International Construction Specialty Conference of the Canadian Society for Civil Engineering (ICSC) (5th : 2015)

Assessment of network-level environmental sustainability in infrastructure systems using service and… Batouli, Mostafa; Mostafavi, Ali Jun 30, 2015

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

Item Metadata

Download

Media
52660-Batouli_M_et_al_ICSC15_102_Network_Level_Environmental_Sustainability.pdf [ 805.14kB ]
52660-Batouli_M_et_al_ICSC15_102_Network_Level_Environmental_Sustainability_slides.pdf [ 1.24MB ]
Metadata
JSON: 52660-1.0076384.json
JSON-LD: 52660-1.0076384-ld.json
RDF/XML (Pretty): 52660-1.0076384-rdf.xml
RDF/JSON: 52660-1.0076384-rdf.json
Turtle: 52660-1.0076384-turtle.txt
N-Triples: 52660-1.0076384-rdf-ntriples.txt
Original Record: 52660-1.0076384-source.json
Full Text
52660-1.0076384-fulltext.txt
Citation
52660-1.0076384.ris

Full Text

5th International/11th Construction Specialty Conference 5e International/11e Conférence spécialisée sur la construction    Vancouver, British Columbia June 8 to June 10, 2015 / 8 juin au 10 juin 2015   ASSESSMENT OF NETWORK-LEVEL ENVIRONMENTAL SUSTAINABILITY IN INFRASTRUCTURE SYSTEMS USING SERVICE AND PERFORMANCE ADJUSTED LIFE CYCLE ANALYSIS Mostafa Batouli1,3, Ali Mostafavi2  1 PhD Candidate, Department of Civil and Environmental Engineering, Florida International University, USA 2 Assistant Professor, OHL School of Construction, Florida International University, USA 3 corresponding_author_sbatouli@fiu.edu Abstract: Managing environmental impacts of civil infrastructure systems is critical for fostering sustainable development. However, despite the growing body of literature, an integrated methodology that captures the specific traits of infrastructure systems for a network-level environmental impact assessment is still missing. The objective of this paper is to propose a novel methodology [called Service and Performance Adjusted Life Cycle Assessment (SPA-LCA)] for addressing the limitations of the traditional LCA in environmental assessment of infrastructure networks. The SPA-LCA methodology adopts a service-based accounting approach to enable aggregation of the impacts pertaining to assets with different functions and service life expectancies at the network level. In the proposed SPA-LCA methodology, first, through conducting traditional asset-level LCA, life cycle inventories for the assets are determined. Second, the life cycle inventories are disaggregated to performance-sensitive and none-sensitive impacts. Then, using a hybrid mathematical/agent-based simulation model, the levels of service and performance are simulated for different assets in the network across the analysis horizon. Finally, the environmental impacts are determined for each year based on the levels of service and performance. The application of the proposed SPA-LCA method is demonstrated in environmental assessment of a road network. The results highlight the capabilities of SPA-LCA in providing better insight regarding environmental performance of infrastructure networks. 1 INTRODUCTION With the growing awareness of environmental protection, decision makers are increasingly interested in careful assessment of the environmental impacts associated with civil infrastructure networks. However, the majority of the existing environmental assessment studies are based on asset-level models, which cannot fully capture the environmental impacts at the network level. There are fundamental differences between asset-level and network-level environmental assessment of infrastructure (Öberg et al. 2012):  1. Asset-level impacts mostly depend on technical solutions (Vanier 2001). For instance, the environmental performance of a single road depends on a technical plan that determines the design specifications and identifies the timing and type of future maintenance/rehabilitation activities. However, at the network level, non-technical considerations (such as financial plans corresponding to allocation of resources among several assets with competing maintenance needs) also affect environmental performance (Zhang et al. 2012). 102-1 2. Environmental impacts of an asset are assessed within the service life of the asset (i.e. from construction to demolition and reconstruction of the asset). However, infrastructure networks do not have a definite life cycle. Instead, at the network level, the objective of network management determines the analysis horizon. Objectives may be related to short-term operational, mid-term tactical, and long-term strategic planning horizons (Vanier 2001). 3. Management of a single asset is mainly consisted of decisions made at the time of asset design. However, the management of infrastructure networks is an ongoing process involving multiple stakeholders whose adaptive decision-making processes and behaviors affect the environmental impacts of these networks over time (Osman and Nikbakht 2014; du Plessis and Cole 2011).  4. Asset-level infrastructure management is solely focused on “objects” (i.e. projects or facilities). However, at the network level, the focus is on objects and their interdependencies (functional, budgetary, and resource) (Öberg et al. 2012; Vanier 2001). Life Cycle Assessment (LCA) is a widely used environmental impact assessment method which was initially developed for assessment of manufactured products (ISO 14040 2006). Due to its unique capability in compilation and evaluation of the potential environmental impacts, LCA has now become a dominant method for assessment of environmental performance in infrastructure. However, LCA has certain methodological limitations that affect its reliability for network-level environmental assessment. First, LCA provides a lump sum assessment of environmental performance by compiling the potential environmental impacts of a product throughout its expected service life (ISO 14040 2006). Such analysis assumes that the performance of the product is uniform throughout the analysis horizon which is contradictory with the varying nature of service and performance in infrastructure networks. The lump sum outcomes of LCA are most useful for manufactured products for which decision making is usually a single task done during the design or procurement of the product. However, in infrastructure networks, decision making is an ongoing process, which has significant implications particularly in use/operation phase when decisions about timing and type of maintenance and rehabilitation of infrastructure are made. Second, LCA is a static method in which the dynamic changes in the level of service and performance are not considered. The level of service and performance in infrastructure networks is affected by the decision-making processes and behaviors of stakeholders. LCA is not capable of capturing these dynamic behaviors affecting the performance and level of service in infrastructure networks. Finally, LCA has been developed based on the premise that a sustainable approach is the one that minimizes the environmental impacts per functional unit for a product. This approach is valid when the life cycle is definite and functional unit is unique. However, infrastructure networks have a continuous service life and serve multiple performance functions. Hence, for infrastructure networks, a sustainable approach is the one that provides the longest service life and greatest performance with the lowest environmental impacts. Hence, an appropriate methodology for assessment of environmental impacts in infrastructure networks should consider the following: (i) Continuous service life of networks; (ii) Dynamic changes in the level of service and performance of networks affecting the environmental impacts; and (iii) The decision making processes of stakeholders affecting the level of service and performance in infrastructure networks. An integrated methodology that captures the requirements of network-level environmental assessment in infrastructure is missing in the existing literature. The objective of this research is to address this gap in the body of knowledge by creating a network-level environmental assessment methodology that considers the specific traits of infrastructure networks. 2 BACKGROUND The majority of the existing environmental assessment methodologies are based on an environmental accounting principle in which the environmental events are recognized at the time when emission occurs or natural resources are depleted. This accounting principle is prone to shifting burdens from one location to another (Hoekstra and Janssen 2006), or from one stage of life cycle to another stage (ISO 14040 2006) [e.g., by postponing the required maintenance of a road and creating a need for earlier reconstruction]. Life cycle assessment has successfully solved this problem for asset/product-level environmental assessment by taking all the impacts related to the entire service life of the asset/product 102-2 into consideration (ISO 14040 2006; ISO 14044 2006). However, when it comes to network-level environmental assessment, individual assets have different start/finish dates of life cycles Hence, when LCA is used for a network-level environmental impact assessment, it only captures the emissions during the analysis horizon and does not capture the emissions related to activities (such as maintenance and rehabilitation activities) prior and after the analysis horizon. Therefore, using LCA for a network-level assessment leads to the same burden shifting problem of emission-based environmental accounting.  