Bus Rapid Transit Practitioner's Guide (2007)

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.

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-1 System Packaging, Integration, and Assessment CHAPTER 5. SYSTEM PACKAGING, INTEGRATION, AND ASSESSMENT INTRODUCTION BRT is an integrated system of services, facilities, and amenities that is designed to improve speed, reliability, and identity. It calls for packaging various components in a coherent and supportive manner that reflects specific needs, resources, and opportunities. This chapter gives guidelines for system packaging and integration that summarize and apply findings from other chapters of the Bus Rapid Transit Practitioner’s Guide. It shows how BRT components can be packaged, it gives parameters and procedures for estimating costs and effects, and it gives examples of estimating BRT performance and impacts for various BRT scenarios. Details of the alternatives analysis process are contained in Chapter 2. CHOOSING THE “BEST” PACKAGE OF COMPONENTS All BRT systems will have running ways, stations, and vehicles. The types of these features, as well as the types of various ITS-related components, will depend upon local needs, conditions, attitudes, and resources. Some guidelines follow. General Guidelines Developing BRT for any community requires the following activities: • Identifying the appropriate corridors • Comparing alternative alignments • Selecting the desired BRT alignments and components Key considerations include the following: 1. Establish the Need. Considerations include (a) slow and unattractive local bus service; (b) peak-period congestion on major roadways; and (c) continued (or anticipated) growth in CBD employment, urban population, and transit ridership. 2. Identify the Market. The nature of the current and future land use and demographic characteristics should be clearly identified. Market segments include riders diverted from local bus and auto and riders making new trips. Similarly, current and future transit—including origin-to- destination patterns, expected BRT ridership, and maximum load section (point) volumes—should be determined. Candidate markets include corridors with sufficient ridership potential to allow frequent all-day service (preferably at intervals not greater than 10 to 12 minutes between buses). A strong CBD (e.g., more than 50,000 jobs) and high-density corridors are supportive of BRT. 3. Select Type of Running Way. Selecting the types of BRT running ways will depend upon (a) availability of right-of-way within the proposed BRT corridors; (b) width, continuity, and operational characteristics of arterial streets; (c) the ability to integrate BRT operation with existing transit service; and (d) proximity to markets. 4. Recognize Public Preferences. Community and agency preferences regarding BRT routes should be taken into account. Selecting special BRT vehicles, All BRT systems have running ways, stations, and vehicles. Developing BRT calls for identifying corridors, selecting components, and assessing options. The Guide provides seven guidelines for packaging BRT components.

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-2 Bus Rapid Transit Practitioner’s Guide for example, should have the support of the transit agency responsible for operating the BRT service. Similarly, operational treatments such as bus lanes, TSP, and queue bypass lanes should have the support of the street transportation agencies. 5. Integrate BRT with Existing Bus Services. It may be desirable to restructure existing bus routes on streets in or serving a BRT corridor. Local routes should feed rather than duplicate the BRT service. Where BRT operates on busways, terminals, or outlying stations, it can serve as a focal point for connecting bus services. 6. Consider Funding Availability. Available resources for capital, operating, and maintenance requirements are essential. The funding available for BRT may influence the type, extent, and staging of BRT features. Where funding is limited, BRT may have to operate on city streets rather than on off-street busways. Similarly, existing vehicles might have to be used initially (although distinctively colored). Resource constraints may also limit the extent of the BRT system, making staging essential. 7. Explore Development Opportunities. Opportunities for land development near BRT stations should be explored. They can have bearing on the (a) extent of the BRT system, (b) location and design of stations, and (c) type of running way selected. Experience suggests that, under the right market conditions, BRT can influence development at major outlying busway stations (e.g., Ottawa) or along rebuilt urban streets with improved landscaping and walks (e.g., Boston). Packaging and Staging Examples BRT can be developed incrementally, with each stage keyed to demand characteristics and the availability of resources. The stages can include the following: • Adding elements or features. • Upgrading key elements such as vehicles, stations, and fare collection systems. • Relocating operations to off-street running ways. • Extending the system. For example, Pittsburgh started out with 4.3 miles of busway in 1977; today there are almost 20 miles of busway. Boston opened a 2.2-mile surface route in 2002 and added a 1.1-mile bus tunnel in 2005. Any modifications of a BRT system should be planned and designed such that the existing bus lanes or busway are not adversely affected by construction. Examples of packaging BRT elements for modest- and high-demand BRT systems are shown in Exhibit 5-1 and Exhibit 5-2, respectively. Exhibit 5-1 illustrates how BRT features could be packaged for a modest- demand and modest-cost system. This system would likely include local and BRT service, standard vehicles in a special livery, and a combination of dedicated bus lanes and mixed-traffic operations, radios, and on-board fare collection. Exhibit 5-2 gives illustrative packaging features for a high-demand, high-cost, and high-performance application. This application includes specially dedicated vehicles, fully dedicated bus lanes (including possible bus-only roadways), and an extensive array of BRT features. BRT can be developed in stages that add or upgrade features, relocate service to off-street alignments, and/or extend the system. Stages are keyed to ridership demand and resource availability.

