimedia Commons) .......... 13 Illustration 2.4 Pavement rehabilitation vs. preventive treatment (adapted from Johnson 2000, © Center for Transportation Studies - University of Minnesota, with permission) ....... 12 Illustration 2.5 Example expenditure streams for the two alternatives over a 45 year                   analysis period ............................................................................................................ 20 Illustration 2.6 Output data with and without uncertainty .................................................................... 25 Illustration 2.7 Triangular fuzzy numbers (60, 80, 100) ...................................................................... 27 Illustration 2.8 Calculating membership of actual cost values to each of the linguistic scales ............ 28 Illustration 2.9 Comparison of environmental and economic index of alternatives being       considered ................................................................................................................... 34 Illustration 3.1 Main screen of Green Proforma ................................................................................... 39 Illustration 3.2 Membership functions for the benchmarks of “Long Life Pavement” criteria        based on values from Table 3.1 (data from Muench et al. 2011) ............................... 43 Illustration 3.3 Intersection of benchmark and input TFNs (data from Muench et al. 2011) ............... 44 Illustration 3.4 Snapshot of tool interface for weighting the criteria (criteria from                      Muench et al. 2011).................................................................................................... 45 Illustration 3.5 Roadway project sustainometer ................................................................................... 46 Illustration 3.6 High-performance ecocentric scenario ........................................................................ 49 Illustration 3.7 High-performance anthropocentric scenario ................................................................ 50 Illustration 3.8 Low-performance ecocentric scenario ......................................................................... 50 Illustration 3.9 Low-performance anthropocentric scenario ................................................................ 51 Illustration 3.10 Sustainability of high-performance ecocentric scenario for the pro-socioeconomic    (left), neutral (middle) and pro-environment (right) preferences of                  decision-maker ........................................................................................................... 51 x  Illustration 3.11 Sustainability ratings of low-performance ecocentric scenario for the    pro-socioeconomic (left), neutral (middle) and pro-environment (right) preferences of decision-maker ........................................................................................................... 52 Illustration 3.12 Sustainability ratings of high-performance anthropocentric scenario for the pro-environment (left), neutral (middle) and pro- socioeconomic (right) preferences      of decision-maker ........................................................................................................... 52 Illustration 3.13 Sustainability ratings of the low-performance anthropocentric scenario for the      pro-environment (left), neutral (middle) and pro- socioeconomic (right)      preferences of decision-maker .................................................................................... 53 Illustration 4.1 Network summary for 118 Canadian municipalities ................................................... 57 Illustration 4.2 Comparison chart for geosynthetically reinforced (TBR & LCR) and conventional pavement design ......................................................................................................... 63 Illustration 4.3 Global warming potential estimates for each alternatives under each scenarios ......... 69 Illustration 4.4   LCCA results for each alternative under each scenario ............................................. 72 Illustration 4.5  Environmental and economic sustainability of alternatives at CBR 5% and AADT          250  vpd ...................................................................................................................... 76 Illustration 4.6 Environmental and economic sustainability of alternatives at CBR 5% and AADT         500  vpd ...................................................................................................................... 76 Illustration 4.7 Environmental and economic sustainability of alternatives at CBR 5% and AADT     1000 vpd ..................................................................................................................... 77 Illustration 4.8 Environmental and economic sustainability of alternatives at CBR 5% and AADT      2000 vpd ..................................................................................................................... 77 Illustration 4.9 Variation of aggregated sustainability index with different traffic levels (neutral) ..... 78 Illustration 4.10 Variation of aggregated sustainability index with different traffic levels                         (pro-environment) ...................................................................................................... 79 Illustration 4.11 Variation of aggregated sustainability index with different traffic levels                        (pro-economic) ........................................................................................................... 79   xi  List of Abbreviations and Acronyms  AADT Average Annual Daily Traffic AASHTO American Association of State Highway and Transportation Officials AC Asphalt Concrete ACPA American Concrete Pavement Association BC Base Course CBR California Bearing Ratio DOT Department of Transportation DST Decision Support Tool ESALs Equivalent Single Axle Loads FSE Fuzzy Synthetic Evaluation FCP Fuzzy Composite Programming Geo BC Geosynthetics Reinforced Pavement for reduced Base Course Geo SL Geosynthetics Reinforced Pavement for long Service Life GHG Greenhouse gas emissions HH Human Health HOV High Occupancy Vehicles IE Impact Estimator IRI International Roughness Index L, M, H Low (or Minimum value), Medium (or Most likely value), High (or Maximum value) xii  LCA Life Cycle Assessment LCCA Life Cycle Cost Analysis LCR Layer Coefficient Ratio LEED ND Leadership in Energy and Environmental Design for Neighborhood Development M&R Maintenance and Rehabilitation MCDM Multicriteria Decision Making PCC Plain Cement Concrete PVI Pavement Vehicle Interaction SDI Sustainable Development Indicator SI Sustainability Index SN Structure Number TBR Traffic Benefit Ratio TFN Triangular Fuzzy Number WinPAS Windows Pavement Analysis Software   xiii  Acknowledgements I would like to express my profound thankfulness to my supervisors Drs. Rehan Sadiq and Dr. Kasun Hewage. Your immeasurable guidance, wisdom, inspiration, and motivation have resulted in a successful and learning experience at UBC. I thank you for your generosity with time, energy, support and understanding during my studies. It is an honor for me to have Dr. Ahmad Rteil and Dr. Lukas Bichler as committee members. I am thankful for their attentive criticism, valuable time, and attention. I am also thankful to Dr. Keekyoung Kim for his valuable time and attention.  I am grateful to Shannon Hohl, Angela Perry and Teija Wakeman for helping out with so many things at School of Engineering. I am deeply indebted to Dr. Solomon Tesfamariam, Dr. Ahmed Idris and Dr. Bahman Nasser for divulging valuable professional and research knowledge for municipal infrastructure planning, design, and management. I am also grateful to Mr. James Kay for providing useful information related to industry practices and data for the research. I am very thankful to all my friends and colleagues in the Project Lifecycle Management Laboratory especially Gyan Kumar Shrestha, Rajeev Ruparathna, Aziz Alghamdi, Muhammad Al Hashmi and Venkatesh Kumar for their joyful and supportive company. My utmost gratitude goes to friends like Muneer Ahmed, Husnain Haider, Hassan Iqbal, Oleg Shabarchin, Taylor Liu and so many more at UBC for sharing space and ideas. I would also like to thank all my friends from Pakistan and abroad, who have always cared about me during my stay in Canada. My sincerest appreciation goes to the School of Engineering for providing the support and equipment I needed for my research study. Most importantly, I am extremely grateful to my parents and siblings for providing invaluable support throughout my studies.  xiv  Dedication    To My Family    1  Chapter 1 Introduction The world urban population is expected to increase by 70% between 2011 and 2050 (United Nations 2012, 2014). As a result of rapid urbanization, there is a growing demand for services from the currently aging, insufficient and vulnerable roadway infrastructure (Adeli 2002). Building, expanding and operating additional roadway infrastructure requires significant use of natural resources and also results in the production of considerable wastes and emissions (Morrissey et al. 2012). Therefore, sustainable practices and policies are needed to minimize the environmental as well as socioeconomic impacts associated with the development of roadway infrastructures (Curwell et al. 2010; Shen and Zhou 2014; Walton 2005). Sustainable development of transportation infrastructures has been given greater importance after the publication of the Brundtland (1987) report (Anderson 2012; Dasgupta and Tam 2005; Martland 2011; Mills and Attoh-Okine 2014; Yigitcanlar and Dur 2010). Subsequently, the necessity to measure and track sustainability of transportation infrastructure has emerged (Anderson 2012; Dasgupta and Tam 2005; Martland 2011; Mills and Attoh-Okine 2014).   1.1 Background There are significant environmental and societal impacts associated with the development of roadway infrastructure (Muench et al. 2011; Rabbani et al. 2014). Sustainable planning and design approaches (e.g. use of recycled material, congestion pricing, bike access) have gained considerable attention of the decision-makers to cope with these impacts (Kennedy et al. 2005; Simpson et al. 2014). In particular, sustainability evaluation encourages decision makers to incorporate alternative technologies and policies for reducing the negative environmental, social and economic impacts. Therefore, engineering and planning experts are increasingly interested in developing transparent methodologies and tools for evaluating sustainability to guide informed decisions (Clevenger 2013; Mcvoy et al. 2011).  The highest utility of such tools and methods is in the early project planning and design phases when the ability to make cost-effective decisions is highest as shown in Illustration 1.1 (Basu et al. 2014; Hunt et al. 2008).  2   Illustration 1.1 Stakeholders’ ability to make changes in a project over time  (Adapted from Hendrickson and Au 1989, © C. Hendrickson, with permission)  However, the prospect is limited due to the lack of certain or reliable information available to make these decisions in early project phases (Atkinson et al. 2006; Serpella et al. 2014). Some uncertainties are associated with the availability of construction materials and methods for sustainable infrastructure development (Atkinson et al. 2006; Verweij et al. 2015). While other uncertainties arise from the applicability of sustainable practices available in green rating systems for a given project due to regional constraints and expert opinion (Atkinson et al. 2006; Verweij et al. 2015). In addition, uncertainties also arise due to the inability to accurately forecast roadway project costs, pavement performance and the timing and extent of maintenance and rehabilitation (M&R) activities (Noshadravan et al. 2013; Swei 2012). Moreover, traditional sustainability evaluation approaches (e.g. green rating systems, LCA and LCCA) are inflexible to such uncertainties and require deterministic values to generate results. The propagation of uncertainties to outputs can improve reliability of results which in turn improves the quality of decisions based on the results. Therefore, sustainability evaluation approaches require certain technique to incorporate and propagate uncertainties in early project phases so that the reliability of final results is transparent.  This study explores the use of fuzzy logic in the sustainability evaluation to deal with uncertainty issues for the roadways (infrastructure systems), and the pavements (infrastructure component). Infrastructure systems imply a combination of multiple and interdependent components (e.g., stormwater, pavements, sidewalks) comprising a whole roadway system (Illustration 1.2). Green  Project Time Ability to influence project plans 100% Construction Costs 100% 3  rating systems are commonly used to evaluate roadway sustainability. Infrastructure components imply a particular part of roadway infrastructure system e.g. pavements. Pavements are one of the most expensive component of a roadway. At component level, life cycle sustainability evaluation techniques such as LCCA and LCA are often used.   Illustration 1.2 Roadway infrastructure (image by AkosSzoboszlay 2005 (Own work) [Public domain], via Wikimedia Commons)  1.2 Research Objectives The goal of this research was to develop a methodology for the sustainability evaluation of roadways (at system level) and pavements (at component level) under uncertainty. Following are the specific objectives of this study: 1. Identify deficiencies in existing rating systems for sustainable roadways; 2. Develop a customizable sustainability evaluation tool for roadways using FSE technique and demonstrate its potential through scenario analysis; 3. Identify the deficiencies in deterministic approach for life cycle economic and environmental sustainability evaluation for pavement alternatives; 4. Apply FCP technique to evaluate and compare life cycle costs and environmental impacts of pavement alternatives1 for different scenarios2 under uncertainty.                                                       1 Specific type of technology or strategy being adopted  2 Specific set of conditions under which alternatives will be tested or analyzed Sidewalks Pavements Bike Lane Traffic Signals and Signs Lighting Infrastructure Vegetation Lane Markings Low Impact Development 4  1.3 Research Outline and Methodology This thesis is arranged into five chapters. Chapter 1 provides introduction to the research. Chapter 2 provides an overview of sustainability evaluation techniques (Section 2.2 and 2.3), identifies the uncertainties involved (Section 2.4) and discusses two different fuzzy based techniques to handle these uncertainties (Sections 2.4.4 and 2.4.5). Chapter 3 provides a detailed discussion on the development and application of fuzzy based sustainability evaluation tool for roadways. Chapter 4 provides a detailed discussion on fuzzy based life cycle sustainability evaluation of pavements. Chapter 5 concludes the research, identifies original contributions and provides recommendations for the future research.   Figure 1.1 outlines the research methodology for the sustainability evaluation of roadways (as systems) and pavements (as components) under uncertainty. The following paragraphs briefly explain the research methodology used to apply these fuzzy techniques for sustainability evaluation with references to the specific sub-headings in relevant chapters.  Fuzzy synthetic evaluation (FSE) technique was applied to a newly developed multicriteria sustainability evaluation framework for roadways. Section 3.1 identifies the main steps involved in applying FSE to the new sustainability evaluation framework. A new multicriteria framework was developed in Section 3.2 based on indicators in GreenroadsTM rating system to evaluate the sustainability of roadways. Section 3.3 discusses the process of applying the sustainability evaluation framework according to the decision-making phases of roadway projects. FSE was used to generate the sustainability indices for roadways in three key steps i.e. fuzzification (Section 3.4), aggregation (Section 3.5) and defuzzification (Section 3.6). A new tool called “Green Proforma for Roadway Infrastructures” was developed based on the new approach and scenario analysis was conducted using the tool in Section 3.7.   Fuzzy composite programming (FCP) technique was applied to evaluate the life cycle sustainability of pavement alternatives. Section 4.1 identifies the main steps involved in applying FCP to evaluate the sustainability of pavement alternatives. At first, pavement alternatives were designed in Section 4.2. Section 4.3.1 identifies specific uncertainties involved in LCA and LCCA of pavements. Pavement M&R schedules were developed in Section 4.3.2. Fuzzy operations were used to conduct LCA and LCCA under uncertainty for each alternative in Sections 4.3.3 and 4.3.4 respectively. The results of LCA and LCCA were further normalized and aggregated in Section 4.4 to represent the overall sustainability of pavements under uncertainty. 5   Figure 1.1 Outline of research methodology 6  Chapter 2 Sustainability Evaluation Techniques There is a growing interest in municipalities and government departments to evaluate the sustainability of alternative technologies and policies for developing sustainable transportation infrastructure. Sustainability evaluation can assist policy makers, developers and construction firms to make informed choices. This chapter discusses sustainability evaluation techniques (which include green rating systems, LCA and LCCA), identifies their deficiencies and explains the utility of fuzzy-based methods to address those deficiencies.   2.1 Sustainability in Civil Engineering Projects “Sustainability”, “sustainable infrastructure” and “sustainable transportation” is often mentioned in the literature but there is no agreement on a clear and pragmatic definition of these terms (Goh and Yang 2014; Mcvoy et al. 2013, 2011; Oswald and McNeil 2009). Several propositions have been made to define, characterize and quantify sustainability. Specific stakeholder’s understanding of the term “sustainability” as well as the context where it is applied, has a major role in determining the procedure for quantifying sustainability (Ramani et al. 2011; Zietsman and Rilett 2002). Sustainability is often characterized with the triple bottom line (TBL) dimensions (i.e. economic, environmental and social) (Dasgupta and Tam 2005; Gil and Duarte 2013; Klotz and Grant 2009; Morrissey et al. 2012; Valentin and Spangenberg 2000). Anderson (2012) asserts that the TBL approach is difficult to apply practically at a project level due to the complexity in characterizing these dimensions with specific objectives, criteria and indicators.   Ramani et al. (2011) described sustainability as identification, assessment and application of practices that mitigate the long term impacts of a development activity. Such a description of sustainability acknowledges certain best practices in engineering and planning initiatives that contribute to the mitigation of future negative impacts. Muench et al. (2011) described sustainability as a quality of a system to support natural laws and human values. The human values pertains to socioeconomic aspects (e.g. human health impacts, traffic delays, costs) while natural laws pertain to environmental aspects (e.g. use of virgin material, emissions). A system based definition enables the system owners, managers, designers and technical operators to identify those practices that support “natural laws” and “human values”. This interpretation also aligns with the TBL approach (Anderson 2012; Muench et al. 2011). More importantly, Anderson (2012) points out that the sustainability initiatives are not something beyond, extra or in addition to the conventional approach, rather, they are fundamental to socially, environmentally and economically responsible development and management of infrastructure.  7   There are several methods proposed to evaluate the TBL sustainability of infrastructure systems and components e.g. system dynamics, multicriteria analysis, cost-benefit analysis, material flow accounting, risk analysis, emergy analysis, etc. (Ness et al. 2007; Reza 2013; Singh et al. 2009). The application of these techniques for civil engineering projects depends on the type of problem, available resources and the context of application (Ness et al. 2007; Singh et al. 2009). This study focuses on the following methods commonly used for evaluating sustainability in transportation projects:   Green Rating System as a multicriteria framework for evaluating the sustainability of roadway infrastructure systems;  Life Cycle Cost Analysis (LCCA) as a cost-benefit analysis technique for evaluating the economic sustainability of pavements;   Life Cycle Assessment (LCA) as a material flow accounting technique for evaluating the environmental sustainability of pavements.  These techniques are further elaborated in the following sections.  2.2 Evaluating Sustainability of Roadway Infrastructure The need of sustaining Canada’s degrading infrastructure has been well documented by the agencies (Federation of Canadian Municipalities 2012) and researchers (Mirza 2007). In order to develop sustainable infrastructure, one of the principal challenge faced by engineers is to develop practical tools for assessing and upgrading the sustainability of infrastructure systems (Rodríguez López and Fernández Sánchez 2011; Sahely et al. 2005). Sustainability evaluation for infrastructure planning and engineering activities is a resource-intensive, time-consuming, iterative and evolving process (Rodríguez López and Fernández Sánchez 2011; Sahely and Kennedy 2007; Sahely et al. 2005). It requires stakeholders to establish, evaluate, and prioritize objectives and indicators that have strongly anthropocentric and ecocentric benefits for infrastructure projects (Mills and Attoh-Okine 2014; Sahely et al. 2005). This process needs to be undertaken from time to time in a multi-stakeholder environment as more research and standards for the best practices can be developed and known (Kevern 2010; Mills and Attoh-Okine 2014).     8  2.2.1 Rating Systems and their Deficiencies Several rating systems were designed to combine the range of best practices into a sustainability evaluation tool for infrastructure projects (Rabbani et al. 2014).  The rating systems provide engineers and constructors guidelines to incorporate the best practices and evaluate the sustainability performance for infrastructure projects (Reeder 2010). The rating systems are an assembly of best practices that are known to enhance sustainability in infrastructure development and maintenance activities. They are a collection of sustainability criteria in the form of credits and achievement points developed by team of experts, practitioners and decision makers in the area. The criteria or credits consist of indicators and benchmarks as two main components in the rating systems. Criteria are the specific sustainability aspect of a roadway infrastructure (e.g. recycled material). Criteria are measured by indicators (e.g., percentage of recycled material used in pavements) and indicators have benchmarks linked to sustainability points (e.g. benchmark of 30% recycled material use qualifies for 1 point and benchmark of 40% recycled material use qualifies for 2 points). Some criteria are non-scoring and obligatory for the project sustainability certification while others are point-scoring voluntary.  