Open Collections

UBC Undergraduate Research

Electrification of the Sunshine Coast Transit System : A Feasibility Study Fu, Carol; Lin, Jason; Marcus, Michelle; Cui, Tom 2020-04-09

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


52966-Fu_C_et_al_ENVR_400_Electrification_2020.pdf [ 1.86MB ]
JSON: 52966-1.0390279.json
JSON-LD: 52966-1.0390279-ld.json
RDF/XML (Pretty): 52966-1.0390279-rdf.xml
RDF/JSON: 52966-1.0390279-rdf.json
Turtle: 52966-1.0390279-turtle.txt
N-Triples: 52966-1.0390279-rdf-ntriples.txt
Original Record: 52966-1.0390279-source.json
Full Text

Full Text

     Electrification of the Sunshine Coast Transit System: A Feasibility Study    ENVR 400 Final Report Carol Fu Jason Lin Michelle Marcus Tom Cui In collaboration with the ​2 Degrees Institute   University of British Columbia ENVR 400: Community Project in Environmental Science  Research Advisor: Tara Ivanochko     April 9, 2020                  Table of Contents  Abstract 5 Author Biographies 6 Introduction 7 The Climate Emergency 7 The Sunshine Coast Transit System 7 Figure 1. 8 Figure 2. 9 Table 1. 9 Project Aims 10 Electric Bus Operations 10 Overview of Trade-offs Between Technologies 10 Table 2. 11 Figure 3. 12 Figure 4. 12 1) Infrastructure requirements 12 Summary: Electric Bus Operations 14 Optimization of Bus Charging Schedules 14 Methodology 14 Figure 5. 15 Figure 6. 16 Results and Discussion 17 Table 3. 17 Figure 7. 18 Figure 8. 19 Summary: Optimization of Bus Charging Schedules 19 Life Cycle Cost Analysis 19 Methodology 19 Table 4. 20 Figure 9. 20 Table 5. 21 Table 6​. 23 Figure 10. 24 2   Table 7. 24 Figure 11. 25 Discussion 25 Summary: Life Cycle Cost Analysis 26 Ridership Strategies 26 Figure 12. 27 Literature Review: Ridership Improvement Strategies 27 Table 8. 30 Figure 13. 30 Figure 14. 30 Results and Discussion 31 Table 9. 31 Summary: Ridership Strategies 33 Conclusion, Limitations and Further Studies 33 Conclusion 33 Limitations 34 Further Studies 34 Acknowledgements 35 References 36 Appendix I: Optimization of Charging Schedules 42 Mathematical Reasoning Behind Charging Schedule Algorithm 42 Algorithm Code in R 42 Appendix II: Ridership Strategies 47 Table 10. 47 Table 11. 48           3   Abstract   Electrifying transportation is a necessary part of reducing greenhouse gas emissions in order to mitigate global climate change. Electric buses are gaining popularity worldwide, in urban and rural communities alike. This research, conducted in partnership with the 2 Degrees Institute, aims to explore pathways towards electrifying the transit system on the Sunshine Coast in British Columbia with the goal of implementing a fully electric fleet earlier than the current deadline of 2040. The project will employ life-cycle cost analysis to determine the most financially feasible charging method and use a numerical algorithm to optimize bus schedules in order to accommodate charging time and minimize demand charges. The report also provides recommendations for potential strategies to increase transit ridership on the Sunshine Coast based on a literature review. The results suggest that: (1) the demand charge for a fully electric fleet will be 625 kW; (2) the charging schedule can be optimized by charging 2 groups of electric buses at 2 different time periods; (3) fast charging has the lowest life-cycle cost due to smaller batteries and reduced number of chargers required; (4) strategies to increase ridership include service expansion, park-and-ride facilities, fare reduction of student and senior monthly passes and the revision of the DayPASS-on-Board program. The report findings may be used by the Sunshine Coast as well as other jurisdictions - especially rural regions - as a model for electric transit implementation.                      4   Author Biographies  Tom Cui A fourth-year Environmental Sciences student major with an area of concentration in land, air, and water at UBC. He is familiar with academic writing and data analysis, and he has experience with programming.   Carol Fu Carol Fu is a fifth-year UBC Science student majoring in Environmental Sciences with a focus on land, air, and water as areas of concentration. Carol has experience with research, data collection and analysis from previous co-op work opportunities at Environment Canada and Agriculture and Agri-Food Canada.   Jason Lin A fourth-year Environmental Sciences student with experience in field-work, data collection and analysis from previous work experience at Agriculture and Agri-Food Canada and Environment Canada.  Michelle Marcus Michelle is a fifth-year Environmental Sciences student enrolled in the land, air, and water concentration. She is a passionate advocate for climate justice having been actively involved in the successful UBC fossil fuel divestment campaign and organizing climate strikes. She enjoys data analysis and modelling.               5   Introduction  The Climate Emergency The transportation sector is the second biggest contributor to greenhouse gas emissions in Canada, as it is predominantly powered by petroleum-based fuels, primarily gasoline and diesel (National Inventory Report, 2017)​. The transportation sector accounts for 37% of British Columbia’s total emissions (Province of British Columbia, n.d.). According to the ​Special Report on Global Warming of 1.5​°​C​ (IPCC, 2018) global net human-caused emissions of carbon dioxide must​ be reduced by approximately 45% from 2010 levels by 2030 to limit warming to a “safe” level of 1.5°C, thus necessitating a rapid transition in energy systems in order to prevent catastrophic climate change impacts.  In November 2018, BC Transit, the regional transit authority of BC outside Greater Vancouver, adopted a ​Low Carbon Fleet Program​ to support provincial 2030 emissions reductions target for greenhouse gas emissions (GHGs) and to align with the provincial ​CleanBC​ plan. BC Transit pledged to start buying only electric heavy duty vehicles in 2023, with a goal of creating a fully electric provincial fleet in all vehicle ​classifications by 2040 (BC Transit, 2019a). Commensurate with purchasing electric heavy duty vehicles, BC Transit will also convert some of its fleet to lower carbon technologies, such as compressed natural gas (BC Transit, 2019a). ​Given the estimated timeline of 10 years to drastically cut global emissions ​(IPCC, 2018)​, the 2 Degrees Institute (2DI) is interested in investigating the feasibility of phasing out diesel buses in favor of zero-emission electric buses sooner than BC Transit’s proposed timeline. This would require developing a plan for BC Transit and the Sunshine Coast Regional District (SCRD) to electrify the Sunshine Coast transit system using the most operationally and financially feasible charging method. The Sunshine Coast Transit System The Sunshine Coast, located on the southern coast of British Columbia (Figure 1), currently operates fourteen diesel buses over five routes, running from Halfmoon Bay to Langdale (Figure 2 & Table 1). Communities north of Halfmoon Bay currently have no access to transit. To evaluate a successful transit system for this region, we must consider context-specific factors such as population growth, climate, topography, and seasonal population fluctuations. These factors will assist in projecting operational feasibility of transportation options, as well as ridership revenue.      6                Figure 1. Sunshine Coast, British Columbia (Google, n.d.)  The Sunshine Coast is a rural regional district with its main transportation corridor stretching across low-density communities. It has a small population of 29,970 people (Statistics Canada, 2017) with an anticipated increase of approximately 12,500 people over 25 years starting in 2014 (BC Transit & SCRD, 2014). With the limited need for road salting, there are minimal impacts to transit service from poor weather conditions and lower maintenance costs (BC Transit & SCRD, 2014). As in other coastal communities, the Sunshine Coast features steep topography, which can increase energy consumption and decrease bus lifetime. Seasonal service changes occur each year to synchronize with the BC Ferries schedule and respond to increased service demand in peak season as popular pursuits include hiking, camping, paddling and other outdoor activities, among tourists (BC Transit, 2019b). The ​Sunshine Coast Integrated Transportation Study​ (ISL Engineering and Land Services, 2011) identifies that 30% of transit rides are linked to the ferry schedule and that regular ferry delays impact transit schedules and reliability. Altogether the infrequent bus service, limited route access and dispersed low density communities have led to a strong dependence on private vehicles in the region.  7    Figure 2. Current Sunshine Coast bus route map (BC Transit, n.d.). Table 1. Sunshine Coast Regional Transit System’s active bus roster. The fleet consists of 8 full-size diesel buses, fleet numbers 4028-4033, and 6 mid-size diesel buses, fleet numbers 9224-9231 (CPTDB, 2019).   