The Impacts of Vehicle Automation on the Public Transportation Workforce (2022)

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.

38 C H A P T E R 4 Although transit vehicle automation would likely affect, directly or indirectly, many differ- ent current transit jobs at all organizational levels and could even create new jobs that do not exist today, the research team focused the analysis on a subset of current transit jobs that would likely experience the effects of transit vehicle automation most directly. Based on the selected use cases, the research team created a taxonomy of likely-to-be-affected jobs: directly affected operations jobs and indirectly affected key jobs. 4.1 Directly Affected Operations Jobs Directly affected operations jobs are those front-line and supervisory transit agency jobs that directly operate, supervise, or maintain revenue vehicles. Obviously, there are other front-line and supervisory positions that would be affected, but the research team and panel selected five jobs to be the focus of this research. Because several of the use cases only affected a specific transit mode (e.g., automated BRT only affects BRT service), the researchers subcategorized the directly affected operations jobs that work directly with revenue service (i.e., operators, dispatchers/controllers, and road/street supervisors) into three modes: fixed route (for tradi- tional fixed-route service), BRT, and demand response. The directly affected operations jobs are listed and defined as follows: • Operators: A person who operates (drives) transit vehicles in revenue service. There are several types of operators of interest to this study: – Fixed-route bus operator: A person whose main job is to operate buses that deliver fixed- route transit service (typically local and commuter bus service). Does not include BRT operators. – BRT operator: A person whose main job is to operate buses that deliver BRT service. Does not include other fixed-route operators. – Demand-response operator: A person whose main job is to operate buses and other vehi- cles that deliver demand-response transit service (e.g., dial-a-ride and ADA paratransit service). Does not include taxi or transportation network company operators. • Dispatchers/controllers: A person who supports daily revenue operations and ensures that scheduled runs have an operator and a vehicle. May also interact with operators throughout the day, giving directions, triaging calls, requesting replacement buses during breakdowns, etc. Includes people who perform this task from a control center or radio room and from a bus garage or other interior area. Does not include call center staff that work mostly with customers. Dispatchers/controllers are separated into three types: – Fixed-route bus dispatcher/controllers: A dispatcher/controller whose job is mainly to support fixed-route bus service (typically local or commuter bus service). – BRT dispatcher/controller: A dispatcher/controller whose job is mainly to support BRT service. May also support other fixed-route bus services. Job Profiles of Targeted Transit Jobs

Job Profiles of Targeted Transit Jobs 39   – Demand-response dispatcher/controller: A dispatcher/controller whose job is mainly to support demand-response transit service (e.g., dial-a-ride and ADA paratransit service). For simplicity, the dispatcher/controller position is referred to as dispatcher in the remain- der of the report. • Road or street supervisors/traffic controllers: A person who supports daily revenue operations by working in the field; responding to incidents, accidents, and breakdowns; and supporting operators in person. May conduct performance checks, route monitoring, or other service monitoring functions. Usually drives a transit agency non-revenue vehicle to travel from task to task. Road/street supervisors are categorized into three types: – Fixed-route bus supervisor: A road/street supervisor whose job is mainly to support fixed- route bus service (typically local or commuter bus service). – BRT supervisor: A road/street supervisor whose job is mainly to support BRT service. May also support other fixed-route bus services. – Demand-response supervisor: A road/street supervisor whose job is mainly to support demand-response transit service (e.g., dial-a-ride and ADA paratransit service). For simplicity, the road or street supervisor/traffic controller position is referred to as supervisor in the remainder of the report. • Bus mechanics/maintenance technicians: A person whose main job is to inspect, main- tain, and repair transit vehicles used in revenue service. May require significant techni- cal expertise and knowledge of vehicle systems and components. For simplicity, the bus mechanic/maintenance technician position is referred to as mechanic in the remainder of the report. • Bus service persons/fuelers/cleaners: A person whose main job is to perform basic tasks that ready vehicles for revenue service. Tasks may include interior cleaning, exterior cleaning, fueling, topping off fluids, and performing basic maintenance tasks like changing the oil or replacing wiper blades. Tasks do not include performing complex maintenance and repair activities. For simplicity, the bus service person/fueler/cleaner position is referred to as service person in the remainder of the report. Appendix B presents each directly affected operation’s job profile, including the job title, typical job functions, and priority KSAs. 4.1.1 Directly Affected Workforce Demographics The research team found that comprehensive demographic profiles of transit jobs do not exist, and there is somewhat limited research on the composition of the transit workforce. However, the U.S. Bureau of Labor Statistics (BLS) has data on the characteristics of bus operators in general. (BLS statistics are not limited to transit bus operators; however, the characteristics of all bus operators in the BLS data are likely similar to the characteristics of transit bus operators.) The preliminary analysis by the USDOT (2021) of the potential workforce effects of driving automation systems provides some helpful information to bet- ter understand the bus operator workforce using BLS data. The USDOT (2021) report’s bus driver demographic data are from the Annual Social and Economic Supplement of the Cur- rent Population Survey (ASEC CPS) 2003–2017, administered by the U.S. Census Bureau. A few highlights from the USDOT report are: • Bus operators are disproportionately located in urban areas when compared to other blue-collar jobs (90 percent reside in urban areas, compared to 81 percent of other blue- collar jobs). • Bus operators are disproportionately women and African Americans (see Figure 5).