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-3 System Packaging, Integration, and Assessment EXHIBIT 5-1 Packaging BRT Elements - Modest-Demand and Modest-Cost BRT System Services Stations Vehicles Running Way Systems Primarily local Simple stops No special treatment Mixed traffic Radios, on- board fare collection Mixed limited- stop, local Super stops Special signage Dedicated arterial curb lanes, com- peting turns allowed AVL for schedule adherence All-stop (local), mixed local/express On-line and off- line stations, significant parking for transit patrons Dedicated vehicles, special livery Dedicated freeway median lanes, merge/weave access/egress ITS passenger information, fare collection Point-to-point express Transfer/transit centers Dedicated vehicles, uniquely specified (e.g., double-articulated buses, hybrid propulsion) Fully dedicated lanes, exclusive freeway access/egress ITS vehicle priority Intermodal transfer/transit center Mechanical or electronic guidance Partial grade separation ITS vehicle lateral guidance Fully electric propulsion system Full grade separation, curbed/striped/cabled for guidance ITS automation, electric power system Overhead power contact system NOTE: Boldface text denotes elements of a modest-demand, modest-cost BRT system. SOURCE: BRT (1), as reproduced in TCRP Report 90 (2) EXHIBIT 5-2 Packaging BRT Elements - High-Demand and High-Cost BRT System Services Stations Vehicles Running Way Systems Primarily local Simple stops No special treatment Mixed traffic Radios, on-board fare collection Mixed limited- stop, local Super stops Special signage Dedicated arterial curb lanes, competing turns allowed AVL for schedule adherence All-stop (local), mixed local/express On-line and off- line stations, significant parking for transit patrons Dedicated vehicles, special livery Dedicated freeway median lanes, merge/weave access/egress ITS passenger information, fare collection Point-to-point express Transfer/ transit centers Dedicated vehicles, uniquely specified (e.g., double- articulated buses, hybrid propulsion) Fully dedicated lanes, exclusive freeway access/egress ITS vehicle priority Intermodal transfer/transit center Mechanical or electronic guidance Partial grade separation ITS vehicle lateral guidance Fully electric propulsion system Full grade separation, curbed/striped/ca bled for guidance ITS automation, electric power system Overhead power contact system NOTE: Boldface text denotes elements of a high-demand, high-cost BRT system. SOURCE: BRT (1), as reproduced in TCRP Report 90 (2)

