Battery Electric Buses—State of the Practice (2018)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

14 This literature review addresses the three overarching topics of BEB deployment. They are planning; service, maintenance, and operations; and costs and benefits associated with deploy- ment. Each category includes its own subcategories, presenting current literature on specific aspects of each topic. For the planning section, this review addresses topics such as a life cycle cost analysis, technical specifications of the BEBs, route selection, charging infrastructure, and scalability. The service, maintenance, and operations section includes training, availability and reliability, and stakeholder involvement. Finally, the costs and benefits section addresses exter- nal funding opportunities, energy costs, and social, environmental, and health benefits. If the current available literature is lacking for a specific topic, the subject is included in a list at the end of the review as a suggestion for further research. The rapid growth of BEB technology and market is reflected in the literature’s progression. Two early BEB studies—the Center for Urban Transportation Research’s Realizing Electric Bus Deployment for Transit Service and the Federal Transit Administration’s Analysis of Electric Drive Technologies for Transit Applications—provide information about BEB experiences and were published in 1998 and 2005, respectively. The latter study evaluated an electric bus produced by a manufacturer that no longer exists and the former study concluded that “perhaps the ulti- mate reality is best expressed [by the statement] ‘there may never be a future for big electric buses because of their power requirements, but it could work well for the smaller ones”’ (Real- izing Electric Bus Deployment for Transit Service, page 32). Judging solely by the number of BEB models currently available and the number of BEB deployments, it is clear that BEB technology has evolved significantly and agency experience with the technology has grown exponentially since these reports were published. As with any developing technology, documented results and information will lag behind actual experience. Although somewhat limited in availability, this literature review has focused on empirical reports published within the last decade for relevance. The literature review relied heavily on an NREL analysis of Foothill Transit’s deployment of BEBs. This evaluation is an appealing reference because it captures the agency’s path toward its goal of becoming 100% battery electric by describing the measured results and lessons learned from its initial deployment of 12 on-route fast charge buses to fully electrify one route. NREL’s evaluation objective was “. . . to provide comprehensive, unbiased evaluation results of advanced technology bus development and performance compared to conventional baseline vehicles” (Eudy et al., 2016, page 6). Foothill Transit, located in California’s San Gabriel and Pomona Valleys, currently operates 361 buses in revenue service. Of those 361 buses, 344 are CNG and the remaining 17 are fast-charge BEBs. Foothill Transit began its conversion to CNG buses in 2003 and has continued to integrate cleaner technology, retiring its last diesel bus in 2013. The Foothill Transit fleet evaluation provides information relevant to many of the subjects that this synthesis was intended to address. Each topic is also supported throughout the report with references to other available literature. C h a p t e r 2 Literature Review

Literature review 15 The literature review will start with brief reviews of international activity to give context to the U.S. market and to trolley buses as a foundational electric drive technology for BEBs. International Activity An estimated 173,000 BEBs were deployed worldwide as of 2015. China has the vast majority of deployments, with more than 170,000 BEBs. The Chinese government has established a policy and program for “new energy buses,” with a goal to produce 1.67 million EVs (including BEBs) and to create 1.2 million jobs annually for the period 2010–2020. Shenzhen City alone currently has 4,887 BEBs in operation. By the end of 2017, all of the city’s buses will be fully electrified, with 16,493 BEBs. Europe follows Asia with more than 956 BEBs delivered or on order. Of those, 64% are over- night charged and 36% are opportunity charged. The United Kingdom has more than 18% of the total European fleet, while the Netherlands, Switzerland, Poland, and Germany each account for about 10%. Europe also has an established electric bus program, called ZeEUS (Zero Emission Urban Bus System), with more than 40 participants and a budget in excess of 22 million euros. The ZeEUS eBus Report is an informative synopsis of the European BEB market and developments. European cities and countries are primarily motivated by the desire to address global warming and make less of an impact on the environment. International government relations, such as the Paris Agreement to limit global warming to 2°C (United Nations Framework Convention on Climate Change 2016) and the U.S.–China Race to Zero Emissions Challenge (Zero Emissions Bus Benefits 2016), are stimulating the zero emissions market overseas. Other worldwide efforts include South Korea’s research, development, and deployment of wireless charging infrastructure and BEBs. Trolley buses with autonomous off-wire operation (equipped with batteries) are being tested throughout Russia, Belarus, Moldova, Kyrgyzstan, and Serbia. Two BEB demonstration projects have taken place in India. A solar-electric bus service has been established in Australia. In Canada, the Societe de transport de Montreal has purchased three BEBs and is installing four fast chargers to test the technology. Trolley Buses First introduced in 1882, trolley buses—also known as trackless trolleys—were the earliest all-electric buses. While trolley bus propulsion is provided by electric motors and accessories are all electric, they are different from BEBs in that they draw power from overhead wires (sus- pended by roadside poles) instead of from energy stored in batteries. Currently around 300 trolley bus systems are in operation around the world and more than 800 systems have existed over time. They are an attractive option for agencies because they are quiet, have “powerful but smooth accel- erations,” and give the public a sense of “permanence of service” (Arieli Associates, n.d., page 7). Their use and development over the last century have contributed to the introduction of BEBs through development of electric components for traction systems and accessories and through public acceptance and familiarity with electric transportation. They share many of the benefits and challenges associated with BEBs, especially BEBs utilizing on-route charging. While trol- leys offer an advantage to BEBs in that they do not carry the weight of the batteries onboard the vehicles, they have significant drawbacks, with more extensive fixed infrastructure and wires that are very expensive to install and maintain and that are often considered to be unsightly. Trolley buses are now being deployed that are equipped for limited off-wire operation. This capability is achieved by adding a small auxiliary power unit such as a diesel engine. This is also known as dual-mode capability. For full electric operation, dual-mode capability is being accomplished