40 The Impacts of Vehicle Automation on the Public Transportation Workforce • Bus operators are generally older than other blue-collar workers (see Figure 6). • Bus operators tend to have less advanced degrees than other blue-collar workers (see Figure 7). These characteristics of the bus operator workforce suggest that, overall, bus operators are more likely to be women, be non-white, and have lower educational attainment than other blue-collar workers. This finding suggests not only that transit automation impacts will “fall disproportion- ately on groups that have historically faced considerable employment discrimination” (USDOT 2021, 49) but also that lower educational attainment may make the potential automation-induced disruption to the bus operator job harder for current employees to adapt to. Another important characteristic from the USDOT’s analysis is the age of current bus opera- tors, which is higher than other blue-collar workers. In fact, “if the age profile of [operators] is maintained, in 10 years, 38 percent of the current bus driving workforce will be 65 years or older, and in 15 years, the number will be 55 percent” (USDOT 2021, 50). With significant transit vehi- cle automation still likely years if not decades away, the bus operator workforce could be signifi- cantly older. This could have two effects. First, the higher-aged bus operators may find it more difficult to translate their skills to changed or new positions, and they may find it more difficult Source: USDOT (2021, 49). Source data are from ASEC CPS data 2003–2017. Figure 5. Location, gender, and race of bus operators compared to other blue-collar workers. Source: USDOT (2021, 50). Source data are from ASEC CPS data 2003–2017. Figure 6. Age of bus operators compared to other blue-collar workers.

Job Profiles of Targeted Transit Jobs 41   to obtain the credentials needed to qualify for new, technology-driven automation-related jobs. Second, some bus operators may opt to retire out of their positions instead of transitioning to new positions. The potential for significant retirements in the wake of automation could be seen as beneficial—using attrition instead of reductions in force to offset the lower need for bus operator jobs—and as a concern—a loss of a highly experienced workforce at a key moment of transition in the transit industry. Although the age profile of bus operators is available, this research did not attempt to calculate the counterbalancing effects of potential operator retirements and potential FTE reductions (or increases) because the timeline for large-scale adoption of each use case is still highly uncertain. 4.1.2 Directly Affected Jobs Current FTE Counts To estimate the effects of transit vehicle automation on the public transit workforce, the research team had to have an estimate of the number of current employees in the directly affected transit jobs. However, the research team found not only that demographic profiles of the current transit workforce do not exist but also that there is no publicly available database of the number of people employed in specific transit jobs. The research team identified two potential sources of data; however, both data sources had shortcomings. The NTD contains full- and part-time employee counts by transit agency, mode, and func- tion (i.e., vehicle operations, vehicle maintenance). However, NTD data have the following limitations: • Data are only available from full reporters. Rural reporters and reduced urban reporters do not report employee counts. • Employee count data are only for directly operated services. Thus, employees in the transit industry that work for private contractors are excluded. • The only directly affected operations job that is directly measured in the NTD is bus opera- tors. There are no data elements that separate out dispatchers, supervisors, mechanics, and service persons. BLS data contain current employment estimates for many industries and positions; how- ever, the BLS data do not completely align with how the transit industry classifies itself or Source: USDOT (2021, 51). Source data are from ASEC CPS data 2003–2017. Figure 7. Bus operator educational attainment compared to other blue-collar workers.