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-4 Bus Rapid Transit Practitioner’s Guide Exhibit 5-3 identifies the allocation of capital costs by BRT component for several existing and under construction BRT systems, illustrating different packaging of components. EXHIBIT 5-3 Allocation of Capital Costs by BRT Component BRT System Total Develop- ment Costs (millions) Land Acqui- sition Run- ning Way Sta- tions Buses ITS/TSP Design/ Adminis- tration/ Super- vision Other Adelaide, Australia $67.9 5.9% 54.5% 6.5% 22.5% — 1.5% design; 9.0% admin. — Boston $37.8 — 60.8% 9.6% 27.6% 2.0% CAD/AVL 2 — Brisbane, Australia $330.1 — 79.6% 2.5% — 2.0% ITS 2 15.9% tunnel Cleveland $168.4 8.1% 26.3% 10.8% 12.8% 5.1% TSP 26.1% 10.8% 3 Hartford $145.0 8.3% 37.1% 19.7% 7.7% 0.7% ITS 22.6% 3.9% 4 Las Vegas $19.2 0% — 23.4% 63.0% 1.5% AVL; 1.3% TSP 2 10.5% 5 Los Angeles: Wilshire- Whittier $5.0 0% 0% 48.7% 0% 51.3% TSP 2 — Los Angeles: Ventura $3.2 0% 0% 48.8% 0% 51.2% TSP 2 — Los Angeles: Phase 2 $101.9 0% 0% 42.9% 0.3% 55.0% TSP 2 1.8% opera- tions support Ottawa $324.0 — 69.0% 27.6% — — 2 3.4% park- and-ride Pittsburgh: East Busway Extension $68.8 14.5% 44.2% 2.9% 0% — 24.4% 13.4% 6 Pittsburgh: West Busway (PAT) $299.1 8.8% 73.9% 0.9% 0% — 2 16.4% 7 Vancouver, BC: 98B (from IBI Group) $41.3 8.9% 22.8%1 6.3% 33.4% 1.0% ITS; 3.9% TSP; 6.3% AVL 6.5% 10.9% garage 1 Includes 3.9% for landscaping 2 Design/administration/supervision costs not itemized in source data 3 1.0% for a maintenance facility, 0.6% for art, and 9.2% for contingencies 4 0.6% for traffic signals, 1.0% for railroad crossings, and 2.3% for a multi-use trail 5 0.6% for dynamic message signs and 9.9% for ticket vending machines 6 7.3% for a linear park and 6.1% for a park-and-ride lot 7 Wabash HOV facility NOTE: CAD = computer-assisted dispatch SOURCE: TCRP Project A-23A Interim Report (3) Major BRT cost components include running ways, stations, design, and (in some cases) vehicles.

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-5 System Packaging, Integration, and Assessment ASSESSING SYSTEM PERFORMANCE Assessing BRT system performance requires estimating travel times, service frequencies, ridership benefits, and development costs. This section brings together key performance and cost parameters and shows how they may be used to assess the effect of a given BRT system. Analysis Parameters BRT costs and travel times were obtained from TCRP Report 90 (2), TCRP Project A-23A Interim Report (3), CBRT (4), and project profiles are set forth in Exhibit 5-4 through Exhibit 5-9. This information can be used as a guide in estimating the costs and effects of various BRT features. Local experience and local information should be used where available. Exhibit 5-4 gives representative unit costs for running ways, including transit priority treatments, stations, vehicles, fare collection, passenger information systems, and ITS. Right-of-way costs have been excluded since they vary widely depending on the running way option and local circumstances. These costs are based on information contained in the BRT component profiles in Chapter 4, the TCRP Project A-23A Interim Report (3), and CBRT (4). Exhibit 5-5 gives typical effects of BRT running way treatments that are keyed to an initial base running time of 6 minutes per mile (10 miles per hour). The off- street travel times and speeds assume wide station spacing, while the on-street treatments are keyed to the initial station spacing. Exhibit 5-6 gives typical effects of station spacing and dwell times on bus travel time rates (in minutes per mile). If the station spacing remains constant and the dwell times change, the changes in the one-way running time represent the changes in dwell time at each station times the number of stations. Exhibit 5-7 shows how the number of door channels and type of fare collection influence passenger service times. For example, pre-payment has a service time of 2.5 seconds per passenger. When two doors are available, the dwell time per passenger per door is 2.5 x 0.6, or 1.5 seconds per passenger. Exhibit 5-8 presents typical cost and travel time effects for various running way options. A partially grade-separated busway would cost $3.0 million per minute of travel time saved. In contrast, TSP would cost $0.4 million per minute of travel time saved. Costs per person-minute saved depend upon the number of buses and the number of passengers per bus along the BRT route. Exhibit 5-9 gives estimated costs and estimated travel time savings for queue bypasses, curb extensions, and TSP. Time savings in seconds per mile ranges from 6 seconds (one queue bypass per mile) to 20 seconds (four TSP treatments per mile). Assessing system performance involves estimating changes in travel times, service frequencies, and ridership as compared with development costs. This guide provides unit costs for BRT components.

