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.

A-1   Description of Methodology to Estimate FTEs per Directly Affected Operations Job The workforce effect calculator in this study required the current number of employees in the service type to which the use case applied. This appendix describes the methods to collect data about the current transit employee counts. The current number of transit employees were estimated by transit agency type (i.e., rural, small urban, and large urban) and by mode (i.e., fixed-route, demand-response, and BRT) for the following directly affected operations jobs: • Operator. • Dispatcher/controller (referred to as dispatcher). • Road/street supervisor or traffic controller (referred to as supervisor). • Bus mechanic/maintenance technician (referred to as mechanic). • Bus service person/fueler/cleaner (referred to as service person). To capture the current count of employees, the research team conducted online surveys of transit agencies, asking them to report their staffing counts. Using these survey results and NTD data, the research team estimated the current number of FTEs. NTD and BLS data were used to assess the quality of the estimation for the number of employees. Details of this process are provided in the following sections. Staffing Count Survey Because transit industry staffing data were not publicly available (neither through NTD nor BLS), the research team conducted a transit agency survey to collect staffing counts from a set of representative transit agencies. (Although NTD and BLS data can be used for the estimation of bus operators, other jobs in this study were not well matched with the NTD or BLS job categories.) The survey, developed in the Qualtrics® online survey platform, collected data on the current number of employees in the directly affected operations jobs and the indirectly affected key jobs (see Table A-1). The research team realized that transit agencies may use different job titles than those listed in Table A-1. To maintain consistency of reporting across agencies, the research team provided brief descriptions of the jobs and requested that transit agencies submit their data for positions that were most closely related to the ones described (regardless of whether the job title matched precisely). For the directly affected operations jobs (i.e., operator, dispatcher, and street super- visor), the survey also asked whether the transit agency separated the positions by bus modes (including fixed-route, demand-response, and BRT). (In this study, fixed-route bus included local bus, trolley bus, and commuter bus.) The research team selected 30 transit agencies for the survey and distributed the online survey link to them via email. The selected transit agencies included 10 rural, 10 small urban, and 10 large A P P E N D I X A

A-2 The Impacts of Vehicle Automation on the Public Transportation Workforce urban agencies, based on the service size [i.e., total vehicles operated in annual maximum service (VOMS)] data obtained from the NTD. The research team also considered each agency’s regional location to ensure the diversity of the sample. Table A-2 shows the list of the selected agencies for the staffing count survey. Among the selected agencies, a total of 14 agencies submitted the data. Of the respondents, four agencies were rural, five were small urban, and five were large urban agencies. The research team summed up the total employees reported by job and mode. Part-time employees were counted as 0.5 FTEs. Table A-3 shows a summary of the survey results. Next, the research team calculated the ratio of revenue hours to FTEs for the job categories of the operator, dispatcher, and supervisor. For the mechanic and service person categories, the Directly Affected Operations Jobs Indirectly Affected Key Jobs • Bus operator • Dispatcher/controller • Street/road supervisor/traffic controller • Bus mechanic/maintenance technician • Bus service person/fueler/cleaner • Transit safety and security personnel • Short-range transit planner/schedule maker • Parts clerk • Facilities maintainer • Bus operations trainer • Bus garage superintendent • Maintenance trainer Table A-1. Jobs expected to be impacted by transit automation. Agency Name Agency Type VOMS a Location Response Ark-Tex Council of Governments Rural 45 Texarkana, TX Yes Grant County Transportation Authority Rural 37 Moses Lake, WA Yes MIDAS Council of Governments Rural 32 Fort Dodge, IA Yes Northern Oklahoma Development Authority Rural 48 Enid, OK Yes Marble Valley Regional Transit District Rural 43 Rutland, VT No Bolivar County Council On Aging, Inc. Rural 53 Cleveland, MS No Jackson County Transportation, Inc. Rural 30 Marianna, FL No Delta Human Resource Agency Rural 34 Covington, TN No Columbiana County/Community Action Rural Transit System Rural 36 Lisbon, OH No KY River Foothills Development Council, Inc. Rural 42 Richmond, KY No Brazos Transit District Small Urban 75 Bryan, TX Yes Champaign-Urbana Mass Transit District Small Urban 113 Urbana, IL Yes City of Gainesville Small Urban 147 Gainesville, FL Yes Duluth Transit Authority Small Urban 68 Duluth, MN Yes Santa Cruz Metropolitan Transit District Small Urban 105 Santa Cruz, CA Yes Ames Transit Agency Small Urban 85 Ames, IA No City of Kenosha Small Urban 55 Kenosha, WI No Erie Metropolitan Transit Authority Small Urban 117 Erie, PA No Greater Lafayette Public Transportation Corporation Small Urban 61 Lafayette, IN No Santa Barbara Metropolitan Transit District Small Urban 93 Santa Barbara, CA No City of Charlotte Large Urban 330 Charlotte, NC Yes Denver Regional Transportation District Large Urban 1,274 Denver, CO Yes Los Angeles County Metropolitan Transportation Authority Large Urban 1,916 Los Angeles, CA Yes The Greater Cleveland Regional Transit Authority Large Urban 430 Cleveland, OH Yes Utah Transit Authority Large Urban 573 Salt Lake City, UT Yes Central Florida Regional Transportation Authority Large Urban 420 Orlando, FL No City of Detroit Large Urban 309 Detroit, MI No City of Phoenix Public Transit Department Large Urban 526 Phoenix, AZ No Dallas Area Rapid Transit Large Urban 639 Dallas, TX No Tri-County Metropolitan Transportation District of Oregon Large Urban 786 Portland, OR No aVehicles operated in annual maximum service (VOMS). Table A-2. List of the selected transit agencies for the survey.

Description of Methodology to Estimate FTEs per Directly Affected Operations Job A-3   ratio of total vehicles available for maximum service (VAMS) to FTEs was calculated. However, due to rural agencies not reporting VAMS in the NTD, revenue hours were initially used instead of VAMS for the mechanic and service person service data. Table A-4 presents the calculated service to employee ratios for the surveyed agencies. To estimate the current number of employees in the overall transit industry, the research team multiplied a service to employee ratio for a specific agency type, job, and mode by the total Rural (n = 4) Small Urban (n = 5) Large Urban (n = 5) Operator Fixed route 24.0 659.5 7,073.0 Demand response 94.0 75.0 226.0 BRT 0.0 0.0 102.5 Dispatcher Fixed route 4.0 13.3 115.1 Demand response 8.5 9.7 20.4 BRT 0.0 0.0 5.0 Supervisor Fixed route 1.5 32.0 85.9 Demand response 3.0 4.0 7.8 BRT 0.0 0.0 2.8 Mechanic Fixed route 2.7 68.8 1,682.9 Demand response 1.3 18.2 71.1 BRT 0.0 0.0 34.9 Service person Fixed route 2.0 42.9 1,008.5 Demand response 0.0 9.6 24.6 BRT 0.0 0.0 21.9 Table A-3. Survey results of the FTE employees for the directly affected jobs. Table A-4. Calculated service to employee ratios for the surveyed agencies. aFor the rural transit agencies, the numbers indicate revenue hours per employee. Rural (n = 4) Small Urban (n = 5) Large Urban (n = 5) Revenue Hours per Employee Operator Fixed route Demand response BRT Dispatcher Fixed route Demand response BRT Supervisor Fixed route Demand response BRT VAMS per Employeea Mechanic Fixed route Demand response BRT Service person Fixed route Demand response BRT 3,076.2 1,301.2 N/A 18,457.3 14,389.9 N/A 49,219.3 37,029.0 N/A 36,914.5 1,538.3 1,629.7 N/A 76,531.9 12,546.5 N/A 31,724.2 30,405.3 N/A 10.6 1,635.7 1,263.1 1,768.9 41,495.4 14,007.4 12,443.0 55,590.3 36,776.2 22,172.9 26,867.4 88,720.6 N/A 6.6 7.9 N/A 2.4 2.1 2.0 4.0 N/A N/A 15.1 N/A 6.2 3.1