This feasibility study aims to support the reduction of regional emissions with the hope that it can also act as a roadmap for other rural communities in BC and around the world to transition to electric public transit. By taking advantage of new technology solutions and transitioning to an electric transit system, each electric bus will reduce carbon emissions by more than 135 tons per year (Sierra Club, 2019). Each year, the Sunshine Coast spends over $100 million dollars on imported fuel (SCRD, 2009). With an electric transit system, a portion of these funds can be redirected to developing local and regional energy systems which support the local economy (SCRD, 2009). 8   Project Aims The goal of this project is to determine the feasibility of electrifying public transit on the Sunshine Coast and investigate the feasibility of phasing out diesel buses in favor of zero-emission electric buses sooner than BC Transit’s proposed timeline. The project also aims to provide recommendations for ridership improvement strategies on the Sunshine Coast in order to increase ridership revenue and improve the feasibility of electric buses compared to diesel buses.  Our objectives are to: 1. Explore and summarize trade-offs between available charging technologies. 2. Optimize bus charging schedules in order to minimize operational costs and maximize operational efficiency. 3. Determine what type of charging infrastructure is most financially feasible for the Sunshine Coast transit system. 4. Develop recommendations for increasing transit ridership on the Sunshine Coast.   Electric Bus Operations  Battery electric buses involve fundamentally different operational requirements than diesel buses due to the very different infrastructure required for refueling (Xylia & Silveria, 2018).​ ​The required charging time is also significantly longer than the refueling time for a conventional diesel bus (Janovec & Koháni, 2019). Furthermore, electric buses are limited in their operational range as they require a minimum state of charge to operate.  As such, electrifying transit fleets requires careful planning of schedules and routes to ensure there is sufficient time for buses to recharge enough in order to complete their routes. Schedules will need to be revised to accommodate for charging time, whether charging occurs overnight or during service hours. Additional buses may be required to accomodate the limited operational range of the buses and dwell time required for charging (Jahic, Eskander, & Schulz, 2019).  Transitioning to electric buses also involves cost considerations. Electric buses have much higher upfront costs due to the newly developed technology as well as the additional costs for batteries and charging equipment (Xylia & Silveria, 2018). It also requires additional infrastructure including upgrading maintenance facilities, installing charging stations and potentially expanding the grid, in addition to decommissioning the old fleet and facilities. Overview of Trade-offs Between Technologies Electric bus charging can be stationary at the bus depot, bus stops or at a terminus station as well as on-road via trolley lines or inductive chargers on the road (Xylia & Silveria, 2018). While electric bus charging at bus depots and terminus stations are becoming increasingly popular, charging at bus stops and on-road are still emerging and expensive technologies. Trolley line 9   charging requires significant infrastructure and is not suitable for rural areas. As such, for this report, we will focus on depot and terminus station charging options. The main technologies currently available for these uses are: ● Plug-in charging​: The bus is charged through a direct current (DC) plug-in station. This is done at low power levels to charge buses for a long period of time overnight or during non-service hours, often at a bus depot. ​We will also refer to this as “overnight charging.” ● Overhead charging​: A bus is charged through a pantograph. This uses higher power levels to rapidly charge buses during service hours. It is usually used at bus stops or end route stations. ​We will also refer to this as “fast charging.” These two technologies have various costs and operational trade-offs which should be considered in the context of the specific location. The main trade-offs include infrastructure requirements, terrain suitability, scheduling constraints and financial costs.  In order to evaluate the feasibility of these two types of charging technologies, we compare two specific bus models: the ​Proterra​ ​Catalyst E2 max​ ​with Duopower Drivetrain (“Catalyst”) ​for overnight charging and the ​New Flyer Xcelsior CHARGE (“Xcelsior”)​ for fast charging (Table 2, Figures 3 and 4). These bus manufacturers were recommended to us by BC Transit. Both buses are 40 feet in length with a seating capacity of 40 and total capacity of 80 (Proterra, 2019a; New Flyer, 2019), which is the same size as the buses currently in use on the Sunshine Coast. Table 2. Bus specifications for plug-in and overhead charging (Proterra, 2019a; New Flyer, 2019 and G. Huber, personal communication, January 21, 2020).   Overnight charging with plug-in charger Fast charging with overhead charger Manufacturer Proterra New Flyer Model Catalyst E2 max with Duopower Drivetrain (“Catalyst”) Xcelsior CHARGE (“Xcelsior”) Battery capacity 660 kWh 160 kWh Charger power 125 kW 450 kW Time for 85% charge 4.5 hours 18 minutes  Using these models we will evaluate the various tradeoffs in the context of the Sunshine Coast transit system. We will expand on scheduling constraints and costs in the following sections. 10    Figure 3. ​Proterra​ ​Catalyst E2 max with Duopower Drivetrain ​and 125 kW plug-in charger (Proterra, 2019c; Proterra, 2019b)  Figure 4. ​New Flyer Xcelsior CHARGE​ bus with ABB overhead pantograph charger (Sustainable Bus, 2019) 1) Infrastructure requirements Both charging methods require physical space at the bus depot or terminal to accommodate the necessary charging infrastructure. Plug-in charging will require more chargers since the buses must charge simultaneously overnight, whereas overhead charging can be completed much more quickly. Overhead chargers also have smaller surface space requirements and thus require less space than plug-in chargers (Jahic, Eskander, & Schulz, 2019). On the other hand, overhead charging uses a much higher power than plug-in charging and thus may require upgrades to the electric grid to accommodate (Jahic, Eskander, & Schulz, 2019).  For the Sunshine Coast transit system, we expect overnight charging to occur at the SCRD work yard, where buses are stationed during non-service hours. Fast charging would occur at both the SCRD work yard and at the Langdale ferry terminal which is the terminus of the main route. Space is limited in the SCRD work yard due to it being a shared space with SCRD fleets and in the ferry terminal due to parking. As for the grid, there is a major powerline at the SCRD work yard which could accommodate fast charging.  11   2) Terrain suitability Because buses with overnight charging require larger batteries than buses with fast charging, the buses have different terrain capabilities. The larger batteries required for overnight charging increase the overall bus weight, thus impeding speed, reducing ability to climb hills and decreasing suitability for locations with less stable ground (Carrilero et al., 2018).  The fast charging ​Xcelsior ​has a curb weight of 13,041 kg (New Flyer, 2019) whereas the overnight charging ​Catalyst ​is heavier with a curb weight of 15,036 kg (Proterra, 2019a). Unfortunately, data on speeds and hill gradeability were unavailable for the ​Xcelsior​ model. However, the ​Catalyst​ bus can traverse hills of up to 24% grade, which should be suitable for the Sunshine Coast given maximum hill grades ranging from 17 to 21% (Proterra, 2019a; S. Sears, personal communication, January 25, 2020). Given that the Catalyst bus is heavier, the ​Xcelsior bus should be suitable as well.  Weight-related terrain considerations are critical to the Sunshine Coast transit system as the transit routes have many steep hills, and some roads are built over peat moss, which is sensitive to weight (S. Sears, personal communication, January 25, 2020). There are also long distances between bus stops, and thus higher speeds are desirable. For these reasons, the lighter weight option may be preferred. 3) Scheduling constraints Scheduling is a key challenge in electrifying bus fleets, regardless of the charging technology. Both technologies require daily charging in order to maintain regular service. Buses using fast charging can charge within minutes due to high charging power but their operational range is limited. Whereas buses using plug-in charging can travel further distances but require several hours to charge. As such, schedules and routes must be adjusted to accommodate sufficient time for buses to achieve a full charge.  According to the Sunshine Coast’s October transit schedules (SCRD, 2019), the transit buses travel between 168-531 km per day with transit routes ranging from 8-70 km (​BC Transit and Sunshine Coast Regional District, 2014)​. The ​Xcelsior​ bus can only travel 120 km on one charge (New Flyer, 2019), while the ​Catalyst​ bus can travel between 373-528 km per charge depending on driver behaviour, weather and terrain (Proterra, 2019a).   Under an overnight charging scenario, schedules should be adjusted to ensure ​Catalyst​ buses can complete daily travel times on a single charge. The fast charging ​Xcelsior​ bus would need to be charged multiple times per day, but can complete at least a single one way trip with a full charge. Thus, fast charging at both terminus would be suitable.  