ADA Paratransit Service Models (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.

1 Transit agencies with ADA complementary paratransit obligations have a choice of the type of service model/design. While smaller to mid-size transit agencies have typically used a single fleet of dedicated vehicles, operated in-house or with a turnkey contractor, large systems have chosen a variety of approaches. Whereas a few have elected to operate service totally in-house or with a single turnkey, many more large agencies have taken a variety of approaches, including: • Establishing a split structure in which the transit agency performs some functions and its contractor(s) provide other functions. • Contracting with a broker to establish a multi-carrier design, retain the service providers, and in some cases, perform some or all of the call and control functions. • Contracting with a call and control center management firm to perform some or all of the call and control functions, while preserving the direct transit agency–service provider contractual relationship. • Utilizing multiple carriers directly or indirectly to provide service, and organizing the work among the carriers based on zone assignments or unzoned packages of work. • Utilizing non-dedicated service providers (NDSPs)—mostly taxis—in an integrated fash- ion to serve ADA paratransit trips that have not been scheduled onto the dedicated fleet(s) and to provide an additional resource for dispatchers. Depending on the service model, these NDSPs can be directly procured by the transit agency or broker, or procured by a service provider contractor. Historically, paratransit service models and designs for ADA paratransit service have formed and evolved organically, with small changes implemented to address specific needs. However, as ADA paratransit demand has continued to grow, there is increasing pressure on transit agencies to be as cost efficient as possible while maintaining service quality standards. As a result, some transit agencies are questioning whether their cur- rent service model is the optimal platform to achieve their desired balance between service quality and cost efficiency, and are reviewing alternative ways to restructure their model. However, there has been a lack of detailed information about current ADA paratransit service models and the underlying reasons why specific transit agencies have opted to keep or change their service model. Accordingly, the Transportation Research Board’s Transit Cooperative Research Program commissioned this synthesis on ADA para transit service models. A sub-focus of the synthesis is how transit agencies with contractors orga- nize the work and pay for the services. S u m m a r y ADA Paratransit Service Models

2 aDa Paratransit Service models The Synthesis Effort The synthesis effort focused first on a comprehensive literature review. This yielded a robust amount of information, which included national studies that focused on, or included, treatises on service model options for paratransit. From the literature review, the synthe- sis team identified specific studies performed for individual transit agencies that included discussions of various applicable service model alternatives for consideration, along with associated benefits and shortcomings. All of this information is synthesized in Chapter 3. Meanwhile, with the help of the SG-14 Panel, 32 transit agencies were identified for a survey. The agencies were chosen based primarily on size, because transit agencies with larger ADA paratransit services tend to have a broader variety of complex service models. A second criterion was that the transit agencies have recently undertaken—or are in the midst of—service model changes. Lastly, a goal of the panel was to achieve geographic and demo- graphic diversity. From the survey, the synthesis team created two- to three-page profiles of each of the 29 transit agencies that responded to the survey. These profiles are included in Chapter 4, and are organized alphabetically based on location. Each profile includes a synopsis of the system, along with ridership, service and cost statistics, and service performance standards. Also included are a background section on the service model and how it works, relevant information on contracting, the outcomes resulting from their service models, and overall lessons learned. The team then grouped the transit agencies into categories of services models. Three groups were established: 1. Systems in which all or some of the primary functions were directly performed by transit agencies; 12 of the transit agency respondents were grouped into this category. 2. Systems that utilized brokers or call and control center managers; nine transit agencies fell into this category. 3. Systems in which service provider contractors were also involved in some or all of the primary call and control center functions; 18 fell into this category. Note that these numbers do not add up to 29, as there are overlaps. Next, specific service models were identified within each of the three groups above, in concert with the prevailing definition of a paratransit service model and its three compo- nents: (1) management structure, (2) how work is divided among multiple carriers, and (3) service mix. Surprisingly, there were 25 specific service models represented among the 29 different respondents, which confirms the paratransit industry adage that “everybody does it a little bit differently.” The synthesis then provides insights from the survey respondents relevant to these service models. Conclusions and Key Findings What Defines a Service Model? ADA paratransit service models can be categorized in three primary elements: Management Structure. The major division of responsibilities relative to the four primary call center functions of reservations, scheduling, dispatching and handling trip status or estimated time of arrival (ETA) calls, and service operation. As illustrated in Figure S-1,

Source: Created by Nelson\Nygaard Consulting Associates. Figure S-1. ADA paratransit management structure archetypes.