We propose that changing the accounting principle for environmental assessment from emission-basis to service-basis accounting can both prevent burden shifting and provide flexibility regarding the length of analysis. To this end, we make an analogy between the environmental accounting and financial accounting. In business and finance literature, two distinctive types of financial accounting are used: i) cash-based accounting in which revenues and expenses are recorded when the cash is transferred. The existing emission-based environmental accounting is similar to cash-based financial accounting in which the release of pollutants or consumption of resources play the role of cash flows; and ii) accrual accounting in which economic events are recognized at the time of transaction rather than when a payment is made (or received). Accrual accounting has shown to be more reflective of the impacts of managerial decisions and the financial conditions of a company because it takes both current and expected future cash flows into consideration (Kwon 1990). Using an analogy to accrual accounting approach (Figure 1), we propose a service-based environmental accounting principle in which the impacts are recognized when the service is provided rather than when emission is made. For example, the huge impacts created during the construction of a new road, are not recognized at the time of construction. Instead, these impacts are attributed to each year of the service life based on proportion of total expected service offered in that year. The service basis principle enables consideration of both current and future impacts of a network on a yearly basis, thus providing a measure for evaluation of the environmental performance of the network during any desired analysis horizon.   Financial Accounting  Environmental Accounting Cash basis Accounting Events are recognized at the time of payment.  Emission basis Accounting Events are recognized at the time of emission. Accrual Accounting Events are recognized at the time of transaction.  Service Basis Accounting Events are recognized at the time of service. Figure 1: The proposed service-basis accounting is analogues to accrual accounting in financial studies. 3 SERVICE AND PERFORMANCE ADJUSTED LIFE CYCLE ASSESSMENT (SPA-LCA) The proposed service-based accounting for environmental assessment of infrastructure networks requires determination of levels of service and performance in a network during the analysis horizon. Hence, a simulation-based approach is needed for dynamic modeling of the service and performance. To this end, the proposed SPA-LCA methodology consists of two modules to capture the specific requirements of network-level environmental assessment in infrastructure systems (Figure 2). The module of network performance simulation is comprised of a hybrid mathematical/agent-based simulation framework to model the complex dynamic interactions between the conditions of assets, the demand pattern, and the managerial decisions regarding preservation of the network. The outcomes of the simulation model include the level of service and performance of each asset. Given an inventory of relevant energy and material inputs and environmental releases, the simulated service and performance of the assets are then used in network environmental assessment module to calculate network-level.The network performance simulation module is based on the methodology proposed by Batouli and Mostafavi (2014) for dynamic modeling of agency-asset-user interactions in infrastructure systems. The level of   102-3 service and performance in infrastructure networks is an emergent property as a result of the interactions between the conditions of assets, the extent of demand, and the decision making processes in administrating agency (Batouli and Mostafavi 2014). The decision making processes are modeled in the asset management component of the proposed framework. The asset management component uses agent-based modeling to abstract and simulate the decisions regarding the timing and type of maintenance and rehabilitation (M&R) activities considering the current condition of the assets as well as the underlying policies and availability of funding (Mostafavi et al. 2013). The service model simulates the level of demand on each asset based on the historical information and the conditions of assets. The conditions of assets are in turn identified in asset performance model using dynamic mathematical simulation. The performance of an asset is a function of the level of service it provides and the preservation (i.e., maintenance or rehabilitation) it receives. Detailed information related to the components of network performance simulation module can be found in Batouli and Mostafavi (2014). The service and performance of all assets are used as inputs to the network environmental assessment module to calculate the network-level environmental impacts. An inventory of relevant energy and material input/output flows is created similar to the traditional LCA (ISO 14040 2006; ISO 14044 2006). However, in the asset-level life cycle inventories (LCI) of the SPA-LCA methodology, the inventory items are divided into two categories based on the sensitivity of the inventory item to the performance of the asset. Disaggregation of LCI to performance sensitive and non-sensitive items is required for translating the lump-sum environmental impacts in traditional LCA into a dynamic environmental profile in SPA-LCA methodology. The items whose quantities vary based on the performance of the asset fall into the performance sensitive category. The flows that impose a fixed amount of impact regardless of the performance of the asset are attributed to performance non-sensitive category. For example, in analyzing the environmental impacts of a pavement network, items related to the use phase of the pavements belong to the performance sensitive category. That is because fuel consumption and pollutant emissions of vehicles depend on the roughness of the pavement as well as the travel time (Yu and Lu 2012). However, the impacts related to the construction of the pavement are independent from the pavement performance and hence they fall into the non-sensitive impact category. In the SPCA-LCA model, a Performance Adjustment Factor (PAF) is defined to modify the quantities of performance-sensitive items based on the simulated levels of performance. The PAF is defined based on historical information regarding the correlation between environmental impacts and performance of infrastructure assets.  The impact values of performance adjusted inventory items are then summed up with the values of non-sensitive items to form an adjusted life cycle inventory. Finally, the adjusted life cycle impacts of each asset are attributed to different years of the asset’s expected service life based on the level of service at each year (obtained from the network performance simulation model). Attributing the impacts based on the level of service is consistent with the service-based accounting principle explained in previous section. Application of the proposed framework is elaborated in next section using a case study pertaining to a road network.   102-4 In Equation 1, PSRi denotes the initial value of PSR for a given link right after construction. This value is 4.5 according to Chootinan et al. (2006) and Lee et al. (1993). Cumulative Equivalent Single Axle Loads per day (CESAL) and STR (existing structure of pavement) capture the impact of traffic load and structural design of the pavement, respectively. CESAL is mathematically calculated in the service model based on the historic traffic data and projections of future traffic growth. An adjustment factor (A.F.) is used for considering the effect of climate conditions. Finally, a,b,c and d are empirically-based coefficients whose values depend on the type of pavement (Lee et al. 1993).  The improvements in pavement condition due to maintenance/rehabilitation activities are incorporated in Equation 1 with the denotation MR. The value of MR is calculated in asset management model. The asset management model uses agent based modeling to simulate the decision making behavior of the administrative agency regarding the timing and type of M&R activities. Four types of M&R activities are considered in this case study: routine maintenance, surface treatment, overlay, and rehabilitation. Each of these activities leads to a certain level of improvement in performance depending on the age of the pavement (Chootinan et al. 2006). For preservation of the network, the administrative agency follows a “worst-first” strategy in which the roads with lowest performance are prioritized for allocation of M&R funding. A maintenance and rehabilitation (M&R) activity is implemented only if it can restore the pavement to an excellent condition; otherwise, if an adequate funding is not available for the required M&R, repair activities are deferred to the next period. Details related to the agent-based modeling of the agency decision processes can be found in Batouli and Mostafavi (2014) and Batouli et al. (2014.  