A-4 The Impacts of Vehicle Automation on the Public Transportation Workforce Agency Type and Position Service to Employee Ratio Ratio Source Service Data (2018 NTD) Rationale Rural Revenue hours per operator FTE NTD small urban reporters’ directly operated service Revenue hours from rural reporters Surveyed rural agency revenue hours per operator ratios were not reasonable. Revenue hours per dispatcher FTE Surveyed rural transit agencies Revenue hours from rural reporters Surveyed rural agency revenue hours per dispatcher ratios were reasonable, and the NTD did not have data on dispatcher counts. Revenue hours per supervisor FTE Surveyed rural transit agencies Revenue hours from rural reporters Surveyed agency revenue hours per supervisor ratios were reasonable, and the NTD did not have data on supervisor counts. Operator Dispatcher Supervisor Mechanic and Service Persona Revenue miles per vehicle maintenance FTE NTD small urban reporters’ directly operated service Revenue miles from rural reporters Surveyed agency revenue miles per mechanic ratios were not reasonable. Therefore, the research team decided to use NTD data from small urban reporters to calculate the rural revenue miles per mechanic and service person ratio. Revenue miles were used in the service data, because the rural agencies did not report VAMS to the NTD. Urban (Small and Large Calculated Separately) Revenue hours per operator FTE NTD reporters’ directly operated service Revenue hours Surveyed agency revenue hours per operator ratios were not reasonable. Revenue hours per dispatcher FTE Surveyed transit agencies Revenue hours Surveyed agency revenue hours per dispatcher ratios were reasonable, and the NTD did not have data on dispatcher counts. Revenue hours per supervisor FTE Surveyed transit agencies Revenue hours Surveyed agency revenue hours per supervisor ratios were reasonable, and the NTD did not have data on supervisor counts. Operator Dispatcher Supervisor Mechanic and Service Personb VAMS per vehicle maintenance FTE NTD reporters’ directly operated service VAMS Surveyed agency VAMS per mechanic ratios were not reasonable when compared to NTD data. Therefore, the research team decided to use NTD data to calculate the VAMS per mechanic ratio. aThe NTD small urban reporters’ revenue miles per vehicle maintenance FTE ratio was used to estimate the total number of vehicle maintenance FTEs at rural agencies; however, the total number of vehicle maintenance FTEs was then split between mechanics and service persons to match to proportions of mechanics and service persons reported by the surveyed rural agencies. bThe NTD small and large urban reporters’ VAMS per vehicle maintenance FTE ratio was used to estimate the total number of vehicle maintenance FTEs at small and large urban agencies, respectively. However, the total number of vehicle maintenance FTEs was then split between mechanics and service persons to match to proportions of mechanics and service persons reported by the surveyed urban agencies. Table A-5. Service to employee ratios and service data used to estimate FTEs. service data reported in the NTD in that agency type and mode. Originally, the research team planned to multiply service data by the service to employee ratios reported by surveyed transit agencies; however, some survey ratios did not make sense (e.g., a rural fixed-route operator could not deliver over 3,000 revenue hours of service in a year). Therefore, the research team used a combination of NTD-based and survey-based ratios for different positions and agency types. The research team used NTD-based ratios for operators and mechanics and used survey- based ratios for dispatchers and supervisors. Table A-5 describes the ratios and service data used for each position and agency type. The actual calculations (and ratios used) are shown in Table A-6 through Table A-11. Thus, the total industry’s revenue hours and VAMS were multiplied by the ratio calculated from the sample to estimate the number of employees in each job category. Table A-12 shows the final number of employees in each job category that the research team has estimated.

