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.

19 Introduction Based on the literature review and the synthesis team’s knowledge of the industry, ADA para- transit service models can be categorized according to three primary elements: • Management structure. This boils down to the 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. The management structure can further be defined by nuances and, in particular, by which entity is responsible for providing various supporting assets such as the call center or operations facilities, vehicles, the in-vehicle communication equipment, the paratransit scheduling software, the radio system, the tele- phone system, and the like. • Division of work among multiple carriers. This element applies to multi-carrier services in cases where there is more than one dedicated fleet, regardless of which entities are responsible for the operation of those fleets. Generally, this element of the service model represents how the service delivery work is split among the carriers and is organized. Is the service area zoned? Is this a factor of the contractors also providing the call center functions? Is the service area unzoned but split into “packages” of work? And does this correlate with the call center func- tions being centralized with the transit agency or another third-party contractor? • Service mix. The third component of service models is whether non-dedicated service pro- viders (NDSPs) are used as a resource and to what extent they are used. More and more, transit agencies have found that the use of taxis, livery operators, chair car carriers, and other providers can help reduce the unit cost per trip and provide a resource for schedulers and dispatchers. (Note: The use of NDSPs to provide ADA paratransit trips is not to be confused with the uses of taxis, transportation network companies, and other non-dedicated resources to pro- vide an alternative service for ADA paratransit customers.) Management Structures for ADA Paratransit From the literature review, management structures for ADA paratransit service can generally be categorized into the six archetype options depicted in Figure 1, and described here: • In-house. All service responsibilities are performed by the transit agency. • Split structure. The agency splits functions between in-house resources and contracted ser- vices. The most common split appears to be the agency performing reservations, scheduling, and handling trip status (ETA) calls with its own employees, while contracting with one or more service providers. In this split arrangement, dispatching could be directly performed by C h a p t e r 3 Service Design Archetypes and Contracting Strategies

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

Service Design archetypes and Contracting Strategies 21 the transit agency or by the contractor(s). And while the more common approach is to use a contractor or contractors to provide service delivery, a few transit agencies also provide a portion of the service. • Administrative broker. The agency retains a contractor to organize and manage the system, with all paratransit service functions (including all four call center functions) performed by the service provider contractors. All service provider contractors are retained by the broker and not by the transit agency. • Operational (or full-service) broker. The transit agency retains a contractor as above, and the broker retains the service providers as above. However, the operational broker also performs some or all of the call center functions. • Call (and control) center manager. The transit agency retains a contractor to perform some or all of the call center functions, and to coordinate and schedule/assign trips to providers contracted by the transit agency. • Turnkey contractor(s). The agency contracts out all ADA paratransit responsibilities to one contractor who performs all responsibilities for the entire service area or for a service zone or region. There are two common subtypes: (1) one turnkey contractor for the entire service area; and (2) multiple-zoned contractors in which each contractor provides all functions within a zone or region. In-House Service Model—Benefits and Challenges The defining characteristic of this service model is that all primary functions, including the call center functions and the operation of the dedicated fleet, are staffed with transit agency employees. The only outsourced exception can be the use of non-dedicated service providers, if any. Benefits Based on the literature review, the major benefit of this service model is control. In theory, changes in policy, particularly with respect to the balance between on-time performance and cost/service efficiency, could be implemented more quickly than through a directive to the operations contractor(s) or via a contract amendment or change order. This could range from geographic and temporal deployment of vehicles and road supervisors to the service mix and the specific strategies employed in the use of taxis. A particular aspect of control also includes the operationalization of conditionally eligible trips. Here, the concept is that agency employees will tend to recognize the conditions and the eligibility of a particular trip in a more objective fashion than a service provider contractor that stands to gain from increased ridership. Challenges and Shortcomings From the literature review, the most significant shortcoming of this model is the additional cost. Most of the additional cost is due to employee compensation and benefits, recognizing that driver wages and fringe benefits represent up to 70% of the variable cost of dedicated paratransit service operations. Another shortcoming is that, if functions are performed in-house, it becomes difficult to outsource them in the future. Also, while TCRP Report 121 (Nelson\Nygaard Consulting Associates, TWJ Consulting, and RLS and Associates 2007) lays out the types of trips that strategically make sense to assign to taxis and other NDSPs (discussed further in Service Mix in this synthesis), there can be a strong internal bias against the use of taxis if service is also operated in-house. This bias can result in assigning so-called unattractive trips to taxis.