Based on an 85% battery discharge per charging cycle (M. Fisher, personal communication, February 12, 2020), the ​Catalyst​ bus takes 4.5 hours to charge its 660 kWh battery with the 125 kW plug-in charger (Proterra, 2019a), whereas the ​Xcelsior​ bus takes 18 minutes to charge its 12   160 kWh battery using the 450 kW overhead charger (New Flyer, 2019). Batteries and chargers are assumed to operate at full efficiency.  For the overnight charging scenario, schedules will need to be adjusted to ensure sufficient time for all buses to charge during non-service hours, while minimizing demand charges from simultaneous charging. For the fast charging scenario, schedules will need to be adjusted to accommodate time for charging in between trips and to distribute charging times among buses. 4) Costs Cost trade-offs between these technologies are complex and highly dependent on the specific operational arrangement. While the buses themselves have similar costs for both charging methods, the costs of batteries, charging infrastructure and charging installation vary. Overnight charging requires higher battery storage and thus more expensive batteries, while fast charging requires higher power and thus more expensive chargers (Xylia and Silveira, 2018). The lifetime of the batteries and number of chargers required for each scenario may also impact these capital costs. Fast charging may require more frequent battery replacements due to more frequent charges whereas overnight charging may require a higher number of chargers to accommodate simultaneous charging overnight. As for operational costs, the basic cost of electricity will be identical, assuming the buses travel the same distances in each scenario. However, fast charging will involve additional demand charges to accommodate the higher charging power. Summary: Electric Bus Operations The main advantages of overnight charging are limited grid capacity needed, a higher operational range, lower charging infrastructure costs and lower demand charges while the main advantages of fast charging are reduced charging space, better terrain suitability, faster charging time and lower battery costs. The following two sections will further explore these tradeoffs to inform the feasibility for the Sunshine Coast.  Optimization of Bus Charging Schedules Methodology  As previously mentioned, overnight charging using the Catalyst model (Proterra, 2019) may be more operationally feasible for regions with smaller demand and numbers of buses. However, due to the time required to charge a bus, charging schedules need to be optimized to fit within bus schedules with minimized demand charges. With overnight charging where buses will be charged at a depot, each bus should already have a defined arrival and departure time. The time in between can thus be used for charging (Figure 5).  13    Figure 5. Example of a charging optimization problem (Jahic, Eskander, & Schulz, 2019). The white blocks represent buses charging at a depot. Bus #27 (light green block) represents a bus that has recently arrived and needs to be charged before departure. In order to reduce the number of buses charging simultaneously, the bus should ideally charge after bus #1 has finished charging.  The objective is to optimize schedules in order to minimize operational costs and maximize operational efficiency.​ An algorithm will be developed using a statistical programming language, R, to minimize the number of buses charging simultaneously given any number of buses. The algorithm will also help calculate the demand charge for the entire charging process. The goal is to minimize demand charge. The smallest demand charge that the algorithm is able to arrange is considered an optimal schedule in terms of reducing energy consumption and electrical grid stress.   It is important to note that the algorithm is made as a reference tool to allow future users to quickly obtain an approximate bus charging schedule and to calculate the demand charge throughout the entire charging period. It is not to be used as a final charging solution as there may be circumstantial variables such as ferry delays, special events, or mechanical and electrical issues.The mathematical setup and reasoning behind the algorithm as well as the R code can be found in Appendix I. A flowchart of the logistical reasoning behind the algorithm is shown in Figure 6.         14     Figure 6. Flowchart of the logical reasoning used in the R code to minimize the number of buses charging simultaneously and reduce demand charge.   15   Results and Discussion The SCRD currently has 14 buses and uses 9 buses regularly (S. Sears, personal communication, January 25, 2020). Therefore, the results of an optimized charging schedule and its corresponding demand charge is based on the assumption that there are 9 buses that need charging from 11pm to 8am the following day such that they can be in operation during service hours. These buses have a charging length of 4.5 hours and a charge requirement of 125 kW based on manufacturer data of the Catalyst E2 Max model from Proterra (Proterra, 2019a).   Based on these parameters, we used the algorithm to produce a final optimized charging schedule for the 10 buses. All buses were fully charged by 8:00am using 5 chargers and a maximum demand charge of 625 kW. Buses #1, #3, #5, #7, and #9 were scheduled to charge simultaneously from 11:00pm to 3:30am; Buses #2, #4, #6, and #8 were scheduled for 3:30am to 8:00am (Table 3; Figure 7). Table 3. Each bus has a charging interval with a start and end time indicated and is assigned a charging depot to be stationed at. The number of required depots and demand charge is also displayed.    16    Figure 7. A visualization of the charging schedule in Table 7 as a gantt chart showing the bus charging schedule. For example, Buses #1, #3, #5, #7, and #9 are charging from Jan 20, 11:00pm to Jan 21, 8:00am.  The demand charge calculated from the algorithm ranged from 500 kW to 625 kW during the charging period. These numbers are based on the ​sum of all the power demands of each bus charging at a certain time. The demand charge is 625 kW from 11:00pm to 3:30am since Buses 1, 3, 5, 7 and 9 are charging simultaneously. This is also the time period with the most number of buses charging simultaneously, thus, 625 kW is the max demand charge (Figure 8). After 3:30am, the demand charge drops to 500 kW and remains constant until 8:00am where all the buses have been fully charged. This is due to there being a maximum number of 4 buses charging concurrently during that time period.  17    Figure 8. Times series graph of total demand charge throughout the implemented charging schedule from Tables 3 and Figure 6. The total demand charge is 625 kW on Jan 20, 10pm and drops to 500 kW on Jan 21, 3:30am. It remains at 500 kW for the rest of the charging duration. Summary: Optimization of Bus Charging Schedules Based on a conservative assumption of 9 buses charging from 11:00pm to 8:00am, each with a 125 kW power demand and charging length of 4.5 hours, the demand charge for a fully electric fleet should be 625 kW. Additionally, the charging schedule can be optimized by having 5 buses charging from 11:00pm to 3:30am and ​4 buses charging from 3:30am 8:00am. These conclusions are derived from the calculations and schedules produced by the R program, data from the bus manufacturer, Prot​erra (Table 4),​ and operation times of the current Sunshine Coast schedules.  Life Cycle Cost Analysis Methodology  The life-cycle cost analysis is a useful tool for comparing costs of projects with varying capital and operational costs, as it incorporates all costs incurred over the lifetime of a project. The main components of lifecycle cost are capital costs, operation costs, and technology replacement costs (Figure 9). A sum of these three factors over a project’s lifetime produces the total lifecycle cost (Lajunen, 2018). In this study, we calculate and compare the total life-cycle costs of the ​Nova LFS Diesel​, overnight charging ​Proterra Catalyst​ and fast charging ​New Flyer Xcelsior​ buses (Table 4).   18   Table 4. Bus specifications for diesel, plug-in and overhead charging (Proterra, 2019; New Flyer, 2019 and G. Huber, personal communication, January 21, 2020).  Diesel bus (current) Overnight charging with plug-in charger Fast charging with overhead charger Manufacturer Novabus Proterra New Flyer Model LFS diesel Catalyst E2 max with Duopower Drivetrain (“Catalyst”) Xcelsior CHARGE (“Xcelsior”) Battery capacity N/A 660 kWh 160 kWh Charger power N/A 125 kW 450 kW Time for full refueling 5 mins 4.5 hours 18 minutes Bus lifetime (years)  14 14 14 Battery lifetime (years) N/A 7 7 Energy consumption 0.451 L/km 1.3 kWh/km 1.3 kWh/km Number of chargers  N/A 5 2  We assume that all 14 buses in the Sunshine Coast fleet are replaced with the new technologies. We calculate the present worth of these costs, which reflects the current valuation of costs that will be spent over time. End of life disposal costs and product financing costs were not considered.          Figure 9. Cost components of total lifecycle cost (Lajunen, 2018).  