Description of Methodology to Estimate FTEs per Directly Affected Operations Job A-5   Position and Mode Revenue Hoursa Revenue Hours per FTEb Estimated FTEsc Operator 27,288,457 17,834 Fixed route 6,908,592 1,418.7d 4,870 Demand response 20,379,865 1,572.0d 12,964 BRTe Dispatcher 27,288,457 1,791 Fixed route 6,908,592 18,457.3 374 Demand response 20,379,865 14,389.9 1,416 BRTe Supervisor 27,288,457 691 Fixed route 6,908,592 49,219.3 140 Demand response 20,379,865 37,029.0 550 BRTe aRevenue hours were obtained from the 2018 NTD, rural reporters only. bUnless otherwise specified, revenue hours per FTE were obtained from transit agencies that responded to TTI’s staffing count survey. cEstimated FTEs were calculated by multiplying the revenue hours by revenue hours per FTE. dRevenue hours per operator FTE were based on data reported to the NTD by small urban transit operators. Rural reporters did not report operator counts (or any other staff count) to the NTD, and the service to employee ratios reported by surveyed transit agencies were not reasonable. eAlthough there was one rural transit agency that operated BRT service, the research team chose to not single out this particular transit agency when generating workforce effect estimates. Therefore, the team did not calculate current FTE estimates for rural BRT service. Table A-6. Rural FTE estimation final calculations for operators, dispatchers, and supervisors. Mode Revenue Milesa Revenue Miles per Vehicle Maint. FTEb Estimated Vehicle Maint. FTEsc Mechanic Proportione Estimated Mechanic FTEsf Estimated Service Person FTEsg Total 481,068,189 2,974 2,479 495 Fixed Route 126,713,472 107,909.7 1,174 58% 680 495 Demand Response 354,354,717 196,883.2 1,800 100% 1,800 0 BRTh a Revenue miles were obtained from the 2018 NTD, rural reporters only. b Revenue miles per vehicle maintenance FTE were obtained from small urban reporters’ NTD data for directly operated service. c Estimated vehicle maintenance FTEs were calculated by multiplying the revenue miles by revenue miles per vehicle maintenance FTE. e Mechanic proportions were based on rural surveyed transit agencies. The values represented the percentage of reported maintenance employees that were designated as mechanics (instead of service persons). f Estimated mechanic FTEs were calculated by multiplying the mechanic proportion by the estimated vehicle maintenance FTEs. g Estimated service person FTEs were calculated by subtracting the estimated mechanic FTEs from the estimated vehicle maintenance FTEs. h Although there was one rural transit agency that operated BRT service, the research team chose to not single out this particular transit agency when generating workforce effect estimates. Therefore, the team did not calculate current FTE estimates for rural BRT service. Table A-7. Rural FTE estimation final calculations for mechanics and service persons.

A-6 The Impacts of Vehicle Automation on the Public Transportation Workforce Position and Mode Revenue Hoursa Revenue Hours per FTEb Estimated FTEsc Operator 19,479,433 13,266 Fixed route 12,720,966 1,418.7 8,967 Demand response 6,758,467 1,572.0 4,299 BRTd Dispatcher 19,479,433 705 Fixed route 12,720,966 76,531.9 166 Demand response 6,758,467 12,546.5 539 BRTd Supervisor 19,479,433 623 Fixed route 12,720,966 31,724.2 401 Demand response 6,758,467 30,405.3 222 BRTd a Revenue hours were obtained from the 2018 NTD, small urban reporters only. b Unless otherwise specified, revenue hours per FTE were obtained from transit agencies that responded to TTI’s staffing count survey. c Estimated FTEs were calculated by multiplying the revenue hours by revenue hours per FTE. d There were no small urban transit agencies that operated BRT service. Table A-8. Small urban FTE estimation final calculations for operators, dispatchers, and supervisors. Mode VAMS a VAMS per Vehicle Maint. FTEb Estimated Vehicle Maint. FTEsc Mechanic Proportiond Estimated Mechanic FTEse Estimated Service Person FTEsf Total 8,233 1,484 926 558 Fixed Route 4,725 4.01 1,178 62% 725 452 Demand Response 3,508 11.45 306 65% 201 106 BRTg a VAMS was obtained from the 2018 NTD, small urban reporters only. b VAMS per vehicle maintenance FTE were obtained from small urban reporters’ NTD data for directly operated service. c Estimated vehicle maintenance FTEs were calculated by multiplying the VAMS by VAMS per vehicle maintenance FTE. d Mechanic proportions were based on small urban surveyed transit agencies. The values represented the percentage of reported maintenance employees that were designated as mechanics (instead of service persons). e Estimated mechanic FTEs were calculated by multiplying the mechanic proportion by the estimated vehicle maintenance FTEs. f Estimated service person FTEs were calculated by subtracting the estimated mechanic FTEs from the estimated vehicle maintenance FTEs. g There were no small urban transit agencies that operated BRT service. Table A-9. Small urban FTE estimation final calculation for mechanics and service persons.