22 aDa paratransit Service Models Split Structure—Benefits and Challenges Under a split structure, the transit agency can perform all or some of the call center func- tions in-house, while vesting the responsibility for the other functions with the service provider contractor(s). In the infancy of the paratransit industry, the conventional approach was to pair the reserva- tions and scheduling function and to pair dispatching and operations. Transit agencies with service provider contractors would perform reservations and scheduling in-house, and leave dispatching to the carriers. However, as the industry has matured, other split structures have emerged. DART in Dallas, for example was the first to pioneer centralized dispatching. Since then, other transit agencies such as Palm Tran in Palm Beach County, FL, have followed suit, while others such as the transit agencies in Atlanta, Las Vegas, and St. Louis have elected to perform only reservations centrally while allowing their contractor(s) to perform the other functions. And while not depicted in the graph, as it is extremely uncommon, there are at least two transit agency (Pierce Transit and the UTA in Salt Lake City) that deliver ADA paratransit service with transit agency drivers, and with drivers of the service provider contractor(s) operating the remaining portion(s). Benefits As discussed, the major advantage of a transit agency performing all or some of the call center functions is to have more direct control over the intake of reservations and ETA calls, as well as direct control over scheduling and dispatching. This can lead to increased productivity and lower unit costs, and also permits the transit agency to have more direct control over the balance between on-time performance and cost/service efficiency than if the contractor(s) performed these functions. In addition, with the transit agency performing the scheduling and dispatching, there would presumably be more objectivity toward optimizing the service mix, if taxis or other non-dedicated service providers (NDSPs) are used, and deciding exactly what types of trips to assign to the NDSPs. With the transit agency performing scheduling and dispatching, it is also easier to shift work from one contractor to another, if more than one contractor is retained during the contract period (see Multi-Carrier Service Designs). Challenges and Shortcomings As with the all in-house service model, the labor costs will be higher for the functions per- formed by the transit agency than if provided by the contractor or contractors. Also, if the transit agency opts to perform reservations and handle the ETA calls, with dispatching being the responsibility of the contractors, responding to ETA calls will be difficult unless the call-takers have the same information that is available to the contractor’s dispatchers. Administrative Broker—Benefits and Challenges As with any type of transportation broker, the distinguishing characteristic is that the transit agency has a contract with a single company to establish the service network, oversee the system, and retain and monitor the network of service providers. An administrative broker does not per- form any of the call and control center functions, choosing instead to vest these responsibilities with the service provider contractors. (Operational brokers, which do perform call and control center functions, are discussed later.) Benefits A transit agency will choose to use a broker when it does not wish to enter into direct con- tracts with service providers; when it believes that a contractor can exercise more leeway in the procurement process (including introducing or home-growing new service providers where/when