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-6 Bus Rapid Transit Practitioner’s Guide EXHIBIT 5-4 Representative BRT Component Development Costs Component Unit Cost/Unit Running Way Off-street busway At-grade Grade-separated Elevated Tunnel Per route-mile Per route-mile Per route-mile Per route-mile $5 million $13 million $50 million $200 million On-street Median arterial busway Bus lane - new construction Bus lane - striping lane Per route-mile Per route-mile Per route-mile $4 million $25 million $100,000 Transit Preferential Treatments Queue bypass Parking removal Use of right-turn lane Added lane Per approach Per approach Per approach Negligible Negligible $300,000 Curb extension Per extension $60,000 TSP Per intersection $30,000 Special transit phase Per intersection $10,000 Stations Typical Basic Enhanced Per station Per station $21,000* $30,000* Major At-grade Grade-separated Per station Per station $150,000 $2.5 million Intermodal center Per station $12.5 million Passing lane Per lane-mile $2.7 million Vehicles Conventional standard Per vehicle $325,000 Stylized standard Per vehicle $350,000 Conventional articulated Per vehicle $570,000 Stylized articulated Per vehicle $780,000 Specialized BRT Per vehicle $1.3 million Fare Collection On-board Magnetic card media Smart media Per vehicle Per vehicle $15,000 $20,000 Off-board Magnetic card media Smart media Per machine Per machine $60,000 $65,000 Passenger Information At-station information Per sign $6,000 On-board information Per vehicle $4,000 Branding Branding Per system Negligible ITS Applications On-board security Per vehicle $10,000 On-board vehicle guidance Optical/magnetic sensors Hardware integration Per mile Per vehicle $20,000 $50,000 On-board precision docking Optical/magnetic sensors Hardware integration Per station Per vehicle $4,000 $50,000 On-board performance monitoring Per vehicle $2,000 AVL Per vehicle $8,000 * One direction NOTE: Values are in 2004 U.S. dollars. SOURCE: TCRP Report 90 (2), TCRP Project A-23A Interim Report (3), CBRT (4), Vehicle Catalog 2005 Update (5), and TCRP Synthesis 40 (6)

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-7 System Packaging, Integration, and Assessment EXHIBIT 5-5 Typical Effects of BRT Running Way Components Component Estimated Effects Savings Compared to Base* Comments Elevated 40 mph, 1.5 min/mi 4.5 min/mi Assumed speed Some grade separation 35 mph, 1.7 min/mi 4.3 min/mi Reflects wide station spacing Off-street At-grade 25 mph, 2.4 min/mi 3.6 min/mi Median arterial busway 13.3 mph, 4.5 min/mi 1.5 min/mi Assumes no change in station spacing On-street Bus lane (new con- struction or striping) 12.2 mph, 4.9 min/mi 1.1 min/mi From TCRP A-23A April-June 2005 Quarterly Progress Report (7) Queue bypass — 6 sec/int Estimated Curb extension — 4 sec/int From TC&QSM (8), Exhibit 4-5, 400 vehicles per hour TSP — 5 sec/int From Los Angeles, Oakland Traffic treat- ments Special signal phase — — Has important safety benefits * Benefits are keyed to a base running speed of 10 mph (6 minutes/mile and 6 stations/mile). NOTE: int = intersection SOURCE: Derived from project profiles EXHIBIT 5-6 Typical Effects of BRT Station Spacing and Dwell Times Condition Before (6 stops per mile) After (2 stops per mile) Change Dwell/stop 15 seconds 15 seconds 0 seconds Same boarding times Minutes/mile 4.8 2.6 +2.2 Dwell/stop 15 seconds 20 seconds -5 seconds Slower boarding times Minutes/mile 4.8 2.7 +2.1 Dwell/stop 15 seconds 10 seconds 5 seconds Faster boarding times Minutes/mile 4.8 2.4 +2.4 NOTE: Excludes traffic delays SOURCE: Transit Capacity and Quality of Service Manual (8), Exhibit 4-6 Reducing the number of stops and reducing dwell times improves BRT speeds.