16 Battery electric Buses—State of the practice with on-board batteries that are charged with the catenary while on wire and then used for off- wire excursions (Trolleybus 2017). Planning Considerations Life Cycle Cost Analysis An important aspect of planning for BEB deployment is analyzing the cost over the total life cycle of the vehicle, including upfront capital costs, component replacement costs, maintenance costs, and electricity costs, as shown in Table 3. Many agencies are hesitant to purchase BEBs because of the higher capital costs compared with diesel buses. The common response to that argument is that BEBs can make up for the higher capital costs in their lower fuel consumption and maintenance costs relative to diesel costs. Due to the nascent stage of BEB development and deployment, accurate BEB capital and operations cost data are limited and difficult to obtain. First, BEB purchase costs are continu- ing to drop and have not yet stabilized. Second, deployments have not seen enough opera- tional time to collect real maintenance data, especially with respect to battery replacement costs. However, one analysis established and compared costs between BEB and diesel buses, as shown in Table 3. Note that these costs were developed to be inputs to a life cycle environ- mental analysis and use manufacturing costs, instead of purchase costs, in order to exclude profit and tax components. The life cycle costs of BEBs were similar to those of diesel buses when the cost of diesel fuel is high (Table 4). However, the assessment shows a wide range of fuel costs for the diesel buses. While the assessment provides a framework for accomplishing a life cycle assessment and provides some details (at 2013 costs), its shortcomings prevent making broad conclusions. The assessment does not take into account variability in electric- ity cost structures (only costs at $0.11/kWh) and sensitivity to power demand while charging. Additionally, there are numerous state and local funding opportunities that a transit agency may be able to utilize to reduce the capital costs of BEBs, which are explored in the Costs and Benefits section of this report. Third, BEB manufacturing costs (identified as Additional Life Cycle Cost Component BEB Cost Compared with Conventional Bus Cost Bus Costs Typically higher Component replacement costs It depends; battery cost replacement costs are high and suspension wear may be higher with increased curb weight, however there are fewer moving components in BEBs, brake components wear slower due to regenerative braking capability. Costs should particularly be compared based on mid-life overhaul expectations where major components such as engines and batteries are designed to be replaced or overhauled. Maintenance labor Comparable, but has potential to become lower once technicians become familiar with electric systems Preventive maintenance Lower; no oil systems, less brake wear Electricity costs Typically lower than conventional fuel costs; BEBs are much more efficient, however electricity rates and rate structures can vary tremendously depending on location; diesel costs fluctuate and are relatively unpredictable Capital Costs Operating Costs Source: Center for Transportation and the Environment. Table 3. Life cycle costs of BEBs.

Literature review 17 Manufacturing costs in Table 4) have continued to show reductions since 2013, particularly with respect to battery costs (Ercan and Tatari 2015). Due to the highly variable degree of electricity costs as well as to the downward progression of the costs for BEB components as technological improvements and economies of scale, it is difficult to make general conclusions regarding BEB life cycle costs. It is imperative to properly evaluate and understand the variables associated with life cycle costs for an individual transit agency, especially when comparing the variables with conventional technologies. In order to better balance the cost components for buses purchased with federal funds, FTA now allows major component capital costs such as batteries to be procured under a lease arrangement in order to reduce the upfront capital costs of BEBs. This financing arrangement allows the purchase of BEBs to be on par with conventional buses with the expectation that ongoing operational costs (plus the battery lease) will be comparable with conventional buses due to savings in other areas of electric bus operations and maintenance. FTA codified this change in the most recent transportation bill, the FAST Act (FAST Act 2017). No industry guidelines or standards were identified for calculating life cycle cost analyses for electric buses, especially for establishing operation and maintenance costs. Bus Technical Specifications, Operational Requirements, and Route Selection APTA released its most recent version of the Standard Bus Procurement Guidelines in 2013 (APTA Standards Development Program 2013). The Guidelines provide guidance for a com- plete procurement for buses from 30′ to 60′ that can be customized to suit an individual agency’s needs. The document is written specifically for conventional drivetrain technologies and does Bus Type Cost Component Cost (2013 $) Total Cost Process-LCA data (if applicable) Manufacturing $300,000 $1,164,728 (using lowest of the FC range) Maintenance $444,000 $1,361,664 (using highest of the FC range) Fuel Consumption (FC) $420,728-$617,664 Manufacturing $300,000 $1,330,462 (using lowest of the RI range) GREET’s Battery Model run for these specifications: Additional Manufacturing (related to the electric drive system) $570,000 $1,332,962 (using highest of the RI range) 12 years lifetime Maintenance $328,560 37,000 annual miles Refueling Infrastructure (RI) (grid mix scenario) $22,500-$25,000 40 ft long; 112,000 lb Fuel Consumption $109,402 112 Wh/kg; 1814 kg (Li-ion) Diesel BEB Table 4. Life cycle inventory summary.

18 Battery electric Buses—State of the practice not currently include considerations for zero emission technologies. A multidisciplinary team that includes bus manufacturers, transit agencies, industry consultants, and other public and private organizations was organized by APTA to update the Guidelines to address the procure- ment of BEBs. The expected to release date for this document is in early 2018. It is important to note the update does not include specifications for procurement of charging infrastructure. Planning BEB deployments requires the transit agency to consider multiple factors, includ- ing specific vehicle capabilities, duty cycle requirements (route speeds and grades), deadhead requirements, service requirements (layovers and bus blocking), environmental conditions, pas- senger loads, and charging schemes. All of these factors are codependent, which can make estab- lishing bus technical specifications, route locations, charging type and locations, and operational planning difficult but important to consider holistically. Table 2 from the Introduction of this report lists current BEB manufacturers and the energy storage options within the product offerings. The examples that follow highlight how different transit agencies have planned for BEB deployments. A Columbia University study about BEB integration into NYC’s transit system explains the importance of battery characteristics when purchasing BEBs: For an electric vehicle, the key battery characteristics are the range (distance) that can be traveled on a full charge and the time required to recharge the battery. However, it is important to understand that these characteristics act differently in the electric vehicle world than they do with gasoline or diesel powered vehicles. For example, most cars have a range of about 400 miles. That range can vary depending on whether the car is being driven primarily on the highway or in the city. Stop and go traffic impacts the fuel economy and therefore the range of the car. In the case of electric vehicles, ambient temperature can influence battery efficiency and therefore fuel economy more than in a gasoline/diesel powered vehicle. The impact will vary by battery type and by the actual ambient temperature in addition to bus load, speed, incline of the bus route, etc. (Aber, 2016, page 23). Variables such as HVAC loads, passenger loads, bus speeds, and route grades can have a significant impact on the energy consumption of a BEB and, depending on the battery size and condition, will have a commensurate impact on range. Many OEMs cite baseline energy consumption based off their Altoona testing results, which occur on flat grades, at seated load weight, and no HVAC loads. The baseline energy consumption estimates are similar between the various bus products. One BEB OEM provides Altoona-tested efficiencies of 1.61 kWh/mile to 1.89 kWh/mile, depending on bus size. However, once duty cycle and HVAC impacts become more demanding, the efficiencies can more than double, effectively halving the range. Modeling and simulation results have shown that expected efficiencies in winter with maximum passenger loads can surpass 3 kWh/mile with the worst case conditions being over 6 kWh/miles (Hanlin 2016). Empirical data for two different BEB fleets operating in cold northern U.S. climates that were collected and analyzed by the SAE J3105 committee also suggest that these variables can have a significant effect on bus efficiency and range, as shown in Figure 3. Chicago Transit Authority (CTA) has two BEBs. Figure 4 depicts the operational requirements of the buses. The agency needed to organize its use of the BEBs in such a way that accounted for available charge time both overnight as well as throughout the day. In addition, they also had to account for range requirements for each bus block (a bus block is the daily schedule of travel for a given bus from depot pull-out to pull-in). For its initial BEB demonstration, Foothill Transit selected Line 291, a 16.1-mile route. Line 291 was the most viable option because of its minimal deadhead distance and suitability to an on-route, fast-charging system because it loops through the transit center in both directions. Line 291 requires seven buses during peak hours; the additional BEBs are used as spares for main- tenance downtime as well as for serving other appropriate routes that go through the Pomona Transit Center such as Line 855. Foothill Transit made minor adjustments to its schedule to