The sustainability points are evaluated using quantitative multicriteria analysis (MCA) methods (Poveda and Lipsett 2011). The obligatory credits are meant to set minimum projects requirements, often based on local or provincial by-laws, to be assessed for sustainability (Muench et al. 2011). The voluntary or optional credits are based on best practices that are not normally considered in project sustainability unlike the obligatory credits which are required for project certification. When a project fulfills the voluntary credits within the rating systems, sustainability points are accumulated and based on the total points achieved, an overall sustainability score is generated. The project sustainability certifications are classified according to the range of overall sustainability points accumulated. As an example, GreenroadsTM rating system developed by Muench et al. (2011) provides the following four certification levels if all obligatory project requirements are met:   Certified: (32-42 Credit Points)  Silver: (43-53 Credit Points)  Gold (54-63 Credit Points)  Evergreen (>64 Credit Points)  Similar methodology is observed in LEED rating system (USGBC 2009). Although, these state-of-the-art rating systems have significantly increased the awareness of sustainability issues, the rationale 9  behind sustainability evaluation process using these systems is often criticized by decision-makers and researchers. These certification systems have increased the point-hunting attitude among decision makers rather than promoting practices that are realistic, achievable and meaningful for a particular project (Fenner et al. 2008). As an example, a bike rack may not be suitable for areas with freezing temperatures where bikes are rarely ever used but the rating systems still award some points of introducing it. Simpson et al. (2014) identified and ranked desired capabilities based on importance levels provided by the professionals from four different US State Departments of Transportation (DOTs). Some of the capabilities that were highly desired by the professionals from DOTs include self-assessment feature, applicability to design phase, customizability of list of criteria, choice of relevant criteria and availability of performance measures toward achieving those credits (Simpson et al. 2014). Curz et al. (2012) developed a qualitative framework to compare four different sustainability rating systems for transportation projects [GreenroadsTM (Muench et al. 2011), INVEST (Sustainability 2011), I-LAST (Knuth and Fortmann 2010), GREENLITES (McVoy et al. 2010)]. Curz et al. (2012) recognized the need for a more holistic transportation rating tool which could be customized to fit the assessment requirement of any project because the transportation projects vary in size and scope significantly as compared to buildings. Clevenger (2013) observed a significant variation in assessment methodology and argued that it is because of the absence of consensus on a common definition of sustainability for highways and infrastructure projects. Some of the features often cited as necessary in the stat-of-the-art rating systems, are listed below:    Work with limited and uncertain data related to indicators (Alsulami and Mohamed 2014; Gasparatos et al. 2009; Hunt et al. 2008; Yoe et al. 2010)  Offer continuous benchmark functions rather than discrete numbers (Rodríguez López and Fernández Sánchez 2011)  Flexible selection, weighting and benchmarking of criteria (Gasparatos et al. 2009; Hacking and Guthrie 2008; Munda 2006; Sharifi and Murayama 2013)   Clear communication of results (Becker 2004; Cole 2005; Oliveira and Pinho 2010; Sharifi and Murayama 2013; Singh et al. 2009; Walton 2005)  The sustainable development and operation of infrastructures as large as roadways requires the cooperation of stakeholders for a suitable course of action (Rodríguez López and Fernández Sánchez 2011). However, a suitable course of action is determined by uncertain and variable number of qualitative and quantitative criteria. Moreover, establishing the benchmarks for the criteria in early project phases can also be vague and imprecise (e.g., % of in-situ waste that can be recycled). 10  Quantitative information can be comfortably represented in the form of estimated ranges while qualitative information is inherently subjective and vague. For this reason, the measurement of sustainability by considering uncertainties due to imprecision and vagueness remains a challenging task (Foxon et al. 2002; Gil and Duarte 2013; Hunt et al. 2008; Yoe et al. 2010). Fuzzy synthetic evaluation (FSE) is an effective techniques that allows the incorporation of such uncertainties while defining benchmarks and inputs for the indicators (Khatri et al. 2011; Sadiq et al. 2005; Sadiq et al. 2004). This technique is elaborated further in Section 2.4 of this chapter.   2.3 Evaluating Sustainability of Pavements  Construction and maintenance of municipal roads necessitates substantial budget, natural resources and human capital. According to Illustration 2.1, Canadian municipal roads have the highest replacement costs, and also have the highest proportion in poor condition as compared to other infrastructure systems. For this reason, effective planning, design and construction strategies are needed to achieve the sustainability of roads. Operationalizing the best strategies for roadway infrastructure in municipal decision-making processes requires transparent quantification of their sustainability benefits. Since policy makers in government departments have been mounting pressure on developers and construction firms to comply with greenhouse gas abatement targets, the need to quantify the sustainability in terms of environmental indicators in addition to life cycle costs is gaining importance. Pavements as part of a roadway infrastructure are the most expensive component that take direct environmental and traffic damage. This section discusses alternative pavement technologies and the use of LCCA and LCA techniques to evaluate their sustainability.  Illustration 2.1  State of Canadian municipal infrastructure  (data from Federation of Canadian Municipalities 2012) 173.1 171.2 121.7 69.1 20.6 2.0 6.3 5.7 050100150200Municipal Roads Drinking Water Wastewater Storm water0.05.010.015.020.025.0Replacement Value in  Billion Dollars (2009-10) % of Assets in Very Poor  and Poor Condition Replacement Value in billion dollars (2009-2010)% of Assets in Very Poor and Poor physical condition11  2.3.1 Pavements Types There are two common types of pavements used in the construction of municipal roads: 1. Flexible or Asphalt Pavements Flexible pavements typically consist of layered materials where quality of material decreases from top to bottom. Figure 2.1 (a) shows a typical cross-section of an asphalt pavement. Huang (1993) identified following types of flexible pavements:  Conventional Pavements: This pavement consists of typical layers of surface course, base course, sub-base course and subgrade.  Full-Depth Ashpalt Pavements: Full-depth ashpalt pavements consist of an asphaltic base between asphalt surface and sub-grade.   2. Rigid or Concrete Pavements Rigid pavement consists of a Portland cement concrete slab placed on base course, sub-base course or subgrade directly. Figure 2.1 (b) shows a typical rigid pavement section. Huang (1993) identified following types of rigid pavements:  Jointed Plain Concrete Pavements: These pavements consist of transverse joints with or without dowels.   Jointed Reinforced Concrete Pavement: These pavements consist of wire mesh in concrete slabs to enable the use of longer span of joint spacings.  Continuous Reinforced Concrete Pavement: These pavements consist of no tranverse joints and is continously reinforced like JRCP.  Prestressed Concrete Pavement: Prestressing the concrete reinforcement results in lesser thickness of concrete needed for same conditions. The reason is that concrete is weak in tension and stronger in compression.         Figure 2.1 Cross-sections of flexible pavement (a) and rigid pavement (b)   Portland Cement Concrete (PCC) Base or Sub-base Course (optional) Sub-grade Asphalt Concrete (AC) Base Course Sub-base Course Sub-grade (a) (b) 12  Rigid pavements are well-known for their durability as well as high construction costs, unlike flexible pavements. On the other hand, flexible pavements are less costly to build and repair but degrade rapidly and require more extensive M&R interventions. Recently, geosynthetic reinforcement has emerged as a possible technology to address some of the deficient qualities of flexible pavements. geosynthetics are planar products manufactured from polymeric materials that are used in combination with engineering structures (Gupta 2009). There are two types of geosynthetic products used in pavement structures.   Geotextiles: Geotextile products are made of certain polymeric textile materials (Gupta 2009). Geotextiles can be either woven or non-woven. In addition to separation, filtration and drainage function (by non-woven geotextiles), woven geotextiles also reinforce the pavement structure for better strength (Zhang 2007). Illustration 2.2 illustrates the woven and non-woven geotextiles. Woven geotextiles are manufactured using traditional weaving methods while non-woven geotextiles are manufactured by punching or melt-bonding (Gupta 2009).   Illustration 2.2  Non-woven (left) and woven (right) geotextiles (by Marilyn475 2008a (Own work) [Public domain], via Wikimedia Commons)  Geogrid: Geogrids are formed by the perpendicular arrangement of linear polymeric elements. The primary function of Geogrid is to enhance strength of the pavement structure. Illustration 2.3 shows a snapshot of a Geogrid. 13   Illustration 2.3  Geogrid made of perpendicular arrangement of linear polymeric elements (by Marilyn475 2008b (Own work) [Public domain], via Wikimedia Commons)  2.3.2 Characterization of Pavement Alternatives The characterization of pavements means identification of the properties and thicknesses of layers and materials that comprise a complete pavement structure. In general, the pavement design is based on four main philosophies:  Experience: The experience of pavement engineer has a major role in determining the size of pavement and specification of materials under different environmental conditions and traffic loadings.  Empirical: Statistical models from road tests (e.g. AASHTO 1993 guidelines developed from different Road tests) relate roadway design with traffic loadings, environmental conditions and other parameters. These models help in determining appropriate design of the pavement structure.  Mechanistic: It is based on mechanistic principles. No empirical approach is considered in this technique.  Mechanistic-Empirical: This methodology combines mechanistic pavement responses (i.e. stresses, strains and deformations) for pavement design with empirically calibrated pavement performance prediction models. The latest Mechnanistic Empirical Pavement Design (MEPD) guidelines by AASHTO (2008) are based on this approach.  Current state-of-the-practice for pavement design in Canada is dominated by AASHTO (1993) standard guidelines (C-SHRP 2002). British Columbia Ministry of Transportation and Infrastructure 14  (BC MoTI) has also issued a technical circular  based on AASHTO 1993 guidelines (BC MoT 2004) for the design of pavement structures. Recently, the pavement design practices are slowly shifting to MEPD approach (AASHTO 2008; BC MoT 2015) but their practical use is still limited due to extensive data requirements for the calibration and analysis of pavement designs. In this research, AASHTO (1993) methodology is adopted for characterizing pavement alternatives due to its wide adoption in the Canadian construction industry (Holt et al. 2011). The AASHTO (1993) guidelines contain detailed steps for the design of rigid and flexible pavements. For the flexible pavements, AASHTO (1993) guidelines provided an empirical equation 2.1 to evaluate an overall structural number required for pavement design based on specified ESALs (𝑊18 ) and other input parameters.   𝑙𝑜𝑔𝑊18 = 𝑍𝑅 × 𝑆𝑜 + 9.26 × 𝑙𝑜𝑔(𝑆𝑁 + 1) − 0.2 +𝑙𝑜𝑔∆𝑃𝑆𝐼2.70.4+1094(𝑆𝑁+1)5.19+ 2.32 × 𝑙𝑜𝑔𝑀𝑅 − 8.07            (2.1) Equation 2.1 where, 𝑊18 = Equivalent Single Axle Loads that a pavement will bear throughout the design life 𝑍𝑅 = Standard normal deviate 𝑆𝑜= Reliability ∆𝑃𝑆𝐼 = Allowable loss in serviceability 𝑀𝑅 = Resilient Modulus of Subgrade 𝑆𝑁 = Structural number representing overall pavement strength  This equation is solved for the required structural number and then equation 2.2 is used to evaluate the thicknesses of layers in pavement structure:  𝑆𝑁 = (𝑎 × 𝑑)𝐻𝑀𝐴 + (𝑎 × 𝑑 ×𝑚)𝐵𝐴𝑆𝐸 + (𝑎 × 𝑑 ×𝑚)𝑆𝑈𝐵−𝐵𝐴𝑆𝐸                            (2.2) Equation 2.2 𝑎 = coefficient of relative strength 𝑑 = thickness of each layer of pavement 𝑚 = drainage coefficient of each layer  Similar equations are available in AASHTO 1993 guidelines for the design of rigid pavements. According to Gupta (2009), the effect of geosynthetic use can be translated into two pavement design modifications: Extended lifespan of pavement or reduced quantity of base course material. In other words, modifications in equations 2.1 and 2.2 can be translated to either higher traffic loadings (W18 15  values) due to increased service life or reduced base course thicknesses (“d” values) due to improved pavement strength. In order to incorporate these changes in design, different ratios have been proposed based on experimental and numerical studies e.g. Traffic Benefit Ratio (TBR), Base Course Reduction (BCR) and Layer Coefficient Ratio (LCR). Traffic Benefit Ratio is evaluated as a ratio of ESAL values for reinforced pavement and unreinforced pavement (Al-Qadi and Yang 2007; Gupta 2009). Maccaferri (2001) proposed a Layer Coefficient Ratio (LCR) (which is similar to BCR) to evaluate reduced thicknesses of asphalt or base layer. The LCR values can be obtained from the graph provided by Maccaferri (2001). According to Gupta (2009), the BCR ratios ranging from 20% to 40% have been reported by various researchers. The modified AASHTO equations presented by Gupta (2009) and Maccaferri (2001) for structural number and thicknesses based on LCR are shown below.  𝑆𝑁 = (𝑎 × 𝑑)𝐻𝑀𝐴 + (𝐿𝐶𝑅 × 𝑎 × 𝑑 ×𝑚)𝐵𝐴𝑆𝐸 + (𝑎 × 𝑑 ×𝑚)𝑆𝑈𝐵−𝐵𝐴𝑆𝐸                   (2.3) Equation 2.3  𝑀𝑜𝑑𝑖𝑓𝑖𝑒𝑑 (𝑑)𝐻𝑀𝐴 =𝑆𝑁−(𝑎×𝑑×𝑚)𝑆𝑈𝐵−𝐵𝐴𝑆𝐸− (𝐿𝐶𝑅×𝑎×𝑑×𝑚)𝐵𝐴𝑆𝐸𝑎                            (2.4) Equation 2.4  OR   𝑀𝑜𝑑𝑖𝑓𝑖𝑒𝑑 (𝑑)𝐵𝐴𝑆𝐸 =𝑆𝑁−(𝑎×𝑑)𝐻𝑀𝐴−(𝑎×𝑑×𝑚)𝑆𝑈𝐵−𝐵𝐴𝑆𝐸 (𝐿𝐶𝑅×𝑎×𝑚)𝐵𝐴𝑆𝐸                                   (2.5) Equation 2.5  These equations can be used to design Geosynthetically reinforced pavements. The characterization of pavements using these methods enables the calculation of the amount and cost of materials required for the construction of pavements.  2.3.3 Maintenance and Rehabilitation of Pavements In terms of life cycle thinking, the maintenance and rehabilitation (M&R) phase is one of the most vital pavement components. The deterioration of pavement serviceability with age is inevitable which is accelerated further by traffic use and environmental effects (e.g. rain, snow etc.). For this reason, certain M&R actions are needed to maintain an adequate level of service. Johnson (2000) identified the following three distinct types of M&R actions needed in a pavement service life:  16   Preventive Maintenance: Extending functional life by retarding progressive failures  Corrective Maintenance: Dealing with materialized damage such as rutting and cracking  Emergency Maintenance: Repairing severe damages such as severe potholes or blowouts   The kind of intervention needed is determined by the serviceability or criticality of existing pavement condition. Table 2.1 identifies some of the preventive, corrective and emergency actions for pavement M&R actions (FCM 2003; Johnson 2000).   Table 2.1 M&R strategies Action Flexible Pavement Rigid Pavement Preventive Rout and Seal Partial Depth Repairs (spot/patch repairs) Seal Coats (Fog Seal, Scrub Seal, Slurry Seal, Chip Seal) Microsurfacing Joint Resealing Partial Depth Repairs (spot/patch repairs) Sub-Sealing Diamond Grinding Corrective and Emergency Full Depth Asphalt Repair Hot-In-Place Recycling Cold-In-Place Recycling Hot mix Overlay Spray Patching Profile Milling Full Depth Concrete Repair Slab Jacking Asphalt/Concrete Overlay  In order to analyze the costs and environmental impacts of pavement M&R activities, it is crucial to determine the type, timing and extent of M&R actions needed. Pavement preservation plans can be developed using a combination of methods and tools such as decision trees, pavement management systems, engineering judgment, past experiences or a simple “worst condition first” strategy (FCM 2003; Hajek et al. 2004; Hicks et al. 1999). The “worst condition first” approach means that the pavements are rehabilitated when a hazard or serious failure has materialized (Hajek et al. 2004). The preventive treatment to rehabilitation cost ratio is estimated to be about 1:4 for the same pavement condition (Johanns and Craig n.d.). The department of Infrastructure and Transportation in Alberta also recommends the application of preventive approach for sustaining the adequate level of pavement condition (Alberta Infrastructure and Transportation 2006). Due to the preventive maintenance, the pavement condition may end up in the preventive or corrective maintenance zone at the end of design life instead of emergency maintenance (see Illustration 2.4). For this reason, preventive approach reduces the severity of damage at the end of design life and ensures an overall 17  better road quality. However, Johnson (2000) identified the lack of federal aid, lack of information on feasibility of preventive approach and the notion that preventive approach causes more user delay costs, as some of the barriers faced by the municipal departments in adopting this approach. The politicians need to see the justification for the investment in preventive approach or they need to have enough money to take the risk of investing ahead of failures that have not materialized yet. At municipality level, both may be the lagging factors.   Illustration 2.4 Pavement rehabilitation vs. preventive treatment (adapted from Johnson 2000, © Center for Transportation Studies - University of Minnesota, with permission)  Selecting appropriate preservation plans also depends on the pavement distresses that contribute to pavement failure e.g. a structural failure may demand a rehabilitation (corrective/emergency repair) while surface deformations can be dealt with a preventive treatment (Johnson 2000). Johanns and Craig (2009) provided a decision matrix for identifying the appropriate maintenance action based on the observed pavement distresses. However, while conducting LCCA in the early phases of project, information on accurate pavement distresses may not be readily available. Darwin-ME software (developed based on Mechanistic Empirical Pavement Design approach) predicts pavement distresses for a particular design strategy. To accurately predict these distresses, Darwin-ME requires significant data for its calibration to the local climatic and traffic conditions. Even if the calibration is performed 22.533.544.50 5 10 15 20 25 30Pavement Condition Pavement Age (Years) Worst Condition First Preventive MaintenanceEmergency Maintenance Corrective Maintenance Preventive Maintenance 18  and pavement distresses are known, determining the extent, timing and costs of M&R activities still relies considerably on engineering judgment and imprecise information (C-SHRP 2000; Li et al. 2008; Morandeira 2012; Peshkin and Hoerner 2005; Swei et al. 2013; TAC 2014; Walls III and Smith 1999; Yildirim 2011).   Hajek et al. (2004) conducted a survey of 56 Canadian municipalities and found that the majority of municipalities identified engineering judgment as one of the methods for developing pavement preservation plans. Especially for smaller municipalities, engineering judgment is the main technique used by the municipality engineer to develop the pavement preservation plans (Hajek et al. 2004). Engineering judgment is based on human understanding, perception and past experiences. Past experiences can also be analyzed from the records kept by the municipal and provincial departments for the preservation activities performed in the past for different roadways. This knowledge is helpful in developing M&R plans for new pavement structures. Similarly, the US state DOTs have developed manuals containing the schemes of M&R plans based on past experiences and state-wide surveys along with other inputs for conducting life cycle economic analysis. Examples for such guidelines include “Life Cycle Cost Analysis Procedures Manual” developed by California Department of Transportation (2010) which contains schedules of M&R activities for different pavements and geographic regions in California.  The concept of M&R scheduling by engineering judgment is based on the deterioration curve as shown in Illustration 2.4. At the end of pavement service life, usually some form of rehabilitation is required (e.g. asphalt mill and fill, concrete slab repairs). For the rest of service life, routine maintenance actions are performed (e.g. rout and seal, partial depth repairs). For conducting LCCA, engineering judgment and available information, however imprecise and vague, are the tools of highest utility in the early project phases (Tinni 2013). Therefore, there are significant ambiguities and uncertainties associated with the estimation of accurate timing, extent and costs for each M&R action (CALTRANS 2007; Johnson 2000; Swei et al. 2013; Tinni 2013). The life cycle analysis techniques need to incorporate and propagate these uncertainties to outputs for transparent results.    2.3.4 Life Cycle Thinking for Pavements Pavement designers and managers often face a challenging task of deciding among the number of pavement alternatives available (Holt et al. 2011). Different types of pavements also differ in the type, extent and timing of M&R activities needed as they are made of different materials and offer 19  different level of resistance to environmental degradation and traffic loadings. For example in case of geosynthetically reinforced pavements, a pavement engineer faces the following key questions that whether:    Increasing the construction cost by incorporating geotextiles and consequently increasing service life i.e. less maintenance costs is beneficial? OR,  Decreasing the construction cost by decreasing base-course thickness, due to the reinforcement action provided by geosynthetics, is beneficial?  Similar questions arise when pavement engineers face the dilemma of deciding between rigid and flexible pavements. The ultimate choice of appropriate pavement design is determined by one of the key question: which pavement combination results in the least cost and environmental impacts over its life cycle? This is in contrast to the current practice whereby life cycle costs are the main decision-making factor for choosing from available pavement alternatives. Life cycle thinking in pavements is a general term used by life cycle experts for evaluating the life-cycle environmental impacts and costs of pavements. Following key phases are involved in pavement life cycle as identified by Masanet and Horvath (2010):  1. Materials Production Phase (e.g. material extraction, transportation, processing) 2. Construction Phase (e.g. onsite use of construction machinery, traffic delay, placement of materials) 3. Use Phase (e.g. pavement vehicle interaction, albedo, carbonation, lighting) 4. Maintenance Phase (construction maintenance and rehabilitation activities as well as associated traffic impacts) 5. End-of-life Phase (e.g. dismantling, disposal, recycling)  There are significant material and energy inflows and outflows with substantial cash flows throughout these phases of pavements. The pavement engineer needs to adopt a life cycle thinking approach and consider the minimization of both costs and environmental impacts for sustainable decision making. The methodology for quantifying both impacts has been termed formally as Life Cycle Cost Analysis (LCCA) for economic evaluation and Life Cycle Assessment (LCA) for environmental evaluation. Recently, there has been a significant growth in published literature related to pavements Life Cycle Assessment due to the increasing realization for “greening” the pavement life cycle. Importance of LCCA and LCA can be appreciated by considering the fact that major sustainability guidelines such 20  as GreenroadsTM (Muench et al. 2011) have included the application of these techniques as sustainability credits. An overview of both LCA and LCCA is provided in the following sections.  Life Cycle Cost Analysis (LCCA) LCCA is a systematic evaluation of total cost of a project by considering the useful life of pavement structure (Walls III and Smith 1998). It is a detailed procedure that aids decision makers in making better investment decisions in early project phases by considering the cost of construction, maintenance and rehabilitation of alternatives as shown in Illustration 2.5 (DoT 2004).   