Service Design archetypes and Contracting Strategies 23 needed); and if the transit agency does not have the expertise in-house to establish, oversee, and modify particular approaches to delivering ADA paratransit service. Compared to directly per- forming these types of administrative tasks, a transit agency may also find that the broker labor costs are less. Using an administrative broker can also be an advantage when the broker is overseeing a coordinated paratransit service. While the transit agency might hold the main contract with the broker for ADA paratransit services and other services for which the transit agency is ultimately responsible, the broker is then authorized to enter into additional sponsorship contracts with human service agencies and/or municipalities. Challenges and Shortcomings Compared to a transit agency directly performing the administrative functions, a transit agency employing an administrative broker does not have as much direct control over planning, policies and procedures, procurement, customer service, provider monitoring, and the day-to-day func- tioning of the system. However, this is exactly why a transit agency vests these responsibilities with a broker. Operational (or Full-Service) Broker—Benefits and Challenges This service model is similar to a split structure in which the transit agency performs all or most of the call center functions. However, with this model, a transit agency opts to use a third-party contractor to perform the call center functions not assigned to the service provider contractors, and for the broker to establish the carrier network, noting that it is the broker that contracts with the service providers and not the transit agency. Indeed, the most common version of this model is to have the call center contractor perform all four of the call center functions with service delivery performed by the service provider con- tractors. The second most common version is similar, except that the broker is not involved in dispatching. While it is possible to have a broker and one service provider contractor, it is much more typical to have a broker in a multi-carrier environment. The reasoning is, if a transit agency has one contractor performing call and control functions and that contractor has only one service provider subcontractor, there are cost efficiencies inherent to one company doing both. Benefits A well-crafted contract, with appropriate provisions, standards, and performance-based incentives and liquidated damages, can help the broker achieve the transit agency’s desired bal- ance of service quality and cost efficiency. Moreover, with the broker being a neutral party (that does not also operate service), the broker potentially would handle trip requests from condition- ally eligible customers in as objective and attentive a manner as a transit agency. This objectivity extends to other functions as well, such as the level and type of trips assigned to taxis and other NDSPs, and shifting work between service provider contractors. In addition, the labor cost and fringe expense of a broker’s staff is likely to be lower than that of in-house labor. The other major advantage, especially in comparison to a call and control center contractor model, is that the broker holds the contracts with the service provider and hence has more direct leverage. Challenges and Shortcomings There will be an additional cost for the broker facility, whether that cost is embedded in the broker cost or provided by the transit agency. There will also be an additional cost to monitor the broker.

24 aDa paratransit Service Models A criticism of operational brokers was that they were typically paid by the trip, and the brokers would turn around and procure less expensive but poorly performing carriers in order to maxi- mize their profit. The relationship between the broker and the service provider contractors must be collab- orative to work well, but the broker also needs to be prescriptive about the run structures and expectations. Call (and Control) Center Manager—Benefits and Challenges This service model is similar to an operational broker, however, with this model, the con- tractor who performs some or all of the call center functions does not contract with the service providers. Instead, the transit agency contracts with the service providers. Thus, there is no contractual relationship between the call center contractor and the service providers. The two most common versions of this model are: 1. The contractor performs all four of the call center functions with service delivery per- formed by the service provider contractors. This is more commonly referred to as a call and control center contractor, as the contractor is controlling the movement of the vehicle via dispatching. 2. The contractor performs all center functions except for the dispatch function, which is assigned to the service providers. Benefits These are essentially the same as with an operational broker, except for the benefits that per- tain to a broker’s direct leverage over the service provider. Challenges and Shortcomings There will be an additional cost for the call center facility, whether it is embedded in the con- tractor cost or provided by the transit agency. There will also be an additional cost to monitor the contractor. As with an operational broker, the call center contractor needs to be prescriptive about the run structures and expectations, especially if the call center functions are being transferred to the call center contractor, i.e., if the transit agency is changing from a completely decentralized system to a centralized system. The transit agency (as the contract holder) needs to hold the service provider contractors accountable for delivering service based on contractual standards. At the same time, if poor ser- vice results from the call center manager doing a bad job with scheduling and dispatching, then the transit agency needs to hold the call center contractor accountable. In some cases, contracts with the two different types of contractors can have shared incentives and liquidated damages, but the difficulty, especially in the transition, is to determine who is at fault, and how to help. Turnkey Contract(s)—Benefits and Challenges The distinguishing characteristic of turnkey contracts—whether with one contractor serving the entire area, with multiple contractors serving trip requests, or trips emanating from their respective “home” zone or region—is that the responsibility for the call center functions (res- ervations, scheduling, dispatching, and handling customers’ ETA calls) is the responsibility of the service provider contractor(s). Under some turnkey contracts, the contractor may be autho- rized to subcontract with other dedicated or non-dedicated service providers. Otherwise, the transit or paratransit agency is responsible for establishing policies, customer service (complaint