Description of Methodology to Estimate FTEs per Directly Affected Operations Job A-7   Operator 203,598,360 136,522 Fixed route 154,505,818 1,508.6 102,419 Demand response 48,025,435 1,436.4 33,435 BRT 1,067,107 1,595.6 669 Dispatcher 203,598,360 7,238 Fixed route 154,505,818 41,495.4 3,723 Demand response 48,025,435 14,007.4 3,429 BRT 1,067,107 12,443.0 86 Supervisor 203,598,360 4,133 Fixed route 154,505,818 55,590.3 2,779 Demand response 48,025,435 36,776.2 1,306 BRT 1,067,107 22,172.9 48 Position and Mode Revenue Hoursa Revenue Hours per FTEb Estimated FTEsc a Revenue hours were obtained from the 2018 NTD, large urban reporters only. b Unless otherwise specified, revenue hours per FTE were obtained from transit agencies that responded to TTI’s staffing count survey. c Estimated FTEs were calculated by multiplying the revenue hours by revenue hours per FTE. Table A-10. Large urban FTE estimation final calculations for operators, dispatchers, and supervisors. Mode VAMSa VAMS per Vehicle Maint. FTEb Estimated Vehicle Maint. FTEsc Mechanic Proportiond Estimated Mechanic FTEse Estimated Service Person FTEsf Total 85,474 33,330 21,207 12,123 Fixed Route 58,734 1.96 30,013 63% 18,767 11,246 Demand Response 26,344 8.42 3,129 74% 2,324 805 BRT 396 2.10 189 61% 116 73 a VAMS was obtained from the 2018 NTD, large urban reporters only. b VAMS per vehicle maintenance FTE were obtained from large urban reporters’ NTD data for directly operated service. c Estimated vehicle maintenance FTEs were calculated by multiplying the VAMS by VAMS per vehicle maintenance FTE. d Mechanic proportions were based on large urban surveyed transit agencies. The values represented the percentage of reported maintenance employees that were designated as mechanics (instead of service persons). e Estimated mechanic FTEs were calculated by multiplying the mechanic proportion by the estimated vehicle maintenance FTEs. f Estimated service person FTEs were calculated by subtracting the estimated mechanic FTEs from the estimated vehicle maintenance FTEs. Table A-11. Large urban FTE estimation final calculation for mechanics and service persons.

A-8 The Impacts of Vehicle Automation on the Public Transportation Workforce Quality Assessment of the Estimation The research team compared the estimated number of current employees with other sources, including BLS values and 2018 NTD data. All BLS occupations were filtered down to include only the following industries: • Urban Transit Systems (485100). • Interurban and Rural Bus Transportation (485200). • Other Transit and Ground Passenger Transportation (485900). • State Government, Excluding Schools and Hospitals [Occupational Employment Statistic (OES) Designation] (999200). • Local Government, Excluding Schools and Hospitals (OES Designation) (999300). Table A-13 summarizes the comparison results. Table A-14 presents NTD job counts reported compared to the estimates. The NTD employee data only included directly operated services for urban full reporters. For bus modes, the agencies reporting employee data (directly operated services from full reporters) operated: • Fifty-seven percent of total bus revenue hours (142,891,618 out of 250,433,629 revenue hours1). • Fifty-two percent of total bus revenue miles (1,710,033,988 out of 3,298,063,095 revenue miles2). Rural Small Urban Large Urban Revenue Hours (All Agencies in NTD) Fixed route 6,908,592 12,720,966 154,505,818 Demand response 20,379,865 6,758,467 48,025,435 BRT 67,379 0 1,067,107 VAMS (All Agencies in NTD) Fixed route N/A 4,725 58,734 Demand response N/A 3,508 26,344 BRT N/A 0 396 Estimated FTEs (All Agencies in NTD) Operator Fixed route 4,870 8,967 102,419 Demand response 12,964 4,299 33,435 BRT 0 0 669 Dispatcher Fixed route 374 166 3,723 Demand response 1,416 539 3,429 BRT 0 0 86 Supervisor Fixed route 140 401 2,779 Demand response 550 222 1,306 BRT 0 0 48 Mechanic Fixed route 680 725 18,767 Demand response 1,800 201 2,324 BRT 0 0 116 Service person Fixed route 495 452 11,246 Demand response 0 106 805 BRT 0 0 73 Table A-12. 2018 NTD data for all agencies and the final FTE estimation. 1 NTD 2018 Sum of Actual Vehicle/Passenger Car Revenue Hours for CB, DR, MB, RB, and TB. 2 NTD Sum of Actual Vehicles/Passenger Car Revenue Miles for CB, DR, MB, RB, and TB.