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-8 Bus Rapid Transit Practitioner’s Guide EXHIBIT 5-7 Typical Effects of Door Channels and Fare Collection Methods on Passenger Service Times Situation Single-Door Boarding Time (seconds/passenger)1 Swipe or dip card 4.5 Exact change 4.0 Smart card 3.5 Single ticket or token 3.5 Pre-payment2 2.5 Situation Single-Door Alighting Time (seconds/passenger)1 Front door 3.3 Rear door 2.1 Situation Proportion of Basic Dwell Time3 1 door channel 1.00 2 door channels 0.60 3 door channels 0.44 4 door channels 0.36 5 door channels 0.24 1 Add 0.5 second/passenger for standees. Subtract 0.5 second/passenger for low-floor buses. 2 Pre-payment includes no fare, bus pass, free transfer, and pay-on-exit. 3 The dwell times in the upper two-thirds of the table are reduced to these percentages as door channels are added. For example, adding a single-channel rear door to a bus that currently has one boarding channel can reduce dwell time to 60% of the current dwell time. These percentages assume fares are prepaid and can be applied to boarding or alighting. SOURCE: Transit Capacity and Quality of Service Manual (8), Exhibits 4- 2 and 4-3 EXHIBIT 5-8 Cost and Travel Time Savings of Various BRT Running Way Options Running Way Option Cost per Mile (millions) Time Savings per Mile (minutes) Cost per Minute Saved (millions) Partially grade-separated busway $13.00 4.30 $3.00 At-grade busway $5.00 3.60 1.40 Median arterial busway $4.00 1.50 2.70 Bus lane (rebuilt) $2.50 1.10* 2.30 Bus lane (re-striped) $0.10 1.10* 0.09 Queue bypass (add lane) $0.30* 0.10 3.00 Curb extension $0.24 0.27 0.90 TSP $0.12 0.33 0.40 * May be 0.5 to 0.7 minutes/mile for higher bus operating speeds NOTE: The base condition is a running speed of 10 mph (6 minutes/mile and 6 stations/mile). SOURCE: Exhibit 5-4 and Exhibit 5-5 EXHIBIT 5-9 Costs and Travel Time Savings of Preferential Treatments Treatment Approaches per Mile Cost/ Installation (millions) Cost/Mile (millions) Time Savings/ Unit (seconds) Time Savings/ Mile (seconds) Queue bypass (with construction) 1 $0.30 $0.30 6 6 Curb extension 4 $0.06 $0.40 4 16 TSP 4 $0.03 $0.12 3 20 SOURCE: Exhibit 5-8 and project profiles

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-9 System Packaging, Integration, and Assessment Analysis Steps and Procedures Key analysis steps in estimating the costs and effects of various BRT options are shown in Exhibit 5-10 and Exhibit 5-11. These steps should be applied, as appropriate, to BRT and local bus services in the same corridor. The steps are the following: 1. Define base conditions. 2. Define future conditions with BRT. 3. Estimate travel time savings. 4. Allocate base corridor ridership to BRT and other services (this reflects diverted riders from other base transit routes). 5. Estimate ridership gains from travel time savings. 6. Estimate ridership effects of improved service frequencies (where applicable). 7. Obtain the total BRT riders by adding the results of Steps 5 and 6. 8. Estimate the additional BRT riders from BRT features (this reflects new BRT riders). 9. Estimate the total base year BRT riders by adding the results of Steps 7 and 8. 10. Develop an initial estimate of BRT fleet requirements. 11. Estimate the effects of growth (including added fleet requirements). 12. Estimate the development costs of various BRT features. The “ridership” shown in Exhibit 5-11 reflects BRT riders diverted from existing bus routes in the BRT corridor. The “enhanced ridership” includes BRT trips diverted from automobiles and new trips. Where comprehensive surveys are not available (or where the BRT route is to be overlaid on local bus service), route ridership data can be used in conjunction with relative BRT and local bus travel times to divert riders to BRT. An on-board rider survey is desirable to provide detailed information on passenger origins and destinations and boarding/alighting patterns. The “new” riders are estimated by increasing ridership estimates based on elasticities to account for special BRT features. The application of these steps in analyzing and comparing BRT options is straightforward. Some general guidelines are as follows. Define Base Conditions The existing conditions in the proposed BRT corridor should be clearly defined. These include route structure, service frequencies, stop spacing and dwell times, travel times, and ridership. Travel times should be developed for each section or segment where operating characteristics differ (e.g., CBD and other). Average daily conditions are a point of departure; where possible, conditions should be defined for peak and off-peak conditions. There are 12 key steps in analyzing the cost and effects of BRT options. On-board rider surveys are valuable for identifying BRT rider travel patterns.