4 aDa Paratransit Service models these functions can be performed by the transit agency itself, a broker or call and control center manager, and/or one or more contractors. The responsibility for providing supporting assets such as the call center or operations facilities, vehicles, the in-vehicle communication equipment, the paratransit scheduling software, the radio system, and the telephone system can also be included under management structure. Division of Work among Multiple Carriers. In the cases where multiple dedicated fleets are used to deliver service, regardless of whether the vehicles are operated by the transit agency and/or one or more contractors, this element generally focuses on whether a service area is organized into zones, with carriers assigned to one or more zones, or the service area is unzoned, with carriers assigned “packages” of work. These options are illustrated in Figure S-2. Service Mix. The third component of service models is NDSPs such as taxis and livery operators, which are used as a resource for serving ADA paratransit trips. The service mix is the split between dedicated and non-dedicated service providers. (Note: The use of NDSPs to provide ADA paratransit trips is not to be confused with the use of taxis, transportation network companies, and other non-dedicated resources to provide an alternative service for ADA paratransit customers via transit-sponsored subsidies.) Groups and Types of Service Models When these three components are considered together, the response to the survey revealed that there are 25 different service models represented among the 29 transit agen- cies that fully or partially completed the survey. The 25 service models, shown in Figure S-3, fall into one or more of the following three groups: • Group 1: Service models in which the transit agency has a direct role in some or all of the four primary call center functions (reservations, scheduling, dispatching, and handling customers’ ETA calls) and, in some cases, operations as well. Based on the survey, 12 tran- sit agencies reflecting 12 service models fall in this group. • Group 2: Service models in which the transit agency retains a call and control center man- ager or an operational broker. Based on the survey, nine transit agencies reflecting eight service models fall in this group. • Group 3: Service models in which the dedicated service provider contractor(s) also perform(s) some or all of the call center functions. Based on the survey, 18 transit agen- cies and 15 different service models fall in this group. Within these three groupings, there were overlaps. For example, nine transit agencies fell into more than one group. Key Findings from the Survey Group 1: Transit Agencies Performing All or Some of the Primary Functions Only one of the 29 respondents (Nashville MTA’s AccessRide in Tennessee) provided all of the dedicated service in-house. The most important benefits of this service model, as cited by the agency, are the flexibility to respond to service quality issues, and direct control over service/cost efficiency. Eleven others (38% of the 29 respondents) perform some or all of the call and control functions. Of these, five perform all four of the primary call and control center functions (reservations, scheduling, dispatching, and handling ETA calls), while the other six only perform some of these functions.

Source: Created by Nelson\Nygaard Consulting Associates. Figure S-2. Division of work among multiple carriers.

6 aDa Paratransit Service models In two cases, Pierce Transit’s Shuttle in Washington state and UTA Paratransit in Salt Lake City, the transit agency also operates a portion of the dedicated service, a somewhat uncommon design. Both agencies mentioned that the in-house call center allows for improved customer service and more direct control over customer service functions. The UTA also stated that by performing reservations in-house, the agency can more effectively operationalize conditional eligibility, leading to reduced demand and costs. Pierce Transit said they are considering eliminating the bifurcated provision of service delivery in the same area because of its inefficiency, and because the cost disparities between the in-house and contracted services create scheduling and budgeting issues. Group Service Model Location Reserv. Sched. Dispatch ETA Operations (#) Zones (#) NDSP (%) 1 1 Nashville TA TA TA TA TA+TC 0 15% 1 2 Pierce County TA TA TA/SP TA TA+SP 0 0 1+3 3 Salt Lake City TA TA TA/SP TA TA+SP (2) 3 0 1 4 Austin TA TA TA TA SP (2) 0 3% 1 5 Houston TA TA TA TA SP (2) 0 11% 1 6 Philadelphia TA TA TA TA SP (7) 5 0% 1+3 7 Ann Arbor TA/SP TA/SP SP SP SP (1) 0 0% 1+3 8 New Jersey TA TA/SP SP TA SP (6) 6 0% 1+3 9 Atlanta TA SP SP SP SP (1) 0 0% 1+3 10 Las Vegas TA SP SP SP SP (1) 0 0% 1+2 11 Kansas City TA CC CC CC SP (2) 2 0% 1+2+3 12 New York City CC TA SP CC SP (13) 17 30% 2 13 Denver CC CC CC CC SP (3) 0 10% 2 Seattle CC CC CC CC SP (2) 0 9% 2 14 Wash, DC CC CC CC CC SP (3) 2 6% 2 15 Portland CC CC CC CC SP (1*) 3 0% 2 16 Oakland/EB OB OB OB OB SP (3) 0 0 2+3 17 Broward Co CC CC SP CC SP (3) 0 0 2+3 18 Chicago SP CC SP CC SP (4) 3 0 3 19 Dallas SP SP SP SP SP (1+TS) 0 70% 3 Orange Co SP/TS SP SP SP SP (1+TS) 0 23% 3 20 San Francisco** SP SP SP SP SP (1+GT) 0 0% 3 21 Milwaukee SP SP SP SP SP (2) 2 0% 3 22 Boston SP SP SP SP SP (3) 3 1% 3 Phoenix SP SP SP SP SP (4) 4 100% 3 23 Los Angeles SP SP SP SP SP (6+ TS) 6 39% 3 24 Pittsburgh** SP SP SP SP SP (6) 6 0% 3 25 Columbus SP SP SP SP SP (1) 0 0% 3 Orlando SP SP SP SP SP (1) 0 0% Legend: TA = Transit Agency, SP = Service Provider Contractor, CC = Call Center Manager, OB = Operational Broker, TC = Taxi Subcontractor(s), GT = Group Trip Contractor. * Same firm provides call center functions and operations under two contracts. ** Overseen by administrative broker. Figure S-3. ADA paratransit service models—division of operational functions.