Description of Methodology to Estimate FTEs per Directly Affected Operations Job A-9   Position Estimates BLS Value Notes (with the Research Team’s Interpretation) Bus Operator 167,623 131,880 BLS occupation: Bus Drivers, Transit and Intercity (SOC 53-3052). Note: Does not include other passenger vehicle drivers in SOC 53-3053 (Shuttle Drivers and Chauffeurs) or SOC 53-3022 (Bus Drivers, School or Special Client), which are part of the 53-3058 aggregation. Dispatcher 9,733 18,270 BLS occupation: Dispatchers, Except Police, Fire, and Ambulance (SOC 43-5032). BLS value would include non-bus modes and other non- transit jobs in state and local governments. Ratio of 17.2 operators per dispatcher and overall count compared to BLS appeared reasonable. Road/Street Supervisor 5,447 23,660 BLS occupation: First-Line Supervisors of Transportation Workers, Except Aircraft Cargo Handling Supervisors (SOC 53-1047). BLS value would include non-bus modes and other non- transit jobs in state and local governments. Ratio of 30.8 operators per supervisor and overall count compared to BLS appeared reasonable. Mechanic 24,612 35,470 BLS occupation: Bus and Truck Mechanics and Diesel Engine Specialists (SOC 49-3031). BLS value included some mechanics that did not work on transit buses (e.g., local or state government fleets). Operator to mechanic ratio was 6.8. Bus Service Person 13,176 7,770 BLS occupation: Cleaners of Vehicles and Equipment (SOC 53-7061). BLS value probably excluded people who performed basic maintenance. Appeared reasonable compared to mechanic counts. Mechanic to service person ratio was 1.9; operator to service person ratio was 12.7. Total Mechanics + Service People 37,788 43,240 Sum of the two BLS occupations above. Total appeared reasonable, albeit perhaps a little high. Table A-13. Comparison between the estimates and BLS values. Position Estimates NTD FTEs Actual NTD Percentage Expected Notes (with the Research Team’s Interpretation) Bus Operator 167,623 95,181 56% 57% NTD field: Operator (Vehicle Operations) [full-time + ½ part-time]. Dispatcher + Road/ Street Supervisor 15,181 14,365 95% 57% NTD field: Non-operator (Vehicle Operations). NTD value included MANY other vehicle-operations positions, including scheduling, fare collection, security, training, clerical work, etc. Difficult to compare values and make judgments. Total Mechanics + Service People 37,788 25,955 67% 52% NTD field: Non-operator (Vehicle Maintenance) [full-time + ½ part- time]. NTD job counts included non- mechanic positions allocated to the vehicle maintenance function, for example, trade specialists (e.g., welders), trainers, managers, and clerks, which would increase the NTD job counts. Table A-14. Comparison between the estimates and 2018 NTD values.

A-10 The Impacts of Vehicle Automation on the Public Transportation Workforce Assuming that the research team’s job counts and NTD job counts represented those same job classifications, the NTD job counts should be about 57 percent of TTI’s estimated job counts for revenue-hour-based positions and 52 percent of TTI’s estimated job counts for revenue- mile-based positions. Of course, NTD’s data elements only aligned exactly with the bus operator position. Other NTD FTE data elements included additional positions beyond the ones targeted by the research team. So, the only step forward was to compare the NTD job counts as a percentage of the research team’s job counts and see if the resulting percentage was within an expected range based on an understanding of what was reported in the NTD.