Service Design archetypes and Contracting Strategies 25 management), contractor procuring and oversight, and, in some cases, providing some or all of the supporting assets, such as facilities, vehicles, and software. Benefits Many transit agencies opt to use this model because it is more cost efficient than operating the service in-house, and because the contractors bring with them expertise that does not exist within the transit agency. The larger transit agencies that elect to contract with a single provider do so because they believe the cost efficiencies that come from economies of scale are greater than those stemming from competition among multiple service providers. It is also administratively easier to have one point of contact and monitor only one contractor. Also, in cases where the transit agency is providing the facility, it is easier to provide one facility than multiple facilities, and easier to swap out contractors at the end of a contract, or sooner if the need arises. With one contractor, the service quality, policies, and procedures, are consistently applied. Contracting with multiple turnkey contractors assigned to specific zones or regions offers cost efficiencies stemming from greater competition and higher productivities. The greater compe- tition comes from offering smaller pieces of work, which tends to attract more proposers. The higher productivities stem from shorter (intra-zone) trips, which in turn typically have greater impact on cost per trip than the economies of scale associated with a single turnkey contractor. In addition, if a contractor defaults in a multiple-carrier service model, it may be easier to expand the role of the other contractor(s) until an emergency procurement to replace the outgoing contractor is undertaken. Using multiple carriers is explored further in the next section. Challenges and Shortcomings Compared to vesting the call center functions with the transit agency or with a call and control center manager or operational broker, a transit agency’s stewardship role in controlling the bal- ance between cost efficiency and service quality is once-removed. Thus, it is imperative that the contract(s) establishes expectations (service standards, incentives/liquidated damages) for both service/cost efficiency metrics and for service quality metrics. In a single contractor environment, the transit agency is “putting all its eggs in one basket” reaping the benefits of a good contractor, as well as the reverse. With more than one contrac- tor, there would clearly be duplication in labor, especially among senior and middle staff. Inconsistency with service quality and even procedures can arise with multiple contractors. More over, the cost of multiple facilities can be higher than a single facility. Compared to service models in which call and control center functions are not provided by the service provider contractor, another potential shortcoming lies with the use of NDSPs. In the case where a service provider is in control of which trips get assigned to NDSPs, the assign- ment will be in the best interest of the service provider contractor and not necessarily the transit agency. Division of Work Among Multiple Carriers Service designs involving more than one dedicated fleet generally fall into two categories, as depicted in Figure 2. The major benefit of having multiple carriers is that the model provides more competition during the procurement. This is because smaller, more manageable pieces of work tend to attract a larger set of proposers, including local and regional carriers, as well as carriers who qualify as disadvantaged business enterprises (DBEs). In contrast, larger single contracts tend to screen out such carriers. This translates into more competitive rates.

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

Service Design archetypes and Contracting Strategies 27 Another major benefit is that there is more resiliency with more than one operations con- tractor. As mentioned, in the event a contractor defaults or suspends service (for whatever reason), a transit or paratransit agency can temporarily expand the responsibilities of the other contractor(s) to include those of the other contractor until a replacement contractor can be secured. Whether or not the service area is divided into zones or regions is often a factor of the entity or entities to which the reservations function is assigned. If the reservations function is decentral- ized with each service provider, there must be a way to funnel trip requests to a specific carrier. This is usually accomplished by creating zones or regions, with customers calling the provider based on where they live (or where the trip is originating). In contrast, if the reservations func- tion is centralized, there is no need for zones or regions, although some systems with centralized reservations do have them as a convenient way to “package” work. Thus, in the case of decentralized reservations, each service provider contractor performs reservations and often some or all of the other primary call center functions as well (schedul- ing, dispatching, and/or handling trip status calls), and customers are typically instructed to call their “home” provider to book a trip, or any of the providers assigned to that zone if there is more than one. In cases where there is an overlapping core area, customers are directed to call a specific provider if they are traveling to a destination beyond the core area, or to call any of the providers for an intra-core trip. One acknowledged benefit of such a zone-based structure is to keep vehicles from straying too far from their base of operations; this helps improve the responsiveness of road supervision to emergencies and breakdowns, and tends to improve productivity (the longer the average trip length, the lower the productivity tends to be, all else being equal). In the case of transfers, however, delays by one carrier can be fairly commonplace and, if unaddressed, can have an adverse impact on productivity. This in turn can negate any gains in productivity that result from trip length reduction. In the case of centralized reservations (and scheduling), the transit agency can still use zones as a way of ensuring service area coverage, especially if providers assigned to a zone are required to locate their operations facility in that zone. But, as with centralized reservations, zones are no longer needed as a means of directing customers to call a certain carrier. In zoneless systems, transit agencies often put out bid packages for percentages of the work. For example, a transit agency wishing to have three providers could approach this in various ways: • Equivalent bid packages (33%/33%/33%); • Unequal packages (50%/30%/20%) to be awarded based on the value of the submitted bids; • Unequal packages with two primary service providers and one smaller provider (40%/ 40%/20%)—this might pertain to a desire to include a smaller local or DBE service provider that has limited capabilities, or to focus such a service provider on ADA paratransit trips going to and from agencies; and • Four 25% packages, with the understanding that a proposer may bid on two packages. Under a multiple-carrier model with centralized scheduling, trips are assigned to a provider based on efficiency. Also, with carriers having unzoned packages, the transit agency, call and control manager, or broker has the ability to shift work between carriers mid-contract based on performance. This can also result in ongoing competition during the contract term. Here, the threat of business being taken away or the incentive to gain more business can not only translate into improved service quality, but possibly reduced cost as well, depending upon the specific circumstances.