Literature Review 19 accommodate additional layover time to allow for connection to the on-route charger. Figure 5 shows how Foothill Transit’s extended layover time allows the batteries to operate between 30% and 80% state of charge (SOC). Figure 5 also highlights how a single duty cycle variable can affect BEB operations. NREL’s fleet evaluation of Foothill Transit’s Route 291 revealed that the SOC of the battery decreased faster between 4:18 p.m. and 4:38 p.m. than between 4:38 p.m. and 4:58 p.m. due to the grade of 0 1 2 3 4 5 6 En er gy C on su m pt io n (k W h/ m ile ) BEB Daily Energy Consumption – 2 Northern US Fleets Case 1 Energy Consumption Case 2 Energy Consumption Figure 3. Empirical BEB efficiency data comparison. Source: Center for Transportation and the Environment. Figure 4. CTA’s operational requirements. Source: Chicago Transit Authority.

20 Battery electric Buses—State of the practice the route changes. The first part of the loop is uphill, and the second part is downhill as the bus returns to the charging station. The Center for Transportation and the Environment (CTE) has developed an approach with tools for holistically accounting for multiple factors when selecting, procuring, and deploying BEBs. The approach is rooted in model development to understand the effects of the different variables on operations to life cycle costs. The first model in CTE’s approach establishes the capabilities and performance of available bus models and charging systems in specific agency environments and duty cycles. Outputs include power capability and energy consumption. This model is augmented to show the effects of various charging options on range. The second model accounts for local electricity rates and evaluates the effect of various rate structure components. The third model accounts for the full costs of a BEB deployment by using results from the previ- ous models and other up-to-date costs. This comprehensive tool enables agencies to make data- driven decisions, including purchasing (buses and charging infrastructure), deployment (route planning and charging system locations), and operations (bus blocking, layovers, and charging schemes). Figure 6 shows the approach that an agency can take when planning BEB deploy- ments utilizing the tools developed by CTE. Charging Infrastructure and Layover Location Characteristics Transit agencies must also be able to support charging for BEB bus deployments, which can be achieved in a variety of ways. There are numerous infrastructure considerations when deploying BEBs. According to the Planning and Optimization of a Fast-Charging Infrastructure for Electric Urban Bus Systems 2014 report, three main factors are key to creating optimal distribution of the charging points: replenishment of energy consumption for the individual buses, local and institutional structural reservations, and intersections of the agency’s other lines with the char- ger so as to optimize the agency’s network. The optimal distribution of charging points is the end result from a number of prior considerations: the grid power, battery type, and battery SOC affect the schedule, which in turn affects the dwell (charging) time, which then affects the level Figure 5. Foothill Transit’s route characteristics. Source: Prohaska et al.

Literature review 21 of replenishment of the energy consumption. If the energy is adequately replenished, then the charging points are optimally distributed, assuming they comply with infrastructure standards and are placed such that other routes can potentially use them in the future (Kunith, 2014, page 44). Another challenge that transit agencies have to address when planning for infrastructure is determining the optimum method of charging to support their particular service needs. Transit agencies have the option of purchasing BEBs that utilize plug-in charging, overhead (conduc- tive) fast charging, or wireless (inductive) charging. Plug-in charging is almost always done at the depot or shop because of its slow rate of charge. Overhead fast charging and wireless charg- ing are typically done on route but can also be installed at the depot. BEBs can be designed and deployed to work with one or a combination of these charging options. Foothill Transit chose to utilize both overhead fast chargers as well as plug-in depot charging: At the end of each day, operators typically charge the BEBs at the Pomona Transit Center (PTC) prior to returning to the depot. A slow charger is used at the operations and maintenance facility for times when a bus needs additional charging. . . . Foothill Transit plans to eventually add a fast charger at this facility (Eudy et al., 2016, page 22). With funding awarded through the second round of the TIGGER program, Foothill Transit replaced the old chargers with two on-route conductive fast chargers (Eaton 500 kW chargers) and purchased 12 more BEBs. The old chargers were replaced because the original manufac- turer, AeroVironment, stopped supplying plug-in chargers to the bus market. The two chargers are co-located at the same station. Both chargers are housed in the same climate-controlled building with charge heads positioned on either side. The two chargers operate as separate units with a dedicated control system for each. A common communication network serves both units with sensors to detect which charge head a bus is approach- ing to enable proper bus-to-charger communication for docking. Emergency shut-off switches for each Life Cycle Cost Modeling Annual Fuel Costs Capital Costs Maintenance Costs 12 Year Cost Analysis Energy Consumption and Charging Profile Bus & Route Modeling Electricity Modeling Route Requirement Proposed Bus/Charger Electricity Rate Schedules Figure 6. Steps an agency can take when analyzing routes. Source: Center for Transportation and the Environment.

22 Battery electric Buses—State of the practice charge head are located on either side of the building. The system is designed to fully charge a bus in under 10 minutes. For Line 291, the BEBs charge an average of 12.5 times a day for an average duration of around five minutes (Eudy et al., 2016, page 23). A major consideration in planning for BEBs is selecting suitable locations for the chargers. Fast chargers “can only be connected to the grid where utilities can provide a dedicated supply line capable of delivering the very high currents demanded” (Air Resources Board, Oct. 2015, page III-5). Furthermore, fast chargers can only be deployed where an agency has access or rights to property to install the infrastructure. These restrictions can significantly limit options for locat- ing on-route charging, particularly in dense urban environments. Foothill Transit chose to locate the fast-charging station at PTC for a number of reasons (charger shown in Figure 7). At a mid-way point in the route (shown in Figure 8), PTC can accommodate fast charging infrastructure for two buses simultaneously and the transformer is within the vicinity. Perhaps most important, Foothill Transit already had rights to use the prop- erty through a 40-year lease with the City of Pomona. The city was supportive of the agency’s efforts to deploy BEB. Finally, the PTC is a transfer point for eight local routes, allowing Foothill Transit the flexibility to expand or modify its BEB service without needing to move or add more charging infrastructure. According to NREL’s analysis of Foothill Transit, Costs for the chargers and installation continue to drop. Installation costs will vary from site to site depend- ing on a number of factors including the distance to a transformer. The total cost for the charging station being installed at the Azusa Intermodal Transit Center was $998,000. The installation includes two 500-kW fast chargers at $349,000 each. The cost to install the chargers was $300,000 (Eudy et al., 2016, page 24). The Society of Automotive Engineers (SAE) International typically establishes the North American standards for electrical connectors for EVs. The standards cover the general physical, electrical, communication protocol, and performance requirements for EV charging systems and couplers. The intent is to define a common EV charging system architecture, including operational, functional, and dimensional requirements for the vehicle and connector interface. Plug-in charging for electric buses is generally based off methods and standards developed for the automotive industry. SAE J1772 chargers are typically used for high power DC charging and SAE J3068 (coming in early 2018) chargers for high power three-phase AC charging. This type of charging requires manual plugs and is typically provided or specified by the bus OEM for purchase and installation at the bus base or depot by a certified electrician. Figure 7. Foothill Transit’s on-route conductive fast charger. Source: Eudy et al.