Illustration 2.5 Example expenditure streams for the two alternatives over a 45 year analysis period  LCCA can be represented using the following economic indicators (Muench et al. 2011; Walls III and Smith 1998):  Net Present Value (NPV) is the difference of discounted present values of benefits and costs of pavement alternative. Usually the benefits of keeping pavement condition above a minimum threshold are the same for all types of pavements under the same conditions. As a result, they are cancelled out when comparing life cycle costs of different alternatives (Walls III and Smith 1998) – see Equation 2.6. Net Present Value (NPV) or Net Present Worth (NPW) method is commonly used to calculate the total life cycle costs discounted to the present time (Walls III and Smith 1998) using Equation 2.7.  𝑁𝑃𝑉 =  𝑃𝑉𝑏𝑒𝑛𝑒𝑓𝑖𝑡𝑠 – 𝑃𝑉𝑐𝑜𝑠𝑡𝑠                                             (2.6) Equation 2.6 0200004000060000800001000001200000 5 10 15 20 25 30 35 40 45Costs (CAD) Years Life Cycle Expenditures Alternative 1 Alternative 221   𝑁𝑃𝑉 = 𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝐶𝑜𝑠𝑡 + ∑ 𝑀&𝑅 𝐶𝑜𝑠𝑡𝑠𝑘1(1+𝑖)𝑛𝑘𝑁𝑘=1                                (2.7) Equation 2.7 Where, 𝑖 = 𝑑𝑖𝑠𝑐𝑜𝑢𝑛𝑡 𝑟𝑎𝑡𝑒 𝑛 = 𝑦𝑒𝑎𝑟 𝑜𝑓 𝑒𝑥𝑝𝑒𝑛𝑑𝑖𝑡𝑢𝑟𝑒 𝑀&𝑅 = 𝑀𝑎𝑖𝑛𝑡𝑒𝑛𝑎𝑛𝑐𝑒 𝑎𝑛𝑑 𝑅𝑒ℎ𝑎𝑏𝑖𝑙𝑖𝑡𝑎𝑡𝑖𝑜𝑛 𝑓𝑜𝑟 𝑖𝑛𝑡𝑒𝑟𝑣𝑒𝑛𝑡𝑖𝑜𝑛 𝑡𝑦𝑝𝑒 "𝑘"   Uniform Equivalent Annual Cost (EUAC) is the equivalent of NPV value such that all benefits and costs were to occur uniformly over the analysis period of the alternative – see Equation 2.8. 𝐸𝑈𝐴𝐶 = 𝑁𝑃𝑉(1+𝑖)𝑛(1+𝑖)𝑛+1                                                        (2.8) Equation 2.8   Benefit/Cost (B/C) ratio is the ratio of discounted benefits and costs of an alternative. B?C values of more than 1.0 indicates benefits are more than the costs.  Walls III and Smith (1998) identified the following sequential steps for conducting the LCCA:  Identify and characterize altervative pavement scenarios;  Establish M&R interventions and timings;  Evaluate agency and user costs;  Determine cashflows and evaluate net present value; and  Assess results and reevaluate alternatives  First two steps have been discussed in the previous sections (2.3.2 and 2.3.3). Determining agency costs requires the estimation of construction and M&R costs while the user costs require data regarding vehicle delay costs, hours of M&R operations etc. For ensuring consistency while conducting LCCA for alternative pavements, it is important to keep the analysis period, traffic parameters and all environmental conditions the same. The analysis period is a span time over which the design alternative is intended to function above a minimum level of service. Walls III and Smith (1998) recommended that analysis period should be longer than the pavement design period and should incorporate at least one rehabilitation strategy. The design period represents the span time over 22  which the pavement is intended tolerate defined environmental and traffic burdens until failure without any M&R activity.  Life Cycle Assessment (LCA) Material Flow Accounting (MFA) is a technique used to evaluate environmental burdens associated with material flows in human activities varying from national economy to smaller specific units such as pavements (Reza 2013). LCA is a type of MFA technique used to quantify environmental burdens of specific units of processes, products or services (e.g. buildings, pavements, concrete, fuel production) throughout the life-cycle phases (Reza 2013). LCA usually includes evaluation of environmental impact indicators such as primary energy use, global warming potential, eutrophication potential, ozone depletion potential, acidification potential, smog potential and ozone depletion potential. Recently, social life cycle assessment (s-LCA) is also gaining importance to complete the TBL life cycle sustainability assessment of products, process and services (Chhipi-Shrestha et al. 2014). The economic life cycle sustainability evaluation is termed as LCCA which has been discussed earlier (Chhipi-Shrestha et al. 2014).  The key phases involved in conducting an LCA are provided by ISO 14040 (2006) as following:  Goal and Scope: Involves establishing the functional unit of alternative, system boundaries, underlying assumptions, impact categories to be used etc. For pavements, it can include identifying pavement design strategies and material flows associated with the life cycle phases (e.g. construction, M&R, end-of-life) under consideration.  Life Cycle Inventory (LCI): Involves identifying, characterising and quantifying material and energy flows in the life cycle of the previously identified alternatives.   Life Cyle Impact Assessment (LCIA): Involves quantification and/or aggregation of each environmental impact category based on LCI results. The environmental impact categories are identified, evaluated and aggregated to summarize the overall impact caused by each alternative.   Interpretation: Involves the evaluation of the sensitivity of results to underlying uncertain data and drawing conclusions from the results. This step also includes identifying areas of concerns based on the results, limitations of study and making recommendations. 23   Figure 2.2 LCA steps based on ISO 14040 (2006)  Life cycle assessment (LCA) can be used for eco-labeling the construction materials, comparing alternatives and developing regulations for environmentally friendly construction of buildings or highways (Reza 2013). Santero et al. (2011) conducted a critical review on life cycle assessment of pavements and emphasized the importance of ensuring consistency in identifying the functional units of alternatives. If the goal and scope are not the same, then it is illogical to compare or aggregate the environmental impacts. Santero et al. (2011) further underlined the importance of using region-specific data as they cannot be translated meaningfully for another area because of difference in electricity mixes, methods for pavement design, materials production and maintenance practices. However, due to considerable imprecise information available in the early decision-making phases of a project, the inputs for life cycle thinking approach are inevitably uncertain. Following sections highlight the importance of considering inputs with uncertainties instead of point estimates.  2.4 Uncertainty in Evaluating Sustainability There are significant uncertainties inherent in the sustainability evaluation approach used by the state-of-the-art rating systems. For example, in case of GreenroadsTM rating systems (Muench et al. 2011), single observed value is needed to calculate overall sustainability points for quantitative sustainability criteria (e.g., % of recycled material used). However, the information available on-site is not always crisp and is reliably estimated as ranges with different levels of certainty (e.g. minimum, most likely, maximum). In early project planning phases, uncertainties are also related to the establishment of benchmarks or targets for these indicators (e.g. how much percentage range of recycled material corresponds to how many points). In such cases, expert knowledge and opinion is needed to modify and adapt these criteria to project specific constraints. The experts imply professionals specializing in the area of expertise related to the sustainability criteria under consideration (e.g. environmental or water resource engineers for assessing runoff flow and quality). There are obvious advantages of the 24  default or pre-set values but the recyclability of materials changes from one project to another. For this reason, the decision-makers and auditors of roadway projects need to be flexible in altering these benchmarks. Therefore, to meaningfully conduct sustainability evaluation, it is crucial to incorporate expert opinion (which is often vague and imprecise) associated with the indicators, benchmarks and inputs of each criterion.   Similarly, traditional LCCA and LCA techniques have used deterministic values (point estimates) to evaluate the total costs and environmental impacts for different alternatives. Using a certain-static data hides the effect of underlying uncertainty of inputs on overall costs and environmental impacts (Santero et al. 2011b; Swei et al. 2013). Some of the key factors contributing to uncertainty in conducting LCCA and LCA studies include the inability to forecast accurate costs of construction and maintenance, pavement performance, and the timing and extent of M&R activities needed during the life cycle of pavements (Atkinson et al. 2006; Noshadravan et al. 2013; Swei et al. 2013; Yoe et al. 2010). Due to these reasons, the pavement preservation programs in Canadian Municipalities are mostly based on engineering judgment (Hajek et al. 2004). Probabilistic methods have generally solved the problem of uncertain inputs. However, often the probabilistic methods require significant data to model input uncertainties and are computationally inefficient and not well equipped to deal with all types of uncertainties (Dyck et al. 2014; Reza et al. 2013). As discussed earlier, project planning and design phases offer significant flexibility to make cost effective decisions but the value of this opportunity is limited because of the significant imprecise and vague information available. For effective decision-making in the early project phases, it is essential to measure and manage such uncertainties in a transparent and computationally efficient manner (Atkinson et al. 2006).   2.4.1 Types of Uncertainties involved in Sustainability Evaluation Uncertainty is an unavoidable quality that exists in decision-making processes. The quality or sureness of a decision can be improved if uncertainties are estimated and propagated to output quantitative measures (e.g. life cycle costs) (Yoe et al. 2010). A simple illustration of this phenomenon can be observed from Illustration 2.6. Without the expression of uncertainty, the alternative “X” clearly seems to be the most expensive one. However, if the uncertainty is mapped with the maximum and minimum LCC values of both alternatives, the dissimilarity becomes indistinct. The distribution of uncertainty for LCC in case of alternative “X” coincides considerably with that of alternative “Y”. Thus, the decision-maker cannot claim that alternative “Y” is the best one with the same confidence level as he could have done without uncertainty. 25   Illustration 2.6 Output data with and without uncertainty  Uncertainties are of various types. International Panel on Climate Change (IPCC 2000) recommended at least three types of uncertainties to be considered while inventorying the data for greenhouse gases which include:   Imprecise definition (vagueness)   Natural variability (randomness)  Inaccurate model and estimation of associated data/parameters: o Measurement o Sampling o Irrelevant reference data o Imprecise understanding and judgement  Dyck et al. (2014) identified two types of uncertainties in the framework for human health risk assessment: (1) Aleatory uncertainty that exists due to natural variation in sampled population/system; and (2) Epistemic uncertainty that exists due to imperfect understanding of the system/model. Probabilistic or Stochastic methods can handle Aleatory uncertainty more effectively than the epistemic uncertainty, because the major source of uncertainty is randomness or variability that is not under human control (Dyck et al. 2014). Moreover, epistemic uncertainty or uncertainties arising as a result of imperfect human understanding/human measurement of data and parameters/human expert opinion/human interpretation of available data and information can be expressed more conveniently in terms of possibilities rather than probabilities (Reza et al. 2013). Sustainability evaluation approaches in early project phases consists more of epistemic uncertainty rather than aleatory uncertainty. Some of the key reasons are listed below: 010203040506070X Y X YLife Cycle Costs  (million dollars) Alternatives HighLowClose26   The definition of sustainability and related attributes depends considerably on the vague human understanding and perception of the term (Baumgärtner and Quaas 2009; Centre et al. 2007; Dovers and Handmer 1992).  Decision-making based on multiple criteria relies on the imperfect human understanding of the qualities and the quantities associated with these criteria (Khan et al. 2002; Sadiq 2001).   In early project phases, available reference data is usually imprecise and inaccurate (Sadiq and Khan 2006; Sadiq 2001; Yoe et al. 2010).  2.4.2 Fuzzy Sets and Fuzzy Numbers Fuzzy logic is one of the most effective technique for handling uncertainty due to vague human understanding and imprecise information (Rajani, Kleiner, and Rehan Sadiq 2006; Reza 2013; Sadiq and Rodriguez 2004; Sadiq, Rajani, et al. 2004). In the real-world, human approximations are best represented with degrees of certainty (fuzzy logic) rather than absolute certitude (classical logic)(Ross 2004). The reason is that all information and models available are based on human understanding, perception, interpretation and measurement, which is inherently imprecise and vague. Zadeh (1965) introduced the fuzzy set theory in his pioneering work which enabled the modelling of imprecise information. Equation 2.9 represents a mathematical form of the fuzzy set theory.  𝐹 = {(𝑥,μ(𝑥))⎸𝑥 ∈ 𝑈}                                                        (2.9) Equation 2.9 𝐹 =  𝑓𝑢𝑧𝑧𝑦 𝑠𝑒𝑡. 𝑈 =  𝑢𝑛𝑖𝑣𝑒𝑟𝑠𝑎𝑙 𝑑𝑖𝑠𝑐𝑜𝑢𝑟𝑠𝑒,  𝑥 =  𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒,  µ(𝑥) =  𝑚𝑒𝑚𝑏𝑒𝑟𝑠ℎ𝑖𝑝 𝑜𝑝𝑒𝑟𝑎𝑡𝑜𝑟 𝑜𝑛 𝑥 (0 − 1 𝑟𝑎𝑛𝑔𝑒),   The returning set “F” is a combination of the series of subsets represented by a variable value and membership value. Membership grades or values represents the belongingness level (on a scale of 0 to 1 in fuzzy logic unlike absolute 0 or 1 in case of classical logic) of a variable value to the linguistic scales that are more comprehensible by human intelligence (Tran et al. 2012). For example, one cannot be sure how much value of tyre pressure is high and low (which is linguistic representation), however, one can say with a certain “degree of confidence” how much value of tyre pressure is high or low. The “degree of confidence” indicates the membership grade shown Equation 2.9. Essentially, fuzzy set theory recognizes the partial truth concept and assigns a degree of truth values between 0 and 1 to the linguistic scales (Elwakil et al. 2014). Therefore, this approach allows the modeling of 27  imprecise information with different linguistic scales of performance (Yager 1977). Fuzzy numbers are the numerical representation of approximations based on human intelligence with different membership grades e.g. variable X can be expressed conveniently as a triangular fuzzy number (TFN) in the form of (minimum value, most-likely value, maximum value). It is clear from the linguistic description of fuzzy numbers that the membership values or degrees of confidence are “0” for the extreme (minimum and maximum) values and “1” for the most-likely value. An example shown in Illustration 2.7 illustrates a TFN (60, 80, 100) plotted for a variable “X”. The range of inputs with a membership value (sometimes called α-cut) above 0.5, in Illustration 2.7, is (70, 90) which means according to the TFN, there is at least 50% possibility that number will be between 70 and 90. Classical logic is not flexible to this interpolation. Fuzzy numbers can also take the form of a trapezoidal fuzzy numbers with two most-likely values (instead of one). The shape of membership functions depends on human understanding, data availability and the complexity and type of problem under consideration.   Illustration 2.7  Triangular fuzzy numbers (60, 80, 100)  As another example, Table 2.2 shows the affordable, moderately expensive, and expensive ranges of project costs as TFNs identified based on imprecise perception of the project owner. Membership functions of the TFNs are plotted as shown in Illustration 2.8. For an actual cost value of 27.5 Million dollars, a line is mapped on the membership functions. The intersection of actual cost line with other TFNs shows the belongingness of the actual cost value to the three linguistic scales. In this case the fuzzy sets can be shown as (µaffordable, µmoderately expensive, µexpensive) = (0.75, 0.25, 0), where “µ” 60, 0 80, 1 100, 0 70, 0.5 90, 0.5 00.10.20.30.40.50.60.70.80.9140 50 60 70 80 90 100 110Membership Value X Minimum X = 60, Most-Likely X = 80, Maximum X = 100 α-cut = 0.5 Fuzzy Number Membership value 28  represents membership grade or belongingness of actual values to the linguistic scales developed based on the imprecise knowledge of project owner.   Table 2.2 TFN for the linguist scale Costs (Million Dollars) Minimum Most Likely Maximum Affordable 10 20 30 Moderately Expensive 20 30 40 Expensive 30 40 50  Illustration 2.8  Calculating membership of actual cost values to each of the linguistic scales  2.4.3 Fuzzy Operations The raw data required for analytical models to calculate project costs (or other indicators from LCA and LCCA) is often vague and imprecise in early project phases (discussed earlier). Such uncertainties can be modelled in the form of fuzzy numbers. The fuzzy operations to handle fuzzy numbers are different from conventional arithmetic operations. Table 2.3 lists the fuzzy arithmetic operations that can be used for evaluating fuzzy outputs from fuzzy inputs.   27.5, 0.25 27.5, 0.75 00.20.40.60.811.20 10 20 30 40 50 60Membership or Possibility values Cost (million dollars) Fuzzy set for Project Cost of 27.5 Million Dollars =  (µaffordable, µmoderately expensive, µexpensive) = (0.25, 0.75, 0) Actual Value Moderately ExpensiveExpensive AffordableMembership with Affordable Range Membership with Moderately Expensive Range29  Table 2.3 Arithmetic operations on fuzzy numbers with only low and high values Arithmetic Operation on Fuzz Inputs Fuzzy Outputs in the form of [low, high] [a, b] + [c, d] [a + c, b + d]  [a, b] – [c, d] [a – d, b – c]  [a, b] × [c, d] [a × c, b × d]  [a, b] ÷ [c, d] [min(a/c, a/d, b/c, b/d), max(a/c, a/d, b/c, b/d)]  B × [a, b] [B × a, B × b] for B > 0, [B × b, B × a] for B < 0  The α-cut (alpha) method is used to evaluate the range of outputs for a range of inputs at a membership value between 0 and 1. Dyck et al. (2014) identified following major steps for developing a complete output membership function using a given input function based on Dong et al. (1985):  Select membership value (α-cut) of inputs and find the corresponding input range  Use fuzzy arithematics to find corresponding output range at the same α-cut  Repeat these steps to build the membership function for the output.  The conversion of fuzzy functions back to crisp number can be achieved by various defuzzification methods such as center of gravity, mean of maxima, weighted fuzzy mean etc. (Chen and Hwang 1992; Ross 2004).  The concept of fuzzy numbers and fuzzy sets has been applied to solve many decision-making problems such as risk assessment, condition ratings, and performance assessments (Abdelgawad and Fayek 2010; Khatri et al. 2011; Rajani et al. 2006; Sadiq et al. 2004; Yan and Vairavamoorthy 2003). To capture inherent vagueness in human judgment, Jato-Espino et al. (2014) combined fuzzy logic with other multi-criteria methods for the selection of urban pervious pavements. Filippo et al. (2007) applied fuzzy-multi-criteria model to prioritize the highways that require environmental restoration. Shen et al. (2011) applied fuzzy set theory in selecting Key Assessment Indicators for sustainability of infrastructure projects. Alsulami & Mohamed (2014) combined fuzzy set theory and cognitive map to develop a hybrid fuzzy sustainability assessment model and applied it to a transportation project. These studies demonstrated the effectiveness of fuzzy logic to handle uncertainties due to subjectivities, vagueness and imprecision in expert opinion and available data.  In this study two types of fuzzy based techniques are discussed and applied for the sustainability evaluation of transportation infrastructure i.e. Fuzzy Synthetic Evaluation and Fuzzy Composite 30  Programming. Overview of the two methods is provided below in the context of roadway infrastructure and pavement sustainability evaluation.  2.4.4 Fuzzy Synthetic Evaluation Roadways, as an infrastructure system, consist of multiple interdependent components, e.g. stormwater system, pavements, sidewalks, etc. Each component has its own sustainability criteria for the low-impact, efficient and long-lasting development and operation of roadway infrastructure system. As a result, the roadways as systems have many sustainability criteria. Evaluating the sustainability of a system requires the comparison of qualitative and quantitative criteria. Sustainability evaluation problems based on multiple criteria also involve significant ambiguity, imprecision and subjectivity due to the vagueness in human understanding and imprecise information (discussed in Section 2.4.1). Such uncertainties can be effectively handled by using fuzzy synthetic evaluation (FSE) technique. FSE develops fuzzy membership functions for each criterion which allows the quantitative modelling of imprecise and linguistic information (Sadiq et al. 2004). FSE produces the outputs in the form of fuzzy sets by comparing the benchmark membership functions and observed inputs for each criterion. Fuzzy sets can be aggregated to determine the generalized sustainability attributes of a system under consideration. The application of FSE technique for assessing any system performance involves the following key steps based on Sadiq et al. (2004):   Framework development: Developing a hierarchical framework involving key attributes and indicators (see Figure 2.);  Fuzzification: Fuzzifying the indicator benchmarks and generating fuzzy sets for the inputs (see Illustration 2.8);  Aggregation: Involves weighting and aggregation of indicators fuzzy sets to more generalized attributes of a system (see Figure 2.). The sustainability indicators can be prioritized/weighted by the experts involved. Prioritization establishes the relative importance or weight of indicators to a given sustainability attribute. It helps in distinguishing the importance of one indicator over the other with respect to a given sustainability objective or attribute (Mendoza and Prabhu 2002; Saaty and Vargas 2012). There are different approaches used to determine the weights of criteria e.g. Analytical Hierarchical Process (AHP), Direct Assignment, Analytic Network Process (ANP) etc. (Hwang and Yoon 1981; Saaty and Vargas 2012; Saaty 2007). In this study, direct assignment method has been used for its simplicity and straightforwardness as compared to 31  other methods. Mathematical equation for weighting using direct assignment is shown below: 𝒘𝒊 =𝒙𝒊∑ 𝒙𝒊𝒏𝒊=𝟏                                                               (2.10) Equation 2.10 𝑤𝑖 = 𝑟𝑒𝑙𝑎𝑡𝑖𝑣𝑒 𝑤𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑖𝑛𝑑𝑖𝑐𝑎𝑡𝑜𝑟 i 𝑥𝑖 = 𝐷𝑖𝑠𝑐𝑟𝑒𝑡𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑐𝑜𝑟𝑟𝑒𝑠𝑝𝑜𝑛𝑑𝑖𝑛𝑔 𝑡𝑜 𝑡ℎ𝑒 𝑔𝑖𝑣𝑒𝑛 𝑙𝑒𝑣𝑒𝑙 𝑜𝑓 𝑙𝑖𝑛𝑔𝑢𝑖𝑠𝑡𝑖𝑐 𝑠𝑐𝑎𝑙𝑒 𝑛 = 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑖𝑛𝑑𝑖𝑐𝑎𝑡𝑜𝑟𝑠 When the weights and fuzzy sets are known for each of the indicators, aggregation can be performed to quantify the fuzzy set of sustainability attributes. Aggregation means the sum-product of weights and fuzzy sets for each of the indicators under each sustainability attribute. The following example illustrates this mathematical process: 𝑊 (𝑤𝑒𝑖𝑔ℎ𝑡 𝑚𝑎𝑡𝑟𝑖𝑥 𝑠ℎ𝑜𝑤𝑖𝑛𝑔 𝑤𝑒𝑖𝑔ℎ𝑡𝑠 𝑜𝑓 3 𝑖𝑛𝑑𝑖𝑐𝑎𝑡𝑜𝑟𝑠) = [𝑤1 𝑤2 𝑤3]=  [0.2 0.4 0.4] 𝐹 (𝑓𝑢𝑧𝑧𝑦 𝑠𝑒𝑡𝑠 𝑜𝑓 3 𝑖𝑛𝑑𝑖𝑐𝑎𝑡𝑜𝑟𝑠; 𝑎, 𝑏, 𝑐 𝑤𝑖𝑡ℎ 1 𝑡𝑜 3 𝑝𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒 𝑙𝑒𝑣𝑒𝑙𝑠) = = [𝑎1 𝑎2 𝑎3𝑏1 𝑏2 𝑏3𝑐1 𝑐2 𝑐3] =  [0.1 0.7 0.10 0.8 00 0.1 0.7] 𝐴𝑔𝑔𝑟𝑒𝑔𝑎𝑡𝑒𝑑 𝑓𝑢𝑧𝑧𝑦 𝑠𝑒𝑡 = 𝑊 𝑥 𝐹 =  [0.2 0.4 0.4]  ×  [0.1 0.7 0.10 0.8 00 0.1 0.7]= [0.02 0.5 0.3]   The weights for sustainability attributes can also be determined by using the direct assignment method. The fuzzy sets for the sustainability attributes can be further aggregated to evaluate an overall sustainability fuzzy set.  Defuzzification: Defuzzifying the aggregated fuzzy sets and generating crisp performance scores of system under consideration (see Figure 2.). Fuzzy sets cannot be interpreted directly by the decision makers. For this reason, defuzzification methods are used to convert the computed aggregated fuzzy sets into crisp values. There are several methods for defuzzification of fuzzy sets, e.g. center of gravity, mean of maxima, weighted fuzzy mean, etc. (Andriantiatsaholiniaina et al. 2004; Chen and Hwang 1992; Khatri et al. 2011). Center of gravity is a commonly used technique in which the centroid of area under the membership graph for a given set is located and the corresponding x-axis value is returned as a defuzzified or crisp value. 