28 aDa paratransit Service Models There are several additional nuances in cases in which a transit agency has implemented a zone- or region-based service area design: • Multiple service provider contractors assigned to more than one zone. In most cases, assign- ments are one-carrier-to-one-zone, although there are examples (in Pittsburgh, New York City, and Washington, DC) where more than one service provider contractor is assigned to more than one zone. In the case of New York City and Washington, DC, where the reserva- tions function is centralized, customers all call the same entity to place a reservation. In the case of Pittsburgh’s ACCESS service, where there is one zone (of six) with two carriers per- forming reservations for customers in that zone, customers in that zone can call one or the other carrier. The premise behind this unusual design was to stimulate the performance of poorly performing providers by adding a competitor to the zone. • In Boston, the Massachusetts Bay Transportation Authority has transfers between zones but has eliminated required transfers for inter-zone trips to a “buffer” area (e.g., trips to destina- tions just across the zone order). • In a decentralized reservations model, one carrier serves the return trip of a round trip to a different zone versus the “home” carrier handing off the return trip request and delivery to the provider assigned to the destination zone. In most cases, it is the “home” provider that also takes and serves the return trip of a round trip in systems without required transfers. How- ever, in Los Angeles, the request and responsibility for delivery of the return trips is handed off to the destination zone carrier. • Overlapping zones. In Boston, where three turnkey, zone-based service providers overlap in a core area, customers living in the core area and traveling to a provider’s home region are directed to call that provider. However, if that same customer is making an intra-core area trip, that customer may call any of the three providers. Service Mix TCRP Report 121 (Nelson\Nygaard Consulting Associates, TWJ Consulting, and RLS and Associates 2007) provides several examples of transit agencies using NDSPs to serve ADA para- transit trips. Generally, the underlying concept is to reduce the transit agencies’ overall cost per trip for ADA paratransit service. Of course, the capacity of the NDSPs and their rates must be a factor in any decision, but they typically result in an overall reduction of cost per trip. As is illustrated in Figure 3 and further detailed in TCRP Report 121, the overall cost per trip can be reduced by shifting to NDSPs any ADA paratransit trips that negatively impact the pro- ductivity of the dedicated fleet. This can include shifting to NDSPs peak period overflow trips, trips from low-demand areas or at low-demand times, and long, out-of-the way trips that are not easily combined with other trips. Scheduling dedicated vehicles to serve such trips can have a significantly adverse effect on productivity. Similarly, if NDSPs are available, dispatchers have a resource available for addressing late-running vehicles by shifting trips to get the late vehicle back on schedule. Dispatchers could also use such a resource to handle an incident or acci- dent, as well as re-emerging no-shows, rather than turning back dedicated vehicles or adding customers as add-ons to another busy run. While technically NDSPs can include taxis, livery vehicles, and private for-hire chair car companies, taxis are the NDSP used most commonly to serve ADA paratransit trips. Figure 3 also indicates that NDSP provision of accessible vehicles for the system, while advan- tageous for both schedulers and dispatchers, is not required. There is a condition to this state- ment not included in Figure 3. Such provision is not required only if customers who require accessible vehicles are getting the same level of service (e.g., same on-time performance) as customers who do not require accessible vehicles, if the system is considered as a whole. That is, if the dedicated portion of the fleet is providing that equivalent level of service throughout to