Literature review 23 Figure 8. Foothill Transit’s BEB route with the blue star showing where the charging station is located. Source: Foothill Transit.

24 Battery electric Buses—State of the practice Standards for overhead conductive and for wireless charging are currently under develop- ment. A committee has been established to develop the SAE J3105 standard for overhead con- ductive charging specifications for transit buses. In addition to ensuring safe, efficient, and effective operation of overhead charging systems, this standard will allow for interoperability of any charging system with any bus that follows the standard. A similar effort is being established for wireless charging systems under SAE J2954/2. Much like using appliances in our homes, establishing charge standards and interoperability for BEBs is an important step toward achiev- ing widespread use of electric buses and full commercialization. Foothill Transit’s depot chargers were $50,000 each, which is consistent with CARB’s analysis of average depot charger cost. When assessing the total cost of charging infrastructure, CARB suggests incorporating the following costs for the analysis: the actual charging station hard- ware; other hardware and materials associated with construction; labor costs; construction time including an initial on-site consultation; and municipal permitting costs (Air Resources Board October 2015). Where, when, and how an agency charges its fleet of BEBs will affect the amount of power consumed that is captured at various utility meters. The usage affects the demand charges and time of use charges from the utility and can significantly affect the cost of electricity consumed. Understanding the electricity rates and utility rate structures is key to optimizing a fleet charg- ing scheme. Undertaking this exercise at that outset of a project or fleet conversion can signifi- cantly lower overall energy costs and total cost of ownership for BEBs (Air Resources Board October 2015). Electricity Rate Structure Utility costs, or electricity rates, contribute to overall BEB operating costs, as addressed later in this literature review. Electricity rates are designed by utilities to be based on the cost of service. These costs are often made up of multiple components including usage, or “energy” charges (cost per kWh) and power or “demand” charges (cost per kW). Energy charges are based on how much electricity a customer uses and demand charges are based on the maximum amount of power a customer draws at once (typically over a 15-minute period). The impact of demand charges is explained by the Union of Concerned Scientist’s BEB analysis: Electricity rates often include an additional “demand charge” related to the maximum power con- sumed during a 15-minute interval for the month. This means that spikes in electricity demand can add significantly to the cost of vehicle charging and erode the savings of electricity compared to other fuels. The impact of demand charges can be most acute when fleets have a small number of electric vehi- cles and charging causes large, relative spikes in electricity demand. With a larger number of vehicles, fleet owners can space out charging over a period of time, minimizing the spikes (Chandler et al., 2016, page 27). As also shown in CALSTART’s Peak Demand Charges and Electric Buses white paper, demand charges can have a significant impact on total energy costs. The white paper also analyzes and illustrates the beneficial effects of using a single charger to charge more than one bus (Gallo et al. 2014). Figure 9 provides a relative comparison of the impact of demand charges (shown in light green) versus the energy costs (in dark green) versus costs for diesel and CNG. The graphs show how demand charges can be reduced on a cost-per-mile basis as more buses utilize the same charger, while other costs stay consistent. The graphs also show the significant impact that demand charges can have on the operational costs for the buses. In the example for demand charges of $20/kW, they are 3.6 times the energy costs on a per-mile-basis when using a single bus per charger.

Literature review 25 More on Demand Charges [Excerpt from Air Resources Board, Electricity Costs for Battery Electric Bus Operation, n.d.] Making an analogy to plumbing, it is comparable to how many gallons of water a person draws at any given moment. In a plumbing example, a user could choose to turn the faucet on low and fill a five gallon bucket over ten minutes (a low demand), or choose to turn the faucet on high and fill the five gallon bucket in one minute. In both cases, the person drew five gallons of water, but in the latter case, the rate of flow was much higher. If the user above wants to fill the bucket in one minute rather than ten minutes, the water utility may need to widen the pipes to the faucet, maintain a higher reserve, and have bigger pumps to deliver the water at the higher rate. Therefore, the water utility may charge the user more to recover the costs of infrastructure and/or reserve needed to deliver the water faster. In both cases the person still has to pay the same amount for the total volume of water used. The total bill for drawing the water faster is higher even though the same amount (five gallons) of water is drawn. This analogy applies to electricity use. A demand charge, simply, is a fee paid based on the rate (think: gallons per minute) at which the customer draws electricity. In the case of electricity, the “gallons per minute” is measured in kilowatts (kW). Kilowatts are useful to think of as kWh/h, where kilowatt-hours (kWh) are equivalent in this analogy to gallons—they’re the total volume of electricity delivered—and kWh/h is how many kWh are delivered in a given amount of time. It is useful to think of demand charges as fees assessed for being able to draw a lot of energy in a short amount of time. It is important to note that this analysis used a specific set of assumptions, including diesel fuel costs ($4/gal), fuel economy (4 miles per gallon), CNG fuel costs, CNG fuel economy, maxi- mum charge power (500 kW), electric bus efficiency (2.5 kWh/mile), energy cost ($0.10/kWh), and demand charges ($10/kW, $20/kW), among other variables. Each of these variables can change considerably over time and, depending on location, by a factor of two or more. For example, as of the writing of this synthesis report, many agencies are paying less than $2 per gallon for diesel fuel, which would bring the cost of diesel down to less than $0.50 per gallon in Figure 9. Thus transit agencies should perform similar detailed analyses for their individual conditions when making purchasing and planning decisions. When comparing diesel, CNG, and electricity costs, it should be noted that over time, electric- ity costs are far more stable than conventional fuels. Figure 10 shows a general 3.5% trend for California electricity costs, while CNG and gasoline can spike or fall 200% to 300% in periods of 2 to 3 years. Electricity costs can be further complicated based on the time of day that the charg- ing occurs (time of use) and/or the season, because increasing power production capability has a direct impact on utilities costs, as shown in Table 5. In the example, on-peak, mid-peak, and off-peak each corresponds to a time period throughout the day. Utility costs can fluctuate signif- icantly depending on when the charging is occurring within 15-minute periods throughout the day. Accounting for these variables and properly planning a charging scheme can significantly reduce energy costs for a BEB fleet.

Figure 9. A comparison of fuel and electricity costs and the impacts of demand charges. Source: Gallo et al.