19   Capital Costs  For diesel buses, capital costs only consist of the cost of the bus whereas capital costs for electric buses also include the cost of the charging infrastructure. Overnight charging requires five chargers to accommodate charging of all vehicles overnight (as per the results from the previous section), while fast charging requires two chargers, one at the SCRD work yard and one at the Langdale ferry terminal. The cost of charging infrastructure installation was included within the cost of the charger. Costs of other infrastructure upgrades such as new maintenance facilities and increased grid capacity were not considered.   Capital costs were calculated by summing the costs of buses, multiplied by the number of buses, and chargers, multiplied by the number of chargers. The bus cost includes the cost of the initial battery. Bus and charger costs were determined by converting values provided by manufacturers from USD to CAD and adding a 10% contingency to account for additional taxes, shipping, insurance costs and other amenities not included in the base price (B. Quite, personal communication, 13 February, 2020).   Operational Costs  Diesel and electricity costs were calculated by multiplying the bus energy consumption per km by the energy cost (electricity or diesel) and the total number of kilometres travelled per year (Table 5).   Demand charges were calculated by multiplying the charger power by the number of chargers and the BC Hydro demand charge rate (Table 4, Table 5). Table 5. Bus route information and energy prices for the Sunshine Coast. Parameter Value  Reference Number of buses 14 S. Sears, personal communication, January 25, 2020 Number of buses in regular service 9 S. Sears, personal communication, January 25, 2020 Total bus distance travelled per year by all buses 1,000,000 km C. Venderford, personal communication, January 21, 2020 Diesel cost $1.432 per L Statistics Canada, 2020 Electricity cost $0.060 per kWh BC Hydro, 2020 20   Demand charge per two month billing cycle $12.22 per kW BC Hydro, 2020 USD to CAD exchange rate 134% March 7, 2020  The annual maintenance cost for the Sunshine Coast diesel fleet was provided by BC Transit (C. Vanderforth, personal communication, January 21, 2020), and the electric bus maintenance cost was assumed to be half of the maintenance cost for diesel buses as per other studies (National Academy of Sciences, Engineering and Medicine, 2018).   Labour costs were assumed to be equivalent and were thus excluded from the analysis.  The present worth of the sum of annual costs is:  i(i+1)n(1+i) −1n  where ​i​ is the discount rate and ​n​ is the product lifetime (Sheth and Sakar, 2019). A discount rate of 1% was applied, and the product lifetime was set to 14 years (Tong et al., 2017).  Technology Replacement Costs  The battery replacement cost for the electric buses was calculated by converting prices provided by manufacturers from USD to CAD and adding a 10% cost adjustment to account for any additional taxes, shipping or insurance costs.   The present worth of the battery replacement costs is calculated by:  1(1+i)n   where ​i​ is the discount rate and ​n​ is the replacement lifetime. A discount rate of 1% was again applied with the battery lifetime being 7 years (Tong et al., 2017).   The present worth of capital, operations and technology replacement costs were then summed to produce the total lifecycle costs for each transportation option.  Cumulative costs were calculated by subsequently adding the present worth of annual costs. Technology replacement costs were added after the seventh year to account for the battery replacement.      21   Results  The life-cycle cost analysis shows that fast charging has the lowest life-cycle costs and overnight charging has the highest life-cycle costs, with diesel in between (Table 6, Figure 10). Overnight charging has the highest capital costs, while diesel has the highest operational costs (Figure 10). Table 6. Present worth of capital, technology replacement, operational and total costs for Novabus diesel, Proterra 6 hour plug-in and New Flyer 10 minute overhead over a 14 year lifecycle (M. Fisher, personal communication, February 12, 2020; M. Bismeyer, personal communication February 5, 2020; Lowell, 2019). All costs are in CAD. Cost Component Diesel Overnight charging with plug-in charger Fast charging with overhead charger Capital costs Bus cost $608,360 $1,104,026 $913,612 Charger cost N/A $98,758 $737,000 Total capital cost $8,517,040 $15,950,154 $14,264,568 Technology replacement cost Battery cost N/A $262,640 $89,536 Present worth of battery replacement cost N/A $3,429,567 $1,169,171 Operational costs Fuel cost per year (diesel or electricity) $645,832 $78,000 $78,000 Demand charge per year N/A $45,825 $65,988 Maintenance cost per year $330,000 $165,000 $165,000 Total operations cost per year $975,832 $288,825 $308,988 Present worth of total operations costs $12,689,430 $3,755,795 $4,017,988 Total cost $21,206,469 $23,135,516 $19,451,727   22    Figure 10. Cost breakdown over 14 year life-cycle of Nova diesel bus, Proterra 6 hour plug-in and New Flyer 10 minute overhead charging buses. Evaluating cumulative costs shows that the costs of fast charging become cheaper than diesel after eleven years (Table 7, Figure 11). The cost difference between diesel and electric buses is shown to decrease over bus lifetime (Figure 11). Table 7. Cumulative annual costs of Nova diesel bus, Proterra overnight plug-in and New Flyer fast overhead charging buses. Costs increase in the sixth year to replace the battery. Costs of fast charging become cheaper than diesel after year 11.  Year Diesel Overnight plug-in charging Fast overhead charging 0 $8,517,040 $15,950,154 $14,264,568 1 $9,483,210 $16,236,119 $14,570,497 2 $10,439,815 $16,519,253 $14,873,396 3 $11,386,947 $16,799,584 $15,173,297 4 $12,324,703 $17,077,139 $15,470,229 5 $13,253,174 $17,351,946 $15,764,220 23   6 $14,172,451 $17,624,032 $16,055,301 7 $15,082,628 $21,322,992 $17,512,670 8 $15,983,792 $21,589,717 $17,798,015 9 $16,876,034 $21,853,801 $18,080,535 10 $17,759,442 $22,115,271 $18,360,258 11 $18,634,103 $22,374,151 $18,637,211 12 $19,500,104 $22,630,469 $18,911,422 13 $20,357,531 $22,884,248 $19,182,919 14 $21,206,469 $23,135,516 $19,451,727 Figure 11. Cumulative annual costs of Nova diesel bus, Proterra overnight plug-in and New Flyer fast overhead charging buses. Costs increase in the sixth year to replace the battery. Costs of fast charging become cheaper than diesel after year 11.  Discussion While the diesel bus has the lowest capital costs due to cheaper bus costs, the electric buses have lower operational costs due to lower cost of electricity compared to diesel and reduced 24   maintenance costs (Figure 10). Compared to the diesel bus scenario, capital costs are 87% and 67% higher for the overnight charging and fast charging scenarios respectively (Table 6). Operational costs, on the other hand, are 238% and 216% higher for the diesel bus than for the overnight charging and fast charging scenarios respectively (Table 6).  Buses that use overnight charging have 10% higher capital costs and 193% higher battery replacement costs than buses that use fast charging due to larger and more expensive batteries (Table 6). Fast charging buses have 7% higher operational costs than overnight charging buses due to higher demand charges (Table 6).   While electric buses have higher upfront costs than diesel buses, these can be offset over time due to lower operational costs. The more years buses are in use, the closer the cumulative costs between electric buses and diesel buses become. The fast charging technology becomes cheaper than the diesel technology after 11 years, thus leading to it being the cheapest option over the whole 14 year lifecycle (Figure 11). While total life-cycle costs of overnight charging buses are higher than diesel buses, the cost difference between the technologies decreases from 87% to 9% over the bus lifetime (Table 7). Summary: Life Cycle Cost Analysis From a financial standpoint, fast charging is recommended for the Sunshine Coast transit system as it has the cheapest life-cycle costs compared with diesel buses and overnight charging. Initial upfront costs to implement fast charging for the Sunshine Coast transit fleet are $14,264,568, with life-cycle costs being $19,451,727. This is 8% and 16% cheaper than the life-cycle costs for diesel and overnight charging options respectively. Ridership Strategies In the ​Transit Future Plan​, BC Transit and the SCRD (2014) have set a transit mode share target of 5.4% for all trips by 2038. To meet or exceed this goal, transit ridership must continue to grow annually. According to the ​Sunshine Coast Integrated Transportation Study​ (ISL Engineering and Land Services, 2011), 70% of the transit ridership are internal trips within the region while 30% are ferry related ridership. When considering all travel modes, approximately 2% of all trips are taken by public transit on the Sunshine Coast (BC Transit & SCRD, 2014). Since 2016, conventional (fixed-route) transit ridership has risen steadily (Figure 12) as riders commit to using transit through the purchase of monthly passes (ISC, 2019a).  25    Figure 12. Transit ridership by month on the Sunshine Coast from 2016 - 2019, obtained from the Infrastructure Services Committee meeting agenda (ISC, 2019a). Most updated data provided by BC Transit.  