Service Design archetypes and Contracting Strategies 29 persons who require access vehicles, then the NDSP contractor does not need to provide acces- sible vehicles. If this condition is not present, then more accessible vehicles need to be infused into the system, whether they are used for dedicated service, for non-dedicated service, or both. As shown in Figure 4, use of NDSPs in this integrated fashion is much more common in models in which the call center functions are centralized with the transit agency or a call and control center manager or (true) broker, because there is no inherent conflict of interest. Such Source: Created by Nelson\Nygaard Consulting Associates. Figure 3. How NDSPs are used to reduce ADA paratransit unit cost. Source: Created by Nelson\Nygaard Consulting Associates. Figure 4. Use of non-dedicated service providers for ADA paratransit service.

30 aDa paratransit Service Models conflicts of interest may exist, depending on how the contractor is paid and other contractual provisions. Relationship Between Service Models, Contractor Procurement Strategies, and Payment Structures The use of contractors to provide turnkey services, service delivery only, or some or all call center functions is fairly common and almost the norm for ADA paratransit, especially among larger systems. As mentioned in Accessible Transit Services for All (FTA Research 2014), “the pri- mary benefit of contracting for . . . ADA paratransit is generally considered to be cost savings, which comes mainly from reduced labor expenses compared to direct public agency operation. For the largest transit agencies . . . 85% use contractors.” With the maturation of the industry, large, national (even multinational) contractors offer additional support services, mostly focused on new technologies, and experience that were not available 20 years ago. In metropolitan areas where the volume of service or the geography or pattern of trips sug- gests that the volume of service can be split into manageable pieces, most transit agencies have implemented a service model that uses multiple providers providing dedicated service, generally because: • It attracts more contractors to the procurement, which often results in more competitive prices. • It provides some insurance to the transit agency in the event one carrier ceases work or sig- nificantly underperforms. • Under certain designs, it provides for mid-contract competition, which can also reduce cost and improve service quality. Based on the literature review and the synthesis team’s experiences, transit agencies report that their procurement processes often do not deliver what the transit agencies seek—their particular, desired balance between service quality and cost efficiency. The reasons can include any of the following: • Prospective proposers don’t learn about procurement. • Prospective proposers elect not to propose if (1) there’s not enough time to put together a credible proposal; (2) there’s not enough time to mobilize if selected; and (3) their thinking that the procurement may be “wired” for the incumbent(s). • Prospective proposers bid high, responding to their perceptions that there is too much risk. This can be due to insufficient data in the RFP and/or because of the payment structure and liquidated damages. Transit agencies also report that the desire for an administratively easy procurement process can backfire. A prime example is a simplified rate structure without requiring supporting line- item costs. While it’s easy to evaluate (to determine the contractor selection), a more complex structure that addresses contractor risks, line-item costs that expose outlier costs, and incorrect assumptions can only help the evaluation and ensure a credible, accurate bid. Another exam- ple is a procurement department’s propensity to adopt a low-bid approach versus a best-value approach, the latter being essential to a paratransit procurement. Also, a payment structure that is mismatched with the service model can lead to an oversupply of service which is costly for the transit agency, or an undersupply of service which negatively impacts service quality (on-time performance, missed trips, etc.). From the literature review and the team’s experiences, the lessons learned regarding contrac- tor procurement include: • Assuming a stable operating environment, use of long contract periods offer a contractor reduced risk, especially if the contractor is providing major assets and can amortize the capital