The outcomes of the network performance simulation model include the level of service and performance of pavement assets, the expected service life of each asset, and the type and timing of M&R activities. The expected service lives of individual pavement assets are determined based on the threshold values of PSR to determine the need for reconstruction. These threshold values are considered to be 2.2 and 2 for urban and rural roads, respectively (Elkins et al. 2013). Once a road reaches this threshold PSR value, it is considered to be irremediable by maintenance activities, and hence, it should be reconstructed. The outcomes of simulation model are used in the module of SPA-LCA to calculate the global warming potential of the network at each year of service. 4.2 Module of network environmental assessment  The cradle-to-grave life cycle inventories related to different types of pavements, based on average conditions in the United States, are obtained from Loijos et al. (2013). The data includes greenhouse gas emissions created during all stages of pavement life cycle from materials production to construction, use, M&R, and end of life. However, the LCI inventory provided in Loijos et al. (2013) is developed using traditional LCA methodology. This inventory information needs to be adjusted to be used in the SPA-LCA model. To this end, the life cycle inventory was divided into three categories of inventory items:  The impacts associated with the materials production, construction and end of life are not sensitive to the level of service and performance. The inventory items related to this category are kept unchanged in the adjusted LCI of SPA-LCA model. The impacts related to M&R activities directly depend on the type and frequency of M&R treatments. For example, a typical preservation plan during the service life of a pavement asset may include 3 routine maintenances, one surface treatment, and one rehabilitation. However, in reality, budget constraints and agency priorities may result in deferring or changing the planned maintenance. Similarly, accelerated deterioration of the pavement may lead to the need for an additional overlay treatment. Hence, the impacts related to M$R activities are determined based on the outcomes of the simulation model, which determines the timing and type of M&R activities. The unit emission for each of these M&R activities is extracted from the LCI inventory and is multiplied by the simulated number of each type of treatment during service life of an asset.  The quantity of GWP generated during use phase of a pavement highly depends on the rate of fuel consumption of vehicles traveling on the pavement. On the other hand, the fuel efficiency of the vehicles is a function of pavement performance. To incorporate the impact of the pavement roughness on the 102-6 overall GHG emissions related to the use phase, the adjustment factors suggested by (Barnes and Langworthy 2003) are used in this study. According to Barnes and Langworthy (2003), when PSR value of a road is between 3 and 3.5 the fuel consumption is 5% greater than fuel consumption on pavements with excellent condition. For PSR values in the range of 2.5-3.0, fuel consumption increases 15%.  The first two categories of inventory items are independent from the level of performance of pavement assets, while the last one is performance sensitive. Calculations of impacts for performance sensitive and non-sensitive impacts are explained below. 1. Calculation of performance non-sensitive impacts The performance adjusted inventory impacts are used in the service basis accounting model to distribute the life cycle impacts of a pavement asset over its service life. For performance non-sensitive items, the impact at each year is calculated using Equation 2:  [2] Xij = NSij*ESALijESALav Where: Xij:𝑇𝑇ℎ𝑒𝑒 𝐺𝐺𝐺𝐺𝐺𝐺 𝑟𝑟𝑒𝑒𝑟𝑟𝑟𝑟𝑟𝑟𝑒𝑒𝑟𝑟 𝑟𝑟𝑡𝑡 𝑝𝑝𝑒𝑒𝑟𝑟𝑝𝑝𝑡𝑡𝑟𝑟𝑝𝑝𝑟𝑟𝑝𝑝𝑝𝑝𝑒𝑒 𝑝𝑝𝑡𝑡𝑝𝑝 − 𝑠𝑠𝑒𝑒𝑝𝑝𝑠𝑠𝑠𝑠𝑟𝑟𝑠𝑠𝑠𝑠𝑒𝑒 𝑠𝑠𝑝𝑝𝑝𝑝𝑟𝑟𝑝𝑝𝑟𝑟𝑠𝑠 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑠𝑠𝑝𝑝 𝑦𝑦𝑒𝑒𝑟𝑟𝑟𝑟 𝑠𝑠 (𝑠𝑠𝑝𝑝 𝑀𝑀𝑀𝑀 𝐶𝐶𝑡𝑡2 𝑒𝑒𝑒𝑒𝑒𝑒𝑠𝑠𝑠𝑠𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟) NSij:𝑇𝑇𝑡𝑡𝑟𝑟𝑟𝑟𝑟𝑟 𝐺𝐺𝐺𝐺𝐺𝐺 𝑟𝑟𝑒𝑒𝑟𝑟𝑟𝑟𝑟𝑟𝑒𝑒𝑟𝑟 𝑟𝑟𝑡𝑡 𝑝𝑝𝑒𝑒𝑟𝑟𝑝𝑝𝑡𝑡𝑟𝑟𝑝𝑝𝑟𝑟𝑝𝑝𝑝𝑝𝑒𝑒 𝑝𝑝𝑡𝑡𝑝𝑝 −𝑠𝑠𝑒𝑒𝑝𝑝𝑠𝑠𝑠𝑠𝑟𝑟𝑠𝑠𝑠𝑠𝑒𝑒 𝑠𝑠𝑝𝑝𝑝𝑝𝑟𝑟𝑝𝑝𝑟𝑟𝑠𝑠 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑟𝑟𝑒𝑒𝑟𝑟𝑠𝑠𝑝𝑝𝑀𝑀 𝑟𝑟ℎ𝑒𝑒 𝑟𝑟𝑠𝑠𝑝𝑝𝑒𝑒 𝑝𝑝𝑦𝑦𝑝𝑝𝑟𝑟𝑒𝑒 𝑟𝑟ℎ𝑟𝑟𝑟𝑟 𝑝𝑝𝑡𝑡𝑝𝑝𝑟𝑟𝑟𝑟𝑠𝑠𝑝𝑝𝑠𝑠 𝑦𝑦𝑒𝑒𝑟𝑟𝑟𝑟 𝑠𝑠 (𝑠𝑠𝑝𝑝 𝑀𝑀𝑀𝑀 𝐶𝐶𝑡𝑡2 𝑒𝑒𝑒𝑒𝑒𝑒𝑠𝑠𝑠𝑠𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟)   ESALij:𝑇𝑇𝑡𝑡𝑟𝑟𝑟𝑟𝑟𝑟 𝑒𝑒𝑒𝑒𝑒𝑒𝑠𝑠𝑠𝑠𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟 𝑠𝑠𝑠𝑠𝑝𝑝𝑀𝑀𝑟𝑟𝑒𝑒 𝑟𝑟𝑎𝑎𝑟𝑟𝑒𝑒 𝑟𝑟𝑟𝑟𝑟𝑟𝑝𝑝𝑝𝑝𝑠𝑠𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑠𝑠𝑝𝑝 𝑦𝑦𝑒𝑒𝑟𝑟𝑟𝑟 𝑠𝑠  ESALav:𝑇𝑇𝑡𝑡𝑟𝑟𝑟𝑟𝑟𝑟 𝑒𝑒𝑒𝑒𝑒𝑒𝑠𝑠𝑠𝑠𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟 𝑠𝑠𝑠𝑠𝑝𝑝𝑀𝑀𝑟𝑟𝑒𝑒 𝑟𝑟𝑎𝑎𝑟𝑟𝑒𝑒 𝑟𝑟𝑟𝑟𝑟𝑟𝑝𝑝𝑝𝑝𝑠𝑠𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑟𝑟𝑒𝑒𝑟𝑟𝑠𝑠𝑝𝑝𝑀𝑀 𝑠𝑠𝑟𝑟𝑠𝑠 𝑠𝑠𝑒𝑒𝑟𝑟𝑠𝑠𝑠𝑠𝑝𝑝𝑒𝑒 𝑟𝑟𝑠𝑠𝑝𝑝𝑒𝑒 In Equation 2, the fraction ESALijESALav is a service adjustment factor that determines what proportion of the total service of road j is provided in year i.  The outcomes of the network performance simulation model is used to determine the level of service, performance condition, number and timing of M&R activities, and service life of individual assets in the network. This information is used in calculating the performance non-sensitive impacts in Equation 2. For example, the simulation model shows that road A reaches its end of life at year 7 of the analysis horizon, and hence, it is reconstructed at year 8. Year 8 is the beginning of a service life for road A. This service life lasts for 41 years. During this service life, based on the simulated conditions and the worst-first preservation strategy, road A will receive two surface treatments, three overlays, and one rehabilitation. Each surface treatment, overlay and rehabilitation of road A creates a total of 24, 71 and 141 Mg of CO2 eq. GWP, respectively. Therefore, during this service life (from year 8 to year 49) a total of 402 Mg CO2 eq. (2×24+3×71+1×141=402 Mg CO2 eq.) impact will be created due to M&R activities. In addition, the materials production, construction and end of life of road A create 6508, 123 and 1175 Mg CO2 eq. global warming potential, respectively. Thus, a total of 8208 Mg CO2 eq. GWP of performance non-sensitive impacts is created during the service life of road A.  For distributing the impacts to each year, total impacts are multiplied by the service adjustment factor. For example, the simulation model shows that the traffic on road A in year 10 is 0.1442 ESAL. The total traffic load of Road A during this life cycle (i.e., year 8 to 49) is 10.70157 ESAL. Therefore 1.3% of the total service is provided in year 10. Based on the service basis accounting principle, 1.3% of the total life cycle impacts of road A (approximately 110.6 Mg CO2 eq. GWP) is due to the service in year 10. 102-7 2. Calculation of performance sensitive impacts  In order to consider the impacts of pavement performance on the fuel consumption of the vehicles, the impacts of use phase are adjusted based on the simulated pavement roughness (measured using in PSR) using Equation 3.   [3] Yij = TSj*ESALijESALav∗ PAF                                                                                                                          Where: Yij:𝑇𝑇ℎ𝑒𝑒 𝐺𝐺𝐺𝐺𝐺𝐺 𝑟𝑟𝑒𝑒𝑟𝑟𝑟𝑟𝑟𝑟𝑒𝑒𝑟𝑟 𝑟𝑟𝑡𝑡 𝑝𝑝𝑒𝑒𝑟𝑟𝑝𝑝𝑡𝑡𝑟𝑟𝑝𝑝𝑟𝑟𝑝𝑝𝑝𝑝𝑒𝑒 𝑠𝑠𝑒𝑒𝑝𝑝𝑠𝑠𝑠𝑠𝑟𝑟𝑠𝑠𝑠𝑠𝑒𝑒 𝑠𝑠𝑝𝑝𝑝𝑝𝑟𝑟𝑝𝑝𝑟𝑟𝑠𝑠 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑠𝑠𝑝𝑝 𝑦𝑦𝑒𝑒𝑟𝑟𝑟𝑟 𝑠𝑠 (𝑠𝑠𝑝𝑝 𝑀𝑀𝑀𝑀 𝐶𝐶𝑡𝑡2 𝑒𝑒𝑒𝑒𝑒𝑒𝑠𝑠𝑠𝑠𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟) TSj:𝐺𝐺𝐺𝐺𝐺𝐺 𝑟𝑟𝑒𝑒𝑟𝑟𝑟𝑟𝑟𝑟𝑒𝑒𝑟𝑟 𝑟𝑟𝑡𝑡 𝑒𝑒𝑠𝑠𝑒𝑒 𝑝𝑝ℎ𝑟𝑟𝑠𝑠𝑒𝑒 𝑠𝑠𝑝𝑝𝑝𝑝𝑟𝑟𝑝𝑝𝑟𝑟𝑠𝑠 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑝𝑝𝑡𝑡𝑝𝑝𝑠𝑠𝑠𝑠𝑟𝑟𝑒𝑒𝑟𝑟𝑠𝑠𝑝𝑝𝑀𝑀 𝑒𝑒𝑎𝑎𝑝𝑝𝑒𝑒𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑝𝑝𝑡𝑡𝑝𝑝𝑟𝑟𝑠𝑠𝑟𝑟𝑠𝑠𝑡𝑡𝑝𝑝𝑠𝑠    PAF:𝐺𝐺𝑒𝑒𝑟𝑟𝑝𝑝𝑡𝑡𝑟𝑟𝑝𝑝𝑟𝑟𝑝𝑝𝑝𝑝𝑒𝑒 𝐴𝐴𝑟𝑟𝑗𝑗𝑒𝑒𝑠𝑠𝑟𝑟𝑝𝑝𝑒𝑒𝑝𝑝𝑟𝑟 𝐹𝐹𝑟𝑟𝑝𝑝𝑟𝑟𝑡𝑡𝑟𝑟 ESALij:𝑇𝑇𝑡𝑡𝑟𝑟𝑟𝑟𝑟𝑟 𝑒𝑒𝑒𝑒𝑒𝑒𝑠𝑠𝑠𝑠𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟 𝑠𝑠𝑠𝑠𝑝𝑝𝑀𝑀𝑟𝑟𝑒𝑒 𝑟𝑟𝑎𝑎𝑟𝑟𝑒𝑒 𝑟𝑟𝑟𝑟𝑟𝑟𝑝𝑝𝑝𝑝𝑠𝑠𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑠𝑠𝑝𝑝 𝑦𝑦𝑒𝑒𝑟𝑟𝑟𝑟 𝑠𝑠  ESALav:𝑇𝑇𝑡𝑡𝑟𝑟𝑟𝑟𝑟𝑟 𝑒𝑒𝑒𝑒𝑒𝑒𝑠𝑠𝑠𝑠𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟 𝑠𝑠𝑠𝑠𝑝𝑝𝑀𝑀𝑟𝑟𝑒𝑒 𝑟𝑟𝑎𝑎𝑟𝑟𝑒𝑒 𝑟𝑟𝑟𝑟𝑟𝑟𝑝𝑝𝑝𝑝𝑠𝑠𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑟𝑟𝑒𝑒𝑟𝑟𝑠𝑠𝑝𝑝𝑀𝑀 𝑠𝑠𝑟𝑟𝑠𝑠 𝑠𝑠𝑒𝑒𝑟𝑟𝑠𝑠𝑠𝑠𝑝𝑝𝑒𝑒 𝑟𝑟𝑠𝑠𝑝𝑝𝑒𝑒 For example, under excellent roughness condition, 2150 Mg CO2 eq. GWP is created due to use of road A during this life cycle (i.e., years 8-49). However, the performance of road A is not excellent in year 10 (PSR=3.36), and hence, impacts will be worse. To account for the additional impacts, the use impact of road A is multiplied by a PAF of 1.05 (associated with PSR of road A in year 10). Thus, the use impact in year 10 is calculated as follows: 2150×1.3%×1.05= 29.35 Mg CO2 eq. GWP. Accordingly, the total environmental impacts of road A in year 10 can be calculated by adding the performance sensitive and non-sensitive impacts. A similar process is conducted for all assets in the network for the 40 year analysis horizon. The results are shown in Figure 3. In Figure 3, the decreases in GWP values are related to reduction of the service due to partial or complete road closures for applying M&R treatments.   Figure 3: Asset-level SPA-LCA GWP impacts.  