32   Figure 2.3 Fuzzy synthetic evaluation process   Figure 2.4 Example of hierarchical sustainability evaluation framework for a system  Sustainability Indicators •Weights = WInd • Indicator Fuzzy Sets = FInd Sustainability Attributes •Weights = WAtt •Attribute Fuzzy Sets = FAtt •FAtt= ∑ WInd x FInd •Defuzzify FAtt   Sustainability Index •FSI= ∑ WAtt x FAtt •Defuzzify FSI  System Indicators System Attributes System Index Sustainability Index Technical Feasibility Indicator 1 Indicator 2 Environmental Impacts Indicator 1 Indicator 2 Costs Indicator 1 Indicator 2 33  2.4.5 Fuzzy Composite Programming The sustainability performance of pavements can be assessed using LCA and LCCA techniques. However, there are considerable uncertainties involved in the estimation of extent, timing and costs of M&R activities needed for each alternative. It is critical to propagate these uncertainties while conducting LCA and LCCA for reliable and transparent outputs. Fuzzy composite programming (FCP) is one of the useful techniques for analyzing multi-objective problems with uncertainties (Lee et al. 1991). This technique enables input uncertainties to be propagated to the output indicators such as life cycle costs, global warming potential, acidification potential etc. (Khan et al. 2002; Sadiq and Khan 2006). FCP involves the application of fuzzy arithmetic operations on fuzzy inputs (e.g. unit costs) to produce fuzzy outputs (e.g. life cycle costs) for each of the basic indicators (Dyck et al. 2014; Sadiq and Khan 2006). The resulting fuzzy outputs can be normalized and aggregated based on the appropriate grouping of indicators to produce overall indices (e.g. economic sustainability index, environmental sustainability index). The normalization process enables the conversion of the indicator values for each alternative to comparable units. In FCP, the aggregation process is the sum product of the normalized fuzzy numbers and the weights of each indicator. This technique is discussed in detail by Sadiq and Khan (2006) for risk-based life cycle assessment. FCP involves following major steps based on Sadiq et al. (2005):  Framework development: Involves grouping of indicators and attributes of the system under consideration (see Figure 2.4);  Normalization: Involves evaluating fuzzy output (indicators) from fuzzy inputs (raw data) using fuzzy operations and then converting the indicators fuzzy values to comparable units for all alternatives. Lee et al. (1991) discussed the methods for converting indicator values to normalized numbers depending on the type of indicators and decision-makers understanding. Following are the two basic approaches used to normalize indicator values based on Sadiq (2001):  If increasing indicator value is not desirable: 𝑁𝑜𝑟𝑚𝑎𝑙𝑖𝑧𝑒𝑑 𝑖𝑛𝑑𝑒𝑥 𝑣𝑎𝑙𝑢𝑒 =𝐴𝑐𝑡𝑢𝑎𝑙 𝐼𝑛𝑑𝑖𝑐𝑎𝑡𝑜𝑟 𝑣𝑎𝑙𝑢𝑒−ℎ𝑖𝑔ℎ 𝑣𝑎𝑙𝑢𝑒𝑙𝑜𝑤 𝑣𝑎𝑙𝑢𝑒−ℎ𝑖𝑔ℎ 𝑣𝑎𝑙𝑢𝑒                         (2.11) Equation 2.11  If increasing indicator value is desirable: 𝑁𝑜𝑟𝑚𝑎𝑙𝑖𝑧𝑒𝑑 𝑖𝑛𝑑𝑒𝑥 𝑣𝑎𝑙𝑢𝑒 =𝐴𝑐𝑡𝑢𝑎𝑙 𝐼𝑛𝑑𝑖𝑐𝑎𝑡𝑜𝑟 𝑣𝑎𝑙𝑢𝑒−𝑙𝑜𝑤 𝑣𝑎𝑙𝑢𝑒ℎ𝑖𝑔ℎ 𝑣𝑎𝑙𝑢𝑒−𝑙𝑜𝑤 𝑣𝑎𝑙𝑢𝑒                          (2.12) Equation 2.12 34   The low and high values are extreme values of a particular indicator for all the alternatives being compared. The normalized values do not have units and therefore, these values can be compared with normalized values of other indicators also for each alternative.  Aggregation: Involves weighting and aggregation of normalized values to more generalized attributes of the system depending on the grouping of indicators (see Figure 2.5).  Ranking: Involves analyzing the overall ranking of alternatives based on the results from the aggregation of sustainability attributes. This step can involve the conversion of aggregated normalized fuzzy index values (e.g. sustainability index) to crisp values using different methods of defuzzification (Chen 1985; Chen and Hwang 1992). If there are two objectives being measured (e.g. economic index and environmental index), then they can also be plotted as 2D fuzzy numbers to to analyze the trade-offs among conflicting objectives under uncertainty (see Illustration 2.9). The 2D illustration of fuzzy numbers was provided by Sadiq and Khan (2006) for risk-based life cycle assessment and decision making.   Figure 2.5 Fuzzy composite programming process  Illustration 2.9 Comparison of environmental and economic index of alternatives being considered  Sustainability Indicators • Input Fuzzy Numbers = Fin •Output Fuzzy Numbers for Indicators  = f (Fin) = FInd •Normalized FInd = Fnorm.Ind •Weights = Windi Sustainability Attributes •Attribute Fuzzy Numbers = FAtt •FAtt = ∑ Wind x Fnorm.Ind •Weights = WAtt  Sustainability Index •Sustainability Fuzzy Number = FSI •FSI = FAtt x WAtt 00.20.40.60.810 0.2 0.4 0.6 0.8 1Economic Index Environmental Index Alternative 1 Alternative 2 Alternative 335  Chapter 3 Sustainability Evaluation of Roadway Infrastructure  There is an urgent need to discern and incorporate best practices in roadway development projects due to rapid urban population growth and climate change. Many best management practices (BMPs) are available in the form of several state-of-the-art sustainability rating tools such as GreenroadsTM (Muench et al. 2011), INVEST (Reid et al. 2015), I-LAST (Knuth and Fortmann 2010), GREENLITES (McVoy et al. 2010; NYSDOT 2008). However, there is a need for user-friendly and flexible sustainability evaluation tools that can facilitate the incorporation of regional priorities based on expert judgment. This chapter presents a detailed discussion on the development of a customizable sustainability evaluation tool called “Green Proforma”. Following sections sequentially discuss the details of each step involved in the development and validation of the “Green Proforma” for the sustainability evaluation of roadway infrastructures.  3.1 Multicriteria Analysis: Sustainability Evaluation of Roadways Sustainability based roadway project planning and design has significant benefits in the promotion of environmental stewardship, protecting infrastructure from rapid degradation, leveraging investments and engaging multi-disciplinary experts to work as teams. Researchers and decision-makers have often highlighted the need of decision-making frameworks that evaluate sustainability metrics of design, construction and maintenance activities for roadway infrastructures (Binney 2010; Sahely et al. 2005; Tighe and Gransberg 2013). The frameworks should support rational decision-making under uncertain and multicriteria environment (Gil and Duarte 2013; Sahely et al. 2005; Sharifi and Murayama 2013). Uncertainty is inevitable as the data and benchmark for indicators are based on imprecise information and vague human understanding (Gil and Duarte 2013; Khan et al. 2002; Khatri et al. 2011; Reza et al. 2013; Sharifi and Murayama 2013).   This study develops a fuzzy-based customizable sustainability evaluation tool that enables the evolution of more robust sustainability evaluation system for roadway project planning, design and construction activities. The sustainability evaluation tool is called “Green Proforma for Roadway Infrastructures”. The Green Proforma evaluates sustainability indices using Fuzzy synthetic evaluation technique. As discussed in Section 2.4.4, FSE involves: (1) development of aggregation framework; (2) fuzzification of indicators benchmarks and inputs; (3) aggregation of indicators; (4) defuzzification of aggregated fuzzy sets to generate sustainability indices. Specific steps in developing the “Green Proforma” for roadway infrastructures are summarized in Figure 3.1. Following sections elaborate further on each of the steps involved in development of Green Proforma.  36   Figure 3.1 Methodology for evaluating roadway infrastructure sustainability  3.2 Green Proforma: Sustainability Evaluation Framework Anderson (2008) described sustainability in civil engineering as a quality of the system that adequately sustains both “human values” and “natural laws”. For reasons discussed in Section 2.1, this study applies this definition to develop a new roadway sustainability evaluation framework. Development of sustainability evaluation frameworks requires characterization of system’s sustainability with specific objectives and criteria that support “natural laws” and “human values”. Each criteria is assessed using specific sustainability indicators or sustainable performance measures which are available from several state-of-the- art guidelines such as GreenroadsTM (Muench et al. 2011). Specific to a roadway infrastructure project, following sustainability objectives and/or attributes were identified using system-based definition of sustainability( Anderson 2012, 2008):  Objectives/attributes related to “Human Values” include:   Accessibility: The roadway infrastructure should provide pedestrian, transit, HOV and bike access through the use of best practices and proper consultation with stakeholders as well as community members.  Safety and Mobility: The roadway infrastructure development should ensure safe and smooth travel of people using sidewalks, bicycles, and vehicles e.g. using intelligent transportation systems.  Economy: The roadway infrastructure should be planned, designed, constructed and maintained for economic efficiency e.g. monitoring pavement condition for effective maintenance, designing long-life pavements.  Develop a sustainability evaluation framework by identifying and linking sustainability objectives, indicators and benchmarks Link the process of applying the framework with decision-making phases of roadway projects Apply Fuzzy Synthetic Evaluation technique for fuzzifying and aggregating indicators Develop an Excel based tool "Green Proforma" and perform scenario analysis for validation 37  Objectives/attributes related to “Natural Laws” include:  Resource Efficiency: The roadway infrastructure development should minimize the use of virgin or natural resources e.g. using regional construction materials, reusing existing pavements.  Environmental Quality: The roadway infrastructure development should enhance the air and acoustic quality of the local environment by implementing appropriate measures e.g. congestion pricing, vegetation planning, using permeable pavements, reducing tire-pavement noise.  Ecological Protection: The roadway infrastructure development should protect existing water bodies, land and ecological systems e.g. protecting banks of the stream along highway with bio-stabilization measures, providing ecological connectivity.  Figure 3.2 illustrates the hierarchical framework linking criteria with sustainability objectives identified earlier for assessing sustainability of roadway infrastructure projects. The indicators and benchmarks for each criteria are provided in Appendix A1. The criteria were obtained from GreenroadsTM rating system (Muench et al. 2011). A single criteria may appear to be related to multiple objectives (or components) at the same time, as is the case with most of the indicators in GreenroadsTM (Muench et al. 2011). Since the criteria were ultimately aggregated to a single index, the most direct relationship is chosen rather than categorizing the criteria to an indirectly related objective. This approach reduces the complexity of sustainability evaluation framework and ensures the much needed simplicity and user-friendliness of the sustainability evaluation process (Andreas et al. 2010). As an example, “Stormwater Cost Analysis” criterion in GreenroadsTM has relationship with ecology and economy at the same time among the sustainability components. But due to the direct economic intent of this criterion, the approach in this study relates it to Economy only, as Ecological sustainability component is only related indirectly.  The developed framework enables the synthesis of a range of criteria and indicators representing best practices to aggregated indices. Aggregated measure or indices signify the relative state of achievements for specific sustainable development goals or objectives (Sahely et al. 2005). Sahely et al. (2005) reasoned that indicators alone are not sufficient for determining state of progress towards sustainability. Evaluating aggregated the measure of sustainability also enables the operationalization of sustainability initiatives. Such measures help in making effective decisions regarding the needs of upgrading sustainability in infrastructure systems.  38   Figure 3.2 Sustainability evaluation framework (criteria from Muench et al. 2011)Aggregated Sustainability Index Accessibility Transit Access Pedestrain Access Bicycle Access Context Sesnitive Solutions Scenic Views Safety and Mobility Intelligent Transportation Systems Safety Audit Economy Stormwater Cost Analysis Pavement Performance Tracking Long Life Pavements Contractor Warranty Resource Efficiency Pavement Reuse Earthwork Balance Recycled Materials Regional Materials Life Cycle Assessment Site Recycling Plan Quality Management System Energy Efficiency Warm Mix Asphalt Ecological Protection Ecological Connectivity Habitat Loss Environmental Management System Runoff Flow Control Runoff Quality Light Polution Environmental Training Permeable Pavement Environmental Quality Traffic Emissions Reduction Fossil Fuel Reduction Site Vegetation Equipment Emission Reduction Paving Emission Reduction Cool Pavements Sustainability Attributes/Objectives Sustainability Criteria 39  3.3  Green Proforma: Sustainability Evaluation Process The sustainability evaluation framework was implemented in an Excel-based decision support tool called the “Green Proforma for roadways.” This tool evaluates sustainability for the planning, design and construction aspects of roadway infrastructure projects. Illustration 3.1 shows the main screen of the tool. Other snapshots of the tool interface are provided in Appendix A2. The mathematical formulation for evaluating sustainability indices using the tool have been discussed in the following sections.  Illustration 3.1  Main screen of Green Proforma   Green Proforma provides a platform for the stakeholders to customize model and input parameters for sustainability evaluation. The sustainability evaluation process using the Green Proforma consists of three phases as shown in Figure 3.3. The proposed sustainability evaluation process is based on Deming’s Wheel or Plan-Do-Study-Act cycle. The PDSA cycle is applied to continuously improve the product or process quality (Taylor et al. 2013).   PHASE 1: Definition of System Boundary System boundary defines the range of criteria that are applicable for sustainability evaluation of a project under consideration. This process is essential to incorporate expert judgment, regional constraints, and project specific priorities. The Green Proforma allows the decision makers or experts to modify the selections, weights and benchmarks of sustainability indicators as part of defining the evaluation criteria that is realistic for a proposed development. The users also provide their preference to sustainability objectives in the Green Proforma e.g. pro-environment user will assign a higher weight to objectives such as resource efficiency, ecological protection, and environmental quality. This phase is crucial for obtaining meaningful results. 40   PHASE 2: Input Field Information and Assess Results When the project begins, users can provide the observed values for each sustainability indicator. When all the relevant inputs are obtained, the Green Proforma computes the sustainability indices and generates a radar diagram. Each edge of the radar chart represents performance across each sustainability objectives. In addition, the thermometer chart shown in Illustration 3.5 also provides the overall sustainability index.  PHASE 3: Evaluation and Decision-Making The third level of sustainability evaluation using Green Proforma relates to decision-making and improving system boundary developed in Phase 1. Depending on the level of performance indices, the decision-making team can take a systematic action to address the limitations that cause deficiencies in obtained performance indices. Since the tool allows the user to change indicators, benchmarks, and weights, there is a flexibility to iteratively evolve the sustainability analysis process based on improved targets (i.e. updated benchmarks) and latest best practices applicable (based on new research and experience). The learned lessons from the undertaken sustainable roadway project and information from the updated literature can be translated to improved system boundary in Phase 1.  Figure 3.3 Iterative sustainability evaluation process 1. DEFINE SYSTEM BOUNDARY  Identify Indicators and Benchmarks  Prioritize Indicators and Objectives  2. INPUT AND OUTPUT  Input field information  Evaluate Sustainability Indices 3. EVALUATION AND DECISION MAKING  Identify Deficiencies  Lessons Learned   Improve System Boundary Project Planning Project Implementation 41  Following sections elaborate more on mathematical formulations and analysis process for generating sustainability indices.  3.4 Fuzzification of Indicators An indicator is a measure of executing best practices to different levels of achievements.  Benchmarking involves establishing reference points for the extent of applicable, and achievable, sustainable best practices (Sharifi and Murayama 2014). For a given indicator, benchmarks illustrate the levels of achievement in numeric or non-numeric representation. Benchmarking allows decision-makers to establish targets for the specific best practices. The benchmark targets can be compared with the observed state of each indicator to assess the achieved sustainability of the system. State-of-the-art sustainability rating systems often differ in types of indicator and benchmarks used for same criteria.   Sharifi & Murayama (2014) conducted a comprehensive review of seven state-of-the-art neighborhood sustainability evaluation tools and suggested that a common indicator and benchmarking system should be established as a fixed baseline for making comparisons. However, it is also true that the benchmarks should be based on the limitations and opportunities offered by a given development context which varies from location to location and time to time. Rodríguez López and Fernández Sánchez (2011) concurred with this notion and mentioned that it is critical to monitor continuously and improve benchmarks in order to validate the ranges assigned. Decision on benchmark values can also differ from one expert to another based on their knowledge and expertise in the relevant area. Therefore, benchmarking or establishing desirable level of achievements for sustainability initiatives is based on uncertainties arising from limitations in human knowledge, available resources and available data. In this study, fuzzy set theory is used to model indicator benchmarks and inputs, and direct assignment method is used to weigh the indicators. This process is based on fuzzy synthetic evaluation (FSE) used by Khatri et al. (2011), Rajani et al. (2006), Sadiq and Rodriguez (2004) and Sadiq et al. (2004). Sustainability evaluation and condition rating using fuzzy synthetic evaluation techniques for infrastructure systems include fuzzification, aggregation and defuzzification as the major steps (Rajani et al. 2006).   3.4.1 Defining and Fuzzifying Benchmarks Fuzzification process for benchmarks entails the application of fuzzy set theory to indicator benchmarking system. Based on the approach used by Khatri et al. (2011), Rajani et al. (2006) and 42  Sadiq and Rodriguez (2004), following major steps were identified for fuzzifying the benchmarking system: I. Defining performance measures or indicators for each criteria II. Establishing a common linguistic scale for developing benchmarks of each indicator III. Establishing indicator values corresponding to each linguistic scale IV. Modeling benchmarks as fuzzy membership functions  In this study, the indicators were identified from GreenroadsTM rating system (Muench et al. 2011). Full list of criteria and indicators from GreenroadsTM rating system (Muench et al. 2011) are attached as Appendix A1. A common linguistic scale can be used to develop benchmarks for all the indicators.  In this study, a five-tuple set of the membership function was used in which indicator values were divided across five levels of understandable linguistic scale as shown in Table 3.1.   Table 3.1 Input for the benchmark of indicators (data from Muench et al. 2011) Criteria Indicators V. Poor Poor Fair Good V. Good Long-Life Pavements % of regularly trafficked lanes designed for long-life (>= 40 years) 0.0 0.2 0.4 0.6 0.7 0.4 0.6 0.7 0.8 1.0  Indicator values were assigned to each of the linguistic scales based on the data in the GreenroadsTM rating system (Muench et al. 2011). Indicator values for each of the linguistic scale (i.e. benchmarking) were assigned by entering values in the cells with white background (as shown in Table 3.1), which are open for input by the users. The top row values in Table 3.1 represent the least values and bottom row values represent the highest values for each linguistic scale. The bottom row is entirely white which means the user can define the highest values possible for each quantitative indicator across the 5-level linguistic scale. The grayed cells represent the numbers calculated using values in white cells. Illustration 3.2 illustrates the five levels of performance (or linguistic scale) with different membership functions.   The range (low to high) of each performance levels (or benchmark scale) is overlapping with immediately surrounding performance levels. The overlap represents the unclear or fuzzy benchmark boundaries due to the vagueness, ambiguity and imprecision in human understanding. Fuzzy logic enables the representation of all quantitative and qualitative indicator in a common numeric form called Fuzzy sets (Rajani et al. 2006). The triangular fuzzy functions for each linguistic scale were 43  developed such that the most likely values (membership = 1) of one linguistic scale coincides with the minimum value (membership = 0)  of next linguistic scale and the maximum value (membership = 0) of the preceding linguistic scale (as shown in Illustration 3.2). The encircled points on Illustration 3.2 represent indicator values provided by the user in white cells of input benchmark field (in Table 3.1). For every membership function, the user only defines one extreme value i.e. on higher side (except for “V. Poor performance level” where minimum value also needs to be defined to close the range of values under consideration).   Illustration 3.2  Membership functions for the benchmarks of “Long Life Pavement” criteria based on values from Table 3.1 (data from Muench et al. 2011)  3.4.2 Fuzzifying Inputs and Generating Fuzzy Sets For large roadway projects, the observed values for the indicators are not crisp and are often characterized by imprecision, ambiguity and subjectivity. In such cases, an uncertainty handling mechanism that allows the user to provide a range of uncertain field information is significantly beneficial. Khatri et al. (2011) applied a fuzzification approach to inputs as well so that uncertain inputs can be mapped onto the pre-defined benchmark membership functions to generate fuzzy sets as a representation of indicator values. In this study, a similar approach was adopted, however with some modifications.   Table 3.2 shows an example of input cells in the Excel tool, which accommodates two values that are linguistically labeled as “low” and “high” field observation for a given indicator. A middle value is calculated as an average of the two provided values as inputs. Then the three values are plotted as a triangular fuzzy number (as shown in Illustration 3.3). The intersection of input fuzzy function with benchmark membership functions returns a fuzzy set (see Illustration 3.3 and Table 3.2). For multiple 00.20.40.60.810% 20% 40% 60% 80% 100%Membership Values % of Regularly trafficked lanes designed for Long-life V. PoorPoorFairGoodV. Good44  intersections with one performance level, the intersection point with the highest membership value is selected as a representative number. When the intersected membership values are obtained, the absolute membership values are further normalized so that the membership grades are relatively distributed across the 5-level linguistic scale.  