Literature review 27 Figure 10. A comparison of the stability of fuel and electricity costs. MMBTU = millions of British thermal units. Source: U.S. Energy Information Agency. Electricity rate structures can be complicated and decisions related to bus type, fleet size, char- ger type, charging scheme, charge locations, route selection, and route planning can all have an effect on electricity costs. Electricity rates structures and costs should be evaluated early in the project in order to minimize these costs. Planning and Support Tools CTE and BEB OEMs are using tools and methods to predict bus range in a variety of con- ditions and a variety of charging options. CTE uses Argonne System Modeling and Control Group’s Autonomie, a tool that can perform powertrain modeling and simulation but that must be adapted to a given situation. By supplying different duty cycles, powertrain configu- rations, and bus components, Autonomie can run a simulated operation of a bus on route to determine how the bus will perform in the given situation. CTE is combining results from Autonomie with outputs from utility rate modeling tools to help agencies make data-driven procurement and operational decisions. Some bus OEMs are using internally developed models designed specifically for their buses to predict operational capabilities. No other planning tools or automated applications were identified to support transit agencies efforts to plan for the deployment of BEBs. Studies specific to an individual agency’s experience are available, such as the Planning and Optimization of a Fast-Charging Infrastructure for Electric Urban Bus Systems 2014 report or the Electric Bus Analysis for New York City Transit 2016 report. Such studies provide insight into topics such as electricity rate modeling or a life cycle cost analysis for a particular agency’s situation, but they lack methodologies that are applicable to the transit industry as a whole.

Rate Structure A B C D E F Allowable Max Demand Range below 20 kW 20 kW-200 kW 200 kW-500 kW 200 kW-500 kW above 500 kW 20 kW-500 kW Fixed Charges Customer Charge [$/Meter/Month] $25.92 $198.79 $441.93 $441.93 $319.93 $198.79 Three Phase Service [$/Month] $18.60 $- $- $- $- $- Demand Charges Facility Demand Charge [$/kW] $- $13.20 $16.37 $16.37 $14.88 $13.20 Time-of-Use Demand Charge [$/kW] Summer On-Peak $- $- $- $18.86 $24.15 $- Summer Mid-Peak $- $- $- $5.53 $6.66 $- Summer Off-Peak $- $- $- $- $- $- Winter On-Peak $- $- $- $- $- $- Winter Mid-Peak $- $- $- $- $- $- Winter Off-Peak $- $- $- $- $- $- Energy Charges [$/kWh] Summer On-Peak $0.24 $0.36 $0.36 $0.14 $0.14 $0.29 Summer Mid-Peak $0.19 $0.15 $0.14 $0.09 $0.08 $0.12 Summer Off-Peak $0.16 $0.07 $0.06 $0.06 $0.06 $0.05 Winter On-Peak Not applicable N/a N/a N/a N/a $0.11 Winter Mid-Peak $0.16 $0.09 $0.09 $0.09 $0.09 $0.09 Winter Off-Peak $0.15 $0.07 $0.07 $0.07 $0.07 $0.06 Source: Center for Transportation and the Environment. Table 5. Example of utility rate structures.

Literature review 29 Scalability While there are approximately 600 BEBs on order or in service, most transit agencies deploy- ing BEBs have a BEB fleet size of less than 10 buses, as shown in Figure 11. There are only four transit agencies currently operating more than 10 BEBs in service and the largest fleet is 21 buses. Recent orders show that larger deployments are planned, but transit agencies are struggling with how to deploy and manage the practical aspects of a larger fleet of BEBs. As the BEB fleet grows, transit agencies will likely need to incorporate different BEB deployment approaches to ensure the buses meet the needs of varying routes. One challenge is how to address deploying infrastructure at scale. Transit agencies will need to consider issues related to land use, space constraints, grid demand impacts, fleet staging for charging, networking on-route charging, labor requirements for making manual connections and maintaining charging equipment, and maintaining operability during power outages. Tran- sit agencies must also account for utility costs associated with scalability and recognize that utility rate schedules (including energy and demand charges) can vary significantly both within a city as well as nationally and can significantly affect the business case for owning BEBs. A compari- son of peak loads alone for different fleets is displayed in Figure 12, which shows that 60 kW of charging a fleet of 50 BEBs simultaneously can lead to peak loads of 3.0 megawatts, which could be infeasible for some agencies given demand charges and utility infrastructure requirements. There is a need for coordinated and well-documented practices and tools that will support transit agencies’ efforts to deploy BEBs. Coordinated and well-documented practices and tools can also help ensure that transit agencies realize all the benefits of the technology as well as understand the risks and challenges associated with BEB fleet scale up. Service, Maintenance, and Operations Considerations Training When deploying a new technology, transit agencies need to ensure they have the necessary resources to train staff on the new technology. In the case of BEBs, drivers and maintenance Figure 11. BEB fleet size. Source: Center for Transportation and the Environment.

30 Battery electric Buses—State of the practice staff need to know how to effectively and efficiently operate and maintain the vehicles. An example of this is highlighted in the NREL report. Foothill Transit encountered low-voltage starter battery issues and had to replace them not because of a technology issue but as a result of a driver training issue (Eudy et al., 2016, page 37). The operators were not turning off the buses at the end of a shift, most likely because they could not hear it running as they could with conventional buses. Compared with diesel or CNG buses, fueling—or charging—is different for BEBs. The case of Foothill Transit’s docking (or positioning the bus correctly for the on-route fast charger to connect) is operationally unique because the driver applies the accelerator instead of the brake to begin the automated process. This has required Foothill Transit to implement extensive and ongoing training in order to educate its drivers. Maintenance training for BEBs differs from conventional buses in that technicians must understand how to work on all-electric propulsion systems and auxiliary systems as well as be concerned with the safe handling of high voltage systems. The 2005 FTA report on electric drive technologies for transit applications is out-of-date in some aspects, but the section on training remains applicable: There is a need for mechanic training in how to service and troubleshoot electric propulsion compo- nents, and understanding how to work with the on-board diagnostics systems. While transit agencies that operate rail systems are familiar with the requirements of operating and maintaining high voltage electrical propulsion systems, there is often no overlap between the maintenance staff for rail and for buses (Analysis of Electric Drive Technologies for Transit Applications 2005). Limited maintenance training experience or guidance was found in this review of the literature. This topic will be further explored in the Survey and Case Example sections of this synthesis report. Operations The NREL report provides operations and maintenance data on the Foothill Transit BEBs from April 2014 to July 2015. The report also compares these data to a baseline fleet of CNG buses. The average monthly operating mileage for the BEBs during the evaluation period is Figure 12. An example of utility rate structures. Source: CALSTART.