Summary 7 A few of the more uncommon models include New York City Transit’s Access-a-Ride, where the transit agency staff perform scheduling, and the service models in Atlanta (MARTA Mobility), Kansas City, Missouri (RideKC Freedom), and Las Vegas (Regional Transporta- tion Commission of Southern Nevada’s RTC Paratransit), where the agency performs reser- vations but does not perform scheduling. For the nine transit agencies that perform all or some call center functions but do not operate any of the service, the most important benefits of this service model are control over service quality, followed by control over service productivity/cost. Both relate to the role of the agency in performing call and control center functions. The greatest challenges or shortcomings involve the use of contractors for service delivery. While the model is less costly, the most common shortcoming mentioned was the perception that level of service quality was not as high as if it were provided in-house. The most common challenges cited were contract compliance issues. Group 2: Contracting with Call and Control Center Managers or Operational Brokers Eight (28%) of the 29 transit agency respondents retain a call and control center manager or operational broker to perform some or all of the four call and control center functions. Seven retain a call and control manager, while one, East Bay Paratransit of California’s Alameda-Contra Costa Transit District and Bay Area Rapid Transit (BART), retains a broker. Of the seven, four perform all four of the functions, while the remaining three leave dis- patching to the service provider contractors. Six of the seven utilize multiple service providers, ranging in number from two to four, except in New York City, where there are 13 service provider contractors. For these six transit agencies, the most important benefit was the transfer of risk away from the transit agency. Also cited by these respondents were cost efficiency through competition and economies of scale of centralizing call and control functions. Mentioned as frequently was enhanced control over service quality with centralized call and control center functions. The most prominent shortcomings were the potentially significant cost of housing and monitoring an additional entity, and that effecting change could be more cumbersome and accountability less clear compared with other models. The one location where there is a call center manager and only one service provider is in Portland, Oregon, TriMet’s LIFT service. Here, the three different service providers, previ- ously assigned to service zones, evolved as a result of acquisitions into one provider serving the entire area. Moreover, the remaining service provider contractor is the same firm that was hired to be the call and control manager. On one hand, this equates to a turnkey service provider contractor; on the other hand, TriMet continues to hold two separate contracts for each, allowing the agency to continue to go out to bid for each function separately in the future, if desired. TriMet states that the current model is effective at creating cost efficien- cies through economies, while enhancing the flexibility to respond to service quality issues, because the same firm has been retained for both call center and service delivery functions. At the same time, TriMet states that this shared responsibility can create conflicts between the two groups. The two most important benefits cited by BART were that the model creates improved cost efficiency by fostering competition among multiple carriers, and centralized scheduling and dispatching enables more direct control over the balance between service quality and productivity. The centralization of dispatch, implemented several years ago, also improved productivity with central access to the entire fleet, after an initial dip and adjustment to