Table 3.2 Example indicator input value (indicator from Muench et al. 2011) Indicator Low High % of Regularly Trafficked lanes designed for long-life 30 40   Illustration 3.3  Intersection of benchmark and input TFNs (data from Muench et al. 2011)  Table 3.3 Fuzzy Set V. Poor Poor  Fair Good V. Good 0.33 0.67  0.00 0.00 0.00  For qualitative indicators, a similar procedure was adopted but with pseudo-numeric values to fuzzify and plot membership functions. Each qualitative representation has an equidistant pseudo-numeric range as benchmarks. When two different qualitative observations are provided, corresponding pseudo-numeric values are generated, and a similar fuzzification process was followed to obtain the fuzzy sets. The cells for fuzzy sets are encoded with the conditional formatting of color intensity with respect to membership values. The higher the fuzzy set values, the more intense is the green color in that zone as shown in Table 3.3. This way, it is convenient for the user to visualize where the field information lies with respect to reference points defined during benchmarking process.  0.67  0.00  0.40  0.80  00.20.40.60.810% 20% 40% 60% 80% 100%Membership Values % of Regularly trafficked lanes designed for Long-life V. PoorPoorFairGoodV. GoodIntersection of Inputwith Benchmark MF45  3.5 Prioritization and Aggregation In this study, a direct assignment procedure based on Hwang and Yoon (1981) was applied for evaluating relative weights of indicators. The weighting process was based on a linear and discrete 1-5 point scale represented by corresponding linguistic descriptions: “Very Low”; “Low”; “Medium”; “High”; and “Very High”.  This technique allows the user to implement the prioritization process using linguistic descriptors that are easier to comprehend (see Illustration 3.4).   Illustration 3.4 Snapshot of tool interface for weighting the criteria (criteria from Muench et al. 2011)  When one of the linguistic scales is selected from the drop-down menu (Illustration 3.4), corresponding numbers on a 1-5 scale is generated. The generated numbers are normalized across the complete list of indicators underneath each category of sustainability objectives. The normalization process results in the relative weights of indicators. After obtaining the weights, aggregation process is followed to determine the fuzzy sets for sustainability objectives which are further aggregated to determine an overall sustainability fuzzy set for a complete roadway system.  3.6 Defuzzification and Result Generation The aggregated fuzzy sets can be converted to crisp numbers using defuzzification methods in order to simplify the representation of system sustainability. In this study, the “center of gravity” method was chosen to defuzzify the aggregated fuzzy sets for sustainability objectives and overall sustainability rating/index. In this method, center of gravity of the fuzzy set is located, and corresponding x-scale value is returned as a crisp value. Subsequently, the crisp values (obtained by 46  the defuzzification of fuzzy sets for each objective of development) are plotted on a radar diagram to illustrate the relative performance across different objectives.   Illustration 3.5 illustrates the radar diagram showing the output indices for the overall sustainability rating and underlying objectives. The diagram fuses all the information related to sustainability criteria in a single meaningful display called the Roadway Project Sustainometer. The performance indices are mapped on color-coded zones representing performance levels similar to those identified earlier for each indicator benchmarks. The zones do not have defined boundaries as the benchmarking process recognizes the fuzziness or vagueness in establishing the specific and non-overlapping extreme performance values for each indicator on a linguistic scale.  Illustration 3.5 Roadway project sustainometer   3.7 Scenario Analysis using Green Proforma Roadway projects vary in scope significantly depending on the location where they are implemented, stakeholder’s priorities, climatic conditions, etc. The Green Proforma enables the incorporation of expert judgment and regional constraints in the benchmarking and prioritization of sustainability objectives and indicators. To demonstrate the flexibility of tool to changes in decision-makers preferences and priorities, this section conducts scenario analysis for the following conditions.   47  Two location- based scenario were assumed:  An ecocentric: Roadway projects that require higher attention to the local ecology, topography, resources, and environment. Potential roads, in this case, can include those passing through environmentally or ecologically sensitive zones e.g. highways in hilly regions or heavily forested zones.  An anthropocentric: Roadway projects that require higher attention to societal benefits such as safety, economy, and accessibility. Potential examples in this category can include urban roads such as locals, collectors, arterials, etc.  Each scenario was further subdivided into two performance levels:  High-Performance: The roadway project adequately fulfills the criteria that require higher attention.  Low-Performance: The roadway project does not fulfill the criteria that require higher attention.  Three priority levels were established for assessing the effect of decision-makers preference on output sustainability rating:  Pro-environment: 80% of weight distribution to the objectives related to natural laws;  Pro-Socioeconomic: 80% of weight distribution to the objectives related to human values;  Neutral: equal weight distribution to all the objectives.   Above scenarios are summarized in Figure 3.4. These scenarios were developed by assuming “reasonable” values for the relevant indicators. Indicator benchmarks, weights, and inputs were provided to the Green Proforma to generate a radar chart illustrative of each scenario. The weights for each indicator were based on points allocated in the GreenroadsTM rating system (Muench et al. 2011). Since the scenarios are hypothetical, benchmarks were arbitrarily assigned to each performance levels based on the information available in GreenroadsTM rating system (Muench et al. 2011). Indicator input values were assumed according to the scenario under consideration e.g. for a high-performance ecocentric scanrio, indicators relevant to the protection of natural laws were assigned values close to “good” and “very good” performance levels. In this way, the radar chart for each scenario was generated.  48  Each scenario was simulated in the tool and Illustration 3.6-9 illustrates the results of each scenario. Illustration 3.10-13 depict the sensitivity of overall sustainability rating based on decision makers’ priority or preference towards sustainability objectives.      Figure 3.4 Potential scenarios  Table 3.4 illustrates the summary of results showing the change in overall sustainability level (based on the closeness of index values to the center of linguistic scales) with respect to the change in decision-makers attitude for the sustainability objectives (i.e., pro-environment, pro-socioeconomic and neutral). As shown in Table 3.4, both high and low performance ecocentric scenarios were rated as “Fair” by the pro-socioeconomic attitude of decision-maker. However, the pro-environment Development of Scenarios Ecocentric High Performance Pro-Environment Pro-Socioeconomic Neutral Low Performance Pro-Environment Pro-Socioeconomic Neutral Anthropocentric High Performance Pro-Environment Pro-Socioeconomic Neutral Low Performance Pro-Environment Pro-Socioeconomic Neutral Priority Level Case Performance Level 49  attitude of the decision-makers changes the rating of High Performance Ecocentric scenario to “Good” and the rating of Low Performance Ecocentric scenario to “Poor” sustainability levels. Clearly, the latter is more realistic. The change in decision-makers attitude can alter the outcome by nearly 15% that is roughly equivalent to skipping one of the sustainability performance levels. It is clear from the scenario analysis, that the overall rating can be misrepresented if prioritization is not adequately considered for sustainability objectives.  Table 3.4 Sensitivity of overall sustainability level based on decision maker’s preference Scenarios Sustainability Level Pro-Socioeconomic Neutral Pro-Environment High-Performance Ecocentric Fair Good Good Low-Performance Ecocentric Fair Poor Poor High-Performance Anthropocentric Good Good Fair Low-Performance Anthropocentric Poor Fair Fair     Illustration 3.6 High-performance ecocentric scenario  50    Illustration 3.7 High-performance anthropocentric scenario     Illustration 3.8 Low-performance ecocentric scenario 51     Illustration 3.9 Low-performance anthropocentric scenario    Illustration 3.10 Sustainability of high-performance ecocentric scenario for the pro-socioeconomic (left), neutral (middle) and pro-environment (right) preferences of decision-maker 52   Illustration 3.11 Sustainability ratings of low-performance ecocentric scenario for the pro- socioeconomic (left), neutral (middle) and pro-environment (right) preferences of decision-maker   Illustration 3.12 Sustainability ratings of high-performance anthropocentric scenario for the pro-environment (left), neutral (middle) and pro- socioeconomic (right) preferences of decision-maker 53    Illustration 3.13 Sustainability ratings of the low-performance anthropocentric scenario for the pro-environment (left), neutral (middle) and pro- socioeconomic (right) preferences of decision-maker  3.8 Summary and Discussion In this study, a customizable sustainability evaluation tool called Green Proforma was developed using FSE technique. The Green Proforma for roadway projects facilitates a team of experts and stakeholders in assessing the level of sustainability considerations achieved while planning, designing, construction and maintenance of existing or new roadways. An iterative sustainability evaluation framework was proposed for improving the parameters for evaluating the sustainability of roadway projects using Green Proforma. The Green Proforma fully integrates the capability to handle imprecise benchmarks and inputs using FSE technique. Scenario analysis was performed using the Green Proforma to demonstrate the influence of decision-maker’s preferences on overall sustainability indices. Although Green Proforma offers significant improvement over existing rating systems, the tool has some limitations. The capabilities and limitations of sustainability evaluation process using the Green Proforma are summarized below:  Capabilities of Green Proforma:  Defining benchmarks on a common and understandable linguistic scale for all qualitative and quantitative indicators is convenient for users.   The application of fuzzy logic allows users to express uncertainty in defining benchmarks for quantitative indicators and providing field information in the tool.  54   Indicators can be selected or removed from sustainability evaluation process depending on the expert opinion and relevance to the given development context. This implies that the tool is adaptable to a given context and expert opinion.  Weighting of indicators is customizable. A five-level qualitative scale and accompanying visual display of relative weights enable the users to prioritize the indicators conveniently.   Classification of overall rating for different achievement levels (“Gold” or “Silver” certification) is not needed as is the case in the state-of-the-art rating systems. This is one of the significant advantages of using fuzzy synthetic evaluation. Defuzzification of aggregated fuzzy sets results in automatic generation of clear and understandable output indices across the pre-defined linguistic scale. The output indices clearly show how well the team met its sustainability goals.  The Green Proforma also possesses the ability to demonstrate the unsustainability of infrastructure projects (unlike other rating systems).  Limitations of Green Proforma:  The input data is needed from multiple sources, departments and firms for a given project. This means that the sustainability evaluation process using the Green Proforma requires a significant commitment of time and resources from multiple stakeholders for meaningful delivery of inputs and realistic evaluation of outputs from the Green Proforma tool.  The tool only considers project planning, design and construction phases as the indicators are taken from GreenroadsTM rating system that is designed primarily for these phases (Muench et al. 2011). Operational phase is also important for effective infrastructure management that has not been included.  The Green Proforma does not possess the capability to model relationship of criteria to multiple objectives or other criteria.  The Green Proforma uses indicators only from GreenroadsTM rating system (Muench et al. 2011). For a comprehensive and holistic evaluation, the Green Proforma should offer a comprehensive list of criteria and indicators available in other guidelines also. This will allow for the consideration of criteria that is not available in the GreenroadsTM rating system (Muench et al. 2011)  The fuzzification and defuzzification of larger number of indicator values within the Excel interface can be computationally intensive.  55  Chapter 4 Sustainability Evaluation of Pavements In this chapter, life cycle economic and environmental evaluation was conducted for the three types of pavements, i.e., Rigid, Flexible and Geosynthetically Reinforced Flexible Pavements. Functional units were designed for each alternative under eight scenarios based on the different combinations of subgrade conditions and traffic parameters. The chosen traffic levels represent local roads with traffic variation in low-medium volume range. AASHTO (1993) guideline was used to design pavements for each scenario. Maintenance and Rehabilitation schedules were adapted from published literature using expert judgment (basics discussed in Section 2.2). Uncertainties were established for estimating life cycle M&R interventions and costs using Fuzzy-based approach (basics discussed in Section 2.4). Life Cycle Cost Analysis was carried out using Net Present worth (NPW) method. Life Cycle Assessment was conducted through the use of Athena Highway Impact Estimator. Fuzzy composite programming technique was used to evaluate overall economic and environmental sustainability (basics discussed in Section 2.4.5). Following sections sequentially discuss the details of each step involved in the sustainability evaluation of pavement alternatives.  4.1 Life Cycle Thinking Approach for Pavements Pavements are one of the most expensive component of roadway infrastructure. Municipalities and government agencies are interested in choosing pavement alternatives that exhibit superior life cycle sustainability performance in terms of both costs and environmental impacts (Nassiri et al. 2013). Life cycle thinking approach requires a sustainable pavement alternative to minimize life cycle economic and environmental burdens. The life cycle economic evaluation can be performed through the estimation of Net Present worth (NPW) using LCCA. The life cycle environmental evaluation can be performed through the estimation of indicators such as Global Warming Potential, Acidification Potential, Human Health Risk, Eutrophication Potential, Ozone Depletion Potential, Smog Potential, and Primary Energy using LCA Tools.   Figure 4.1 illustrates an overview of methodology used in this study to compute life cycle environmental and economic sustainability indices for the pavement alternatives. The process adopted for life cycle sustainability evaluation was based on fuzzy composite programming (FCP) used by Sadiq (2001), Khan et al. (2002) and Sadiq et al. (2005). GreenPro-I methodology developed by Khan et al. (2002) also uses FCP for risk-based life cycle assessment and multi-criteria decision making. In this study, FCP incorporates weights of computed indicators using the values available in published literature for environmental indicators. Furthermore, different weight distribution settings 56  were created to represent the influence of decision makers’ preference for environmental and economic concerns on the overall sustainability index values for each scenario.    Figure 4.1  Life cycle thinking methodology under uncertainty  As discussed in Section 2.4.5, FCP involves the (1) identification of alternatives, (2) definition and evaluation of indicators, and (3) normalization and aggregation of indicators for the final ranking of alternatives. Detailed steps for the computation of sustainability indices have been discussed in the following sections.  4.2 Development of Scenarios In this section, pavement analysis and design was carried out to characterize each of the alternatives under different scenarios. There were basically eight different scenarios under which each of the pavement alternatives will be modelled. The scenarios were based on different subgrade conditions and traffic levels. Illustration 4.1 identifies the 32 Alternatives for sustainability evaluation in this study. California Bearing Ratio (CBR) represents subgrade condition. CBR values were assumed as 5% and 10% representing low-high range of subgrade conditions. Average Annual Daily Traffic (AADT) is the level of average traffic anticipated on the pavements. The range of AADT assumed in this study was 250-2000 vehicles per day (vpd) for Low-Medium Volume Municipal Roadways based on Ahmed et al. (2006) and FHWA (2013). Moreover, this range represents the municipal local roadways which constitute majority of roads in Canadian Municipalities (Illustration 4.1). Four AADT values were assumed for the design and analysis purpose: 250, 500, 1000 and 2000 vpd.   Characterize Alternatives for each Scenario Characterize Uncertain Inputs as Fuzzy Numbers for LCCA and LCA Conduct LCCA and generate Fuzzy Numbers for Construction, Maintenance & Rehab Costs Conduct LCA and generate Fuzzy Numbers for Environmental Indicators Normalize and Aggregate Indicators to generate environmental and economic sustainability indices for ranking alternatives 57   Illustration 4.1  Network summary for 118 Canadian municipalities  (data from Federation of Canadian Municipalities 2012)  For each combination of CBR and AADT, four pavement types were designed. Plain Cement Concrete “PCC” is used to denote rigid pavements. Asphalt Concrete “AC” is used to denote flexible pavements. GeoSL AC is used to represent Geosynthetically reinforced flexible pavement designed for longer service life (SL). GeoBC AC is used to represent Geosynthetically reinforced flexible pavement designed for reduced base-course (BC) thickness. Two (2) CBR values, four (4) traffic levels and four (4) pavement types means eight (8) scenarios with four (4) pavement types each. In total, there were thirty two (32) Alternatives to model.    Illustration 4.2 Scenarios for pavement life-cycle thinking approach  2% 22% 17% 50% 8% Highways / ExpresswaysArterialCollectorLocalAlleysScenario Development CBR 5% AADT 250 PCC AC GeoSL AC  GeoBC AC AADT 500 PCC AC GeoSL AC  GeoBC AC AADT 1000 PCC AC GeoSL AC  GeoBC AC AADT 2000 PCC AC GeoSL AC  GeoBC AC CBR 10% AADT 250 PCC AC GeoSL AC  GeoBC AC AADT 500 PCC AC GeoSL AC  GeoBC AC AADT 1000 PCC AC GeoSL AC  GeoBC AC AADT 2000 PCC AC GeoSL AC  GeoBC AC 58  Each scenario of pavements was modelled using WinPAS tool to design pavements. WinPAS is a pavement analysis and design tool that has been developed by American Concrete Pavement Association (ACPA). The tool is based on AASHTO (1993) guide for the design of pavement structures (ACPA 2012). The WinPAS software aids in designing both flexible and rigid pavements. Also, the software is capable of conducting LCCA for comparing various design alternatives. A user can make use of the WinPAS tool that displays the following key areas of functionalities(ACPA 2012):  1. Project: To input project specific information such as name, location, agency, design engineer etc. 2. Estimate ESALs: To evaluate ESALs for both flexible and rigid pavements by providing data such as design start year traffic volume, vehicle distribution, traffic growth rate etc. 3. Design/Analysis: To provide data specific to flexible and/or rigid pavements to evaluate the each of the pavement structural layer thicknesses. 4. Overlays: To design the overlay thickness for appropriate rehabilitation of selected pavement designs 5. Life Cycle Costing: To conduct economic evaluation by providing data such as discount rate, inflation rate, construction costs, M&R costs etc. 6. Reports: To generate different reports for design, overlays, traffic etc.  4.2.1 Input Parameters for Characterizing Alternatives The input parameters that were common to all the pavement designs are shown in Table 4.1 while the alternative specific input parameters are identified in Table 4.2.  Table 4.1  Common input parameters for pavement alternatives Input Parameters Units Values References/ Assumptions Average Annual Daily Traffic (AADT) Vehicles per day 250, 500, 1000, 2000 Low-Medium Volume Municipal Roadway Service Life Years 20 (BC MoT 2004) 59  Input Parameters Units Values References/ Assumptions Vehicular Distribution -  % Axles and Load values Cars 95 Two 2 kip single axles Commercial Vehicles 4 12 kip single axle and 34 kip tandem axle Buses 1 10 kip single axle and 24 kip single axle  Percentage distribution adapted from ARA (2007) for Collector/Local Roads. Axle Loads based on WinPAS defaults Directional Distribution % 50 Assumed Lane Distribution Factor % 100 One lane in one direction Traffic Growth Rate % 3.0 Assumed Terminal Serviceability - 2.5 (BC MoT 2004) Subgrade California Bearing Ratio (CBR) % 5, 10 Assumed Thickness of layer above subgrade Inches (mm) 8 (200) Assumed as constant for all pavement types Lane width m 4 Assumed Lane length KM 1 Assumed  Table 4.2  Alternative specific pavement design inputs and assumptions Pavement Type Input Parameter Units Values References / Assumptions Rigid or Concrete Pavement Reliability % 75 (BC MoT 2004) Overall Standard Deviation - 0.35 Assumed Flexural Strength Psi (MPa) 667 (4.6) (C-SHRP 2002) 60  Pavement Type Input Parameter Units Values References / Assumptions Modulus of Elasticity ksi (GPa) 4206 (29) (C-SHRP 2002) Load Transfer Coefficient - 3.8 No dowels and No edge support, PCC Modulus of Subgrade Reaction Psi/in (MPa/m) 302.8 (82)  440.2 (119.2) Estimated from WinPAS for 5%  and 10% CBR Drainage Coefficient - 0.80 (C-SHRP 2002) Initial Serviceability - 4.5 (C-SHRP 2002) Terminal Serviceability - 2.50 (C-SHRP 2002) Depth of Rigid Foundation ft >10 Assumed Resilient Modulus of the Sub-base Psi (MPa) 15000 (103.4) (BC MoT 2004) Flexible or Asphalt Pavement Reliability % 75 (BC MoT 2004) Overall Standard Deviation - 0.45 (BC MoT 2004) Subgrade Resilient Modulus Psi (MPa) 5842 (40.3) 9388.7 (64.7) Estimated from WinPAS for 5%  and 10% CBR Initial Serviceability - 4.2 (BC MoT 2004) Terminal Serviceability - 2.50 (BC MoT 2004) Structural Layer Coefficient for Asphalt - 0.4 (BC MoT 2004) Structural Layer Coefficient for Granular Base - 0.14 (BC MoT 2004) Structural Layer Coefficient for Granular Sub-base - 0.10 (BC MoT 2004) Drainage Coefficient Granular Base - 0.95 (BC MoT 2004) Drainage Coefficient Granular Sub-base - 0.80 (BC MoT 2004)    61  4.2.2 Pavement Section Designs WinPAS Software was used to generate thicknesses for Plain Cement Concrete (PCC), Asphalt Concrete (AC) and Base Course (BC) layers. Table 4.3 and Table 4.4 show the final results for Rigid and Unreinforced Flexible Pavements. The minimum threshold values for thicknesses were established as 100 mm for base course, 100 mm for PCC and 50 mm for AC layers for low volume roadways as per AASHTO (1993) guidelines.  Table 4.3  Unreinforced flexible pavement design  CBR (%) AADT (vehicles per day) 250 500 1000 2000 5 Asphalt concrete (mm) 75 75 75 75 Aggregate Base Course (mm) 125 175 225 300 10 Asphalt Concrete (mm) 50 75 75 75 Aggregate Base Course (mm) 125 100 125 200  Table 4.4  Rigid pavement design  CBR (%) AADT (vpd) 250 500 1000 2000 5 PCC (mm) 150 150 175 200 10 PCC (mm) 125 150 175 200   TBR and LCR methods were used to design geosynthetically reinforced flexible pavements. A detailed methodology for designing Geosynthetically reinforced pavements has been provided by Berg & Associates (2000) and overviewed in Chapter 2.  