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-10 Bus Rapid Transit Practitioner’s Guide EXHIBIT 5-10 Key BRT Assessment Steps Step Items to Analyze 1. Define base conditions. A. Existing bus services B. Existing travel times C. Existing ridership 2. Define future conditions. A. Type of running ways B. Station types and spacing C. Vehicle type and door configuration D. Method of fare collection E. Transit priority treatments 3. Estimate travel time savings. A. BRT B. Other bus services 4. Allocate base corridor riders to BRT and local services. A. Rider survey to identify origin-to-destination patterns and preferences B. Relative travel times of various services 5. Estimate ridership gains from travel time savings (for BRT and other services). A. Effects of running way type B. Effects of station spacing and dwell times C. Effects of priority treatments 6. Estimate ridership gains from improved frequency. A. Greater frequency on BRT routes B. BRT riders who save time by taking first bus on combined BRT-local route 7. Subtotal ridership from Steps 5 and 6. 8. Estimate additional ridership from BRT features. A. Features on BRT route 9. Estimate total base year riders (Step 7 + Step 8). 10. Estimate BRT fleet requirements. A. Peak-hour peak direction riders in maximum load section B. Vehicle type, size, and passenger capacity C. Round-trip vehicle travel time (including schedule recovery) D. Provision for spares 11. Estimate effects of growth. A. Population and employment growth in BRT corridor 12. Estimate development costs of BRT components (features). Define Future Conditions The proposed BRT alignment and types of running ways should be established. The station types, station locations, frequencies, vehicle types and door configurations, methods of fare collection, and (in turn) station dwell times should be identified. Transit priority treatments that may improve BRT, and, in some cases, local bus performance, should be indicated. Initial BRT service frequencies should be established (generally not more than 10- to 12-minute headways). Changes in route structure in the corridor should be identified. Estimate Travel Time Savings The travel times for BRT service should be estimated taking into account the anticipated service pattern, future station spacing and dwell times, running way types, and transit preferential treatments. Travel time savings for local buses using priority lanes and other treatments also should be estimated. The travel time savings should be estimated based on the following order: • Type of running way • Station spacing/dwell times • Transit priority treatments BRT travel times are based on station spacing, dwell time, type of running way, and transit preferential treatments.

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-11 System Packaging, Integration, and Assessment SERVICE FREQUENCY Number/type Dw ell time - Fare collection - Door conf iguration - Floor height Type (on/off street) Type of priorities RUNNING TIME RIDERSHIP + Additional ridership gains from special BRT features ENHANCED RIDERSHIP STOPS RUNNING WAY NOTE: Additional ridership from population and economic growth EXHIBIT 5-11 BRT Ridership Analysis Concept Allocate Base Ridership to BRT and Local Bus Service The method of allocating existing bus riders to BRT and local bus service in the BRT corridor will depend on (1) where BRT will operate and (2) whether it replaces an existing service. When BRT replaces a single local service, all base ridership can be allocated to the BRT service. Where BRT will operate in the same corridor as local service, the existing ridership can be allocated based on boarding and alighting patterns. An approximately equal division between the two services provides a reasonable default value. Alternatively, the ridership allocation can be based on judgment, based on the division of ridership equally between both services, or (preferably) based upon origin-to-destination surveys, boarding and alighting patterns, market research, and/or relative travel times. (See Chapter 3 and Exhibit 3-18 for further discussion.) Where BRT will operate on a new alignment, the allocation should be based on a conventional demand modeling process wherever data are available. Existing experience with BRT and limited-stop bus service on city streets indicates that both services often operate at about the same frequencies. Thus, equal headways are desirable at major boarding points such as downtown areas. An initial allocation of riders between BRT and other services is necessary.