Attracting and retaining riders is a primary challenge faced by many rural transit systems (Whitaker & Derk, 2018). Like other rural communities, the Sunshine Coast is characterized by long distances between communities, low population densities, infrequent bus service, and limited route access. This has led to high personal vehicle usage, making it a challenge to increase and retain transit ridership. There are a variety of reasons for increasing ridership including improving performance and reliability, more frequent service, flexibility of work hours, road congestion mitigation and environmental benefits (Lewis, 2012).  Furthermore, sustainable long-term revenue sources serve as a basis of funding and are fundamental to the ability of transit providers to deliver and improve transit systems such as the transition to zero-emission electric buses. In order to maintain and increase government support, continuing public demand for transit services in the form of ridership must be demonstrated (Auditor General of British Columbia, 2013). The main purpose of this component of the project is to provide a comprehensive literature review of potential strategies to attract and maintain transit ridership on the Sunshine Coast.  Literature Review: Ridership Improvement Strategies Service Expansion  Transit service in the region is one of the most important factors in attracting new riders (ISC, 2019b). This strategy includes increasing service frequencies as well as additional routing to new neighbourhoods and areas to increase the value proposition to riders. The elasticity of transit ridership per capita with respect to transit frequency and route density is 1.175 and 0.947 26   respectively (Lyons et al., 2017). An elasticity value describes how the change in one variable can be used to predict the change in another. This means that a 10% increase in transit frequency could be expected to produce a 11.75% increase in transit ridership.  According to the ​Transit Future Plan Sunshine Coast​ (2014), public feedback from locals of the Sunshine Coast suggested that increased frequency on all routes, particularly service between Sechelt, Gibsons and Langdale is in great demand. SCRD’s current proposed sequence of upgrades is to first increase service frequency between Sechelt, Gibsons and Langdale Ferry Terminal to 30 minute intervals at peak times and eventually all-day service. Secondly, increase service frequency to West Sechelt to 30 minute intervals at peak times and eventually all-day service (BC Transit & SCRD, 2014).  Another major theme in public feedback is new and improved service coverage in areas including Elphinstone, Sandy Hook, Tuwanek, and Pender Harbor (BC Transit, 2014). A survey aimed at identifying gaps in the Sunshine Coast’s transit system was conducted by the Voice Lab, a group interested in innovative models for alternative transportation, with nearly 200 respondents. The results demonstrated that 2 of the most popular solutions were to increase transit frequency by 21% and a combination of improved or new services including taxis, shuttles, increased transit services, and ride-hailing by 34% (Coast Reporter, 2019).  Park-and-Ride Facilities Park and ride facilities offer commuters the option to park their private vehicles at designated parking locations to conveniently transition to public transit. This allows travellers to use public transportation as a portion of their overall trip, reducing traffic congestion, vehicle emissions and potential travel costs as gas prices increase. Due to the Sunshine Coast’s widely distributed network of communities along the corridor and therefore low population densities, the establishment of park and ride facilities along the Highway 101 corridor can help concentrate potential transit service users (ISL Engineering and Land Services, 2011). BC Transit and the SCRD (2014) have identified 2 potential locations including the Raven’s Cry Theatre in Sechelt and Upper Gibsons Transit Exchange in Gibsons. These facilities have potential to attract riders from a wider area such as communities north of Sechelt and west of Halfmoon Bay and create awareness of public transit as a viable alternative to private vehicle use (ISL Engineering and Land Services, 2011). In Victoria, established facilities have been proven to support a large portion of ridership and will likely continue to maximize transit ridership and availability (VTPI, 2019).  DayPass-on-Board Program Revision Another possible strategy is to adjust the DayPASS program. In the Sunshine Coast, the DayPASS offers riders the ability to request a transfer for a one-way travel. It is currently sold as a prepaid fare product for $5 at designated ticket vendors and is not valid for return trips. As an improvement recommended by BC Transit, the DayPASS program should be updated by providing unlimited travel for one transit service day and offering on-board purchase at two times the base fare, or $4 (ISC, 2019b). By improving convenience for passengers, primarily tourists, this strategy would substantially increase ridership. This program has been demonstrated 27   to be successful in the transit systems of Nanaimo, Victoria, Squamish, Prince George and Kamloops where the day pass has resulted in increased casual ridership, increased monthly pass product purchases, and increased revenue as a result of eliminating opportunities for riders to fraudulently use expired transfers (ISC, 2019b).  Pricing Strategies Pricing strategies are incorporated as part of broader marketing initiatives that offer reduced fares and price concessions for targeted user groups with the goal of encouraging the use of transit services. 1)​      ​Fare Reduction of Monthly Passes for Youth/Students and Seniors Approximately 42% of the Sunshine Coast’s population consists of youth (18 and under) and seniors (65+), forming the region’s key transit markets (Statistics Canada, 2017; BC Transit, 2014). With an aging population, it is important for the district to shift their focus towards attracting new long-term riders, students or youth. Based on the assessment completed by BC Transit, a fare reduction of monthly passes for youth and seniors is expected to decrease the expected revenue increase, offset by an anticipated increase in ridership (ISC, 2019b).  2)​      ​Complimentary Youth Transit There is growing interest in exploring complimentary transit for youth on the Sunshine Coast. A member-based group that advocates for sustainable transportation on the Sunshine Coast, Transportation Choices – Sunshine Coast (TraC)​, ​has recently started a petition addressed to the SCRD for free public transit for students on the basis that it will reduce CO​2​ emissions and make transit equally accessible regardless of income (TraC, 2020). The petition has gathered 352 signatures as of March 9, 2020.  Student Transit Education Program To further increase transit awareness among students (grades K - 12), student transit education programs can help to change the perception of transit long-term. In a transit promotion and orientation project initiated by the City of Kingston and its local school board, transit staff identified several barriers that were preventing students from using public transit such as not knowing how to use the transit system, not knowing where ticket vendors are, and having no experience riding the bus (FCM, 2019). Part of the solution was to provide transit orientation sessions which teach students practical skills for riding the bus along with explaining the associated environmental and economic benefits (FCM, 2019). This program enables young riders to gain confidence in using public transit and increases their likelihood of becoming regular paying and committed passengers after completing schooling. Bus Stop Amenities Bus stops are key access points that link riders to public transit service. According to BC Transit’s Transit Shelter Program Report (2018), poor quality shelters, inadequate lighting and 28   other poorly designed infrastructure have often been cited as a barrier to transit use. To attract new transit riders, recent market analysis has found that shelters and improvements at bus stops were one of the top enhancements needed (BC Transit, 2018). In 2014, it was reported that only 20% of all bus stops on the Sunshine Coast had transit shelters (Table 8) which are concentrated on bus routes 1 and 2 (BC Transit & SCRD, 2014). Based on in person observations, many bus stops on the Sunshine Coast also serve as a stop sign and those that do not have seating, often have lawn chairs next to them (Figures 13 and 14).  Table 8. Summary of bus stop conditions on the Sunshine Coast (Adapted from BC Transit & SCRD, 2014). Stops are concentrated on routes 1 & 90.           Figure 13. A typical bus stop on the Sunshine Coast with a lawn chair.     Figure 14. A bus stop at the intersection of Lower Road and Agnes Road, Roberts Creek on the Sunshine Coast. The bus stop also serves as a stop sign. 29   According to a University of Minnesota Study (Fan et al., 2016), amenities such as shelters, benches, lighting and information on schedules at bus stops make the wait a more agreeable experience and help riders feel safer (TransitCenter, 2018). The study found that on average, riders at stops without amenities perceived their wait time to be more than double, where 10 minutes felt like 21 minutes. Stops with amenities reduced the perceived wait time for the same wait to 13 minutes. Other research indicates that by improving public transit waiting conditions, transit ridership increases as riders are particularly sensitive to waiting time (VTPI, 2019).  