Layer Coefficient Ratio (LCR) values for different subgrade conditions were provided by Maccaferri (2001) for Geogrids. Traffic Benefit Ratio (TBR) for different ESAL values were provided by Al-Qadi et al. (1997) (Al-Qadi and Yang 2007). The results of the Geosynthetically reinforced flexible pavement design are summarized in the Table 4.5 and Table 4.6. 62  Table 4.5  Geosynthetically reinforced flexible pavement design based on TBR values from Al-Qadi et. al (1997)  Basic Data Layer Coefficient Ratio Method CBR AADT (vpd) Original Service Life (years) Original Structure Number, SN Asphalt Concrete (mm) Base Course (mm) LCR (from Graph) Improved Service Life (years) BC Reinforced (mm) Improved SN 5    250 20 62.63 75 125 1.45 37 100 70.1 500 20 69.28 75 175 1.45 42 125 79.8 1000 20 75.93 75 225 1.45 43 175 89.5 2000 20 85.90 75 300 1.45 51 225 103.9 10    250 20 52.63 50 125 1.43 39 100 59.8 500 20 59.30 75 100 1.43 37 100 65.0 1000 20 62.63 75 125 1.43 32 100 69.8 2000 20 72.60 75 200 1.43 43 150 84.0  Table 4.6  Geosynthetically reinforced flexible pavement design based on LCR values from Maccaferri (2001) Basic Data Traffic Benefit Ratio Method CBR AADT (vpd) Original Service Life (years) Original Structure Number, SN Asphalt Concrete (mm) Base Course (mm) TBR  (from Graph) Improved Service Life (years) BC Reinforced (mm) Design SN for New BC  5    250 20 62.63 75 125 2.55 37 100 51.9 500 20 69.28 75 175 2.60 38 100 57.3 1000 20 75.93 75 225 2.63 38 150 64.0 2000 20 85.90 75 300 2.59 38 200 71.6 10    250 20 52.63 50 125 2.55 37 100 43.1 500 20 59.30 75 100 2.60 38 100 47.8 1000 20 62.63 75 125 2.63 38 100 53.5 2000 20 72.60 75 200 2.59 38 125 60.1 63   Illustration 4.3 Comparison chart for geosynthetically reinforced (TBR & LCR) and conventional pavement design  There was a significant improvement in service life and reduction in base course at higher levels of traffic for the approach based on Layer Coefficient Ratio (LCR) Method (Illustration 4.3). However, LCR method gives lower base course reduction as compared to TBR. Moreover, both LCR and TBR based methods provided results that were close in terms of service life improvements (majority within ±7 years) and reduction in the base course thickness (majority within ±25 mm). This finding is in line with the assessment of Berg and Associates (2000) that geosynthetic products generally do not differ significantly in terms of base course reduction or increased performance lives when the reinforced sections are compared with unreinforced sections of pavement. For the life cycle thinking approach, a 35 year estimate of improved service life was chosen for all the geosynthetically reinforced alternatives. Moreover, for base course thickness, a conservative approach was adopted and highest value obtained from either of the two methods was chosen (i.e. one which gives the least reduction from original).  02550751001251501752002252502753003250510152025303540455055CBR = 5,AADT =250CBR = 5,AADT =500CBR = 5,AADT =1000CBR = 5,AADT =2000CBR = 10,AADT =250CBR = 10,AADT =500CBR = 10,AADT =1000CBR = 10,AADT =2000Base Course Thickness (mm) Service Life (years) Scenarios Original Service Life (years) LCR - Service Life (years) TBR - Service Life (years)Original BC (mm) LCR - BC (mm) TBR  - BC (mm)64  4.3 Life Cycle Thinking Approach In this section, the proposed pavement alternatives for each scenario were analyzed over their life cycle phases. The life cycle phases considered for each alternatives were material production, construction and maintenance of pavements. In order to conduct LCA and LCCA, it is crucial to determine the life cycle maintenance and rehabilitation activities for a suitable analysis period. The analysis period is different from design period. Design period represents the time to reach terminal serviceability without any maintenance. Analysis period represents a span of time across which pavement design alternatives should function above a minimum level of service.  For both LCCA and LCA, this study establishes a forty (40) year analysis period. Rangaraju et al. (2008) surveyed LCCA practices for the US states and Canadian provinces and recommended a 40 year LCCA period instead of 30 year period based on their findings. Muench et al. (2011) recommended at least 40 year period for LCCA of alternative pavement types in their GreenroadsTM rating system. AASHTO (1993) recommended analysis period in the range of 30-50 years for high volume urban roads and suggested a lower duration of analysis periods for the low-volume road cases. Therefore, for the roadway scenarios considered in this study, 40 years of analysis period was deemed to be the optimal time period that aligns well with the recommendation of standard guidelines and best practices.   4.3.1 Uncertainty Analysis The uncertainty in life cycle thinking approach is inherent due to the imprecise estimates of cost, timing and extent of pavement M&R activities needed. There is no standard or accurate method to establish these parameters. For LCCA, the M&R schedules are usually developed and adapted based on information available from previous studies and guidelines. The adaptation and modification involves considerable epistemic uncertainty due to the dubious precision and applicability of the available data/information and the vagueness in human judgment and understanding. The accuracy of uncertainty depends on the precision and applicability of data/models used and the available expertise in the field of pavement engineering. Therefore, the assigned uncertainties levels can vary significantly from one expert to another, one project to another and one place to another.  Such uncertainties are effectively handled by modelling the uncertain inputs as fuzzy numbers with different possibility levels. In this study, a triangular distribution was used for modelling fuzzy numbers in the form of minimum, most likely and maximum values. The assumed uncertainties for the key LCA parameters are shown in Table 2.1. The uncertainty in M&R activities was generally 65  higher due to the dubious accuracy and applicability of available information, erroneous guesstimate of pavement performance and vague expression of human understanding. On the other hand, a lower value for uncertainty in construction costs can be used because latest construction cost data is usually available. Table 4.7 shows arbitrary assignment of uncertainties for the construction costs and M&R activity timing and extent. The word “arbitrary” is most appropriate as it means something based on expert opinion and human understanding which is often vague and imprecise. For M&R costs and discount rates, the source of available information is also cited in the Table 4.7.   Table 4.7  Uncertainty values used for characterizing fuzzy inputs Item Uncertainty References/Assumptions Construction Cost ±10% Arbitrary M&R Costs ±20% (BC MoT 2013) M&R Timing ±3.00 Arbitrary M&R Extent ±20% Arbitrary Discount Rate 3%, 4%, 5% (C-SHRP 2002; Muench et al. 2011)  4.3.2 Maintenance and Rehabilitation Schedules Developing pavement maintenance and rehabilitation schedules involves the estimation of the timing and extent of pavement preservation activities over an analysis period. Timing and extent of Pavement M&R activities depends on the type of pavement, level of traffic, climatic conditions, accuracy and applicability of information used and the resources available to implement the M&R activities. Table 4.8 to Table 4.10 present the most likely assessed values for maintenance and rehabilitation schedules modified from information provided by Holt et al. (2011) for different roadway cases. In general, the pavement requires some routine maintenance activities (e.g. rout and seal, partial depth repairs) and rehabilitation activities at the end of service (Full depth repair i.e. Mill and Inlay or slab repairs). In this case, the extent of M&R activities was assigned ± 20% of epistemic uncertainty and the timing of M&R activities was assigned ± 3 years of epistemic uncertainty.   66  Table 4.8  Pavement maintenance schedule for flexible pavements (adapted from Holt et al. 2011, © Transportation Association of Canada, with permission) Flexible AADT 250-500 Yr. % Flexible AADT 1000 - 2000 Yr. % Rout and Seal 10 15 Rout and Seal 10 25 Partial Depth Repair 10 2 Partial Depth Repair 10 2 Full Depth Repair 20 100 Full Depth Repair 20 100 Rout and Seal 27 30 Rout and Seal 27 50 Partial Depth Repair 27 4 Partial Depth Repair 27 5 Full Depth Repair 35 100 Full Depth Repair 35 100 Rout and Seal 40 40 Rout and Seal 40 50  Table 4.9  Pavement maintenance schedule for rigid pavements (adapted from Holt et al. 2011, © Transportation Association of Canada, with permission) Rigid AADT 250 -500 Yr. % Rigid AADT 1000-2000 Yr. % Reseal Joints 10 10 Reseal Joints 10 10 Partial Depth PCC repair 20 2 Partial Depth PCC repair 20 2 Full Depth PCC repair 20 3 Full Depth PCC repair 20 5 Reseal Joints 20 15 Reseal Joints 20 20 Partial Depth PCC repair 35 3 Partial Depth PCC repair 35 5 Full Depth PCC repair 35 5 Full Depth PCC repair 35 10 Reseal Joints 35 20 Reseal Joints 35 20  Table 4.10  Pavement maintenance schedule for geosynthetically reinforced flexible pavements  (adapted from Holt et al. 2011, © Transportation Association of Canada, with permission) Flexible GeoSL AADT 250 - 500 Yr. % Flexible GeoSL AADT 1000 - 2000 Yr. % Rout and Seal 10 15 Rout and Seal 10 25 Partial Depth Repair 10 2 Partial Depth Repair 10 2 Rout and Seal 20 30 Rout and Seal 20 50 Partial Depth Repair 20 4 Partial Depth Repair 20 5 Full Depth Repair 35 100 Full Depth Repair 35 100 Rout and Seal 40 30 Rout and Seal 40 50    67  4.3.3 Implementation of LCA Athena Highway Impact Estimator was used for LCA of alternative pavement scenarios. Each pavement scenario was modelled with all the fuzzy inputs to produce minimum, most-likely and maximum value for output environmental indicators. Arbitrarily assigned transportation distances are provided in Table 4.11. The material compositions for asphalt layers, plain cement concrete, base course and sub-base courses were assumed as following for simplifying the modelling of pavement systems in the Impact estimator:   Plain Cement Concrete: Contains 4% ground granulated blast-furnace slag, 30% fine aggregate crushed stone, 48% coarse aggregate crushed stone, 11.5% portland cement and 6.5% water by weight  Asphalt Concrete: Contains 93% natural aggregates (49% coarse and 44% fine), 1% mineral filler natural and 6% bitumen by weight.  Base Course: Contains 57% fine aggregate natural, 0.05% mineral filler natural and 42.95% coarse aggregate natural by weight  Sub-base Course: Contains 74.67% fine natural aggregate, 0.33% mineral filler natural, and 25% coarse natural aggregate by weight  Table 4.11 Transportation distances Item Distance Bitumen 300 km Crushed Stone Aggregate (Coarse/Fine) 20 km Natural Aggregate (Coarse/Fine) 30 km Portland Cement 107 km Avg. Distance Site to Stockpile 30 km Avg. Distance Plant (PCC/HMA) to Site 30 km Avg. Distance Equipment to Site 30 km  The data for Geosynthetics was not found in the material library of Athena Highway Impact Estimator (v. 2.0). However, the LCA of Geosynthetics was conducted by Frischknecht et al. (2012) for roadway foundations. The study by Frischknecht et al. (2012) found zero emissions related to ozone depletion and at least 1% increase in acidification potential, eutrophication, global warming potential, photo-chemical oxidation, and energy use indicators for the pavement case specific to their study (Frischknecht et al. 2012). The absolute values for geosynthetic LCA results for each indicators 68  were obtained from Stucki et al. (2011) and compared with the results obtained without the use of Geosynthetics from Athena Highway Impact Estimator. Majority of percentage difference in indicator values varied from 0.2% to 1.5%. Since, the LCA of geosynthetic was not representative of Canadian case, a conservative value of triangular fuzzy percentage difference values were assigned for each of the indicators.   Table 4.12 Fuzzy numbers for % increase in environmental indicators % Increase in Environmental Impacts Low Medium High Global Warming Potential (kg CO2 eq) 0.5 1 1.5 Acidification Potential (kg SO2 eq) 0.5 1 1.5 HH Particulate (kg PM2.5 eq) 0.5 1 1.5 Eutrophication Potential (kg N eq) 0.5 1 1.5 Ozone Depletion Potential (kg CFC-11 eq) 0 0 0 Smog Potential (kg O3 eq) 1 2 3 Total Primary Energy (MJ) 1 1.5 2  The environmental indicators obtained from life cycle assessment using IE include global warming potential, acidification potential, particulate matter affecting human health, eutrophication potential, ozone depletion potential, smog potential and energy consumption. Illustration 4.3 shows the estimated ranges of global warming potential for each case in all the pavement scenarios. Similarly, other indicators were also obtained from Life Cycle Assessment using Athena Highway Impact estimator. Results for all other indicators are summarized in Appendix B1.    69   Illustration 4.4 Global warming potential estimates for each alternatives under each scenarios1.00E+061.20E+061.40E+061.60E+061.80E+062.00E+062.20E+062.40E+062.60E+062.80E+063.00E+06HMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCCBR 5 AADT 250 CBR 10 AADT 250 CBR 5 AADT 500 CBR 10 AADT 500 CBR 5 AADT 1000 CBR 10 AADT 1000 CBR 5 AADT 2000 CBR 10 AADT 2000Global Warming Potential (kg CO2 eq) Scenarios Global Warming Potential (kg CO2 eq) Minimum Most Likely Maximum70  4.3.4 Implementation of LCCA In this study, the LCCA was conducted based on a forty year analysis period for consistency. Table 4.13 to Table 4.15 present the most likely estimated costs from available data in Canadian dollars. For reasons already discussed, the assumed uncertainty in costs for construction items was ±10% and for the M&R items was ±20%. Table 4.16 and Table 4.17 present the cost calculation sheet for pavement construction, maintenance and rehabilitation. Illustration 4.4 shows the complete results of Fuzzy based LCCA conducted using the identified uncertainties.  Table 4.13  Construction costs Item Unit Costs (CAD) Units Asphalt Concrete 25.31 sq. m / 75 mm Tack Coat 1.27 sq. m Prime Coat 2.53 sq. m Plain Cement Concrete 151.8 sq. m / 100 mm Base Course 68.31 cum Geosynthetics 2 sq. m Select Granular Sub-base 50.61 cum Excavation 35.42 cum  Table 4.14  Flexible pavement maintenance and rehabilitation unit costs Item Unit Costs (CAD) Units Rout and Seal 8 m Partial Depth Repair  40 sq. m Full Depth Repair  45 sq. m  Table 4.15  Rigid pavement maintenance and rehabilitation unit costs Item Unit Costs (CAD) Units Reseal Joints 14 m Partial Depth PCC repair  130 sq. m Full Depth PCC repair  110 sq. m  71  Table 4.16 Asphalt pavement construction costs   Table 4.17 Asphalt pavement maintenance and rehabilitation costs   Length (m) Width (m) Depth (mm) Qty. Units Low Medium High Units Low Medium High1 Asphalt Concrete 1000 8 75 8000sq. m per 75 mm22.78 25.31 27.84sq. m per 75 mm182232.0 202480.0 222728.02 Tack Coat 1000 8 4000 sq. m 1.14 1.27 1.40 sq. m 4572.0 5080.0 5588.02 Prime Coat 1000 8 4000 sq. m 2.28 2.53 2.78 sq. m 9108.0 10120.0 11132.03Plain Cement Concrete1000 0 150 0sq. m per 100 mm136.62 151.80 166.98sq. m per 100 mm0.0 0.0 0.04 Ba e Course 1000 8 125 500 cum 61.48 68.31 75.14 cum 30739.5 34155.0 37570.55 Geosynthetics 1000 0 0 sq. m 1.80 2.00 2.20 sq. m 0.0 0.0 0.06Select Granular Sub-base1000 8 200 800 cum 45.55 50.61 55.67 cum 36439.2 40488.0 44536.87 Excavation 1000 8 400 3200 cum 31.88 35.42 38.96 cum 102009.6 113344.0 124678.4$365,100.3 $405,667.0 $446,233.7Initial Construction Cost =Item # ItemQuantities Unit Costs (CAD) Amount ($)Low Medium High Low Medium High Low Medium High Low Medium High7 10 13 Rout and Seal 12 15 18 64.00 80.00 96.00 677.47 1021.61 1443.037 10 13Partial Depth Repair1.6 2 2.4 1280.00 1600.00 1920.00 1806.59 2724.29 3848.0917 20 23 Full Depth Repair 80 100 100 1600.00 2000.00 2400.00 94389.62 144955.93 174476.702 27 30 Rout and Seal 24 30 36 64.00 80.00 96.00 999.16 1554.14 2280.1124 27 30Partial Depth Repair3.2 4 4.8 1280.00 1600.00 1920.00 2664.43 4144.37 6080.2932 35 38 Full Depth Repair 80 100 100 1600.00 2000.00 2400.00 72144.22 113865.06 141719.1237 40 40 Rout and Seal 32 40 48 64.00 80.00 96.00 1055.39 1680.98 2646.60$173,736.9 $269,946.4 $332,493.9Activity% Area / % LengthYears Cost Per % Length / % AreaMaintenance and Rehabilitation Cost =NPW Costs72   Illustration 4.5 LCCA results for each alternative under each scenario  0.00E+005.00E+051.00E+061.50E+062.00E+062.50E+063.00E+063.50E+06HMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCCBR 5 AADT 250 CBR 10 AADT 250 CBR 5 AADT 500 CBR 10 AADT 500 CBR 5 AADT 1000 CBR 10 AADT 1000 CBR 5 AADT 2000 CBR 10 AADT 2000Cost (Canadian Dollars) Scenarios Total cost per kilometer of Pavement Minimum Most Likely Maximum73  4.4 Normalization and Aggregation of Indicators The results of life cycle assessment and life cycle cost analysis of pavement alternatives are based on several indicators with different units (e.g. costs in dollars, global warming potential in KgCO2). The absolute fuzzy values of each indicators can be normalized for each scenarios to enable the comparison of values in different units. The normalization process is based on the approach provided by Lee et al. (1991). In general, higher values of evaluated indicators were undesirable. The normalization process uses the following equation to generate comparable indices across all the alternatives in a particular scenario for a given indicator X:  𝑆 (𝑋) =(𝑋−𝑋𝑚𝑎𝑥)(𝑋𝑚𝑖𝑛−𝑋𝑚𝑎𝑥)                                                         (4.1) Equation 4.1 𝑊ℎ𝑒𝑟𝑒, 𝑆 (𝑋) = 𝑁𝑜𝑟𝑚𝑎𝑙𝑖𝑧𝑒𝑑 𝑖𝑛𝑑𝑒𝑥 𝑣𝑎𝑙𝑢𝑒  𝑋 = 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑎𝑛 𝑖𝑛𝑑𝑖𝑐𝑎𝑡𝑜𝑟,  𝑋𝑚𝑎𝑥/𝑚𝑖𝑛 = 𝑚𝑎𝑥𝑖𝑚𝑢𝑚 𝑎𝑛𝑑 𝑚𝑖𝑛𝑖𝑚𝑢𝑚 𝑣𝑎𝑙𝑢𝑒𝑠 𝑜𝑓 "𝑋" 𝑖𝑛 𝑎 𝑠𝑐𝑒𝑛𝑎𝑟𝑖𝑜 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑎𝑙𝑡𝑒𝑟𝑛𝑎𝑡𝑖𝑣𝑒𝑠  The normalization process results in indicator index values that can be compared and aggregated. The framework for aggregating the normalized indices for each scenario is illustrated in the Figure 4.3.   Figure 4.3 Framework for implementing FCP Sustainability Index Environmental Sustainability Index Global Warming Potential Acidification Potential Eutrophication Potential Smog Potential Ozone depletion Potential Human Health Particulate Total Primarly Energy Economic Sustainability Index Life Cycle Costs 74  Before aggregation, it is important to prioritize indicators so that their contribution to higher level attributes (i.e. environmental SI, economic SI and overall SI) can be differentiated. Although, MCDM techniques can be used to elicit weights, the aggregation of environmental indicators in this study was conducted using the weights provided by Goedkoop and Spriensma (2001) as shown in Table 4.18 and Table 4.19. Table 4.20 shows the normalized values for economic and environmental sustainability index.  Table 4.18 Hierarchist weighting of environmental impact indicators (Goedkoop and Spriensma 2001) Category Weights based on Hierarchist Category Human Health 40 Ecosystem Quality 40 Resource 20  Table 4.19 Individual weights distribution of environmental indicators based on Goedkoop & Spriensma (2001) Environmental Indicators Wt. (%) Global Warming Potential (kg CO2 eq) 10 Acidification Potential (kg SO2 eq) 20 HH Particulate (kg PM 2.5 eq) 10 Eutrophication Potential (kg N eq) 20 Ozone Depletion Potential (kg CFC-11 eq) 10 Smog Potential (kg O3 eq) 10 Total Primary Energy (MJ) 20  Table 4.20 Normalized values for economic and environmental sustainability index values  Normalized LCCA Indicator Normalized and Aggregated LCA Indicators HMA PCC GeoSL GeoBC HMA PCC GeoSL GeoBC CBR 5 AADT 250 H 0.953 0.245 1.000 0.952 0.279 0.493 1.000 0.276 M 0.873 0.124 0.949 0.872 0.042 0.461 0.921 0.025 L 0.813 0.000 0.905 0.811 0.035 0.403 0.905 0.000 CBR 10 AADT 250 H 0.945 0.248 1.000 0.943 0.276 0.494 1.000 0.274 M 0.855 0.126 0.944 0.853 0.038 0.461 0.922 0.024 L 0.788 0.000 0.897 0.787 0.026 0.402 0.907 0.000 CBR 5 AADT 500 H 0.953 0.248 1.000 0.959 0.273 0.507 1.000 0.279 M 0.871 0.125 0.947 0.878 0.034 0.474 0.921 0.026 L 0.808 0.000 0.901 0.816 0.022 0.415 0.904 0.000 75   Normalized LCCA Indicator Normalized and Aggregated LCA Indicators HMA PCC GeoSL GeoBC HMA PCC GeoSL GeoBC CBR 10 AADT 500 H 0.953 0.243 1.000 0.945 0.285 0.489 1.000 0.274 M 0.874 0.123 0.949 0.865 0.050 0.457 0.922 0.025 L 0.815 0.000 0.906 0.805 0.038 0.399 0.906 0.000 CBR 5 AADT 1000 H 0.961 0.246 1.000 0.967 0.273 0.213 1.000 0.279 M 0.892 0.125 0.954 0.897 0.036 0.148 0.993 0.027 L 0.838 0.000 0.914 0.844 0.022 0.074 0.913 0.000 CBR 10 AADT 1000 H 0.962 0.241 1.000 0.961 0.279 0.185 1.000 0.276 M 0.895 0.122 0.957 0.894 0.043 0.121 0.922 0.026 L 0.845 0.000 0.920 0.844 0.030 0.047 0.902 0.000 CBR 5 AADT 2000 H 0.967 0.240 1.000 0.976 0.344 0.125 1.000 0.356 M 0.904 0.122 0.958 0.915 0.129 0.067 0.928 0.128 L 0.856 0.000 0.922 0.868 0.117 0.000 0.910 0.104 CBR 10 AADT 2000 H 0.967 0.236 1.000 0.971 0.363 0.121 1.000 0.368 M 0.908 0.120 0.961 0.912 0.155 0.065 0.931 0.147 L 0.862 0.000 0.927 0.868 0.143 0.000 0.913 0.123  After obtaining the normalized fuzzy values for environmental and economic impacts in each scenario, the results were plotted on a 2D display of fuzzy numbers similar to that used by Khan et al. (2002). The 2D fuzzy plots for each alternative are shown in Illustration 4.6 - 4.9 which enables the visualization of trade-off involved among conflicting objectives under uncertainty. The x-axis represents the economic sustainability index whereas the y-axis shows the environmental sustainability index. The rectangles represent the uncertainty distribution for each alternative across the two objectives. The borderlines of the rectangles represent the minimum and maximum values of economic and environment sustainability indices for each alternatives. The points within the rectangles represent the most-likely values The top right corner of the graph represents the most ideal condition where the economic and environmental sustainability index values are the highest i.e. 1. The plots show that GeoSL alternative is showing highest economic and environmental sustainability index values. GeoBC and AC alternatives displayed very similar performance and both were least in environmental performance for lower traffic levels (<= 500 vpd). PCC alternative resulted in the least economic performance but its environmental performance was better than AC and GeoBC for lower traffic levels (i.e. <= 500 vpd). 76   Illustration 4.6 Environmental and economic sustainability of alternatives at CBR 5% and AADT 250 vpd   Illustration 4.7 Environmental and economic sustainability of alternatives at CBR 5% and AADT 500 vpd 00.10.20.30.40.50.60.70.80.910 0.2 0.4 0.6 0.8 1Environmental Sustainability Index Economic Sustainability Index CBR 5% AADT 250 vpd GeoBCGeoSLPCCHMAHMAPCCGeoSLGeoBC00.10.20.30.40.50.60.70.80.910 0.2 0.4 0.6 0.8 1Environmental Sustainability Index Economic Sustainability Index CBR 5% AADT 500 vpd GeoBCGeoSLPCCHMAHMAPCCGeoSLGeoBC77   Illustration 4.8 Environmental and economic sustainability of alternatives at CBR 5% and AADT 1000 vpd   Illustration 4.9 Environmental and economic sustainability of alternatives at CBR 5% and AADT 2000 vpd 00.10.20.30.40.50.60.70.80.910 0.2 0.4 0.6 0.8 1Environmental Sustainability Index Economic Sustainability Index CBR 5% AADT 1000 vpd GeoBCGeoSLPCCHMAHMAPCCGeoSLGeoBC00.10.20.30.40.50.60.70.80.910 0.2 0.4 0.6 0.8 1Environmental Sustainability Index Economic Sustainability Index CBR 5% AADT 2000 vpd GeoBCGeoSLPCCHMAHMAPCCGeoSLGeoBC78   The results of conflicting objectives can be aggregated to a single sustainability index value. It is important to establish relative weights of economic and environmental sustainability indices so that their contribution to overall sustainability index can be estimated. In this case, three different scenarios were created to analyze the effect of decision makers concerns towards environmental and economic impact of pavement alternatives. Following scenarios were identified for evaluating the effect of priority levels on the overall sustainability index values:  1. Pro-Environment: 80% weightage to environmental sustainability index 2. Pro-Economy: 80% weightage to economic sustainability index 3. Neutral: Equal weightage to environmental and economic sustainabilty index  The impact of these weights on overall sustainability indices were quantified and plotted against traffic levels for each CBR value. Illustration 4.9 - 4.11 represent the variation of sustainability indices against traffic levels at 5% CBR. For CBR 10%, similar results are provided in Appendix B3. The solid lines in graphs represent the most-likely values. Again the GeoSL alternative was the best alternative in all scenarios followed by nearly similar performance of GeoBC and AC. PCC displayed the least sustainable performance except in the pro-environment scenario and at traffic levels less than 500 vpd when its performance overlapped significantly with AC and GeoBC alternatives.  