Another advantage of SPA-LCA for the network-level environmental assessment is that it facilitates aggregation of asset-level impacts into network-level environmental performance. The results of traditional LCA are based on the aggregation of the life cycle of individual assets. However, the aggregated lump-sum impacts of individual assets could be misleading. For example, a planning strategy might lead to lower aggregated lump-sum impacts at the network level; however, it might reduce the service life of the network and raise the need for major reconstruction projects after the analysis horizon. However, in the SPA-LCA method, the impacts are not presented as a lump sum value for the total life cycle of an asset. Instead they are calculated for each year, and thus, the impacts at network-level are obtained by aggregating the asset-level impacts in each year during the analysis horizon. Hence, using this approach, the need for defining a definite service life for the network is eliminated, and the analysis could be performed for any desired analysis horizons. Figure 4 depicts the network-level impacts for the case study network. As shown in the figure, the network-level impacts increase with time. This is due to 102-8 two reasons. First, the traffic growth rate is positive in this network which basically means there will be accelerated deterioration and increased fuel consumption in future. Second, with the current preservation strategy on this network, the average PSR of the network decreases over time from an initial value of 4.1 to a final value of 3.55. This results in increased fuel consumption at the later stages of the life cycle of these assets, which leads to greater GWP impacts. This implies that, for networks with expected demand growth, a sustainable approach is to adopt more durable materials and pavement types to slow down the rate of deterioration and eliminate the need for frequent M&R treatment.  Figure 4: Network-level SPA-LCA GWP impacts.    4. CONCLUSION This study proposed a new methodology for assessment of environmental impacts of infrastructure systems at the network level. From a theoretical aspect, the underlying premise of the proposed methodology introduces a new paradigm for assessment of environmental sustainability in infrastructure networks. While the traditional LCA fails to recognize the level of service and performance in conceptualizing sustainability in networks, the proposed method defines a sustainable approach to be the one that provides the longest service life and greatest performance with the lowest environmental impacts. From a methodological perspective, the proposed methodology addresses the limitations of the traditional LCA methods for network-level assessment of infrastructure systems by: (i) capturing the complex interactions affecting the level of service, performance, and environmental performance of infrastructure networks; (ii) eliminating burden shifting through recognizing environmental impacts at the time of service; and (iii) considering the continuous service life of networks by determining the environmental impacts for each year, and thus, eliminating the need for defining a global life cycle for the network. From a practical aspect, the proposed method enables evaluation of environmental impacts associated with various operational (e.g., prioritizing M&R activities for funding allocation) and strategic (e.g., corridor planning) decisions.  REFERENCES Barnes, G., and Langworthy, P. (2003). "The per-mile costs of operating automobiles and trucks." Batouli, M., and Mostafavi, A. (2014a). "A hybrid simulation framework for integrated management of infrastructure networks." Proceedings of the 2014 Winter Simulation Conference, IEEE Press, 3319-3330. Chootinan, P., Chen, A., Horrocks, M. R., and Bolling, D. (2006). "A multi-year pavement maintenance program using a stochastic simulation-based genetic algorithm approach." Transportation Research Part A: Policy and Practice, 40(9), 725-743. 102-9 du Plessis, C., and Cole, R. J. (2011). "Motivating change: shifting the paradigm." Build.Res.Inf., 39(5), 436-449. Elkins, G. E., Thompson, T. M., Groeger, J. L., Visintine, B., and Rada, G. R. (2013). Reformulated Pavement Remaining Service Life Framework, . Haas, R. (2008). "The ICMPA7 investment analysis and communication challenge for road assets." Prepared for the 7th Int. Conf. on Managing Pavement Assets, Calgary, . Hoekstra, R., and Janssen, M. A. (2006). "Environmental responsibility and policy in a two-country dynamic input–output model." Econ.Syst.Res. 18(1), 61-84. ISO 14040, I. (2006). "14040 Environmental management-life cycle assessment-principles and framework." London: British Standards Institution. ISO 14044, I. (2006). "14044: environmental management—life cycle assessment—requirements and guidelines." International Organization for Standardization. Kwon, Y. K. (1990). "Accrual versus cash-basis accounting methods: An agency-theoretic comparison." J.Account.Public Policy, 8(4), 267-281. Lee, Y., Mohseni, A., and Darter, M. I. (1993). "Simplified pavement performance models." Transp.Res.Rec. 7-7. Loijos, A., Santero, N., and Ochsendorf, J. (2013). "Life cycle climate impacts of the US concrete pavement network." Resour.Conserv.Recycling, 72: 76-83. Mostafavi, A., Abraham, D., and DeLaurentis, D. (2013). "Ex-ante policy analysis in civil infrastructure systems." J.Comput.Civ.Eng. 28(5). Ness, B., Urbel-Piirsalu, E., Anderberg, S., and Olsson, L. (2007). "Categorising tools for sustainability assessment." Ecol.Econ. 60(3): 498-508. Öberg, C., Huge-Brodin, M., and Björklund, M. (2012). "Applying a network level in environmental impact assessments." Journal of Business Research, 65(2): 247-255. Osman, H., and Nikbakht, M. (2014). "A game-theoretic model for roadway performance management: A socio-technical approach." Built Environment Project and Asset Management, 4(1): 40-54. Vanier, D. “. (2001). "Why industry needs asset management tools." J.Comput.Civ.Eng., 15(1): 35-43. Yu, B., and Lu, Q. (2012). "Life cycle assessment of pavement: Methodology and case study." Transportation Research Part D: Transport and Environment, 17(5): 380-388. Zhang, H., Keoleian, G. A., and Lepech, M. D. (2012). "Network-Level Pavement Asset Management System Integrated with Life-Cycle Analysis and Life-Cycle Optimization." J Infrastruct Syst, 19(1): 99-107. 102-10  5th International/11th Construction Specialty Conference 5e International/11e Conférence spécialisée sur la construction    Vancouver, British Columbia June 8 to June 10, 2015 / 8 juin au 10 juin 2015   ASSESSMENT OF NETWORK-LEVEL ENVIRONMENTAL SUSTAINABILITY IN INFRASTRUCTURE SYSTEMS USING SERVICE AND PERFORMANCE ADJUSTED LIFE CYCLE ANALYSIS Mostafa Batouli1,3, Ali Mostafavi2  1 PhD Candidate, Department of Civil and Environmental Engineering, Florida International University, USA 2 Assistant Professor, OHL School of Construction, Florida International University, USA 3 corresponding_author_sbatouli@fiu.edu Abstract: Managing environmental impacts of civil infrastructure systems is critical for fostering sustainable development. However, despite the growing body of literature, an integrated methodology that captures the specific traits of infrastructure systems for a network-level environmental impact assessment is still missing. The objective of this paper is to propose a novel methodology [called Service and Performance Adjusted Life Cycle Assessment (SPA-LCA)] for addressing the limitations of the traditional LCA in environmental assessment of infrastructure networks. The SPA-LCA methodology adopts a service-based accounting approach to enable aggregation of the impacts pertaining to assets with different functions and service life expectancies at the network level. In the proposed SPA-LCA methodology, first, through conducting traditional asset-level LCA, life cycle inventories for the assets are determined. Second, the life cycle inventories are disaggregated to performance-sensitive and none-sensitive impacts. Then, using a hybrid mathematical/agent-based simulation model, the levels of service and performance are simulated for different assets in the network across the analysis horizon. Finally, the environmental impacts are determined for each year based on the levels of service and performance. The application of the proposed SPA-LCA method is demonstrated in environmental assessment of a road network. The results highlight the capabilities of SPA-LCA in providing better insight regarding environmental performance of infrastructure networks. 1 INTRODUCTION With the growing awareness of environmental protection, decision makers are increasingly interested in careful assessment of the environmental impacts associated with civil infrastructure networks. However, the majority of the existing environmental assessment studies are based on asset-level models, which cannot fully capture the environmental impacts at the network level. There are fundamental differences between asset-level and network-level environmental assessment of infrastructure (Öberg et al. 2012):  1. Asset-level impacts mostly depend on technical solutions (Vanier 2001). For instance, the environmental performance of a single road depends on a technical plan that determines the design specifications and identifies the timing and type of future maintenance/rehabilitation activities. However, at the network level, non-technical considerations (such as financial plans corresponding to allocation of resources among several assets with competing maintenance needs) also affect environmental performance (Zhang et al. 2012). 102-1 2. Environmental impacts of an asset are assessed within the service life of the asset (i.e. from construction to demolition and reconstruction of the asset). However, infrastructure networks do not have a definite life cycle. Instead, at the network level, the objective of network management determines the analysis horizon. Objectives may be related to short-term operational, mid-term tactical, and long-term strategic planning horizons (Vanier 2001). 