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-12 Bus Rapid Transit Practitioner’s Guide Estimate BRT Ridership BRT ridership for the base (and future) years should be estimated through the traditional trip generation, trip distribution, mode split, and trip assignment process wherever BRT will run on a new alignment such as a busway. The alternative is to apply elasticity methods or use an incremental logit model (pivot- point procedure). For situations where BRT operates on the same street with local bus service, elasticity methods may be appropriate. Typical travel time elasticity values of -0.3 to -0.5 can be used. (A value of -0.4 is used in subsequent examples. Similarly, a service frequency elasticity value of +0.4 was used.) Ridership based upon travel time elasticities should be calculated first, followed by application of service frequency elasticities as appropriate. Calculations should use the midpoint arc elasticity equation (Equation 3-8) shown in Chapter 3. (See Chapter 3 for more details.) Additional BRT ridership may result from providing various BRT features such as busways, specially delineated bus lanes, attractive stations, modern stylized vehicles, and passenger information systems. Accordingly, the base ridership should be increased up to 25% depending upon the extent of these features. Where a conventional modeling process is used, a travel time bias constant up to 10 minutes of in-vehicle travel time can be used. These adjustments account for the new, previously non-transit, trips that would be attracted to BRT. Estimate Fleet Requirements Peak BRT vehicle requirements (an input to cost estimates) can be obtained by converting the daily line ridership to peak-hour peak-direction riders at the maximum load section. These computations will depend upon the nature of the route, the likely turnover of passengers, the round-trip running times, vehicle size, and established loading standards. The basic relationships are as follows: DirectionPeakinHourPeakin Turnover RidersDaily P %% ××= (5-1) In Equation 5-1, P = peak-hour peak-direction passengers in the maximum load section. Turnover ranges from about 1.2 to 2.0 passengers per bus depending on the route structure and areas served. (Turnover is essentially the inverse of the ratio between the riders at the maximum load section and the daily ridership on a bus route.) The peak-hour peak-direction factor is about 0.05 to 0.07 (with 0.06 assumed for the purposes of this chapter). The number of buses in the maximum load section needs to carry the maximum volume of peak-hour passengers, n, which is equivalent to: BusperSpacesPassenger Pn = (5-2) In Equation 5-2, 50 passenger spaces per bus is assumed for regular buses and 60 passenger spaces per bus is assumed for articulated buses. The peak headway is 60 divided by the number of buses per hour (n). The fleet requirement equals: Spares HeadwayPeak TimeoveryecRLayoverTimeRunningTripRound + + (5-3) Daily ridership should be translated into peak-hour peak-direction ridership in the maximum load section. From this number, the necessary headway and fleet requirements can be computed.

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-13 System Packaging, Integration, and Assessment There usually should be at least two to three spare vehicles, or 10 to 20 percent of the fleet, whichever is greater. Estimate Effects of Growth BRT ridership can be expected to grow in future years as a result of population and employment growth and greater acceptance of the BRT service. Ridership estimates, therefore, are desirable for (1) several years after BRT ridership has stabilized and (2) future years, especially where major capital investments are involved. Fleet requirements should be adjusted accordingly. These estimates should be based on each community’s experience and projections. Estimate System Development Costs BRT development costs should be estimated based upon (1) local experience and (2) the values shown in Exhibit 5-4. They will, of course, vary depending upon the type and extent of BRT features. They should be estimated for each BRT feature and then aggregated. Example applications of these features are set forth in the following section. EXAMPLE BRT DEVELOPMENT SCENARIOS This section analyses the effects and costs of six different BRT scenarios. It describes each scenario, analyzes its travel time and ridership changes, and estimates its costs. Finally, the results of the six scenarios are compared and assessed. In many respects, the six scenarios represent an alternatives analysis of the candidate corridor. Costs and effects are based on the values contained in earlier sections of the chapter. Agencies should use locally observed values wherever available. Context and Assumptions The following assumptions underlie the analysis of each scenario. 1. The BRT route is 15 miles long—1 mile in the CBD and 14 miles in outlying areas. Buses operate in mixed traffic unless otherwise specified. 2. The existing bus speeds are 6 mph in the CBD (1 mile) and 10 mph elsewhere (14 miles). This translates into a 94-minute one-way travel time. 3. Existing daily ridership in the corridor for the six scenarios were assumed to range from 16,000 to 20,000. 4. Where the BRT and local service would run on the same street, the initial base ridership was allocated equally between BRT and local services. Initial allocations between BRT and local bus only include diverted riders. 5. The BRT headway does not exceed 10 minutes. 6. The BRT layover is 10 minutes. 7. The daily riders in the maximum load section were estimated by dividing the daily ridership by the “turnover.” The turnover ranged from 1.2 to 1.8. (Thus, a daily ridership of 12,000 would have 6,700 to 10,000 passengers at the maximum load section.) The lower values were used for busways with the CBD at one end of the route; the higher values were used for the CBD centrally located along an arterial street. 8. The peak-hour peak-direction ridership was assumed to be 6% of the daily riders in the maximum load section. This guide presents six BRT development scenarios to show how costs and effects of BRT can be estimated. Agencies should use locally observed values in assessing their BRT services. Several assumptions are inherent in the six scenarios.