8 aDa Paratransit Service models software parameters. BART also mentioned that using a broker has lowered administra- tive costs. The most prominent shortcomings or challenges cited were the conflicts that have arisen between broker and service providers as a result of centralized scheduling and dispatching. Group 3: Use of Turnkey Contractors that Perform All or Some Call and Control Center Functions Eighteen (62%) of the 29 respondents use single or multiple contractors that perform all or some of the call and control center functions. Within this group, there is a wide variety of service models. Of these 18, half have only one service provider. In seven of these agen- cies, one turnkey contractor performs all functions. In two of these—Dallas, Texas, DART’s Mobility Management, and Orange County, California, OCTA’s Access—each of the contrac- tors has a taxi subcontractor that plays a prominent role. The other nine systems have multiple contractors. Of these, five assign carriers to zones, while the others utilize service packages. Generally, the primary cited benefit of service models in the third grouping is the lower cost, as wages and fringe benefits associated with in-house operations are perceived to be higher, and in consideration that driver wages/benefits is the single highest cost element. However, a few transit agencies mentioned this can be “a double-edged sword,” as wage competition with other companies and industries, especially in a rising economy, can pose a significant challenge to driver recruitment and retention, as well as recruitment and reten- tion of call and control center staff and managers. For the transit agencies in this group that have a single turnkey contractor, the most important benefits cited are cost efficiency through economies of scale and one-stop accountability. Some of these transit agencies also mentioned that competition fostered during re-procurement was sufficient to keep prices down. At the same time, one of the transit agencies mentioned the possibility of getting stuck with a “lemon” when a transit agency “puts all its eggs in one basket.” There is nowhere to turn if that contractor defaults or simply provides poor service, despite the rendering of performance-liquidated damages. For the transit agencies in this group that have multiple turnkey contractors, the most important benefits cited include cost efficiency through the fostering of competition; more control over cost and service quality, with the flexibility to adjust service; and greater resil- iency. The most prominent shortcomings cited included the additional duplication of costs and more difficult role of monitoring and ensuring consistency, and the potential for inconsistent service quality if not closely monitored. Zoned and Unzoned Service Providers Eighteen (62%) of the 29 respondents use multiple dedicated service contractors. With the exception of New York City, which has 13 dedicated service provider contractors, the number of contractors ranges from two to six. Of the 18 transit agencies using multiple dedicated service provider contractors, 13 (or 72%) assign their carriers to zones. For these transit agencies, the most important benefits of using multiple carriers focused on how zones can turn a very large service area into more man- ageable pieces, and can increase productivity if the zones are based on trip patterns. The use of zones was also cited as enhancing competition at procurement time, as smaller pieces attract more proposers. The transit agencies also said the simplicity of a zone system improved accountability and simplified access from the customer’s perspective. The most prominent shortcomings mentioned were the inefficiencies created by unproductive dead- heading and the potential for inconsistent service.

Summary 9 The other five (28%) do not use zones and instead bid the work out in packages. For these transit agencies, the most important benefits cited were fostering competition by attracting more proposers—to allow for economies-of-scale bid alternatives involving that combina- tion of packages—and to enhance cost and service quality control through performance- based adjustments during the contract period. Interestingly, Washington, DC’s, Metro Access service design incorporates both zoned and unzoned elements. The service has one zone shared by two carriers, and a second zone to which only one service provider is assigned. Use of Non-Dedicated Service Providers for ADA Paratransit Ten (34%) of the 29 transit agencies responding to the survey use NDSPs to serve ADA paratransit trips. In all but one case, taxis are used. The exception is in New York City, where livery operators (black cars) are used. In five of the nine locations, the NDSPs are taxi subcontractors. In four of the nine locations, the transit authority, broker, or call center manager contracts directly with and assigns trips to the taxi subcontractors. In New York, the livery operators are accessed via two NDSP brokers under contract to New York City Transit. In seven of these cases, the use of NDSPs is more modest, where the NDSPs serve fewer than 15% of the ADA paratransit trips. The three exceptions are New York City (30%), Los Angeles (39%), and Phoenix (nearly 100%). The use of NDSPs in Los Angeles and New York City is on the rise, while a new service model involving more dedicated service is about to be implemented in Phoenix. Where taxis are used, schedulers and dispatchers in only one location (Oakland/East Bay) do not have access to accessible taxicabs. Among the 13 transit agencies that use taxis and other NDSPs, the most important benefit cited is the reductions in unit cost use that result from using taxis to serve peak overflow and longer trips and to address day-of-service needs. The most prominent shortcomings are the degradation in service quality and knowing the identity and location of a particular vehicle to which a trip has been assigned, in order to better respond to ETA calls. Another issue commonly mentioned was increased opportunities for fraud, especially if payment is paid based on mileage. Use of Taxis and Transportation Network Companies for Alternative Services Transit agency subsidy mobility options, such as taxi subsidy and, more recently, trans- portation network company (TNC) subsidy programs, that are offered by a transit agency to ADA paratransit customers are being referred to more often as simply “alternative services.” While alternative services must be compliant with the ADA (in general), they are not gov- erned by the service criteria that govern ADA complementary paratransit. This is because (1) the decision to use the alternative service is totally the customer’s; (2) while the transit agency can offer/suggest the alternative service option, the customer may still choose to use the ADA paratransit service; (3) a customer choosing to use the alternative service does not impact the customer’s ADA paratransit eligibility or right to continue to request trips on the ADA paratransit service; and (4) none of the vehicles used are owned, operated, or controlled by the transit agency. These alternative services are not part of the ADA paratransit service model, as trips served through these subsidy programs are not considered to be ADA paratransit trips and hence do not contribute to a transit agency meeting its ADA paratransit obligation. That said, a transit agency’s decision to offer an alternative service may be influenced by its