3. Management of a single asset is mainly consisted of decisions made at the time of asset design. However, the management of infrastructure networks is an ongoing process involving multiple stakeholders whose adaptive decision-making processes and behaviors affect the environmental impacts of these networks over time (Osman and Nikbakht 2014; du Plessis and Cole 2011).  4. Asset-level infrastructure management is solely focused on “objects” (i.e. projects or facilities). However, at the network level, the focus is on objects and their interdependencies (functional, budgetary, and resource) (Öberg et al. 2012; Vanier 2001). Life Cycle Assessment (LCA) is a widely used environmental impact assessment method which was initially developed for assessment of manufactured products (ISO 14040 2006). Due to its unique capability in compilation and evaluation of the potential environmental impacts, LCA has now become a dominant method for assessment of environmental performance in infrastructure. However, LCA has certain methodological limitations that affect its reliability for network-level environmental assessment. First, LCA provides a lump sum assessment of environmental performance by compiling the potential environmental impacts of a product throughout its expected service life (ISO 14040 2006). Such analysis assumes that the performance of the product is uniform throughout the analysis horizon which is contradictory with the varying nature of service and performance in infrastructure networks. The lump sum outcomes of LCA are most useful for manufactured products for which decision making is usually a single task done during the design or procurement of the product. However, in infrastructure networks, decision making is an ongoing process, which has significant implications particularly in use/operation phase when decisions about timing and type of maintenance and rehabilitation of infrastructure are made. Second, LCA is a static method in which the dynamic changes in the level of service and performance are not considered. The level of service and performance in infrastructure networks is affected by the decision-making processes and behaviors of stakeholders. LCA is not capable of capturing these dynamic behaviors affecting the performance and level of service in infrastructure networks. Finally, LCA has been developed based on the premise that a sustainable approach is the one that minimizes the environmental impacts per functional unit for a product. This approach is valid when the life cycle is definite and functional unit is unique. However, infrastructure networks have a continuous service life and serve multiple performance functions. Hence, for infrastructure networks, a sustainable approach is the one that provides the longest service life and greatest performance with the lowest environmental impacts. Hence, an appropriate methodology for assessment of environmental impacts in infrastructure networks should consider the following: (i) Continuous service life of networks; (ii) Dynamic changes in the level of service and performance of networks affecting the environmental impacts; and (iii) The decision making processes of stakeholders affecting the level of service and performance in infrastructure networks. An integrated methodology that captures the requirements of network-level environmental assessment in infrastructure is missing in the existing literature. The objective of this research is to address this gap in the body of knowledge by creating a network-level environmental assessment methodology that considers the specific traits of infrastructure networks. 2 BACKGROUND The majority of the existing environmental assessment methodologies are based on an environmental accounting principle in which the environmental events are recognized at the time when emission occurs or natural resources are depleted. This accounting principle is prone to shifting burdens from one location to another (Hoekstra and Janssen 2006), or from one stage of life cycle to another stage (ISO 14040 2006) [e.g., by postponing the required maintenance of a road and creating a need for earlier reconstruction]. Life cycle assessment has successfully solved this problem for asset/product-level environmental assessment by taking all the impacts related to the entire service life of the asset/product 102-2 into consideration (ISO 14040 2006; ISO 14044 2006). However, when it comes to network-level environmental assessment, individual assets have different start/finish dates of life cycles Hence, when LCA is used for a network-level environmental impact assessment, it only captures the emissions during the analysis horizon and does not capture the emissions related to activities (such as maintenance and rehabilitation activities) prior and after the analysis horizon. Therefore, using LCA for a network-level assessment leads to the same burden shifting problem of emission-based environmental accounting.  We propose that changing the accounting principle for environmental assessment from emission-basis to service-basis accounting can both prevent burden shifting and provide flexibility regarding the length of analysis. To this end, we make an analogy between the environmental accounting and financial accounting. In business and finance literature, two distinctive types of financial accounting are used: i) cash-based accounting in which revenues and expenses are recorded when the cash is transferred. The existing emission-based environmental accounting is similar to cash-based financial accounting in which the release of pollutants or consumption of resources play the role of cash flows; and ii) accrual accounting in which economic events are recognized at the time of transaction rather than when a payment is made (or received). Accrual accounting has shown to be more reflective of the impacts of managerial decisions and the financial conditions of a company because it takes both current and expected future cash flows into consideration (Kwon 1990). Using an analogy to accrual accounting approach (Figure 1), we propose a service-based environmental accounting principle in which the impacts are recognized when the service is provided rather than when emission is made. For example, the huge impacts created during the construction of a new road, are not recognized at the time of construction. Instead, these impacts are attributed to each year of the service life based on proportion of total expected service offered in that year. The service basis principle enables consideration of both current and future impacts of a network on a yearly basis, thus providing a measure for evaluation of the environmental performance of the network during any desired analysis horizon.   Financial Accounting  Environmental Accounting Cash basis Accounting Events are recognized at the time of payment.  Emission basis Accounting Events are recognized at the time of emission. Accrual Accounting Events are recognized at the time of transaction.  Service Basis Accounting Events are recognized at the time of service. Figure 1: The proposed service-basis accounting is analogues to accrual accounting in financial studies. 3 SERVICE AND PERFORMANCE ADJUSTED LIFE CYCLE ASSESSMENT (SPA-LCA) The proposed service-based accounting for environmental assessment of infrastructure networks requires determination of levels of service and performance in a network during the analysis horizon. Hence, a simulation-based approach is needed for dynamic modeling of the service and performance. To this end, the proposed SPA-LCA methodology consists of two modules to capture the specific requirements of network-level environmental assessment in infrastructure systems (Figure 2). The module of network performance simulation is comprised of a hybrid mathematical/agent-based simulation framework to model the complex dynamic interactions between the conditions of assets, the demand pattern, and the managerial decisions regarding preservation of the network. The outcomes of the simulation model include the level of service and performance of each asset. Given an inventory of relevant energy and material inputs and environmental releases, the simulated service and performance of the assets are then used in network environmental assessment module to calculate network-level.The network performance simulation module is based on the methodology proposed by Batouli and Mostafavi (2014) for dynamic modeling of agency-asset-user interactions in infrastructure systems. The level of   102-3 service and performance in infrastructure networks is an emergent property as a result of the interactions between the conditions of assets, the extent of demand, and the decision making processes in administrating agency (Batouli and Mostafavi 2014). The decision making processes are modeled in the asset management component of the proposed framework. The asset management component uses agent-based modeling to abstract and simulate the decisions regarding the timing and type of maintenance and rehabilitation (M&R) activities considering the current condition of the assets as well as the underlying policies and availability of funding (Mostafavi et al. 2013). The service model simulates the level of demand on each asset based on the historical information and the conditions of assets. The conditions of assets are in