Results and Discussion  Based on the literature review of 7 potential transit strategies that may be appropriate for the Sunshine Coast, strategies were evaluated and prioritized based on expected cost, expected increase in ridership and public interest. Strategies with reasonable costs, medium to high expected increase in ridership and high public interests are recommended as higher priorities for the SCRD transit system to increase bus ridership (Table 9).  Table 9. Potential strategies to increase transit ridership evaluated based on cost, expected impact on ridership, and public interest. Refer to Appendix II, Table 10. for quantitative costs and ridership impacts.                30    Service expansion is expected to involve extensive funding beyond what is available through provincial funding, local property tax, passenger fares and advertising revenue (BC Transit, 2014). This would require additional or alternative funding sources to support necessary investment in infrastructure, operations and maintenance such as increased number of buses and associated operator costs, bus capacity upgrades, and greater maintenance expenses (ISC, 2019b). Despite high investment costs, service expansion is deemed a high priority (Table 9) as greater service frequency and coverage has been a common theme in public feedback. Thus, it is expected that service expansion would result in a high increase in ridership and revenue but at a higher capital cost.  Costs of Park and Ride facilities depend heavily on the availability of existing parking lots, which can potentially reduce capital investment costs, versus new land which is associated with much higher capital investment costs (Gris Orange Consultant, 2012). Considering only surface parking as it is significantly cheaper than parking garages, the cost ranges from $2,760 to $13,800 per stall (City of Edmonton, 2018; CTG, 2014). With limited parking availability, transit ridership targeting private vehicle users is expected to increase moderately (Table 9) but to verify this, a survey is recommended to evaluate the degree of public interest. It is recommended that Park and Ride facilities are provided on an eligibility basis to focus on diverting private vehicles and prevent those who work in the area from taking advantage of the program. An improved DayPASS on Board program is also evaluated as a high priority (Table 9) as a result of low costs and high anticipated annual increase in ridership. The estimated annual impact on revenue is an increase of $39,000 (5%) and an increase of 24,000 (5%) in ridership, primarily tourists (ISC, 2019b). Therefore, the DayPASS program is a viable strategy that can increase transit ridership.  Fare reduction for students and seniors is considered a high priority (Table 9) as a result of the costs offset by an anticipated increase in annual ridership of 3,000 (or 1%) (ISC, 201). It is also a common theme in public feedback as voiced through signing of petitions and letters addressed to the SCRD. It is also important to note that anticipated increase in ridership will not completely offset the entire amount of lost revenue from the reduction of student and senior monthly fares (ISC, 2019b).  A more ambitious pricing strategy is complimentary student transit which is considered a medium priority (Table 9) due to the significant financial implications. An assessment conducted by BC Transit projects that the implementation of complimentary youth transit will result in an annual impact of $175,000 (23%) reduction in revenue and an increase of $173,000 to $398,000 (5 to 11%) in expenses (ISC, 2019b). The projected impact on annual ridership is an increase of 33,000 to 52,000 (7 to 10%) (ISC, 2019b). Further analysis of alternative approaches to achieve the desired increase in youth ridership is recommended. It is also important to consider that the use of fare free transit does not guarantee continued increases in ridership. A student transit education program is expected to involve little administrative and operational costs and an unknown impact on annual transit ridership among youth. The student transit 31   education program can take form as an informative and entertaining presentation and tour provided at no cost to local schools from Kindergarten through Grade 12. This initiative may be more effective in increasing transit awareness and ridership among students when implemented in combination with fare reductions or complimentary youth transit passes. With a low investment cost, a student transit education program has the potential to attract new youth ridership and increase their potential to continue as committed passengers after the completion of secondary school. Thus, it is evaluated as a medium priority (Table 9).  Evaluated as a medium priority (Table 9), bus stop amenities present variable costs and expected increases in ridership. Based on our observations on the Sunshine Coast, it appears that many locals put out lawn chairs next to bus stops that have no seating available. This indicates a high interest in bus stop amenities. Depending on bus stop characteristics and levels of daily passenger boarding, rural shelter product models can range from $13,613 to $25,324 (Figure 11, Appendix III) while solar lighting kits cost $4,640 each at full costs (BC Transit, 2018). Potential issues that may arise as a result of installing bus stop shelters include the need for regular maintenance to keep facilities clean and functional (Gris Orange Consultant, 2012).  Summary: Ridership Strategies Based on the literature review considering the expected cost, expected ridership and public opinion, several ridership improvement strategies are recommended to be prioritized for further investigation and analysis. These strategies include service expansion, park-and-ride facilities, fare reduction of student and senior monthly passes, and the revision of the DayPASS-on-Board program.  Conclusion, Limitations and Further Studies Conclusion The Sunshine Coast Regional District currently runs 9 buses daily. Based on this number, if electric buses with overnight charging were to be used, the maximum demand charge for a fully electric fleet should be 625kW. This demand charge is calculated from buses that have a 125 kW power demand and charging length of 4.5 hours (Catalyst E2 Model from Proterra). Additionally, the charging schedule can be optimized by having 5 buses charging from 11:00pm to 3:30am; and ​4 buses charging from 3:30am 8:00am.   Therefore, overnight charging would work for a small number of buses, but bus scheduling would not be feasible if SCRD were to increase fleet size. From the life-cycle cost analysis, fast charging also has reduced costs due to smaller batteries and a smaller number of required chargers. Therefore, fast charging is recommended as it has the cheapest overall costs. Regardless, conversion to a fully electric fleet would require financing to cover upfront costs. BC Transit will need $13,822,368 to finance capital costs versus $8,517,040 for diesel. To the aid transition to a fully electric fleet, ridership increases could help pay off the high initial investments. Possible ridership increase strategies include service expansion, park-and-ride 32   facilities, fare reduction of student and senior monthly passes, and the revision of the DayPASS-on-Board program. Further feasibility analysis, public surveys, and ridership projections should be conducted on each strategy in order to better inform land use and transportation developments with an overarching goal of improving transit ridership. Limitations Due to the short time period of the study, we were unable to assess multiple bus manufacturer options or optimize our data to fit the Sunshine Coast routes and driving styles. We were also limited in evaluating the accurate energy efficiency of electric buses, which was given as a range by manufacturers as it depends on climate, driving style and bus weight. Our cost estimates are fairly rough as some costs were provided as ranges due to company confidentiality, and the ultimate purchase costs will be determined through negotiations between BC Transit and the bus manufacturers. Rapidly changing technologies and pricing may also affect the usefulness of this study. Costs are quickly changing and compatibility of current technologies with future technologies may be limited.  As for transit ridership, this research is limited in that we were unable to find projections of ridership and more detailed information for transit ridership in the Sunshine Coast. In addition, the evaluation of ridership improvement strategies is somewhat subjective, which limits accuracy. Further Studies We limited our schedule optimization to only depot charging and did not assess the impact that fast charging would have on the schedules. A future study could generate new transit schedules for each season that are optimized to maximize ridership and minimize power demand for the number of buses required. Furthermore, a more in-depth schedule optimization algorithm could be produced that would take into account individual scheduling times for each bus, rather than assuming the same arrival and departure times for all buses. A future study could also gather data from more bus manufacturers to determine the cheapest options. As a final step, an implementation plan for transit electrification on the Sunshine Coast including a timeline and financing strategy should be developed.   