Illustration 4.10 Variation of aggregated sustainability index with different traffic levels (neutral) 00.10.20.30.40.50.60.70.80.910 500 1000 1500 2000 2500Aggregated SI AADT (vpd) Neutral - CBR 5% PCC-HPCC-LPCCHMA-HHMA-LHMAGeoSL-HGeoSL-LGeoSLGeoBC-HGeoBC-LGeoBC79   Illustration 4.11 Variation of aggregated sustainability index with different traffic levels (pro-environment)   Illustration 4.12 Variation of aggregated sustainability index with different traffic levels (pro-economic)  00.10.20.30.40.50.60.70.80.910 500 1000 1500 2000 2500Aggregated SI AADT (vpd) Pro-Environment - CBR 5% PCC-HPCC-LPCCHMA-HHMA-LHMAGeoSL-HGeoSL-LGeoSLGeoBC-HGeoBC-LGeoBC00.10.20.30.40.50.60.70.80.910 500 1000 1500 2000 2500Aggregated SI AADT (vpd) Pro-Economic - CBR 5% PCC-HPCC-LPCCHMA-HHMA-LHMAGeoSL-HGeoSL-LGeoSLGeoBC-HGeoBC-LGeoBC80  4.5 Summary and Discussion This study applied fuzzy composite programming (FCP) technique for analyzing economic and environmental sustainability of alternatives under uncertainty. Fuzzy arithmetic operations were used for the processing of uncertain fuzzy inputs to produce fuzzy outputs for costs and environmental indicators. The fuzzy absolute numbers for each indicators were normalized and aggregated to represent economic and environmental sustainability index. The environmental and economic indices were further aggregated to produce overall sustainability indices for each scenario. The effect of decision maker’s attitude towards environmental and economic concerns on the overall sustainability indices of alternatives was also analyzed.   The 2D illustration of fuzzy environmental and economic sustainability indices is shown in Illustration 4.6 – 4.9 for 5% CBR (and Appendix B2 for 10% CBR). These illustrations helped to analyze the environmental-economic trade-off under uncertainty for the pavement alternatives under different scenarios. The results show that GeoSL alternative exhibited the best performance in terms of both environmental and economic sustainability. This is because GeoSL is the only alternative that needs rehabilitation once over the forty year analysis period. Moreover, the uncertainty in economic and environmental sustainability index of GeoBC alternative overlapped considerably with AC in all the scenarios. This implies that the economic and environmental sustainability of AC alternative may not be significantly different if geosynthetics are used to reduce pavement thicknesses. Similarly, PCC pavement had higher environmental sustainability index values compared to AC at lower traffic levels (<=500 vpd). At higher traffic levels, the uncertainty distribution in environmental sustainability index of PCC and AC/GeoBC alternatives overlapped. However, AC was economically more sustainable when compared with PCC.   In addition, single sustainability index vales were calculated using different weight distributions to the economic and environmental sustainability indices. The results show that GeoSL is the best alternative in all scenarios. Again, the uncertainties in sustainability index values of AC and GeoBC overlapped considerably. PCC generally exhibited the least sustainability index values except in the pro-environment scenario and low traffic levels (<=500 vpd) where its uncertainty distribution overlapped considerably with that of AC.   Without the expression of uncertainty, the similitude in sustainability indices of conflicting objectives for the alternatives would not have been observable which may result in false understanding of the results. The fuzzy techniques improved the reliability of results by propagating input uncertainties 81  due to imprecise information and vague human judgement in the early project phases. Therefore, this study demonstrated the effectiveness of fuzzy composite programming (FCP) for analyzing environmental-economic trade-offs under uncertainty for different pavement alternatives. However, this study had following limitations, mainly due to the lack of available data, resources and time:   The pavement vehicle interaction (PVI) effect was not considered even though Athena Impact Estimators offered this capability. PVI effect requires data on International Roughness Index (IRI) values which was not available due to limitations in data, resources and time.  The life cycle costs considered only the agency costs. User costs were ignored due to non-availability of data on vehicle delay and cost associated with those delays in adition to other parameters.  The pavement preservation plans were adapted from roadway cases with different traffic levels.   Accurate data on geosynthetic LCA was not available.    82  Chapter 5 Conclusion and Recommendations In the early phases of project, decision-makers, planners and engineers often face the challenge of deciding among variety of initiatives and policies for the development of sustainable transportation infrastructure. Sustainability evaluation techniques are used to quantify the benefits of adopting alternatives or strategies. Sustainability evaluation techniques often rely on accurate information to assist decision-making in the early project phases. However, these phases of project are characterized with significant uncertainties due to considerable imprecise information and vague human understanding. For this reason, sustainability evaluation techniques require methods to handle uncertainties and assist decision-making processes with more reliable results in early project phases.   In particular, decision-makers and researchers have often criticized sustainability evaluation process using the green rating systems because of their inflexibility to allow prioritization of indicators, uncertainty of data and modification of benchmarks. This research applied Fuzzy synthetic evaluation (FSE) technique to evaluate sustainability of roadway projects under uncertainty by using indicators and benchmarks available from GreenroadsTM rating systems (Muench et al. 2011). An Excel-based tool called Green Proforma was developed using FSE technique to provide a platform for expert-based iterative sustainability evaluation of roadway infrastructure. The expert opinion can be incorporated by Green Proforma in the form imprecise benchmarks, imprecise inputs and different prioritization of indicators and attributes depending on the roadway projects. The sustainability evaluation results of the tool were summarized in the form of a “sustainometer” which shows sustainability indices across the sustainability objectives and also sums up the overall sustainability of roadway projects using a single sustainability index.  Moreover, engineers and roadway administrators often face a major challenge when deciding among a number of available pavement alternatives. Sustainability evaluation of pavements involves conflicting sustainability objectives and significant uncertainties in available data to analyze those objectives. Majority of previous approaches have applied deterministic LCCA or LCA techniques independently for evaluating the sustainability of pavement alternatives. This study has successfully expanded a narrow economic or environmental view point to an integrated evaluation of environmental and economic impacts under uncertainty for the selection of pavement alternatives. Fuzzy composite programming (FCP) technique was used in the life cycle thinking approach to analyze the environmental-economic trade-offs under uncertainty for different pavement alternatives. There are significant uncertainties involved in identifying the timing, extent and cost of pavement M&R activities needed. FCP enabled the propagation of such uncertainties to outputs in the form of 83  2-dimensional environmental and economic sustainability indices that enhanced the quality of observations from the results.   The results of this thesis delivered compelling evidence on the significant utility of fuzzy-based techniques for sustainability evaluation under uncertainty. These techniques are more important in early project because the ability to make sustainable decisions is highest in these phases. Therefore, for practical decision-making, fuzzy-based techniques should be used in conjunction with sustainability evaluation approaches based on multicriteria frameworks and life cycle thinking approach. The limitations of roadway and pavement evaluation studies were highlighted in their individual chapters (i.e. Chapter 3 and Chapter 4, respectively). Specific research contributions made by this research are identified below:   Developed a novel framework for the sustainability evaluation of roadway infrastructures using FSE technique  Developed a highly customizable Excel tool that allows users to incorporate expert opinion and tailor the tool evaluation criteria to match a given development context and priorities. The tool was also tested through scenario analysis to demonstrate the utility of these qualities  Analyzed the environment-economic trade-offs under uncertainty for the pavement alternatives under eight different scenarios using FCP technique  Aggregated the LCCA and LCA results into single sustainability index and compared the effect of decision-makers’ attitude towards environmental and economic concerns on the sustainability indices of all alternatives under each scenario  Although, the proposed objectives of this research have been achieved, but the outcome has opened more directions that need to be explored further. These horizons are identified below as specific recommendations for further research:   The Green Proforma tool can be further developed for a broader-scale development or other infrastructure systems (e.g. urban water systems, neighborhood developments) with additional indicators in a more sophisticated programmable interface (e.g. visual basic).   The Green Proforma can be developed as an ArcGIS application to assist in spatial based sustainability evaluation of roadway infrastructures. 84   Pavement Vehicle Interaction effect and User costs can be included for more comprehensive LCA and LCCA. In addition to environmental and economic objectives, other objectives (e.g. technical feasibility, social impacts) may also be included for comprehensive evaluation.   The sustainability evaluation of pavements can be extended to a high volume roadways as this study considered a low-medium volume roadway cases.   85  References Abdelgawad, M., and Fayek, A. R. (2010). "Risk management in the construction industry using combined fuzzy FMEA and fuzzy AHP." J.Constr.Eng.Manage., 136(9), 1028-1036.  American Association of State Highway and Transportation Officials, and Knovel Civil Engineering & Construction Materials Library - Academic Collection. (2011). Mechanistic-empirical pavement design guide: a manual of practice. American Association of State Highway and Transportation Officials, Washington, D.C.  American Association of State Highway and Transportation Officials, and National Cooperative Highway Research Program. (1988). AASHTO guide for design of pavement structures. AASHTO, Washington, D.C.  ACPA. (2012). Windows Pavement Analysis Software (WinPAS) Guide. Adeli, H. (2002). "Sustainable Infrastructure Systems and Environmentally-Conscious Design-A View for the Next Decade." J.Comput.Civ.Eng., 16(4), 231-233.  Ahmed, Z., Eng, P., and Maher, M. (2006). “Context Sensitive Pavement Design for Low Volume Road Applications.” Low-Volume Roads - Beyond the Boundaries: Design and Policy Issues Session of the 2006 Annual Conference of the Transportation Association of Canada, Transportation Association of Canada, Charlottetown, Prince Edward Island, 17. AkosSzoboszlay. (2005). “Page Mill Road in Palo Alto, CA, USA.” <http://commons.wikimedia.org/wiki/File:PageMillRoad.png> (Mar. 18, 2015). Alberta Infrastructure and Transportation. (2006). “Guidelines For Assessing Pavement Preservation Treatments And Strategies.” Al-Qadi, I. L., Brandon, T. L., and Bhutta, S. A. (1997). “Geosynthetics stabilized flexible pavements.” 2. Al-Qadi, I. L., and Yang, S.-H. (2007). “Cost-effectiveness of using geotextiles in flexible pavements.” Geosynthetics International, 14(1), 2–12. Alsulami, B., and Mohamed, S. (2014). “Hybrid Fuzzy Sustainability Assessment Model: A Case Study of a Regional Infrastructure Transport Project.” Bridges, 2, 400–408. Anderson, J. (2012). Measuring sustainability in civil engineering: development, testing and implementation of the Greenroads (TM) Rating System. University of Washington. Anderson, J. L. (2008). Sustainability in civil engineering. University of Washington. Andreas, G., Allen, J., and Farley, L. (2010). “Towards the development of a rating system for sustainable infrastructure: A checklist or a decision-making tool?” Proceedings of the Water Environment Federation, 2010 (2), 379–391. 86  Andriantiatsaholiniaina, L. a, Kouikoglou, V. S., and Phillis, Y. a. (2004). “Evaluating strategies for sustainable development: fuzzy logic reasoning and sensitivity analysis.” Ecological Economics, 48(2), 149–172. ARA. (2007). Estimation of Road Cost Allocation Between Light Vehicles and Heavy Vehicles in Canada. Toronto. Atkinson, R., Crawford, L., and Ward, S. (2006). “Fundamental uncertainties in projects and the scope of project management.” International Journal of Project Management, 24(8), 687–698. Basu, D., Misra, A., and Puppala, A. J. (2014). “Sustainability and geotechnical engineering : perspectives and review.” Canadian Geotechnical Journal, 52(1), 96–113. Baumgärtner, S., and Quaas, M. F. (2009). “Ecological-economic viability as a criterion of strong sustainability under uncertainty.” Ecological Economics, 68(7), 2008–2020. BC MoT. (2004). Pavement Structure Design Guidelines. BC MoT. (2013). British Columbia Ministry of Transportation and Infrastructure: Construction and Rehabilitation Cost Guide. BC MoT. (2015). Pavement Structure Design Guidelines - Technical Circular T-01/15. Becker, J. (2004). "Making sustainable development evaluations work." Sustainable Dev., 12(4), 200-211.  Berg, R., and Geosynthetic Materials Association. (2000). Geosynthetic reinforcement of the aggregate base/subbase courses of pavement structures. Geosynthetic Materials Association. Binney, P. D. (2010). “Transitional Approaches to the Development of Sustainable Civil Engineering Infrastructure.” World Environmental and Water Resources Congress 2010, American Society of Civil Engineers, Reston, VA, 504–513. Brundtland, G. H. (1987). Report of the World Commission on environment and development:"Our common future.". United Nations. California Department of Transportation. (2010). LIFE-CYCLE COST ANALYSIS PROCEDURE MANUAL. CALTRANS. (2007). LIFE-CYCLE COST ANALYSIS PROCEDURES MANUAL. Council, US Green Building. (2006). "LEED for Neighborhood Developments." Washington DC: US Green Building Council. Chen, S.-H. (1985). “Ranking fuzzy numbers with maximizing set and minimizing set.” Fuzzy Sets and Systems. 17(2), 113-129 87  Chen, S.-J. J., and Hwang, C. L. (1992). Fuzzy Multiple Attribute Decision Making: Methods and Applications. (M. J. Beckmann and W. Krelle, eds.), Springer-Verlag New York, Inc., Secaucus, NJ, USA. Chhipi-Shrestha, G. K., Hewage, K., and Sadiq, R. (2014). “‘Socializing’ sustainability: a critical review on current development status of social life cycle impact assessment method.” Clean Technologies and Environmental Policy, 1-18. Clevenger, C. (2013). “Review of sustainability rating systems used for infrastructure projects.” 49th ASC Annual International Conference Proceedings, 10-13. Cole, R. J. (2005). “Building environmental assessment methods: redefining intentions and roles.” Building Research & Information, 33(5), 455–467. C-SHRP. (2000). PAVEMENT DESIGN AND PERFORMANCE : Current Issues and Research Needs. C-SHRP. (2002). “Pavement structural design practices across Canada.” Curwell, S., Yates, A., Howard, N., Bordass, B., and Doggart, J. (1999). "The green building challenge in the UK." Build.Res.Inf., 27(4-5), 286-293.  Curz, R., Kim, J., and Cha, H. S. (2012). "Using a Thematic Framework to Compare Sustainability Rating Systems Applicable to Transportation Projects." Bridges, 10 9780784412329.200.  Dasgupta, S., and Tam, E. K. (2005). "Indicators and framework for assessing sustainable infrastructure." Canadian Journal of Civil Engineering, 32(1), 30-44.  Dong, W., Shah, H., and Wongt, F. (1985). "Fuzzy computations in risk and decision analysis." Civ.Eng.Syst., 2(4), 201-208.  DoT. (2004). “Life-Cycle Cost Analysis RealCost User Manual.” 46. Dovers, S. R., and Handmer, J. W. (1992). “Uncertainty, sustainability and change.” Global Environmental Change, 2, 262–276. Dyck, R., Sadiq, R., Rodriguez, M., Simard, S., and Tardif, R. "A comparison of membership function shapes in a fuzzy-based fugacity model for disinfection byproducts in indoor swimming pools." International Journal of System Assurance Engineering and Management, 1-13.  Elwakil, E., Eweda, A., and Zayed, T. (2014). "Modelling the effect of various factors on the condition of pavement marking." Structure and Infrastructure Engineering, 10(1), 93-105.  FCM. (2003). Timely Preventive Maintenance for Municipal Roads. Felio, G., Canadian Construction Association, and Canadian Public Works Association. (2012). Canadian Infrastructure Report Card: Volume 1: 2012 Municipal Roads and Water Systems. Canadian Construction Association. 88  Fenner, R. A., & Ryce, T. (2008). "A comparative analysis of two building rating systems Part 1: Evaluation." Proceedings of the ICE-Engineering Sustainability, 161(1), 55-63. FHWA. (2013). Highwar Functional Classfication Concepts, Criteria and Procedures. Filippo, S., Ribeiro, P. C. M., and Ribeiro, S. K. (2007). "A Fuzzy Multi-Criteria Model applied to the management of the environmental restoration of paved highways." Transportation Research Part D: Transport and Environment, 12(6), 423-436.  Foxon, T. J., McIlkenny, G., Gilmour, D., Oltean-Dumbrava, C., Souter, N., Ashley, R., Butler, D., Pearson, P., Jowitt, P., and Moir, J. (2002). "Sustainability criteria for decision support in the UK water industry." J.Environ.Plann.Manage., 45(2), 285-301.  Frischknecht, R., Stucki, M., Büsser, S., Itten, R., and Wallbaum, H. (2012). "Comparative life cycle assessment of geosynthetics versus conventional construction materials." Ground Engineering, .  Gasparatos, A., El-Haram, M., and Horner, M. (2009). “The argument against a reductionist approach for measuring sustainable development performance and the need for methodological pluralism.” Accounting Forum, 33(3), 245–256. Gil, J., and Duarte, J. P. (2013). "Tools for evaluating the sustainability of urban design: a review." Proceedings of the ICE-Urban Design and Planning, 166(6), 311-325.  Goedkoop, M., and Spriensma, R. (2001). “The Eco-indicator99: a damage oriented method for life cycle impact assessment: methodology report.” 1–144. Goh, K. C., and Yang, J. (2014). “Importance of Sustainability-Related Cost Components in Highway Infrastructure: Perspective of Stakeholders in Australia.” Journal of Infrastructure Systems, 20(1). Gupta, R. (2009). “A study of geosynthetic reinforced flexible pavement system.” The University of Texas at Austin. Gwo-Hshiung, T. (2010). "Multiple attribute decision making: methods and applications." Multiple Attribute Decision Making: Methods and Applications Hacking, T., and Guthrie, P. (2008). "A framework for clarifying the meaning of Triple Bottom-Line, Integrated, and Sustainability Assessment." Environ.Impact Assess.Rev., 28(2), 73-89. Hajek, D. J., Hein, D., and Chris, O. (2004). "Decision making for maintenance and rehabilitation of municipal pavements." 2004 Annual Conference of the Transportation Association of Canada. Hendrickson, C., and Au, T. (1989). Project Management for Construction: Fundamental Concepts for Owners, Engineers, Architects, and Builders. Chris Hendrickson. Hicks, R. G., Lundy, J. R., and Epps, J. A. (1999). Life cycle costs for asphalt-rubber paving materials. Rubber Pavements Association,  89  Holt, A., Sullivan, S., and Hein, D. K. (2011). “Life Cycle Cost Analysis of Municipal Pavements in Southern and Eastern Ontario.” 2011 CONFERENCE AND EXHIBITION OF THE TRANSPORTATION ASSOCIATION OF CANADA. TRANSPORTATION SUCCESSES: LET’S BUILD ON THEM. 2011 Congress et Exhibition de l'Association des Transports du Canada. Les Succes en Transports: Une Tremplin vers l'Avenir. Huang, Y. (1993). Pavement analysis and design. Hunt, D. V., Lombardi, D. R., Rogers, C. D., and Jefferson, I. (2008). "Application of sustainability indicators in decision-making processes for urban regeneration projects." Proceedings of the ICE-Engineering Sustainability, 161(1), 77-91.  Hwang, C.-L., and Yoon, K. (1981). Multiple Attribute Decision Making: Methods and Applications, A State-of-the-art Survey. IPCC. (2000). “Conceptual Basis for Uncertainty Analysis.” IPCC Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories, A1.1 – A1.22. International Organization for Standardization. (1997). Environmental Management: Life Cycle Assessment: Principles and Framework. ISO. Jato-Espino, D., Rodriguez-Hernandez, J., Andrés-Valeri, V. C., and Ballester-Muñoz, F. (2014). “A fuzzy stochastic multi-criteria model for the selection of urban pervious pavements.” Expert Systems with Applications, 41(15), 6807–6817. Jensen, K. K. (2010). "Sustainability and uncertainty: Bottom-up and top-down approaches." Italian Journal of Animal Science, 6(1s), 853-855.  Johanns, M., and Craig, J. (n.d.). Pavement Maintenance Manual. Johnson, A. (2000). Best Practices Handbook on Asphalt Pavement Maintenance (No. MN/RC-2000-04). Minnesota T2/LTAP Program, Center for Transportation Studies, University of Minnesota, Minneapolis, MN. Kennedy, C., Miller, E., Shalaby, A., Maclean, H., and Coleman, J. (2005). “The Four Pillars of Sustainable Urban Transportation.” Transport Reviews, 25(4), 393–414. Kevern, J. (2010). "Green building and sustainable infrastructure: Sustainability education for civil engineers." J.Prof.Issues Eng.Educ.Pract., 137(2), 107-112.  Khan, F. I., Sadiq, R., and Husain, T. (2002). "GreenPro-I: a risk-based life cycle assessment and decision-making methodology for process plant design." Environmental Modelling & Software, 17(8), 669-692.  Khatri, K. B., Vairavamoorthy, K., and Akinyemi, E. (2011). "Framework for computing a performance index for urban infrastructure systems using a fuzzy set approach." J Infrastruct Syst, 17(4), 163-175.  90  Klotz, L., and Grant, D. (2009). “A balanced view of sustainability in civil engineering and construction.” Proc., 2009 Construction Research Congress, 1338–1347. Knuth, D., and Fortmann, J. (2010). “The development of I-LAST Illinois—Livable and sustainable transportation.” Lee, Y. W., Bogardi, I., and Stansbury, J. (1991). “Fuzzy Decision Making In Dredged Material Management.” Journal of Environmental Engineering, 117(5), 614–630. Li, Z., Lee, S. H., and Madanu, S. (2011). Highway Transportation Project Evaluation and Selection Incorporating Risk and Uncertainty. INTECH Open Access Publisher, .  Marlow, D., Tran, D., May, R., and Mashford, J. (2012). "Development of a Fuzzy Risk Ranking Model for Prioritizing Manhole Inspection." J.Comput.Civ.Eng., 26(4), 550-557.  Maccaferri. (2001). Technical Reference GRID-DE-2 Design of Flexible Pavements with Maccaferri Macgrid(TM) EB Series. Marilyn475. (2008a). “Geotextiles.” Wikimedia Commons - Public Domain, <http://commons.wikimedia.org/wiki/File:Geotextiles1.jpg> (Feb. 18, 2015). Marilyn475. (2008b). “Geogrids.” Wikimedia Commons - Public Domain, <http://commons.wikimedia.org/wiki/File:Geogrids.JPG> (Feb. 18, 2015). Martland, C. D. (2011). “Toward More Sustainable Infrastructure: Project Evaluation for Planners and Engineers.” Construction Management and Economics, 1–8. Mcvoy, G. R., Nelson, D. A., Krekeler, P., Kolb, E., and Gritsavage, J. S.  (2010). “Moving Towards Sustainability: New York State Department of Transportation’s GreenLITES Story.” Proceedings of the Green Streets and Highways 2010 Conference, ASCE, 461–479. McVoy, G., Gunasekera, K., Sousa, L. R., and Schaffner, P.  (2013). “AN ANALYTICAL FRAMEWORK FOR SUSTAINABILITY ANALYSIS OF TRANSPORTATION INVESTMENTS ACROSS THE TRIPLE BOTTOM LINE USING.” Mendoza, G. A., and Prabhu, R. (2002). “Qualitative multi-criteria approaches to assessing indicators of sustainable forest resource management $.” 174(2003), 329–343. Mendoza, G. A., and Prabhu, R. (2003). "Qualitative multi-criteria approaches to assessing indicators of sustainable forest resource management." For.Ecol.Manage., 174(1), 329-343.  Mills, L. O., and Attoh-Okine, N. (2014). "SUSTAINABILITY OF CIVIL INFRASTRUCTURE SYSTEMS: THE PAST, THE PRESENT AND THE WAY FORWARD." Sustainable Development of Critical Infrastructure, 79.  Mirza, S. (2007). Danger ahead: The coming collapse of Canada's municipal infrastructure. Federation of Canadian Municipalities. 91  Morandeira, A. D. (2012). Training development for pavement preservation : chip sealing and fog sealing. Morrissey, J., Iyer-Raniga, U., McLaughlin, P., and Mills, A. (2012). “A Strategic Project Appraisal framework for ecologically sustainable urban infrastructure.” Environmental Impact Assessment Review, Elsevier Inc., 33(1), 55–65. Muench, S., Anderson, J., Hatfield, J., Koester, J., and Söderlund, M. (2011). "Greenroads Manual v1. 5." Seattle, WA: University of Washington.  Munda, G. (2006). “Social multi-criteria evaluation for urban sustainability policies.” Land Use Policy, 23(1), 86–94. Nassiri, S., Bayat, A., Skirrow, R., and Transportation, A. (2013). “Integrated Road Research Facility ( IRRF ): An Albertan Research Initiative.” 2013 CONFERENCE AND EXHIBITION OF THE TRANSPORTATION ASSOCIATION OF CANADA-TRANSPORTATION: BETTER-FASTER-SAFER, 1–18. Ness, B., Urbel-Piirsalu, E., Anderberg, S., and Olsson, L. (2007). “Categorising tools for sustainability assessment.” Ecological Economics, 60(3), 498–508. Noshadravan, A., Wildnauer, M., Gregory, J., and Kirchain, R. (2013). “Comparative pavement life cycle assessment with parameter uncertainty.” Transportation Research Part D: Transport and Environment, 25, 131–138. Oliveira, V., and Pinho, P. (2010). "Evaluation in urban planning: Advances and prospects." Journal of Planning Literature, 24(4), 343-361.  Oswald, M. R., and McNeil, S. (2009). "Rating sustainability: Transportation investments in urban corridors as a case study." J.Urban Plann.Dev., 136(3), 177-185. Peshkin, D. G., and Hoerner, T. E. (2005). Pavement Preservation: Practices, Research Plans, and Initiatives. Report. Peshkin, D. G., and Hoerner, T. E. (2005). Pavement preservation: Practices, research plans, and initiatives. American Association of State Highway and Transportation Officials Poveda, C. a, and Lipsett, M. (2011). “A Review of Sustainability Assessment and Sustainability/Environmental Rating Systems and Credit Weighting Tools.” Journal of Sustainable Development, 4(6), 36–55. Rabbani, K., Collins, F., and Benfica, R. (2014). “A Framework for Assessing Transportation Sustainability Rating Systems for Implementation in U.S. State Departments of Transportation.” 1–19. Rajani, B., Kleiner, Y., and Sadiq, R. (2006). “Translation of pipe inspection results into condition ratings using the fuzzy synthetic evaluation technique.” Journal of Water Supply: Research and Technology - AQUA, 55(1), 11–24. 92  Ramani, T. L., Zietsman, J., Gudmundsson, H., Hall, R. P., and Marsden, G. (2011). “Framework for Sustainability Assessment by Transportation Agencies.” Transportation Research Record: Journal of the Transportation Research Board, 2242(-1), 9–18. Ramani, T. L., Zietsman, J., Knowles, W. E., and Quadrifoglio, L. (2011). "Sustainability enhancement tool for state departments of transportation using performance measurement." J.Transp.Eng., 137(6), 404-415. Rangaraju, P. R., Amirkhanian, S. N., Guven, Z., and Carolina, S. (2008). Life Cycle Cost Analysis for Pavement Type Selection. FHWA-SC-08-01. Reid, L., Bevan, T., Davis, A., Neuman, T., Penney, K., Seskin, S., VanZerr, M., Anderson, J., Muench, S., Weiland, C., Ramani, T., Zietsman, J., Crossett, J., Crocker, C., and Schulz, J. (2015). INVEST Version 1.1 Sustainable Highways Self-Evaluation Tool Sustainable Highways Self-Evaluation Tool. January. Reza, B. (2013). Emergy-based life cycle assessment (Em-LCA) for sustainability appraisal of built environment. University of British Columbia (Okanagan). Reza, B., Sadiq, R., and Hewage, K. (2013). “A fuzzy-based approach for characterization of uncertainties in emergy synthesis: An example of paved road system.” Journal of Cleaner Production, 59, 99–110. Rodríguez López, F., and Fernández Sánchez, G. (2011). “Challenges for Sustainability Assessment by Indicators.” Leadership and Management in Engineering, 11(4), 321–325. Ross, T. J. (2009). Fuzzy logic with engineering applications. John Wiley & Sons. Saaty, T. L. (2007). “Time dependent decision-making; dynamic priorities in the AHP/ANP: Generalizing from points to functions and from real to complex variables.” Mathematical and Computer Modelling, 46(7-8), 860–891. Saaty, T. L., and Vargas, L. G. (2012). Models, Methods, Concepts & Applications of the Analytic Hierarchy Process. Springer Science & Business Media. Sadiq, R. (2001). Drilling waste discharges in the marine environment: a risk based decision methodology, Memorial University of Newfoundland. Sadiq, R., Husain, T., Veitch, B., and Bose, N. (2004). “Risk-based decision-making for drilling waste discharges using a fuzzy synthetic evaluation technique.” Ocean Engineering, 31, 1929–1953. Sadiq, R., and Khan, F. I. (2006). “An integrated approach for risk-based life cycle assessment and multi-criteria decision-making: Selection, design and evaluation of cleaner and greener processes.” Business Process Management Journal, 12, 770–792. Sadiq, R., Khan, F. I., and Veitch, B. (2005). “Evaluating offshore technologies for produced water management using GreenPro-I - A risk-based life cycle analysis for green and clean process selection and design.” Computers and Chemical Engineering, 29, 1023–1039. 93  Sadiq, R., Rajani, B., and Kleiner, Y. (2004). "Fuzzy-based method to evaluate soil corrosivity for prediction of water main deterioration." J Infrastruct Syst, 10(4), 149-156.  Sadiq, R., and Rodriguez, M. J. (2004). “Fuzzy synthetic evaluation of disinfection by-products--a risk-based indexing system.” Journal of Environmental Management, 73(1), 1–13. Sahely, H. R., and Kennedy, C. A. (2007). "Water use model for quantifying environmental and economic sustainability indicators." J.Water Resour.Plann.Manage., 133(6), 550-559.  Sahely, H. R., Kennedy, C. A., and Adams, B. J. (2005). "Developing sustainability criteria for urban infrastructure systems." Canadian Journal of Civil Engineering, 32(1), 72-85.  Santero, N. (2010). "Life cycle assessment of pavements: A critical review of existing literature and research." Lawrence Berkeley National Laboratory, .  Santero, N. J., Masanet, E., and Horvath, A. (2011). "Life-cycle assessment of pavements. Part I: Critical review." Resour.Conserv.Recycling, 55(9), 801-809.  Serpella, A. F., Ferrada, X., Howard, R., and Rubio, L. (2014). “Risk Management in Construction Projects: A Knowledge-based Approach.” Procedia - Social and Behavioral Sciences, 119, 653–662. Sharifi, A., and Murayama, A. (2014). "Neighborhood sustainability assessment in action: Cross-evaluation of three assessment systems and their cases from the US, the UK, and Japan." Build.Environ., 72 243-258.  Sharifi, A., and Murayama, A. (2013). "A critical review of seven selected neighborhood sustainability assessment tools." Environ.Impact Assess.Rev., 38 73-87.  Shen, L., Wu, Y., and Zhang, X. (2010). "Key assessment indicators for the sustainability of infrastructure projects." J.Constr.Eng.Manage., 137(6), 441-451.  Shen, L., and Zhou, J. (2014). "Examining the effectiveness of indicators for guiding sustainable urbanization in China." Habitat International, 44 111-120.  Simpson, S. P., Clevenger, C. M., Ozbek, M. E., Rabbani, E. R. K., and Atadero, R. (2014). "A Framework for Assessing Transportation Sustainability Rating Systems for Implementation in US State Departments of Transportation." Transportation Research Board 93rd Annual Meeting, .  Singh, R. K., Murty, H., Gupta, S., and Dikshit, A. (2009). "An overview of sustainability assessment methodologies." Ecol.Ind., 9(2), 189-212.  Stucki, M., Büsser, S., Itten, R., Frischknecht, R., and Wallbaum, H. (2011). "Comparative Life Cycle Assessment of Geosynthetics versus Conventional Construction Material." ESU-Services Ltd.Commissioned by European Association for Geosynthetic Manufacturers (EAGM), Uster and Zürich, CH, .  94  Sustainability, V. E. (2011). Integrated VicRoads Environmental Sustainability Tool (INVEST), VicRoads, .  Swei, O. A. (2012). Incorporating uncertainty in the Life Cycle Cost Analysis of pavements. Swei, O., Gregory, J., and Kirchain, R. (2013). “Probabilistic Characterization of Uncertain Inputs in the Life-Cycle Cost Analysis of Pavements.” Transportation Research Record: Journal of the Transportation Research Board, 2366 (1), 71–77. TAC. (2014). TAC Primer: Pavement Asset Design and Management Guide. Ottawa. Taylor, M. J., McNicholas, C., Nicolay, C., Darzi, A., Bell, D., & Reed, J. E. (2013). "Systematic review of the application of the plan–do–study–act method to improve quality in healthcare". BMJ Quality & Safety, bmjqs-2013. Tighe, S., and Gransberg, D. "Demonstrating How Sustainability Can Effectively Be Incorporated into Pavement Engineering and Management." Airfield and Highway Pavement 2013@ sSustainable and Efficient Pavements, ASCE, 76-91.  Tinni, A. (2013). Cautions with Life Cycle Cost Analysis for Optimal Pavement Selection. Tran, D., Mashford, J., May, R., and Marlow, D. (2012). “Development of a Fuzzy Risk Ranking Model for Prioritizing Manhole Inspection.” Journal of Computing in Civil Engineering, 26(4), 550-557. United Nations. Department of Economic and Social Affairs.Population Division. (2012). "World urbanization prospects: the 2011 revision." United Nations, New York. United Nations, W. U. P. (2014). World Urbanization Prospects. USGBC. (2014). LEED 2009 for Neighborhood Development Rating System. Valentin, A., and Spangenberg, J. H. (2000). "A guide to community sustainability indicators." Environ.Impact Assess.Rev., 20(3), 381-392.  Verweij, S., van Meerkerk, I., and Korthagen, I. A. (2015). "Reasons for contract changes in implementing Dutch transportation infrastructure projects: An empirical exploration." Transp.Policy, 37 195-202.  Walls III, J., and Smith, M. R. (1999). Life-Cycle Cost Analysis in Pavement Design. Washington. Walton, J. S., El-Haram, M., Castillo, N. H., Horner, R. M. W., Price, A. D. F., and Hardcastle, C. (2005). "Integrated assessment of urban sustainability." Engineering Sustainability, 158(2), 57-65.  Yager, R. R. (1977). "Multiple objective decision-making using fuzzy sets." International Journal of Man-Machine Studies, 9(4), 375-382.  95  Yang, S., and Al-Qadi, I. (2007). "Cost-effectiveness of using geotextiles in flexible pavements." Geosynthetics International, 14(1), 2-12. Yan, J., and Vairavamoorthy, K. (2003). “Fuzzy approach for pipe condition assessment.” In Pipeline Engineering and Construction International Conference 2003, (No. Volume 1). Yigitcanlar, T., and Dur, F. (2010). “Developing a Sustainability Assessment Model: The Sustainable Infrastructure, Land-Use, Environment and Transport Model.” Sustainability, 2(1), 321–340. Yildirim, Y. (2011). Texas Pavement Preservation Center Two-Year Summary Report. Yoe, C. E., Burks-Copes, K., Schultz, M. T., & Suedel, B. C. (2010). "Addressing risk and uncertainty in planning ecological restoration projects." (No. ERDC-TN-EMRRP-ER-13). ENGINEER RESEARCH AND DEVELOPMENT CENTER VICKSBURG MS COASTAL AND HYDRAULICS LAB. Zadeh, L. A. (1965). "Fuzzy sets." Information and Control, 8(3), 338-353.  Zhang, Y. (2007). Investigation of geosynthetic-soil confinement using asphalt pavement analyzer. University of Kansas. Zietsman, J., and Rilett, L. (2002). Sustainable transportation: Conceptualization and performance measures. (No. SWUTC/02/167403-1). Southwest University Transportation Center, Texas Transportation Institute, Texas A & M University.    96  Appendices Appendix A1: Sustainability Criteria, Indicator and Benchmark List based on Greenroads TM v 1.5 Manual (Muench et al. 2011) Category # Criteria Indicator V. Poor Poor Fair Good V. Good Reference Value Environment and Water 1 Environmental Management System (EMS) Existence of a formal Environmental Management Process/System Contractor/Designer/Management Firm No Such System Exists -- Informal EMS exists -- Organizations have ISO Certification & EMS meeting ISO requirements Modified   2 Runoff Flow Control Ratio of post vs. pre - development stage Volume/Flow rate  for 90th Percentile average annual rainfall event 5 3.5 2 1 0.95 <1.20   2 1 0.95 0.9 0    3 Runoff Quality % of 90th percentile average annual rainfall event post-construction runoff volume - treated 0 0.2 0.4 0.7 0.8 80%-90%+   0.4 0.7 0.8 0.9 1    3 Stormwater Cost Analysis Conduct Lifecycle Cost Analysis (LCCA) according to NCHRP Report 565: Evaluation of BMPs for Highway Runoff Control Guidelines Not Performed -- Performed but no proper record or document exists -- Complete Application with adequate Documentation Modified   5 Site Vegetation Use non-invasive and native plants No Consideration -- Partial Fulfilment --- Full Compliance Modified   6 Habitat Restoration % of required mitigation area restored 0 0.25 0.5 0.9 1 100%-105%   0.5 0.9 1 1.01 1.5    7 Ecological Connectivity Development of recommended wildlife mobility and protective structures No Consideration -- Partial Consideration -- Full Consideration Modified   8 Light Pollution Provision of Dark-Sky compliant or equivalent lighting fixtures No Consideration -- Partial Consideration -- Full Consideration Modified Access & Equity 9 Safety Audit Road safety audit (RSA) on the project roadway in accordance with FHWA’s Road Safety Audit Guidelines.  No Consideration -- Partial Consideration -- Full Consideration Modified   10  Intelligent Transportation Systems No. of Category where ITS is employed for 1 or more application (categories identified in Greenroads rating system) 0 0 0 1 2 1 to 5 categories   0 1 2 3 10    11 Context Sensitive Solutions Consideration to Context Sensitive Solutions in the project design No Consideration -- Partial Consideration -- Full Consideration Modified 97  Category # Criteria Indicator V. Poor Poor Fair Good V. Good Reference Value   12 Traffic Emissions Reduction Use of congestion pricing and demonstrate the reduction in GHGs and criteria pollutants No Consideration -- Partial Consideration -- Full Consideration Modified   13 Pedestrian Access Develop new or existing facilities for enhancing pedestrian access No Consideration -- Partial Consideration -- Full Consideration Modified   14 Bicycle Access Develop new or existing facilities for enhancing bicycle access  No Consideration -- Partial Consideration -- Full Consideration Modified   15 Transit and HOV Access Develop new or existing facilities for enhancing transit and HOV access No Consideration -- Partial Consideration -- Full Consideration Modified   16 Scenic Views Provide scenic access or viewpoints No Consideration -- Partial Consideration -- Full Consideration Modified   17 Cultural Outreach Install informational infrastructure to explain the site or direct roadway users to the site with special cultural value  No Consideration -- Partial Consideration -- Full Consideration Modified Construction Activities 18 Quality Management System The prime contractor, design builder or construction management firm shall have a documented quality management system (QMS)  No Consideration -- Partial Consideration -- Full Consideration Modified   19 Environmental Training Provide an environmental training plan that is customized to the project No Consideration -- Partial Consideration -- Full Consideration Modified   20 Site Recycling Plan Establish, implement, and maintain a formal Site Recycling Plan as part of the Construction and Demolition Waste Management Plan (CWMP) No Consideration -- Partial Consideration -- Full Consideration Modified   21 Fossil Fuel Reduction % Fossil Fuel reduction by nonroad construction equipment (using biofuels or equivalent) by contractors 0 0.5 1 10 15 15%-25%   1 10 15 20 30    22 Equipment Emissions Reduction % of the nonroad construction equipment fleet operating hours for the project, accomplished on equipment with installed emission reduction exhaust retrofits and add on fuel efficiency technologies that achieve the EPA Tier 4 emission standard.  0 5 10 20 50 50%-75%   10 20 50 60 100    23 Paving Emissions % of the hot mix asphalt (HMA) placed using a paver that 0 5 10 20 30 90%+ 98  Category # Criteria Indicator V. Poor Poor Fair Good V. Good Reference Value   Reduction is certified to have met National Institute for Occupational Safety and Health (NIOSH) emission guidelines 10 20 30 50 100    24 Water Tracking Create a spreadsheet that records total water use during construction. No such activity considered -- Irregular or Incomplete monitoring --- Full Implementation and Recording activities Modified   25 Contractor Warranty Years of Contractor warranty for constructed portions of pavements including surfacing and underlying layers 0 0 0 1 2 3 years   0 1 2 3 4  Materials & Resources 26 Life Cycle Assessment (LCA) Conduct a detailed process based lifecycle assessment (ISO LCA) or hybrid economic input output lifecycle assessment (Hybrid EIO) according to the ISO14040 standard frameworks for the final roadway design alternative. No Consideration -- Partial or Incomplete Application of LCA --- Full Application Modified   27 Pavement Reuse % Reuse of Existing Pavement Materials or Structural Elements 0 5 10 50 70 50%-90%+   10 50 70 90 100    28 Earthwork Balance % difference between cut and fill with respect to the average total volume of material moved 100 60 20 15 10 <10%   20 15 10 5 0    29 Recycled Materials % of recycle material by weight used in pavement surfacing and underlying layers along with any other structures 0 0 0 5 10 10%-60%   0 5 10 50 100    30 Regional Materials % of these basic materials by weight have traveled less than the maximum haul distances (225 miles) 0 5 10 70 80 84%+   10 70 80 90 100    31 Energy Efficiency % of total luminaires installed on the project with energy efficient fixtures that are 2009 Energy Star compliant 0 0.5 1 20 40 20%-100%   1 20 40 80 100  Pavement Technologies 32 Long-Life Pavement % of the total new or reconstructed pavement surface area for regularly trafficked lanes of  pavement to meet long life pavement design criteria 0 5 10 50 60 75%+   10 50 60 70 100    33 Permeable Pavement % of the 90th percentile average annual rainfall event post construction runoff volume treated to 25 mg/L concentration of total suspended solids (TSS) or less. 0 5 10 50 60 50%   10 50 60 70 100    34 Warm Mix Asphalt (WMA) % of the total project pavement (hot mix asphalt or Portland cement concrete) by weight built using WMA 0 5 10 50 60 50%+   10 50 60 70 100    35 Cool Pavement % of the  total project pavement surfacing by area built 0 5 10 50 60 50%+ 99  Category # Criteria Indicator V. Poor Poor Fair Good V. Good Reference Value   with minimum albedo of 0.3 (measured using ASTM E 903) 10 50 60 70 100    36 Quiet Pavement % of the total regularly trafficked pavement surface area designed to reduce tire pavement noise levels at or below certain standards 0 20 40 70 80 50%+   40 70 80 90 100    37 Pavement Performance Tracking Spatially located and correlation of data from construction quality and long term pavement performance measurements  No Consideration -- System needs Improvement --- Effective System Exists Modified    100  Appendix A2: Excel Interface for the Green Proforma      Main Sheet 101       Benchmarking Sheet 102       Prioritization Sheet 103       Input Sheet 104      Results Sheet 105  Appendix B1: Life Cycle Assessment Results showing fuzzy numbers for Environmental Impact Indicators for each Scenario   1.00E+061.20E+061.40E+061.60E+061.80E+062.00E+062.20E+062.40E+062.60E+062.80E+063.00E+06HMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCCBR 5 AADT 250 CBR 10 AADT 250 CBR 5 AADT 500 CBR 10 AADT 500 CBR 5 AADT 1000 CBR 10 AADT 1000 CBR 5 AADT 2000 CBR 10 AADT 2000Global Warming Potential (kg CO2 eq) Scenarios Global Warming Potential (kg CO2 eq) Minimum Most Likely Maximum106   9.00E+031.10E+041.30E+041.50E+041.70E+041.90E+042.10E+042.30E+042.50E+04HMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCCBR 5 AADT 250 CBR 10 AADT 250 CBR 5 AADT 500 CBR 10 AADT 500 CBR 5 AADT 1000 CBR 10 AADT 1000 CBR 5 AADT 2000 CBR 10 AADT 2000Acidification Potential (kg SO2 eq) Scenarios Acidification Potential (kg SO2 eq) Minimum Most Likely Maximum107    8.00E+021.00E+031.20E+031.40E+031.60E+031.80E+032.00E+03HMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCCBR 5 AADT 250 CBR 10 AADT 250 CBR 5 AADT 500 CBR 10 AADT 500 CBR 5 AADT 1000 CBR 10 AADT 1000 CBR 5 AADT 2000 CBR 10 AADT 2000HH Particulate (kg PM2.5 eq) Scenarios HH Particulate (kg PM2.5 eq) Minimum Most Likely Maximum108    6.00E+027.00E+028.00E+029.00E+021.00E+031.10E+031.20E+031.30E+031.40E+031.50E+031.60E+03HMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCCBR 5 AADT 250 CBR 10 AADT 250 CBR 5 AADT 500 CBR 10 AADT 500 CBR 5 AADT 1000 CBR 10 AADT 1000 CBR 5 AADT 2000 CBR 10 AADT 2000Eutrophication Potential (kg N eq) Scenarios Eutrophication Potential (kg N eq) Minimum Most Likely Maximum109     0.00E+005.00E-041.00E-031.50E-032.00E-032.50E-033.00E-033.50E-034.00E-034.50E-03HMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCCBR 5 AADT 250 CBR 10 AADT 250 CBR 5 AADT 500 CBR 10 AADT 500 CBR 5 AADT 1000 CBR 10 AADT 1000 CBR 5 AADT 2000 CBR 10 AADT 2000Ozone Depletion Potential (kg CFC-11 eq) Scenarios Ozone Depletion Potential (kg CFC-11 eq) Minimum Most Likely Maximum110     3.00E+053.50E+054.00E+054.50E+055.00E+055.50E+056.00E+056.50E+057.00E+057.50E+058.00E+05HMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCCBR 5 AADT 250 CBR 10 AADT 250 CBR 5 AADT 500 CBR 10 AADT 500 CBR 5 AADT 1000 CBR 10 AADT 1000 CBR 5 AADT 2000 CBR 10 AADT 2000Smog Potential (kg O3 eq) Scenarios Smog Potential (kg O3 eq) Minimum Most Likely Maximum111  1.50E+071.70E+071.90E+072.10E+072.30E+072.50E+072.70E+072.90E+073.10E+073.30E+07HMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCHMAPCCGeoSLGeoBCCBR 5 AADT 250 CBR 10 AADT 250 CBR 5 AADT 500 CBR 10 AADT 500 CBR 5 AADT 1000 CBR 10 AADT 1000 CBR 5 AADT 2000 CBR 10 AADT 2000Total Primary Energy (MJ) Scenarios Total Primary Energy (MJ) Minimum Most Likely Maximum112  Appendix B2: Comparison of Environmental and Economic Sustainability Indices of Alternatives for CBR 10%     00.10.20.30.40.50.60.70.80.910 0.2 0.4 0.6 0.8 1Environmental Sustainability Index Economic Sustainability Index CBR 10% AADT 250 vpd GeoBCGeoSLPCCHMAHMAPCCGeoSLGeoBC00.10.20.30.40.50.60.70.80.910 0.2 0.4 0.6 0.8 1Environmental Sustainability Index Economic Sustainability Index CBR 10% AADT 500 vpd GeoBCGeoSLPCCHMAHMAPCCGeoSLGeoBC113      00.10.20.30.40.50.60.70.80.910 0.2 0.4 0.6 0.8 1Environmental Sustainability Index Economic Sustainability Index CBR 10% AADT 1000 vpd GeoBCGeoSLPCCHMAHMAPCCGeoSLGeoBC00.10.20.30.40.50.60.70.80.910 0.2 0.4 0.6 0.8 1Environmental Sustainability Index Economic Sustainability Index CBR 10% AADT 2000 vpd GeoBCGeoSLPCCHMAHMAPCCGeoSLGeoBC114  Appendix B3: Aggregated Sustainability Indices of Alternatives for CBR 10%    00.10.20.30.40.50.60.70.80.910 500 1000 1500 2000 2500Aggregated SI AADT (vpd) Neutral - CBR 10% PCC-HPCC-LPCCHMA-LHMAGeoSL-HGeoSL-LGeoSLGeoBC-HGeoBC-LHMA-HGeoBC00.10.20.30.40.50.60.70.80.910 500 1000 1500 2000 2500Aggregated SI AADT (vpd) Pro-Environment - CBR 10% PCC-HPCC-LPCCHMA-HHMA-LHMAGeoSL-HGeoSL-LGeoSLGeoBC-HGeoBC-LGeoBC115    00.10.20.30.40.50.60.70.80.910 500 1000 1500 2000 2500Aggregated SI AADT (vpd) Pro-Economic - CBR 10% PCC-HPCC-LPCCHMA-HHMA-LHMAGeoSL-HGeoSL-LGeoSLGeoBC-HGeoBC-LGeoBC

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0074439/manifest

Comment

Related Items