turn identified in asset performance model using dynamic mathematical simulation. The performance of an asset is a function of the level of service it provides and the preservation (i.e., maintenance or rehabilitation) it receives. Detailed information related to the components of network performance simulation module can be found in Batouli and Mostafavi (2014). The service and performance of all assets are used as inputs to the network environmental assessment module to calculate the network-level environmental impacts. An inventory of relevant energy and material input/output flows is created similar to the traditional LCA (ISO 14040 2006; ISO 14044 2006). However, in the asset-level life cycle inventories (LCI) of the SPA-LCA methodology, the inventory items are divided into two categories based on the sensitivity of the inventory item to the performance of the asset. Disaggregation of LCI to performance sensitive and non-sensitive items is required for translating the lump-sum environmental impacts in traditional LCA into a dynamic environmental profile in SPA-LCA methodology. The items whose quantities vary based on the performance of the asset fall into the performance sensitive category. The flows that impose a fixed amount of impact regardless of the performance of the asset are attributed to performance non-sensitive category. For example, in analyzing the environmental impacts of a pavement network, items related to the use phase of the pavements belong to the performance sensitive category. That is because fuel consumption and pollutant emissions of vehicles depend on the roughness of the pavement as well as the travel time (Yu and Lu 2012). However, the impacts related to the construction of the pavement are independent from the pavement performance and hence they fall into the non-sensitive impact category. In the SPCA-LCA model, a Performance Adjustment Factor (PAF) is defined to modify the quantities of performance-sensitive items based on the simulated levels of performance. The PAF is defined based on historical information regarding the correlation between environmental impacts and performance of infrastructure assets.  The impact values of performance adjusted inventory items are then summed up with the values of non-sensitive items to form an adjusted life cycle inventory. Finally, the adjusted life cycle impacts of each asset are attributed to different years of the asset’s expected service life based on the level of service at each year (obtained from the network performance simulation model). Attributing the impacts based on the level of service is consistent with the service-based accounting principle explained in previous section. Application of the proposed framework is elaborated in next section using a case study pertaining to a road network.   102-4 In Equation 1, PSRi denotes the initial value of PSR for a given link right after construction. This value is 4.5 according to Chootinan et al. (2006) and Lee et al. (1993). Cumulative Equivalent Single Axle Loads per day (CESAL) and STR (existing structure of pavement) capture the impact of traffic load and structural design of the pavement, respectively. CESAL is mathematically calculated in the service model based on the historic traffic data and projections of future traffic growth. An adjustment factor (A.F.) is used for considering the effect of climate conditions. Finally, a,b,c and d are empirically-based coefficients whose values depend on the type of pavement (Lee et al. 1993).  The improvements in pavement condition due to maintenance/rehabilitation activities are incorporated in Equation 1 with the denotation MR. The value of MR is calculated in asset management model. The asset management model uses agent based modeling to simulate the decision making behavior of the administrative agency regarding the timing and type of M&R activities. Four types of M&R activities are considered in this case study: routine maintenance, surface treatment, overlay, and rehabilitation. Each of these activities leads to a certain level of improvement in performance depending on the age of the pavement (Chootinan et al. 2006). For preservation of the network, the administrative agency follows a “worst-first” strategy in which the roads with lowest performance are prioritized for allocation of M&R funding. A maintenance and rehabilitation (M&R) activity is implemented only if it can restore the pavement to an excellent condition; otherwise, if an adequate funding is not available for the required M&R, repair activities are deferred to the next period. Details related to the agent-based modeling of the agency decision processes can be found in Batouli and Mostafavi (2014) and Batouli et al. (2014.  The outcomes of the network performance simulation model include the level of service and performance of pavement assets, the expected service life of each asset, and the type and timing of M&R activities. The expected service lives of individual pavement assets are determined based on the threshold values of PSR to determine the need for reconstruction. These threshold values are considered to be 2.2 and 2 for urban and rural roads, respectively (Elkins et al. 2013). Once a road reaches this threshold PSR value, it is considered to be irremediable by maintenance activities, and hence, it should be reconstructed. The outcomes of simulation model are used in the module of SPA-LCA to calculate the global warming potential of the network at each year of service. 4.2 Module of network environmental assessment  The cradle-to-grave life cycle inventories related to different types of pavements, based on average conditions in the United States, are obtained from Loijos et al. (2013). The data includes greenhouse gas emissions created during all stages of pavement life cycle from materials production to construction, use, M&R, and end of life. However, the LCI inventory provided in Loijos et al. (2013) is developed using traditional LCA methodology. This inventory information needs to be adjusted to be used in the SPA-LCA model. To this end, the life cycle inventory was divided into three categories of inventory items:  The impacts associated with the materials production, construction and end of life are not sensitive to the level of service and performance. The inventory items related to this category are kept unchanged in the adjusted LCI of SPA-LCA model. The impacts related to M&R activities directly depend on the type and frequency of M&R treatments. For example, a typical preservation plan during the service life of a pavement asset may include 3 routine maintenances, one surface treatment, and one rehabilitation. However, in reality, budget constraints and agency priorities may result in deferring or changing the planned maintenance. Similarly, accelerated deterioration of the pavement may lead to the need for an additional overlay treatment. Hence, the impacts related to M$R activities are determined based on the outcomes of the simulation model, which determines the timing and type of M&R activities. The unit emission for each of these M&R activities is extracted from the LCI inventory and is multiplied by the simulated number of each type of treatment during service life of an asset.  The quantity of GWP generated during use phase of a pavement highly depends on the rate of fuel consumption of vehicles traveling on the pavement. On the other hand, the fuel efficiency of the vehicles is a function of pavement performance. To incorporate the impact of the pavement roughness on the 102-6 overall GHG emissions related to the use phase, the adjustment factors suggested by (Barnes and Langworthy 2003) are used in this study. According to Barnes and Langworthy (2003), when PSR value of a road is between 3 and 3.5 the fuel consumption is 5% greater than fuel consumption on pavements with excellent condition. For PSR values in the range of 2.5-3.0, fuel consumption increases 15%.  The first two categories of inventory items are independent from the level of performance of pavement assets, while the last one is performance sensitive. Calculations of impacts for performance sensitive and non-sensitive impacts are explained below. 1. Calculation of performance non-sensitive impacts The performance adjusted inventory impacts are used in the service basis accounting model to distribute the life cycle impacts of a pavement asset over its service life. For performance non-sensitive items, the impact at each year is calculated using Equation 2:  [2] Xij = NSij*ESALijESALav Where: Xij:𝑇𝑇ℎ𝑒𝑒 𝐺𝐺𝐺𝐺𝐺𝐺 𝑟𝑟𝑒𝑒𝑟𝑟𝑟𝑟𝑟𝑟𝑒𝑒𝑟𝑟 𝑟𝑟𝑡𝑡 𝑝𝑝𝑒𝑒𝑟𝑟𝑝𝑝𝑡𝑡𝑟𝑟𝑝𝑝𝑟𝑟𝑝𝑝𝑝𝑝𝑒𝑒 𝑝𝑝𝑡𝑡𝑝𝑝 − 𝑠𝑠𝑒𝑒𝑝𝑝𝑠𝑠𝑠𝑠𝑟𝑟𝑠𝑠𝑠𝑠𝑒𝑒 𝑠𝑠𝑝𝑝𝑝𝑝𝑟𝑟𝑝𝑝𝑟𝑟𝑠𝑠 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑠𝑠𝑝𝑝 𝑦𝑦𝑒𝑒𝑟𝑟𝑟𝑟 𝑠𝑠 (𝑠𝑠𝑝𝑝 𝑀𝑀𝑀𝑀 𝐶𝐶𝑡𝑡2 𝑒𝑒𝑒𝑒𝑒𝑒𝑠𝑠𝑠𝑠𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟) NSij:𝑇𝑇𝑡𝑡𝑟𝑟𝑟𝑟𝑟𝑟 𝐺𝐺𝐺𝐺𝐺𝐺 𝑟𝑟𝑒𝑒𝑟𝑟𝑟𝑟𝑟𝑟𝑒𝑒𝑟𝑟 𝑟𝑟𝑡𝑡 𝑝𝑝𝑒𝑒𝑟𝑟𝑝𝑝𝑡𝑡𝑟𝑟𝑝𝑝𝑟𝑟𝑝𝑝𝑝𝑝𝑒𝑒 𝑝𝑝𝑡𝑡𝑝𝑝 −𝑠𝑠𝑒𝑒𝑝𝑝𝑠𝑠𝑠𝑠𝑟𝑟𝑠𝑠𝑠𝑠𝑒𝑒 𝑠𝑠𝑝𝑝𝑝𝑝𝑟𝑟𝑝𝑝𝑟𝑟𝑠𝑠 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑟𝑟𝑒𝑒𝑟𝑟𝑠𝑠𝑝𝑝𝑀𝑀 𝑟𝑟ℎ𝑒𝑒 𝑟𝑟𝑠𝑠𝑝𝑝𝑒𝑒 𝑝𝑝𝑦𝑦𝑝𝑝𝑟𝑟𝑒𝑒 𝑟𝑟ℎ𝑟𝑟𝑟𝑟 𝑝𝑝𝑡𝑡𝑝𝑝𝑟𝑟𝑟𝑟𝑠𝑠𝑝𝑝𝑠𝑠 𝑦𝑦𝑒𝑒𝑟𝑟𝑟𝑟 𝑠𝑠 (𝑠𝑠𝑝𝑝 𝑀𝑀𝑀𝑀 𝐶𝐶𝑡𝑡2 𝑒𝑒𝑒𝑒𝑒𝑒𝑠𝑠𝑠𝑠𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟)   ESALij:𝑇𝑇𝑡𝑡𝑟𝑟𝑟𝑟𝑟𝑟 𝑒𝑒𝑒𝑒𝑒𝑒𝑠𝑠𝑠𝑠𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟 𝑠𝑠𝑠𝑠𝑝𝑝𝑀𝑀𝑟𝑟𝑒𝑒 𝑟𝑟𝑎𝑎𝑟𝑟𝑒𝑒 𝑟𝑟𝑟𝑟𝑟𝑟𝑝𝑝𝑝𝑝𝑠𝑠𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑠𝑠𝑝𝑝 𝑦𝑦𝑒𝑒𝑟𝑟𝑟𝑟 𝑠𝑠  ESALav:𝑇𝑇𝑡𝑡𝑟𝑟𝑟𝑟𝑟𝑟 𝑒𝑒𝑒𝑒𝑒𝑒𝑠𝑠𝑠𝑠𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟 𝑠𝑠𝑠𝑠𝑝𝑝𝑀𝑀𝑟𝑟𝑒𝑒 𝑟𝑟𝑎𝑎𝑟𝑟𝑒𝑒 𝑟𝑟𝑟𝑟𝑟𝑟𝑝𝑝𝑝𝑝𝑠𝑠𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑟𝑟𝑒𝑒𝑟𝑟𝑠𝑠𝑝𝑝𝑀𝑀 𝑠𝑠𝑟𝑟𝑠𝑠 𝑠𝑠𝑒𝑒𝑟𝑟𝑠𝑠𝑠𝑠𝑝𝑝𝑒𝑒 𝑟𝑟𝑠𝑠𝑝𝑝𝑒𝑒 In Equation 2, the fraction ESALijESALav is a service adjustment factor that determines what proportion of the total service of road j is provided in year i.  