Some additional factors that could be included in future studies include environmental impacts (such as greenhouse gas emissions, land use, waste and pollution from technology production) and the impact on the local economy. Future studies could also assess and compare the viability of other technologies including battery swapping, on-route inductive charging technology, hybrid buses and hydrogen fuel cell technology.     33   Acknowledgements  We would like to express our gratitude to the following people who have provided our team with guidance, insight and expertise that greatly assisted our research: Elizabeth Lytviak, our community partner and Director of the 2 Degrees Institute, for proposing this research opportunity; Steven Sears, Transportation Superintendent at Sunshine Coast Regional District, for providing information regarding the transit system and for his hospitality during our visit to the Sunshine Coast; Geoff Huber, Environmental Services Supervisor at BC Transit, for providing data used in the cost-benefit analysis; Mike Bismeyer, Regional Sales Director at Proterra and Mark Fisher, Director of National Sales at Newflyer, for providing electric bus specifications and costs; Dr. Tara Ivanochko and Dr. Michael Lipsen, instructors of ENVR 400 at UBC, for providing valuable feedback and guidance throughout our progress.                        34   References BC Hydro. “General Service Business Rates.” Retrieved from BC Transit. (2018).​ Transit Shelter Program.​ Retrieved from BC Transit. (2019a). ​BC Transit’s Low Carbon Fleet Program supports provincial targets to reduce greenhouse gas emissions​. Retrieved from  BC Transit. (2019b). ​Seasonal service changes on the Sunshine Coast.​ Retrieved from​.  BC Transit. (n.d.). ​Schedules and Maps.​ Retrieved from  BC Transit and Sunshine Coast Regional District. (2014). ​Transit Future Plan​. Retrieved from​.  Carrilero, I., González, M., Anseán, D., Viera, J. C., Chacón, J., & Pereirinha, P. G. (2018). Redesigning European Public Transport: Impact of New Battery Technologies in the Design of Electric Bus Fleets. ​Transportation Research Procedia​, ​33​, 195–202. doi: 10.1016/j.trpro.2018.10.092 Chen, Z., Yin, Y., & Song, Z. (2018). A cost-competitiveness analysis of charging infrastructure for electric bus operations. ​Transportation Research Part C, 93​, 351-366. doi:10.1016/j.trc.2018.06.006 City of Edmonton. (2018). ​Park & Ride Guidelines​. Retrieved from​. Coast Reporter. (2019). ​Online survey shows demand for ride-sharing, increased bus service. Retrieved from CPTDB. (2019). ​Sunshine Coast Regional Transit System​. Retrieved from  35   CTG. (2014). ​Technical Memorandum #5 Financial Analysis and Management Options​. Retrieved from  Damart, S., & Roy, B. (2009). The uses of cost-benefit analysis in public transportation decision-making in France. ​Transport Policy, 16​(4), 200-212. doi:10.1016/j.tranpol.2009.06.002 Fan, Y., Guthrie, A., & Levinson, D. (2016). Waiting time perceptions at transit stops and stations: Effects of basic amenities, gender, and security. ​Transportation Research Part A: Policy and Practice​, ​88​, 251–264. doi: 10.1016/j.tra.2016.04.012 Federation of Canadian Municipalities. (2019).​ Engaging Students to Increase Public Transit Ridership​.  Retrieved from  Gao, Z., Lin, Z., Laclair, T. J., Liu, C., Li, J.-M., Birky, A. K., & Ward, J. (2017). Battery capacity and recharging needs for electric buses in city transit service.​ Energy, 122​, 588–600. doi:10.1016/  Google. (n.d.). [Sunshine Coast, BC]. Retrieved from  Gris Orange Consultant. Improving bus service: modest investments to increase transit ridership, Improving bus service: modest investments to increase transit ridership (2012). Ottawa: Minister of Transport.  Häll, C. H., Ceder, A., Ekström, J., & Quttineh, N. (2019). Adjustments of public transit operations planning process for the use of electric buses.​ Journal of Intelligent Transportation Systems, 23​(3), 216-230. doi:10.1080/15472450.2018.1488131 IPCC, 2018: Global warming of 1.5°C. An IPCC Special Report on the impacts of global warming of 1.5°C above pre-industrial levels and related global greenhouse gas emission pathways, in the context of strengthening the global response to the threat of climate change, sustainable development, and efforts to eradicate poverty [V. Masson-Delmotte, P. Zhai, H. O. Pörtner, D. Roberts, J. Skea, P.R. Shukla, A. Pirani, W. Moufouma-Okia, C. Péan, R. Pidcock, S. Connors, J. B. R. Matthews, Y. Chen, X. Zhou, M. I. Gomis, E. Lonnoy, T. Maycock, M. Tignor, T. Waterfield (eds.)]. In Press. 36   Infrastructure Services Committee. (2019a). ISC Agenda, Sunshine Coast Regional District, 3 Oct 2019, SCRD Boardroom, Sechelt, BC. Meeting.  Infrastructure Services Committee. (2019b). ISC Agenda, Sunshine Coast Regional District, 3 Dec 2019, SCRD Boardroom, Sechelt, BC. Meeting.  ISL Engineering and Land Services (2011) ​Integrated Transportation Study.​ Retrieved from Jahic, A., Eskander, M., & Schulz, D. (2019). Charging Schedule for Load Peak Minimization on Large-Scale Electric Bus Depots. ​Applied Sciences​, ​9(9)​, 1748. doi:10.3390/app9091748. Janovec, M., & Koháni, M. (2019). Exact approach to the electric bus fleet scheduling. Transportation Research Procedia​, ​40​, 1380–1387. doi: 10.1016/j.trpro.2019.07.191 Lajunen, A. (2018). Lifecycle costs and charging requirements of electric buses with different charging methods. ​Journal of Cleaner Production,​ ​172​, 56-67. Lewis, Carol Abel, et al. (2012).​ ​Reviewing, Analyzing and Updating Marketing Strategies to Increase Public Transit Ridership​. Southwest Region University Transportation Center. Retrieved from​.  Lin, L.-T., Yeh, C.-f., Chen, S., & Huang, C.-H. (2017). Role of governance in the achievement of 20-fold increase in bus ridership – A case study of Taichung City. ​Transportation Research Part A: Policy and Practice, 98​, 64-76. doi:10.1016/j.tra.2017.01.025 Liu, Y.-H., Liao, W.-Y., Li, L., Huang, Y.-T., Xu, W.-J., & Zeng, X.-L. (2019). Reduction measures for air pollutants and greenhouse gas in the transportation sector: A cost-benefit analysis. ​Journal of Cleaner Production, 207​, 1023-1032. doi:10.1016/j.jclepro.2018.10.094  Liu, Z., Song, Z., & He, Y. (2019). Economic analysis of on-route fast charging for battery electric buses: A case study in Utah. ​Transportation Research Record​. doi: 10.1177/0361198119839971  Lowell, D. (2019)​. Electric Bus 101: Economics, politics, myths & facts.​ Retrieved from 37    Lyons, T., Ewing, R., & Tian, G. (2017). ​Coverage Vs Frequency: Is Spatial Coverage Or Temporal Frequency More Impactful On Transit Ridership?​ Utah Department of Transportation Research Division . Menninger, H., & Werly, V. (2014). Americans with Disabilities Act Cost Savings and Increased Fixed-Route Ridership through Transit Agency Travel Training. ​Transportation Research Record: Journal of the Transportation Research Board​, ​2469​(1), 89–99. doi: 10.3141/2469-10  Mohamed, M., Farag, H., El-Taweel, N., & Ferguson, M. (2017). Simulation of electric buses on a full transit network: Operational feasibility and grid impact analysis. ​Electric Power Systems Research​, ​142​, 163–175. doi: 10.1016/j.epsr.2016.09.032  National Academies of Sciences, Engineering, and Medicine. (2018). ​Battery Electric Buses  State of the Practice​. Washington, DC: The National Academies Press. doi:10.17226/25061.  National Inventory Report. (2017). ​National Inventory Report 1990–2015: Greenhouse Gas Sources and Sinks in Canada – The Canadian Government’s Submission to the UN Framework Convention on Climate Change, 2017​. Retrieved from​. New Flyer. (2019). ​Xcelsior Charge​. Retrieved from Nikolai, G., & Ksaibati, K. (2019). Development of Benefit Cost Analysis Tools for Evaluating Transportation Research Projects. ​Transportation Research Record, 2673​(1), 123-135. doi:10.1177/0361198118821675 Office of the Auditor General of British Columbia. (2013). ​Shaping Transit's Future in British Columbia​.​ Retrieved from​.  Paul, T., & Yamada, H. (2014). Operation and charging scheduling of electric buses in a city bus route network. ​17th International IEEE Conference on Intelligent Transportation Systems (ITSC)​, 2780-2786. doi: 10.1109/ITSC.2014.6958135 Proterra. (2019a). ​Catalyst 40 foot Performance Specifications - Canada​. Retrieved from 38   Proterra. (2019b). Proterra launches multi-dispenser charging solution for heavy-duty electric vehicle fleets. Retrieved from Proterra. (2019c). Vehicles. Retrieved from Province of British Columbia. (n.d.). ​Climate Action - Transportation. ​Retrieved from  Santoso, D. S., Yajima, M., Sakamoto, K., & Kubota, H. (2012). Opportunities and strategies for increasing bus ridership in rural Japan: A case study of Hidaka City. ​Transport Policy, 24​, 320-329. doi:10.1016/j.tranpol.2012.09.005 Sheth, A., & Sarkar, D. (2019). Life cycle cost analysis for electric vs. diesel bus transit in an indian scenario.