10 aDa Paratransit Service models current ADA paratransit service design model. Similarly, switching to a new model may be influenced by a desire to introduce an alternative service. Thus, these alternative services are discussed tangentially in this synthesis. Of the 29 survey respondents, 12 or 41% of the transit agencies currently offer taxi- based alternative subsidy programs to their ADA paratransit customers. Three of these 12 (the transit agencies in Boston, Dallas, and New York City) also have implemented pilot alternative subsidy programs using TNCs, while two others (Broward County, Florida, and Washington, DC) were planning to implement pilot TNC-based alternative services in 2017. The Las Vegas transit agency also planned to implement a pilot alternative service for its ADA paratransit customers by the end of 2017. For this alternative service, the transit agency will partner initially with both a TNC and a taxi company. Thus, in total, 13 or 45% of transit agency survey respondents have implemented or are planning to implement taxi- and/or TNC-based alternative services for their ADA paratransit customers. Contracting for Dedicated Service Providers Four times as many transit agencies pay for dedicated service based on revenue vehicle hours or gate-to-gate service hours as those that use a per-trip rate. However, for those transit agencies that use contractors with taxi subcontractors, virtually all use a per-trip rate. This is the only contracting strategy that seems tied to specific service models. Forty percent of the transit agencies use a split rate, which includes a set monthly fixed amount that covers fixed costs, with an hourly or per-trip rate that covers variable costs. More than twice as many transit agencies reimburse contractors separately for fuel than agencies that include fuel in the rate or variable rate. More than 80% of the respondents utilize performance-based incentives and/or liquidated damages. The most common metrics that trigger incentives and liquidated damages are, in descending order, on-time performance, missed trip percentage, complaint frequency ratio, productivity, excessively long trips, call hold times, and no-show rate. Other Conclusions The responses from the survey unveiled some other interesting connections while also dis- pelling suspected connections. These include the following: • Among the 18 systems that use multiple providers, there does not appear to be a common strategy about balancing the work among the providers. About half have made attempts, either with zone boundaries or the size of the package, to have their service providers deliver about the same volume of work. Meanwhile, the other half have one or two domi- nant contractors with others playing a smaller role. • Choice of ADA paratransit service models among larger transit agencies appears not to have a correlation with the organizational status of the transit agency, nor with what other public transportation services are provided or whether unions represent the operators of those services. Future Research This synthesis focuses on ADA paratransit service models and related contracting strat- egies. With the advent of TNCs, there has been a significant increase in the number of alternative services being planned and implemented. While TCRP Project J-11/Task 26, “Collaborations and Partnerships between Public Transportation and Transportation

Summary 11 Network Companies (TNCs),” from 2017/2018, is researching transit-TNC partnerships (including both alternative services and general public subsidy programs), additional research will likely be needed to better understand (1) the extent to which ADA paratransit customers are using alternative services (including taxi subsidy programs) for replacement trips on the ADA paratransit system versus new trips, and why, and (2) the extent to which these alternative services successfully contribute to reducing overall paratransit costs, and what related strategies are being utilized. Also, as there are currently no cases of transit agencies partnering with TNCs as a resource for non-dedicated service providers to serve ADA paratransit trips, there is a need to research the related issues, obstacles, and solutions, if such research is not already being performed in the context of the TCRP J-11/Task 26 project and/or in the context of some of the FTA’s Mobility on Demand Sandbox projects.