The outcomes of the network performance simulation model is used to determine the level of service, performance condition, number and timing of M&R activities, and service life of individual assets in the network. This information is used in calculating the performance non-sensitive impacts in Equation 2. For example, the simulation model shows that road A reaches its end of life at year 7 of the analysis horizon, and hence, it is reconstructed at year 8. Year 8 is the beginning of a service life for road A. This service life lasts for 41 years. During this service life, based on the simulated conditions and the worst-first preservation strategy, road A will receive two surface treatments, three overlays, and one rehabilitation. Each surface treatment, overlay and rehabilitation of road A creates a total of 24, 71 and 141 Mg of CO2 eq. GWP, respectively. Therefore, during this service life (from year 8 to year 49) a total of 402 Mg CO2 eq. (2×24+3×71+1×141=402 Mg CO2 eq.) impact will be created due to M&R activities. In addition, the materials production, construction and end of life of road A create 6508, 123 and 1175 Mg CO2 eq. global warming potential, respectively. Thus, a total of 8208 Mg CO2 eq. GWP of performance non-sensitive impacts is created during the service life of road A.  For distributing the impacts to each year, total impacts are multiplied by the service adjustment factor. For example, the simulation model shows that the traffic on road A in year 10 is 0.1442 ESAL. The total traffic load of Road A during this life cycle (i.e., year 8 to 49) is 10.70157 ESAL. Therefore 1.3% of the total service is provided in year 10. Based on the service basis accounting principle, 1.3% of the total life cycle impacts of road A (approximately 110.6 Mg CO2 eq. GWP) is due to the service in year 10. 102-7 2. Calculation of performance sensitive impacts  In order to consider the impacts of pavement performance on the fuel consumption of the vehicles, the impacts of use phase are adjusted based on the simulated pavement roughness (measured using in PSR) using Equation 3.   [3] Yij = TSj*ESALijESALav∗ PAF                                                                                                                          Where: Yij:𝑇𝑇ℎ𝑒𝑒 𝐺𝐺𝐺𝐺𝐺𝐺 𝑟𝑟𝑒𝑒𝑟𝑟𝑟𝑟𝑟𝑟𝑒𝑒𝑟𝑟 𝑟𝑟𝑡𝑡 𝑝𝑝𝑒𝑒𝑟𝑟𝑝𝑝𝑡𝑡𝑟𝑟𝑝𝑝𝑟𝑟𝑝𝑝𝑝𝑝𝑒𝑒 𝑠𝑠𝑒𝑒𝑝𝑝𝑠𝑠𝑠𝑠𝑟𝑟𝑠𝑠𝑠𝑠𝑒𝑒 𝑠𝑠𝑝𝑝𝑝𝑝𝑟𝑟𝑝𝑝𝑟𝑟𝑠𝑠 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑠𝑠𝑝𝑝 𝑦𝑦𝑒𝑒𝑟𝑟𝑟𝑟 𝑠𝑠 (𝑠𝑠𝑝𝑝 𝑀𝑀𝑀𝑀 𝐶𝐶𝑡𝑡2 𝑒𝑒𝑒𝑒𝑒𝑒𝑠𝑠𝑠𝑠𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟) TSj:𝐺𝐺𝐺𝐺𝐺𝐺 𝑟𝑟𝑒𝑒𝑟𝑟𝑟𝑟𝑟𝑟𝑒𝑒𝑟𝑟 𝑟𝑟𝑡𝑡 𝑒𝑒𝑠𝑠𝑒𝑒 𝑝𝑝ℎ𝑟𝑟𝑠𝑠𝑒𝑒 𝑠𝑠𝑝𝑝𝑝𝑝𝑟𝑟𝑝𝑝𝑟𝑟𝑠𝑠 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑝𝑝𝑡𝑡𝑝𝑝𝑠𝑠𝑠𝑠𝑟𝑟𝑒𝑒𝑟𝑟𝑠𝑠𝑝𝑝𝑀𝑀 𝑒𝑒𝑎𝑎𝑝𝑝𝑒𝑒𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑝𝑝𝑡𝑡𝑝𝑝𝑟𝑟𝑠𝑠𝑟𝑟𝑠𝑠𝑡𝑡𝑝𝑝𝑠𝑠    PAF:𝐺𝐺𝑒𝑒𝑟𝑟𝑝𝑝𝑡𝑡𝑟𝑟𝑝𝑝𝑟𝑟𝑝𝑝𝑝𝑝𝑒𝑒 𝐴𝐴𝑟𝑟𝑗𝑗𝑒𝑒𝑠𝑠𝑟𝑟𝑝𝑝𝑒𝑒𝑝𝑝𝑟𝑟 𝐹𝐹𝑟𝑟𝑝𝑝𝑟𝑟𝑡𝑡𝑟𝑟 ESALij:𝑇𝑇𝑡𝑡𝑟𝑟𝑟𝑟𝑟𝑟 𝑒𝑒𝑒𝑒𝑒𝑒𝑠𝑠𝑠𝑠𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟 𝑠𝑠𝑠𝑠𝑝𝑝𝑀𝑀𝑟𝑟𝑒𝑒 𝑟𝑟𝑎𝑎𝑟𝑟𝑒𝑒 𝑟𝑟𝑟𝑟𝑟𝑟𝑝𝑝𝑝𝑝𝑠𝑠𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑠𝑠𝑝𝑝 𝑦𝑦𝑒𝑒𝑟𝑟𝑟𝑟 𝑠𝑠  ESALav:𝑇𝑇𝑡𝑡𝑟𝑟𝑟𝑟𝑟𝑟 𝑒𝑒𝑒𝑒𝑒𝑒𝑠𝑠𝑠𝑠𝑟𝑟𝑟𝑟𝑒𝑒𝑝𝑝𝑟𝑟 𝑠𝑠𝑠𝑠𝑝𝑝𝑀𝑀𝑟𝑟𝑒𝑒 𝑟𝑟𝑎𝑎𝑟𝑟𝑒𝑒 𝑟𝑟𝑟𝑟𝑟𝑟𝑝𝑝𝑝𝑝𝑠𝑠𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑡𝑡𝑝𝑝 𝑟𝑟𝑡𝑡𝑟𝑟𝑟𝑟 𝑗𝑗 𝑟𝑟𝑒𝑒𝑟𝑟𝑠𝑠𝑝𝑝𝑀𝑀 𝑠𝑠𝑟𝑟𝑠𝑠 𝑠𝑠𝑒𝑒𝑟𝑟𝑠𝑠𝑠𝑠𝑝𝑝𝑒𝑒 𝑟𝑟𝑠𝑠𝑝𝑝𝑒𝑒 For example, under excellent roughness condition, 2150 Mg CO2 eq. GWP is created due to use of road A during this life cycle (i.e., years 8-49). However, the performance of road A is not excellent in year 10 (PSR=3.36), and hence, impacts will be worse. To account for the additional impacts, the use impact of road A is multiplied by a PAF of 1.05 (associated with PSR of road A in year 10). Thus, the use impact in year 10 is calculated as follows: 2150×1.3%×1.05= 29.35 Mg CO2 eq. GWP. Accordingly, the total environmental impacts of road A in year 10 can be calculated by adding the performance sensitive and non-sensitive impacts. A similar process is conducted for all assets in the network for the 40 year analysis horizon. The results are shown in Figure 3. In Figure 3, the decreases in GWP values are related to reduction of the service due to partial or complete road closures for applying M&R treatments.   Figure 3: Asset-level SPA-LCA GWP impacts.  Another advantage of SPA-LCA for the network-level environmental assessment is that it facilitates aggregation of asset-level impacts into network-level environmental performance. The results of traditional LCA are based on the aggregation of the life cycle of individual assets. However, the aggregated lump-sum impacts of individual assets could be misleading. For example, a planning strategy might lead to lower aggregated lump-sum impacts at the network level; however, it might reduce the service life of the network and raise the need for major reconstruction projects after the analysis horizon. However, in the SPA-LCA method, the impacts are not presented as a lump sum value for the total life cycle of an asset. Instead they are calculated for each year, and thus, the impacts at network-level are obtained by aggregating the asset-level impacts in each year during the analysis horizon. Hence, using this approach, the need for defining a definite service life for the network is eliminated, and the analysis could be performed for any desired analysis horizons. Figure 4 depicts the network-level impacts for the case study network. As shown in the figure, the network-level impacts increase with time. This is due to 102-8 two reasons. First, the traffic growth rate is positive in this network which basically means there will be accelerated deterioration and increased fuel consumption in future. Second, with the current preservation strategy on this network, the average PSR of the network decreases over time from an initial value of 4.1 to a final value of 3.55. This results in increased fuel consumption at the later stages of the life cycle of these assets, which leads to greater GWP impacts. This implies that, for networks with expected demand growth, a sustainable approach is to adopt more durable materials and pavement types to slow down the rate of deterioration and eliminate the need for frequent M&R treatment.  Figure 4: Network-level SPA-LCA GWP impacts.    4. CONCLUSION This study proposed a new methodology for assessment of environmental impacts of infrastructure systems at the network level. From a theoretical aspect, the underlying premise of the proposed methodology introduces a new paradigm for assessment of environmental sustainability in infrastructure networks. While the traditional LCA fails to recognize the level of service and performance in conceptualizing sustainability in networks, the proposed method defines a sustainable approach to be the one that provides the longest service life and greatest performance with the lowest environmental impacts. From a methodological perspective, the proposed methodology addresses the limitations of the traditional LCA methods for network-level assessment of infrastructure systems by: (i) capturing the complex interactions affecting the level of service, performance, and environmental performance of infrastructure networks; (ii) eliminating burden shifting through recognizing environmental impacts at the time of service; and (iii) considering the continuous service life of networks by determining the environmental impacts for each year, and thus, eliminating the need for defining a global life cycle for the network. From a practical aspect, the proposed method enables evaluation of environmental impacts associated with various operational (e.g., prioritizing M&R activities for funding allocation) and strategic (e.g., corridor planning) decisions.  REFERENCES Barnes, G., and Langworthy, P. (2003). "The per-mile costs of operating automobiles and trucks." Batouli, M., and Mostafavi, A. (2014a). "A hybrid simulation framework for integrated management of infrastructure networks." Proceedings of the 2014 Winter Simulation Conference, IEEE Press, 3319-3330. Chootinan, P., Chen, A., Horrocks, M. R., and Bolling, D. (2006). "A multi-year pavement maintenance program using a stochastic simulation-based genetic algorithm approach." Transportation Research Part A: Policy and Practice, 40(9), 725-743. 