Service Design archetypes and Contracting Strategies 31 expenses over a long period. Even in an unstable environment, carrier risk can be somewhat mitigated by adding renegotiation clauses to long-term commitments or by shortening the base contract periods. Also, in cases in which the transit agency supplies most of the capital assets (facility, vehicles, software), there is less need for longer contract periods, especially when ridership is fluctuating. • Transit agencies must make sure that procurements are well advertised. Good practices include notifying national, regional, and local carriers well in advance that a procurement is coming up and notifying them again when the notices and RFPs are issued. This should include placing the notices and RFPs on the transit agency’s website and including all impor- tant dates in the notices and the RFPs. It is not uncommon for transit agencies to personally and proactively invite prospective proposers to attend the pre-bid conference. • Most transit agencies do not leave enough time to recraft the RFP (to address short comings or opportunities) and get the RFP approved. Many transit agencies also do not provide suf- ficient time in the procurement process for the prospective carriers to do their due-diligence and local homework. It is also common for transit agencies to not leave enough time to: – Thoroughly review the proposals – Check references for each firm and proposed staff – Analyze and evaluate first the technical and then the cost proposals – Interview the candidate contractors – Seek clarifications and best and final offer responses – Prepare a justification memo to the board, noting the lead time necessary for some boards – Undergo contract negotiation and execution. Many transit agencies also do not leave enough time (6 months) for mobilization in order to generate a “level playing field” between the incumbent and new proposers. In the case of major changes, the entire process can take as long as 1½ years from start to finish. When the procurement period “sneaks up” on staff, time contractions ensue, often to the detriment of the procurement process. • Some transit agencies have two completely different committees evaluate the technical and cost proposals; this is generally a bad idea because the cost evaluators have little context. • ADA paratransit RFPs can be badly organized, making it difficult for prospective proposers to find key information and requirements. Key points/key changes should be included in the introduction/summary. Well-constructed RFPs will have: – Description of service and policies. – Historic and up-to-date service statistics and characteristics. – Scope of work, including a division of responsibilities, expectations by functional area, and staff and training expectations. – Service performance expectations for each key performance indicator (KPI), especially if they also trigger incentives and/or liquidated damages. – Safety requirements and emergency preparedness. – Mobilization and contracts transition requirements. – The rate structure; if the rate structure is split (fixed monthly fee plus a variable rate), the RFP should explain what costs are covered by each. – The supporting cost forms/spreadsheets that specify line-item costs for the mobilization period, fixed versus variable costs for each year of the contract, rolling up to a split rate. This will include, for example, wages and fringe for each staffing position, and detail about employee versus employer contributions for healthcare benefits. – Evaluation process and criteria; in a best-value approach, most transit agencies weigh the technical proposal as 60%–70% of the total score and the cost as 30%–40% of the total score. – Instructions on what a proposal should look like and include. • Many transit agencies split up contractor payment into a fixed monthly rate that covers fixed costs that do not change as the level of service increases or decreases, and a variable per-trip