42 The Impacts of Vehicle Automation on the Public Transportation Workforce its positions. Federal statistical agencies rely on standard taxonomies to classify industry and occupation, the North American Industry Classification System (NAICS) and the BLS’s Stan- dard Occupational Classification (SOC) system, respectively. While this common taxonomy assists statistical agencies in presenting data in standard formats, it often does not align with a specialized analysis of more nuanced industry and occupation classifications, as were needed in this case. For example: • Public transit is not an industry in the BLS data. In fact, to attempt to capture the public transit industry from BLS data, several industries have to be included, some of which contain posi- tions that likely fall outside of the boundaries of providing public transit. • The five directly affected positions are not unique positions in BLS data. The BLS occupation “Bus Drivers, Transit and Intercity” (SOC 53-3052) is the closest direct match to the bus operator job; however, other directly affected jobs are only approximately matched by different BLS occupations. • BLS data do not disaggregate positions or industries according to the same transit agency categories needed in this research (i.e., rural, small urban, and large urban) or according to transit mode. Therefore, the research team had to produce an original methodology to estimate the current number of employees in the directly affected operations jobs. The methodology had to create FTE counts for each directly affected job, transit mode, and transit agency type (rural, small urban, and large urban). Appendix A contains the full details of the methodol- ogy used by the research team; however, in summary, the research team surveyed 30 transit agencies (10 rural, 10 small urban, and 10 large urban) with bus service as of 2018. The transit agencies received an invitation email and an online survey link that asked for directly affected operations job staffing counts by mode (see Attachment 1 for the full survey). Staffing counts at agencies that did not separate counts by mode were allocated to operated modes using NTD service data for those agencies. The research team compared staffing count survey data with the amount of service provided by each surveyed agency to generate average staffing ratio estimates (in FTEs) for each position and agency type (e.g., the average number of fixed-route dispatcher jobs per revenue hour at large urban transit agencies). If survey-based staffing ratio estimates appeared unreasonable or out of line with NTD data, the research team used NTD-based staffing ratios to the extent possible. Then, the research team expanded staffing ratios across all NTD reporting agencies to estimate the number of jobs in each position and agency type. Last, the results of the staffing counts were compared with other sources (e.g., the NTD and BLS) to validate the reasonableness of each estimate. BRT operator job counts were based on actual NTD data because the NTD data accu- rately reflected the number of BRT operators. Table 8 shows the final results of the staffing count estimation. The FTE estimates in Table 8 were used by the research team when estimating the potential workforce effects of each transit automation use case. 4.2 Indirectly Affected Key Jobs Indirectly affected key jobs are those front- and second-line and supervisory transit agency jobs that work directly with operators, transit planning and scheduling, and vehicle and yard O&M. These jobs will be indirectly affected due to their proximity to directly affected operations jobs. The indirectly affected key jobs are listed and defined as follows: • Bus garage superintendent: A person who manages a bus garage, division, or yard and super- vises operations at the garage, including the operators, dispatchers, timekeepers, or others

Job Profiles of Targeted Transit Jobs 43   who are assigned to his or her garage. May be responsible for ensuring that all open runs are filled and pull-out is made daily. Includes both fixed-route bus and demand-response bus garages. • Bus operations trainer: A person whose main job is to train bus operators—including training new hires and retraining or recertifying current operators. Includes both fixed-route bus and demand-response trainers. • Maintenance trainer: A person whose main job is to train bus mechanics/technicians— including training new hires and retraining or recertifying current mechanics/technicians. • Parts clerk: A person whose main job is to inventory, manage, and distribute bus parts to mechanics/technicians. Usually works in the bus garage, immediately proximal to bus mechanics/technicians. Does not include general storeroom personnel who provide parts and inventory support to multiple, non-bus-maintenance business units. • O&M facilities maintainer: A person whose main job is to maintain bus O&M facilities, includ- ing maintaining electrical, HVAC (or heating, ventilation, and air conditioning), plumbing, and communications systems as well as buildings and yard infrastructure. • Short-range transit planner/schedule maker: A person whose main job is to prepare transit schedules (trips, blocks, runs, and rosters) for fixed-route bus services—may also include creating and scheduling demand-response runs. Also includes a person whose main job is to prepare, evaluate, and analyze short-range transit plans for fixed-route bus services. Short- range planning usually encompasses the next 1 to 5 years. • Transit safety and security personnel: A person whose main job is to protect transit agency assets, employees, and passengers from security incidents and crimes. Includes sworn police Agency Type Position Rural Small Urban Large Urban Grand Total Bus Operator (Total) 17,834 13,266 136,522 167,623 Fixed-Route Operator 4,870 8,967 102,419 116,256 Demand-Response Operator 12,964 4,299 33,435 50,698 BRT Operator 0 0 669 669 Dispatcher (Total) 1,791 705 7,238 9,733 Fixed-Route Dispatcher 374 166 3,723 4,264 Demand-Response Dispatcher 1,416 539 3,429 5,384 BRT Dispatcher 0 0 86 86 Supervisor (Total) 691 623 4,133 5,447 Fixed-Route Supervisor 140 401 2,779 3,321 Demand-Response Supervisor 550 222 1,306 2,079 BRT Supervisor 0 0 48 48 Mechanic (Total) 2,479 926 21,207 24,612 Fixed-Route Mechanic 680 725 18,767 20,172 Demand-Response Mechanic 1,800 201 2,324 4,324 BRT Mechanic 0 0 116 116 Service Person (Total) 495 558 12,123 13,176 Fixed-Route Service Person 495 452 11,246 12,193 Demand-Response Service Person 0 106 805 910 BRT Service Person 0 0 73 73 Note: BRT FTE estimates did not include BRT service offered by the Roaring Fork Transportation Authority. Table 8. FTE estimates for the directly affected operations jobs.

44 The Impacts of Vehicle Automation on the Public Transportation Workforce officers working for the transit agency or for local/state governments and civilian security personnel (e.g., security guards). Responds to safety and security incidents, acts as a first responder to emergencies, and works to prevent crime and other violations on the transit system. Officers who also work to solve crimes. Appendix C presents each indirectly affected job profile, including the job title, typical job functions, and priority KSAs. Because the job gains or losses on the indirectly affected key jobs were not estimated, the research team did not prepare FTE estimates for these jobs.