Literature review 31 2,333 miles, which is consistent with the requirements of the route on which they operated (Line 291). “The average runtime per day is 13.2 hours with an average of 13 charges per day. Each charge averages 20 kWh of energy delivered” (Eudy et al., 2016, page 35). Availability (defined as the number of days that the buses are actually available compared with the number of days the buses are planned for operation) for the BEB fleet was 93% during the period (excluding one outlier that was out for extended periods) compared with 94% for baseline CNG buses. “The majority of the issues were for general bus problems—repair of accident damage and the air conditioning system—and not due to any advanced technology component” (Eudy et al., 2016, page 27). Through 399,663 miles of use, “the BEBs had an overall average efficiency of 2.15 kWh per mile, which equates to 17.48 miles per diesel gallon equivalent (DGE). The CNG buses had an average fuel economy of 4.04 miles per gasoline gallon equivalent (GGE), which equates to 4.51 miles per DGE” (Eudy et al., page 19). It should be noted that accessory loads for the CNG buses contribute to lower fuel economy and lower range capability, as more than 50% of “system on” time is spent at a speed of zero miles per hour where lighting and HVAC loads are still required. Ultimately, the BEB fuel economy was almost four times higher than that of CNG buses. Despite this improvement, NREL reported that the battery electric bus fuel cost was $0.39 per mile compared with the $0.23 per mile for the CNG buses during the evaluation period. The utility rate was reported to be $0.18 per kWh during this period. The report goes on to say that “Foothill Transit is working with [their utility] on a new agreement to set a rea- sonable rate charge. This will be a major challenge for any fleet looking to deploy electric buses that charge during peak times. The industry needs to work on a permanent solution for all BEB adopters to keep costs reasonable in the future” (Eudy et al., 2016, page 36). Resiliency and Emergencies In 2013, FTA released the Innovative Safety, Resiliency, and All-Hazards Emergency Response and Recovery Demonstrations program. In one of the selected projects, the Center for Trans- portation and the Environment partnered with the University of Texas Center for Electro- mechanics and Hagerty Consulting to develop a Bus Exportable Power Supply (BEPS) system that will give electric and hybrid-electric buses the capability to act as on-demand, mobile, electrical-power generators. This technology will be especially useful in emergency disaster response and recovery when traditional power supplies are not reliable. Emergency response involves a variety of organizations with different core objectives in addition to the general public. In order to capture the knowledge of emergency response professionals, a team of cross-industry experts has been organized to investigate technologies, methods, practices, and techniques for utilizing the BEPS system. The project team is responsible for system design, demonstration, and a documented recommended methodology for implementation in real-world applications (Center for Transportation and the Environment, University of Texas at Austin 2017). Equipment Longevity and Risk Mitigation As shown in Figure 2, current growth of the electric bus market did not begin in earnest until 2010. Buses and equipment have not yet realized sufficient operation in the field to allow for full assessment of equipment longevity. Component replacement data for a new technology are generally limited in the early stages of deployment, whether because there is only one manufacturer or because the products evolve quickly in their early stages, resulting in design changes. Foothill Transit encountered the latter situation with its first charging infrastructure that became no longer available after the

32 Battery electric Buses—State of the practice manufacturer ceased production. Eventually they were unable to find replacement parts and determined to replace the chargers all together (Eudy et al., 2016, page 11). One component of concern is the traction battery due to its cost and relatively unknown life in a transit bus application. To alleviate this concern for transit agencies, most BEB manufactur- ers are offering a standard 6-year warranty for the batteries to get operators through the midway point of bus life and offering extended warranties up to 12 years to mitigate further risk (Proterra 2017). One BEB manufacturer has “no doubt they will outlast the life of the bus” and is provid- ing a 12-year unconditional warranty for the batteries (BYD Motors, Inc. 2015). Alternatively, manufacturers are also offering battery lease programs to help mitigate potential risk. Technology for Managing, Charging, and Dispatching In 2017, the California Energy Commission awarded nearly $2 million to Prospect Silicon Valley and an innovative Silicon Valley collaborative including the Santa Clara Valley Transpor- tation Authority (VTA) to research, develop, and demonstrate an advanced energy management and grid services system for electric transit bus fleets. The system is intended to reduce costs for charging electric buses, minimize the impact of bus charging on the grid, and provide valuable services that assist the integration of intermittent renewables like solar and wind. The project will integrate systems to reduce charging costs by managing demand, demand response, and wholesale ancillary services such as frequency regulation. These features will be integrated with commercial fleet management tools for what is expected to be the first fully integrated energy management in a heavy-duty fleet. VTA has acquired innovative smart networked charging sta- tions and will provide engineering services, fleet management requirements, in field testing, and collection of charging/energy usage data from the fleet. Working with its Clever Devices, VTA dispatch software provider, VTA will be updating the dispatch software to improve EV fleet management and coordinating with utility (Pacific Gas and Electric) on rate usage and inter- action with the VTA one megawatt solar installation (Valley Transportation Authority 2017). Stakeholder Involvement At the end of NREL’s Foothill Transit report, the agency lists the lessons learned from the experience, and the first bullet is regarding stakeholder involvement. Foothill Transit advises planning ahead to identify stakeholders that need to be engaged at specific points in the plan- ning process. Stakeholder engagement is addressed in more detail later in the synthesis report. Costs and Benefits External Funding Opportunities The availability of external funding sources can drive the adoption of new technology. Transit agencies benefit from federal, state, and local financial support to help offset the higher incre- mental capital costs associated with advanced technologies such as BEBs. Funding to support deployment has been vital to the growth of the BEB market. Federal funding through FTA for the purchase of buses typically covers 80% of the purchase costs to help offset the higher capital costs of BEBs and associated infrastructure; transit agencies have taken advantage of other federal and state funding opportunities. Foothill Transit utilized a 2009 American Recovery and Reinvestment Act grant to purchase its first three BEB that were deployed in 2011. It purchased the next 12 buses through a TIGGER II grant for $10.2 million (Foothill Transit Business Plan and Budget 2015). Foothill Transit also used California Hybrid and Zero-Emission Truck and Bus Voucher Incentive Project funds to further reduce the pur- chase cost of the bus.