102-9 du Plessis, C., and Cole, R. J. (2011). "Motivating change: shifting the paradigm." Build.Res.Inf., 39(5), 436-449. Elkins, G. E., Thompson, T. M., Groeger, J. L., Visintine, B., and Rada, G. R. (2013). Reformulated Pavement Remaining Service Life Framework, . Haas, R. (2008). "The ICMPA7 investment analysis and communication challenge for road assets." Prepared for the 7th Int. Conf. on Managing Pavement Assets, Calgary, . Hoekstra, R., and Janssen, M. A. (2006). "Environmental responsibility and policy in a two-country dynamic input–output model." Econ.Syst.Res. 18(1), 61-84. ISO 14040, I. (2006). "14040 Environmental management-life cycle assessment-principles and framework." London: British Standards Institution. ISO 14044, I. (2006). "14044: environmental management—life cycle assessment—requirements and guidelines." International Organization for Standardization. Kwon, Y. K. (1990). "Accrual versus cash-basis accounting methods: An agency-theoretic comparison." J.Account.Public Policy, 8(4), 267-281. Lee, Y., Mohseni, A., and Darter, M. I. (1993). "Simplified pavement performance models." Transp.Res.Rec. 7-7. Loijos, A., Santero, N., and Ochsendorf, J. (2013). "Life cycle climate impacts of the US concrete pavement network." Resour.Conserv.Recycling, 72: 76-83. Mostafavi, A., Abraham, D., and DeLaurentis, D. (2013). "Ex-ante policy analysis in civil infrastructure systems." J.Comput.Civ.Eng. 28(5). Ness, B., Urbel-Piirsalu, E., Anderberg, S., and Olsson, L. (2007). "Categorising tools for sustainability assessment." Ecol.Econ. 60(3): 498-508. Öberg, C., Huge-Brodin, M., and Björklund, M. (2012). "Applying a network level in environmental impact assessments." Journal of Business Research, 65(2): 247-255. Osman, H., and Nikbakht, M. (2014). "A game-theoretic model for roadway performance management: A socio-technical approach." Built Environment Project and Asset Management, 4(1): 40-54. Vanier, D. “. (2001). "Why industry needs asset management tools." J.Comput.Civ.Eng., 15(1): 35-43. Yu, B., and Lu, Q. (2012). "Life cycle assessment of pavement: Methodology and case study." Transportation Research Part D: Transport and Environment, 17(5): 380-388. Zhang, H., Keoleian, G. A., and Lepech, M. D. (2012). "Network-Level Pavement Asset Management System Integrated with Life-Cycle Analysis and Life-Cycle Optimization." J Infrastruct Syst, 19(1): 99-107. 102-10  Assessment of Network-level Environmental Sustainability in Infrastructure Systems Using Service and Performance Adjusted Life Cycle AnalysisMostafa Batouli1, Ali Mostafavi21 PhD Candidate, Department of Civil and Environmental Engineering, Florida International University, USA2 Assistant Professor, OHL School of Construction, College of Engineering and Computing, Florida International University, USA2015 Canadian Society for Civil Engineering Construction Specialty ConferenceProblem Statement Proposed Framework Numerical Case Results Ongoing Work1IntroductionCivil Infrastructure systems greatly influence the sustainability of urban areas1Problem Statement Proposed Framework Numerical Case Results Ongoing Work2Problem StatementOngoing Decision MakingDynamic Service & Performance Continuous Service LifeThe conventional approaches do not address the requirements of sustainability assessment at network level1Problem Statement Proposed Framework Numerical Case Results Ongoing Work31 Problem StatementThe complex adaptive nature of infrastructure systems requires an evolutionary approach toward sustainability assessment• Agency Decision-Making Processes• Degradation of performance in network links• Dynamic changes in the level demand/service3Problem Statement Proposed Framework Numerical Case Results Ongoing Work4Step 1: Simulate agency/network behaviors and interactions. Step 3:  Create dynamicasset level life cycle inventories.Step 2: Calculate unit impact of environmental events.Step 4: Allocate impacts using a Service-based environmental accounting model.Step 5: Calculate asset-level service and performance adjusted impacts.Step 6: Aggregate  environmental impacts at network-level.2 Framework for Assessing Network Environmental Sustainability4Problem Statement Proposed Framework Numerical Case Results Ongoing Work52 Framework for Assessing Environmental Sustainability EvolutionStep 1: Simulating dynamic behaviors of agency and degradation of network conditionNetwork ConditionAgency BehaviorTiming and Type of M&R Level of PerformanceAgent-Based ModelingDynamic Mathematical Simulation(Batouli and Mostafavi 2014)5Problem Statement Proposed Framework Numerical Case Results Ongoing Work62 Framework for Assessing Environmental Sustainability EvolutionStep 2: Calculating unit impacts of environmental eventsService Life of AssetEnvironmental EventsConstructionM1 M1M2 M2M3 M3 M2EOLUnit Impact of Mi Number of Occurrence6Problem Statement Proposed Framework Numerical Case Results Ongoing Work72 Framework for Assessing Environmental Sustainability EvolutionStep 3: Creating the dynamic life cycle inventories of network linksFixed ImpactM&R Impact * # of occurrencePerformance Adjusted Use ImpactDynamic LCI7Problem Statement Proposed Framework Numerical Case Results Ongoing Work82 Framework for Assessing Environmental Sustainability EvolutionStep 4: Allocating impacts using a service-based environmental accounting Model.Cash Basis: Events are recognized at the time of payment.Accrual Basis: Events are recognized at the time of transaction.Financial AccountingEmission Basis: Events are recognized at the time of emission.Service Basis: Events are recognized at the time of service.Environmental Accounting8Problem Statement Proposed Framework Numerical Case Results Ongoing Work92 Framework for Assessing Environmental Sustainability EvolutionStep 5: Calculating asset-level service-adjusted Impacts.𝑋𝑋𝑖𝑖𝑗𝑗 = 𝐿𝐿𝐿𝐿𝐿𝐿𝑗𝑗*𝑆𝑆𝑖𝑖𝑖𝑖𝐶𝐶𝑆𝑆𝑖𝑖Impact at year i for Asset jDynamic life cycle impacts for asset j Cumulative service of  asset j during its service lifeService level of  asset j in year i9Problem Statement Proposed Framework Numerical Case Results Ongoing Work102 Framework for Assessing Environmental Sustainability EvolutionStep 6: Aggregating environmental impacts at network-Level.𝑁𝑁𝑁𝑁𝐿𝐿𝑖𝑖 = �𝑗𝑗=1𝑛𝑛𝑋𝑋𝑖𝑖𝑗𝑗𝑁𝑁𝑁𝑁𝐿𝐿𝑖𝑖= Total Environmental Impact at Year i𝑋𝑋𝑖𝑖𝑗𝑗: Impact at Year i for Asset jn:Number of  Assets in the NetworkEITimeAsset 1EITimeAsset2EITimeAsset n......EITimeTotal Network Impact10Problem Statement Proposed Framework Numerical Case Results Ongoing Work11Section Name A B C D E F G H I J K LRoad Type R I I I R R I R R I I IPavement F C F F F CP CP CP CP C F FLength 1.6 0.5 0.7 0.2 0.4 2.7 0.6 1.1 2.8 1.4 1.7 0.6Width (feet) 36.1 37.4 41.0 37.4 42.7 41.3 46.6 53.8 39.0 40.7 38.7 54.5No. of Lanes 4 4 4 4 4 4 4 6 4 4 4 6ESAL/ Day 224 1185 1645 1756 864 688 1142 1785 1785 1185 1479 1756Numerical case3Data obtained from Haas (2007)Application of the proposed framework in assessing environmental impacts in a pavement network.Problem Statement Proposed Framework Numerical Case Results Ongoing Work12Numerical caseAgency follows a “worst-first” policy for M&R decision-making3𝑃𝑃𝑃𝑃𝑃𝑃𝑗𝑗 = 𝑃𝑃𝑃𝑃𝑃𝑃𝐼𝐼 − 𝐴𝐴.𝐹𝐹 ∗ 𝑎𝑎 ∗ 𝑃𝑃𝑆𝑆𝑃𝑃𝑏𝑏 ∗ 𝐴𝐴𝐴𝐴𝐴𝐴𝑗𝑗𝐶𝐶 ∗ 𝐿𝐿𝑁𝑁𝑃𝑃𝐴𝐴𝐿𝐿𝑗𝑗𝑑𝑑+𝑀𝑀𝑃𝑃𝑗𝑗 (Lee et al. 1993) Problem Statement Proposed Framework Numerical Case Results Ongoing Work131.82.32.83.33.84.30 10 20 30 40Asset-Level PSRYearRoad A Road B Road C Road D Road E Road FRoad G Road H Road I Road J Road K Road L422.533.544.50 10 20 30 40Average Weighted Network PSRYearSimulation ResultsNetwork performance reaches an equilibrium if agency decisions, demand, and other variables remain unchanged over the analysis horizon.Base scenario: $500k in M&R Budget and no traffic demand growth .Problem Statement Proposed Framework Numerical Case Results Ongoing Work14R² = 0.93463.33.353.43.453.53.553.63.650 200 400 600 800 1000PSRMR Budget (Thousand $)Scenario AnalysisThe impact of funding increase on performance improvement and environmental impact reduction diminishes after a certain thresholdR² = 0.946128002900300031003200330034000 500 1000 1500GWP (Mg. CO2 eq.)MR Budget (Thousand $)4Problem Statement Proposed Framework Numerical Case Results Ongoing Work1527002900310033003500370039000 200 400 600 800 1000GWP (Mg. CO2 eq.)MR Budget (Thousand $)No Growth 5% Traffic Growth4 Scenario AnalysisThe impact of decision variables (e.g., level of funding) are more significant for environmental impact reduction when the network experiences demand growthProblem Statement Proposed Framework Numerical Case Results Ongoing Work1628003000320034003600380040003.3 3.35 3.4 3.45 3.5 3.55 3.6GWP (Mg CO2 eq.)PSRBase Scenario 5% Demand GrowthScenario AnalysisNetwork average annual environmental impacts decrease with improvement in network performance The sensitivity of network environmental impacts to performance is greater when demand is growing4Problem Statement Proposed Framework Numerical Case Results Ongoing Work175 Contributions Theory: Evolutionary network sustainability: Network environmental assessment based on evaluation of the dynamic evolution of infrastructure systemsMethodology: Simulation-based approach for network sustainability assessment A service-based environmental accounting to eliminate burden shifting in network-level environmental assessment Informed Decision-Making:  Actionable science: Prescriptive knowledge of network-level sustainability for decision-making and policy formulationProblem Statement Proposed Framework Numerical Case Results Ongoing Work18Ongoing WorkIntegration of environmental, economic, and social impacts in the network sustainabilityReal world testing using Miami-Dade Expressway network dataInvestigation of various factors influencing network sustainability6Thanks!Questions?Infrastructure System-of-Systems (I-SoS) Research GroupProblem Statement Proposed Framework Numerical Case Results Ongoing Work20Road NameMg CO2 eq.Total Fixed ImpactSurface Treatment Overlay RehabilitationAverage Annual Use ImpactA 7805 24 71 141 55B 3408 10 29 59 47C 4686 13 40 81 64D 1278 4 11 22 17E 2185 7 20 39 16F 13737 41 124 248 97G 4260 12 37 56 57H 5307 16 48 124 37I 14049 42 127 254 98J 9372 27 81 161 127K 8929 33 99 198 155L 3307 12 37 73 583Step1: Calculating Unit Impacts of Environmental EventsNumerical case20Problem Statement Proposed Framework Numerical Case Results Ongoing Work213 Numerical casePast impacts Analysis horizon21

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:
http://iiif.library.ubc.ca/presentation/dsp.52660.1-0076384/manifest

Comment

Related Items