​ International Journal of Technology, 10​(1), 105-115. doi:10.14716/ijtech.v10i1.1958 Sierra Club. (2019). ​Zero Emission Electric Buses​. Retrieved from  Statistics Canada. (2017). ​Sunshine Coast, RD, British Columbia and British Columbia ​(table). Census Profile​. 2016 Census. Statistics Canada Catalogue (No. 98-316-X2016001). Retrieved from  Statistics Canada. (2020). “Monthly average retail prices for gasoline and fuel oil, by  geography.” Retrieved from Sunshine Coast Regional District. (n.d.). ​Fares, Routes & Schedules​. Retrieved from​. Sunshine Coast Regional District. (2009). ​Sunshine Coast Community Energy and Emissions Plan. ​Retrieved from 39  Sunshine Coast Regional District. (2019). Weekday Bus Book Out - October 2019 Sign-up [Excel spreadsheet].   Sustainable Bus. (2019). ​Toronto brings to 20 the number of New Flyer electric buses in the fleet.  Retrieved from Tang, L., & Thakuriah, P. V. (2012). Ridership effects of real-time bus information system: A case study in the City of Chicago. ​Transportation Research Part C: Emerging Technologies​, ​22​, 146–161. doi: 10.1016/j.trc.2012.01.001 Tong, et. al. (2017). Life cycle ownership cost and environmental externality of alternative fuel  options for transit buses. ​Transportation Research Part D: Transport and Environment 57​, 287-302. doi:10.1016/j.trd.2017.09.023 TransitCenter. (2018). ​From Sorry to Superb: Everything You Need to Know about Great Bus Stops.​ Retrieved from  Victoria Transport Policy Institute (VTPI). (2019). ​Transit Station Improvements Improving Public Transit Waiting Conditions.​ Retrieved from  Whitaker, B. and Derk, M. (2018) ​Best Practices and Marketing to Increase Rural Transit Ridership and Investment.​ No. 20-65 Task 73. Transportation Research Board. Retrieved from​.  Xylia, M., & Silveira, S. (2018). The role of charging technologies in upscaling the use of electric buses in public transport: Experiences from demonstration projects. Transportation Research Part A: Policy and Practice, 118​, 399-415. Yaw, S., & Mumey, B. (2017). Scheduling Non-Preemptible Jobs to Minimize Peak Demand. Algorithms​, ​10(4)​, 122. doi:    40     Appendix I: Optimization of Charging Schedules Mathematical Reasoning Behind Charging Schedule Algorithm  Each bus (​b​) has an arrival time ​a​b​ ​, departure time ​d​b​ , charging start ​s​b​ , charging length ​l​b​ , and power demand ​p​b ​. Bus can charge in the interval of [​a​b​ ​, ​d​b​] as long as:  d​b ​ ​- ​a​b​ ​≥ ​l​b ​≥ ​0  d​b ​ ​- ​a​b​ ​- ​l​b​  is defined as shifting time  δ​b ​ ​where if ​d ​b ​ ​- ​a ​b​ = l ​b​ then the shifting time becomes zero. If this is the case only one charging interval for the bus exists. If ​δ​b​ > 0 then multiple charging intervals are possible.   Charging interval can then be defined as [​ s​b​ , s​b​ ​+ ​l​b​ ] ​so that is an element of ​[​a​b​ ​, ​d​b​]​ :  [​ s​b​ , s​b​ ​+ ​l​b​ ] ∈ ​[​a​b​ ​, ​d​b​]   Each charging interval will then be assigned to a time slot of length ​l​b​.  The initial charging interval will start at ​s​b​. The algorithm will then continue automatically placing each charging interval, [​ s​b​ , s​b​ ​+ ​l​b​ ​], within the confines of ​[​a​b​ ​, ​d​b​] while trying to avoid overlapping intervals.   The total demand charge, or power demand ​P​,​ ​is the sum of all the power demand of each bus charging at a certain time ​t ​:   P (t) = ∑ p​b   The goal is to minimize ​P​max​ ​by assigning one charging interval to each bus from the possible charging intervals.  The smallest ​P​max​ ​that the algorithm is able to schedule all the buses to charge is considered an optimal schedule in terms of reducing energy consumption and electrical grid stress.   Algorithm Code in R  #open required libraries library(lubridate) library(rowr) library(ggplot2) 41   library(reshape2) library(dplyr) library(tidyr) ##create list of variables #number of buses to be scheduled nbus = 10 #arrival time arrival = dmy_hms("20-01-2020 22:00:00", tz = "America/Los_Angeles") #departure time departure = dmy_hms("21-01-2020 06:00:00", tz = "America/Los_Angeles") #charging length clength = 2.5 #charging depot number ndepot = 1 #charging rate of bus (kW) crate = 80 #required charge for each bus (kWH) #charge = clength*crate  #amount of time available to charge tmdiff = as.numeric(difftime(departure, arrival))  #create list of buses buslist = list() #create list of schedules schedlist = list()  #create a bus number bus = 0 #create a charge start time starttm = arrival   while (nbus != 0){   bus = bus + 1   #add to bus list   buslist = append(buslist, paste0("Bus #", bus))     if (bus > 1){     starttm = endtm   }     #create end charge time   endtm = starttm + clength*3600     #ensures that charging dues not go past departure time   if (endtm > departure){     #add one to charging depot     ndepot = ndepot + 1     starttm = arrival     endtm = starttm + clength*3600   }     #creates schdule time for each bus 42     schedule = interval(starttm, endtm)   schedule = paste0(schedule, "--", ndepot)   #adds schedule to bus number   schedlist = append(schedlist, toString(schedule))     nbus = nbus - 1   }  #splitting time interval into start and end times for charge  #create list for charge start and end time tmlist = list()  #create a loop that will append the times to tmlist for (i in 1:length(buslist)){   #splits each of the times to two separate strings   tmlist = append(tmlist, strsplit(schedlist[[i]], "--")) }   #combines bus numbers with the charge times and creates a table yeet =, Map('c', buslist, tmlist))  #add required depot to dataframe yeet = cbind.fill(yeet, ndepot, fill = " ")  #maximum demand charge to dataframe demand = ndepot*crate yeet = cbind.fill(yeet, demand, fill = " ")  #name the columns colnames(yeet) <- c("Bus #", "Charge Start Time", "Charge End Time", "Depot #", "Required Depots", "Demand Charge (kW)")  #export schedule to indicated location write.csv(yeet, "C:/Users/Jason Lin/Documents/yeet.csv", row.names = FALSE)  #create new dataframe with just bus, charge start and end times dfr <- data.frame(   name = (yeet[1]), = (yeet[2]), = (yeet[3])   )  #total number of depots and maximum demand charge totals = c(ndepot, demand) names(totals) <- c("# of Required Depots","Maximum Demand Charge")  #rename the columns names(dfr) <- c("task", "start", "end")  #create factors and convert variables to POSIXct 43   dfr$task <- factor(dfr$task, levels = dfr$task) dfr$start <- as.POSIXct(dfr$start) dfr$end <- as.POSIXct(dfr$end) #melt the dataframe into one  dfr_melted <- melt(dfr, measure.vars = c("start", "end"))  #create a start date start_date <- as.POSIXct("2020-01-20 22:00:00")  #create gantt chart using dfr yeet_gantt <- ggplot(dfr_melted, aes(value, task, colour = "Red")) +    geom_line(size = 10) +   labs(x = '', y = '', title = 'Timeline of Bus Charging Schedules') +   scale_x_datetime(limits = c(start_date, NA), date_labels = "%b %d %H:%M", date_breaks = "1 hour") +   xlab("Date & Time") +   ylab("Bus #") +   theme(axis.text.x = element_text (angle = 50, hjust =1))  #create a new schedule newsched = substr(schedlist, 1, nchar(schedlist)-3)  #create a dataframe with only time intervals and frequency each interval test =  #create empty list for demand charge chargelist = list()  #calculate demand charge for each time interval for(j in 1:length(test$Freq)){   charge = as.numeric(test$Freq[j] * crate)   chargelist = append(chargelist, charge) }  #remove levels from factors test = levels(droplevels(test$newsched))  #create empty list for time tmlist = list()  #split each of the time intervals into start and end times for (k in 1:length(test)){   #splits each of the times to two separate strings   tmlist = append(tmlist, strsplit(test[k], "--")) }  #combine time intervals and chargelist test =, chargelist))  #rename columns  colnames(test) <- c("time", "dcharge")  #separate the charge start and end times then make it into a row slap = separate_rows(test, "time", sep = "--") 44    #Convert time from string into a date format slap$time <- as.POSIXct(slap$time) slap$dcharge <- as.numeric(slap$dcharge)  yeet_dcharge <- ggplot(slap, aes(time, dcharge)) +   geom_line() +   labs(x = '', y = '', title = 'Demand Charge Throughout Charging Schedule') +   scale_x_datetime(limits = c(start_date, NA), date_labels = "%b %d %H:%M", date_breaks = "1 hour") +   xlab("Date & Time") +   ylab("Demand Charge (kW)") +   theme(axis.text.x = element_text (angle = 50, hjust =1)) +   ylim(0,350)                         45   Appendix II: Ridership Strategies Table 10. Cost value and expected impacts on ridership of each potential ridership strategy.   1​ City of Edmonton. (2018). Park & Ride Guidelines. Retrieved from​. 2​ ​ CTG. (2014). Technical Memorandum #5 Financial Analysis and Management Options. Retrieved from​. 3​ ​ Infrastructure Services Committee. (2019b). ISC Agenda, Sunshine Coast Regional District, 3 Dec 2019, SCRD Boardroom, Sechelt, BC. Meeting. 4​ ​ BC Transit. (2018). Transit Shelter Program. Retrieved from ​​.  46   Table 11. Typical base costs (not including taxes) different types of E-series transit shelters, a series designed for rural communities.   47 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items