Literature review 33 Other transit agencies in California are using state funds to support their purchase of BEBs. Antelope Valley Transit Authority (AVTA), close to Foothill Transit, recently unanimously voted to fully convert its entire bus fleet to BEBs by the end of fiscal year 2018. The agency is sup- porting the deployment through multisource funding, including a $24.4 million grant from the California State Transportation Agency (CalSTA) and additional federal funds (AVTA 2017). FTA’s Low or No Emission Vehicle Program (Low-No) supports adoption of technologically advanced vehicles to help the transit industry become cleaner and more energy efficient. FTA’s 2016 Low or No Emission Vehicle Program provided $55 million for supporting transit agen- cies’ transitions to “the lowest polluting and most energy efficient transit vehicles” (“Low or No Emission Program,” 2016). The Low-No program is included in the current transportation bill, FAST Act, and there are provisions for annual funding opportunities through FY2020. More than 30 agencies have purchased buses with Low-No funding. The Low-No program encourages transit agencies to use the funding to cover the incremental costs associated with the techno- logically advanced vehicles and allows for a higher federal share—85% for buses and 90% for infrastructure. FTA’s Clean Fuels Grant Program has also supported BEB deployments, including Central Contra Costa Transit Authority, Denver’s Regional Transportation District, Nashville Metro- politan Transit Authority, Transit Authority of River City, and Worcester Regional Transit Authority, among others. In addition to discretionary funding opportunities, regulations also drive the adoption of new technology. Foothill Transit incorporated BEBs into its fleet in response to CARB’s “Fleet Rule for Transit Agencies.” The 2000 rule required that urban buses meet stricter California exhaust emission standards and that 85% of a transit agency’s annual urban bus purchases be alternatively fueled (Fact Sheet: Fleet Rule for Transit Agencies Urban Bus Requirements). Further- more, agencies with more than 200 buses must include zero-emission buses as 15% of new bus purchases. Foothill Transit deployed its initial fleet of BEBs with the “goal of evaluating the tech- nology to determine if it could meet service requirements” implemented through the rule (Eudy et al., page 7). The regulation is one of the primary drivers for demonstration and deployment of advanced technology buses in the state of California. CARB has also proposed the Advanced Clean Transit Fleet Rule (California Air Resources Board 2017). Recognizing the role public transit will play in reducing emissions from the mobile sector, the rule would require transit agencies to transition their entire fleet to zero emission vehicles by 2040. Transit agencies nationwide have responded to more stringent national and state air quality regulations by deploying alternative fuel vehicles in their fleets. Areas classified as nonattain- ment by the U.S. Environmental Protection Agency for exceeding the National Ambient Air Quality Standards often benefit from the deployment of cleaner transit technologies (“FAQ’s about Attainment and Nonattainment” 2008). The U.S. Department of Transportation asserts that implementing one BEB will eliminate “10 tons of nitrogen oxides and 350 pounds of diesel particulate matter [over its lifespan], improving air quality in the communities that they serve.” More and more areas classified as nonattainment are looking to their local transit agencies to help meet their attainment goals (“Zero Emissions Bus Benefits” 2016). Public Opinion While there are plenty of anecdotal statements and observations that the public enjoys riding on BEBs due to the clean, smooth, quiet operation of the buses, actual rider surveys were dif- ficult to find. One quantitative measure of ride quality is noise, and documentation of the reduced noise associated with BEBs is available through Altoona Test results. Data are available for New Flyer’s 40′ bus platform using four different propulsion technologies. Figure 13 shows the recorded

34 Battery electric Buses—State of the practice exterior noise levels for each bus model when accelerated from standstill at full throttle. The CNG bus produced the highest noise levels and the electric bus produced the lowest noise levels consistently on both sides of the vehicle. For reference, 80 dB is typically the level of an airplane at one mile while 60 dB is typically the level of conversational speech. Noise levels were also measured in the interior during acceleration from 0 to 35 mph. Fig- ure 14 displays the recorded noise levels for each vehicle at each measurement location. The 60 62 64 66 68 70 72 74 76 78 Driver's Seat Front Passenger Seats Middle Passenger Seats Rear Passenger Seats M ea su re d So un d Le ve l (d B( A) ) Measurement Location Interior Noise Comparison Accelerating 0 to 35 mph-New Flyer 40' Electric Bus CNG Diesel Hybrid Figure 14. Interior noise data comparison. Source: The Altoona Bus Research and Testing Center. 58 60 62 64 66 68 70 72 74 76 78 Curb (Right) Side Street (Left) Side A ve ra ge o f T w o H ig he st A ct ua l N oi se L ev el s ( dB (A )) Measurement Location Accelerating from Constant Speed Electric Bus CNG Diesel Hybrid Figure 13. Exterior noise data comparison. Source: The Altoona Bus Research and Testing Center.

Literature review 35 BEB and diesel bus had the same measured sound level for the driver’s seat location, but for all other measures BEBs were noticeably more quiet than CNG, diesel, and hybrid models. Environmental and Health Benefits Heavy-duty buses and trucks are major contributors of pollutants. In California, 7% of all global warming emissions are from heavy-duty vehicles and that is predicted to rise over the next 30 years. In 2012, heavy-duty vehicles emitted more anthropogenic particulate matter (2.5 micrometers and smaller) than all of California’s power plants combined (23 tons per day versus 7 tons per day, respectively). Trucks and buses also contributed to more than 30% of the nitrogen oxides emitted across the state (Chandler et al. 2016). BEBs have no tailpipe emissions. Their “well-to-wheel” emissions depend solely on how the electricity is produced. Using 100% renewable energy to generate the electricity would elimi- nate emissions entirely from transit bus operations. In a 2016 report, the Union of Concerned Scientists and Life Cycle Associates adapted models from Argonne National Laboratory and California Air Resources Board to analyze transit bus emissions on a life cycle analysis basis (Figures 15 and 16). Life cycle global warming emissions are almost 75% less than CNG and diesel buses. BEB life cycle NOx emissions are significantly lower than diesel (approximately 80% lower) and CNG buses, including those using the new Near Zero NOx CNG engines (less than 0.02 NOx/brake horsepower-hour). Life cycle particulate matter emissions can be reduced by over 20% (CA electricity mix) when replacing diesel buses with BEBs. When using an energy mix of 50% renewables/50% natural gas, particulate matter emissions are reduced relative to both diesel and CNG buses. As electricity energy production technology continues to develop nationwide, the energy grid will become cleaner and the particulate matter emissions levels will continue to decrease. Additionally, meeting emissions requirements for diesel buses requires sophisticated and sensitive add-on emissions control equipment. Maintaining these emissions control sys- tems can be cumbersome and expensive for transit agencies, while utilizing battery electric Note: CO2e stands for carbon dioxide equivalent. Figure 15. Buses powered by low-carbon fuel blends produce fewer global warming emissions. Source: Chandler et al.

36 Battery electric Buses—State of the practice technology inherently provides an emissions reduction strategy that does not rely on addi- tional vehicle systems. Improvements to power generation are helping reduce the life cycle emissions associated with BEBs. Air pollution control equipment and changes in electricity sources have reduced acid-rain-causing SO2 emissions by 73% from 2006 to 2015 and continue to lower emissions (DeVilbiss and Ray 2017). BEBs allow the United States to extend emissions benefits gained in energy production to the transportation sector. Tools are available to assess life cycle environmental effects attributed to transportation. Argonne National Laboratory’s Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation (GREET) Model was sponsored by the U.S. Department of Energy’s Office of Energy Efficiency and Renewable Energy and “allows researchers and analysts to evalu- ate various vehicle and fuel combinations on a full fuel-cycle/vehicle-cycle basis” (GREET Model 2017). An analysis of deployment of electric buses in New York City documented potential social cost savings. The analysis utilized the U.S. Environmental Protection Agency’s Diesel Emissions Quantifier tool to assess the health benefits of implementing BEBs. The tool “considers the cost of hospitalization, cost of emergency room visits, and the cost of absence from work.” The analy- sis concluded that if New York City Transit switches to a fully electric fleet, the “total health care savings is roughly $100 per NYC resident per year.” The analysis also concludes that the “social cost of carbon” savings over the lifetime of a BEB is “a little over $36,000” (Aber, 2016, page 18). Capital and Operational Costs A common hesitation with investing in any new technology is the upfront costs and the unknowns associated with ongoing operating expenses. According to NREL’s Foothill Transit analysis: Figure 16. Emissions decrease. Source: Chandler et al.