32 aDa paratransit Service Models or per-hour rate (and for call center contractors, sometimes a per-call rate) that covers the costs that vary with levels of service. The idea behind this split rate is to protect the transit agency if ridership were to suddenly increase significantly beyond the forecast levels, and to protect the contractor in case ridership were to decrease significantly below forecast levels. Thus, in the costs proposal, there are forms, if not spreadsheets, which roll up fixed and vari- able costs into the respective rates of payment. • Paying dedicated service providers for all costs or variable costs via revenue-per-vehicle-hour rate (RVH) appears to be more common than a per-trip rate. The concept is that because most of the contractors’ expenses (labor wage and fringe) are a function of hours, an hourly based rate will offer less risk and, therefore, result in more accurate bids and more competition. However, under a per-RVH rate, the number of RVHs must be closely controlled by the transit agency. • In contrast, a per-trip rate provides a built-in incentive for a contractor to “deliver the goods.” That is, a contractor is incentivized to serve trips if the contractor gets paid by the number of trips served. In the case of a RVH rate, a contractor is paid for operating a vehicle, regardless of whether a particular trip is served or not. Per-trip rates are often used in a turnkey contrac- tor setting, i.e., where the contractor is also responsible for all call center functions; in such a model, it is important to have balancing liquidated damages for late and missed trips; other- wise, a contractor will be tempted to jam trips into runs to maximize revenue at the expense of on-time performance. Per-trip rates are also often used when contractors have a NDSP component or NDSP subcontractors. • Where transit agencies have a direct contract with NDSPs, the current practice appears to use distance-based rates (a mileage-based or zone rate). In the case of mileage-based rates, a best practice is a flat rate for each trip based on the GIS system underlying the paratransit software (or equivalent). In this way, the transit agency knows exactly what the trip will cost in advance, and can even generate a draft invoice for the NDSP contractor. • Most brokers and call center managers are paid by the trip, although some are paid based on a cost-plus basis with an annual revisiting of the budget, and with carrier costs passed through, in the case of brokers. Transit agencies that retain contractors only for reservations and ETA calls, pay their contractors based on a per-call rate. • It is generally recognized that conflicts of interest arise when brokers or call center managers are allowed to also operate service, and so this is generally a prohibition. • In cases where a transit agency is retaining a call and control center manager and service pro- vider contractors, those procurements should be staggered and not run at the same time. This not only preserves stability in the operation, but also allows a firm that is unsuccessful with a service provider bid to bid on a call center manager proposal, for example. • In the case of zoned service provider contractors, staggering these procurements if a carrier is prohibited from bidding on more than zone also allows for a second try at another zone or region. Conversely, if the transit agency is seeking alternative, multi-zone bids to determine what economies of scale might present themselves, the procurements should happen at the same time. If there are no zones, carrier packages should be bid out at the same time. Alternative Services for ADA Paratransit Customers As reported in TCRP Report 121, some transit agencies have opted to offer their ADA para- transit customers an “alternative” service and, specifically, a taxi subsidy program. While alter- native services must be compliant with the ADA (in general), they are not governed by the service criteria that govern ADA Complementary Paratransit. This is because: • The decision to use the alternative service is totally the customer’s. • While the transit agency can offer/suggest the alternative service option, the customer may still choose to use the ADA paratransit service.

Service Design archetypes and Contracting Strategies 33 • A customer choosing to use the alternative service does not impact the customer’s ADA para- transit eligibility or right to continue to request trips on the ADA paratransit service. • None of the vehicles used are owned, operated, or controlled by the transit agency. The transit agency can cap the number of trips taken by a particular customer on this ser- vice; establish whatever fare level and structure it desires; and not hold drivers to the same requirements as the drivers of the ADA paratransit service. At the same time, such service must meet Title VI requirements and equal access requirements, especially if federal funds are used. The benefits of these types of programs fall into two categories. For the customer, the taxis provide an on-demand mobility option (compared to the ADA paratransit’s next-day/advance reservations), and the transit agencies enhance their ADA paratransit customers’ financial ability to use these resources by defraying the fare. A customer opting to use this service instead of the ADA paratransit service for a particular trip potentially means a reduction in demand and hence cost for the transit agency, if: • The trip diversion does not result in significantly decreasing the productivity of the dedi- cated fleet; for example, if there is a significant diversion of short trips, the productivity of the dedicated fleet could be reduced to the point where the overall per-trip cost actually increases. • The savings from the diverted trips (i.e., the difference between the higher subsidy of the ADA paratransit service and the lower subsidy of the subsidy program) is more than the additional subsidies associated with any new trips generated (that would not have been taken on the ADA paratransit service). This second point is illustrated in Figure 5. In this graphic, the $30 in savings from a diverted trip is wiped out if two new trips are generated. Thus, if the transit agency generates more than two new trips for every one diverted, its costs increase. Conversely, if the transit agency is able to divert more than one trip for each two new trips generated, its paratransit costs decrease. For a transit agency to determine whether or not savings have been achieved, it is essential that some tracking take place to better understand the level of diverted trips versus the level of new trips for each customer participating in the subsidy program. In many such programs, transit Source: Nelson\Nygaard, 2014a. Figure 5. How transit agencies reduce costs by shifting ADA trips to alternative services.

34 aDa paratransit Service Models agencies look at the before-and-after levels of ADA paratransit ridership to estimate those levels. Customer surveys, while less accurate, have also been used. Note, too, that some transit agencies have recently implemented pilot alternative services involving TNCs in tandem with their taxi subsidy programs. These include the MBTA in Boston, DART in Dallas, and New York City Transit. Three other transit agencies—in Broward County, Las Vegas, and Washington, DC—planned to implement pilot alternative services involving TNCs in 2017.