Literature review 37 The capital costs for BEBs are currently higher than that of conventional technology, although the costs have dropped considerably over the last few years as orders for the buses have increased. The increase in orders allows the manufacturers to take advantage of economies of scale to reduce the pro- ductions costs (Eudy et al., 2016, page 18). From 2009 to 2015, Foothill Transit’s per bus purchase cost dropped from $1.2 million to $789,000. For comparison, a CNG bus cost $575,000 in 2015 according to the NREL report (Eudy et al., 2016, page 8). Differences in operational costs for BEBs are primarily driven by fuel efficiency, electric- ity costs, component replacement costs, and maintenance labor reductions. Foothill Transit’s analysis supports claims that BEBs generally have lower maintenance costs. When calculating the maintenance cost per mile, Foothill Transit included the price of parts and labor rates at $50 an hour. Foothill Transit determined that the total BEB cost per mile was $0.08 and $0.09 for scheduled maintenance and unscheduled maintenance, respectively. The scheduled main- tenance cost of CNG was higher at $0.14 per mile for scheduled and lower at $0.04 per mile for unscheduled. Cumulatively, the BEB maintenance cost per mile was 11% lower than CNG buses. Maintenance costs for both bus fleets were low due to the fact that they were still under warranty during the period. Challenges and Lessons Learned Challenges from the Foothill Transit BEB deployment included BEB operations and mainte- nance training, the learning curve that comes along with a new technology, and availability of parts. The report on Foothill Transit’s experience also summarized the lessons learned, which included encouraging strong working relationships between the agency and bus OEM, deploying the BEBs on routes that accommodated their capabilities, adjusting the route schedules instead of trying to fit the BEBs’ needs into the existing schedule, and ensuring that the charging station is readily available (Eudy et al., 2016, pages 36–38). The ZeEUS project partners identified five challenges that must be addressed in Europe. These recommendations are consistent with the needs of the U.S. BEB market and the findings of this synthesis report. The challenges include addressing each of the following: 1. The higher upfront cost of electric buses and their charging infrastructure compared with conventional vehicles. 2. The importance of identifying suitable technology solutions for specific local operational contexts. 3. The necessity to review current procurement and contractual frameworks. 4. The requisite to standardize charging interfaces to ensure the interoperability of e-buses, which allows multibrand fleets to recharge with multibrands infrastructures. 5. The need to develop trust and cooperation with the electricity power generation and distribu- tion sector, as well as with grid owners and energy regulators. Summary An estimated 173,000 electric buses have been deployed worldwide, with more than 170,000 deployed in China. BEB technology is certainly not new and is arguably commercialized in other parts of the world. The planning phase of any purchase starts with developing the business case. Capital costs of BEBs are higher than conventional buses but by all indications are continuing to fall (21% from 2009 to 2015 in Foothill Transit’s case) as a result of technology improvements and economies

38 Battery electric Buses—State of the practice of scale. Transit agencies have several options to address how to offset remaining incremental costs. First, the transit agency may be able to take advantage of federal and state funding oppor- tunities. Second, given that there are potential operations and maintenance cost reductions associated with BEBs, costs should be evaluated and compared on a life cycle or total-cost-of- ownership basis. Maintenance costs for Foothill Transit’s BEBs were reported to be 11% better than CNG costs. With an efficiency that was approximately four times better than diesel and CNG buses, the fuel costs of BEB are typically less. However, variability of electricity costs, and demand charges in particular, can significantly affect these operational costs. Utility rates vary tremendously throughout the country and eligible rate plans vary significantly within a utility. More work is required to understand the true impact of utility rates on BEB fleet operational costs as well as how to analyze rate structures and obtain the most reasonable rates. Ideally, the industry can establish rate plans specifically suited to fleets of BEBs that bring operations on par with conventional buses while still allowing utilities to cover the costs of electricity. In many areas of the country, the costs of charging a BEB are already lower than fueling diesel or CNG buses. New provisions that allow for leasing traction batteries (as opposed to purchasing traction batteries) can help shift some or all of the incremental capital costs to operational costs, creating a scenario that more closely resembles the total cost of ownership and competes with the cost structure of conventional buses. Regulations are also driving BEB deployment. This is particularly evident in California, where the Air Resources Board Fleet Rule for Transit Agencies requires that buses meet strict emissions standards and that, for larger agencies, a percentage of new bus purchases be zero emission. The decision process is complex when planning for bus technical specifications, charging methods, charger locations, and route selection due to a vast set of co-dependent agency and environmental variables. Optimizing a fleet-charging scheme at the outset of a project can significantly lower overall energy costs and the total cost of ownership for BEBs. Given that these decisions can have a significant effect on the cost, performance, and operations for a transit agency, access to evaluation tools and established methodologies is important. Limited tools and technical support are available but more development is needed in this area. While information was provided on why Foothill Transit chose their charging location and made adjustments to schedules to accommodate the buses, it is more anecdotal in nature. Little guidance or information was found in the literature on how agencies should plan for an overall BEB deployment. Foothill Transit identified that scaling up a BEB fleet presents challenges not encountered with the deployment of a small number of vehicles and noted these challenges need to be addressed by the BEB industry. Only three transit agencies are currently operating more than 10 BEBs within their fleets. While larger orders have been placed, transit agencies are trying to understand how to deploy and manage the practical aspects of a larger fleet of BEBs. Challenges are primarily related to managing charging at scale and having the available space and power for large fleets as well as how to make charger connections in an efficient manner. In addition, transit agencies need to ensure that drivers are adequately trained on the operational differ- ences between BEBs and conventional technologies and that recurring training is available. Availability and reliability of BEBs was comparable to that of CNG buses during the Foothill Transit evaluation. Transit agencies are concerned about the durability of the traction battery due to its cost and limited amount of data regarding life expectancy for a BEB application. How- ever, most BEB manufacturers are offering 6-year warranties on the batteries and one manufac- turer is offering a 12-year unconditional warranty.

Literature review 39 BEBs emit no emissions at the tailpipe. When considering electricity production in the total emissions profile, life cycle emissions for BEBs have 75% less global warming emissions and significantly lower NOx emissions than CNG and diesel buses when considering a California energy production mix. Life cycle particulate matter emissions can be reduced by approximately 20% when replacing diesel buses with BEBs. Transitioning bus fleets to electric drive can have a significant positive impact on local, regional, and global emissions. Overall, the current literature provides information regarding early stage BEB deployment; however, published data and summaries on nationwide experience are still needed to assist transit agencies in their efforts to procure BEBs.