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-14 Bus Rapid Transit Practitioner’s Guide 9. Sixty spaces per bus were assumed for each scenario (articulated buses). 10. Anticipated BRT ridership and fleet requirements for each scenario were developed for the base year. They should be adjusted, as appropriate, to reflect likely future growth. 11. Cost estimates were based on the values set forth in Exhibit 5-4. The scenarios are straightforward and enable comparison of costs and effects of various BRT features. They use several simplifying assumptions to facilitate computations: • The scenarios assume a single daily value of travel time and ridership. In practice, it is desirable to look at peak- and off-peak travel times and ridership in assessing BRT. The effects of traffic signal timing on bus speeds could be used to refine the travel time savings and their relation to ridership. Both BRT and local bus ridership would be built from the ground up on a segment-by-segment basis. • The scenarios use service elasticities. Incremental logit models can be used where detailed travel patterns and network information is available. • Assumptions were made regarding the allocation of base ridership between local service and BRT. The assumptions differed by scenario. The effects of other allocations can be assessed by iterating the procedure. • It was assumed that BRT stations would be placed at the locations of major passenger attraction, thus accounting for a large portion of existing ridership. Experience indicates that relatively few stations can account for most of a line’s ridership. Daily BRT riders reflect diverted and new trips. Rider surveys in conjunction with street, bus route, and land use patterns will influence the assignment of diverted riders to the BRT system. Stated preference surveys could provide further insight into desirable BRT features and their ridership impacts. In most cases, local bus service parallels BRT to serve short trips and trips where the origin or destination is beyond a convenient walking distance from a BRT station. For each scenario, the following exhibits were prepared: • A diagram that describes the scenario • A table that provides the assumptions for the scenario • A table that describes each step in the analysis procedure • A table that gives estimated development costs for each BRT feature The following six scenarios were analyzed: • Grade-separated busway (Exhibit 5-12) • At-grade busway (Exhibit 5-16) • At-grade busway and median arterial busway (Exhibit 5-20) • Bus lanes and TSP (Exhibit 5-24) • Bus lanes only (Exhibit 5-28) • TSP only (Exhibit 5-32) The six BRT development scenarios show the effects of various running way types and station spacings.

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-15 System Packaging, Integration, and Assessment Scenario 1: Grade-Separated Busway Connecting CBD to Park-and- Ride Lot The description and assumptions for this scenario are shown in Exhibit 5-12 and Exhibit 5-13. Exhibit 5-14 describes the detailed analysis and Exhibit 5-15 gives the estimated costs by BRT feature. A 14-mile grade-separated busway with 10 stations connects with a 1-mile curbside bus lane in the CBD with three stations. Specialized, articulated BRT vehicles would be used. Fare collection would be off the vehicle. The local bus route would remain on city streets. The BRT route would have the following effects: • Reduce the number of stops from 90 to 13 • Reduce the one-way running time from 94 to 29 minutes • Increase the daily BRT ridership from 10,000 to almost 18,000 Development costs for Scenario 1 were estimated at $242.0 million. CBD PARK-AND-RIDE LOT LOCAL BUS ROUTE BUSWAY (14 MILES, 10 STATIONS) BUS LANE (1 MILE, 3 STATIONS) VACANT LAND 1,000 SPACES EXHIBIT 5-12 BRT Development Scenario 1: Grade-Separated Busway EXHIBIT 5-13 Key Assumptions of BRT Development Scenario 1 Computed Results Feature Existing Service (Base) Local BRT Daily Ridership 20,000 10,000 17,661 Stops 90 90 13 Dwell/Stop 15 sec 15 sec 20 sec1 10-15 sec2 Frequency 8 min 10-min minimum 4 min Speed 6 mph CBD; 10 mph elsewhere 9.8 mph 31 mph2 One-Way Travel Time 94 min 92 min 29 min Fare Collection On board On board On board Vehicle Conventional Conventional Specialized Passenger Information Stations Vehicles No No No No Yes Yes Branding No No Yes Development Potential N/A At outer end At outer end Development Costs N/A N/A $242.0 million 1 The 20-second dwell time for BRT assumes no change in fare collection or door channels available, in comparison to the existing service. The 20-second dwell time for BRT is higher than the dwell time for local bus because BRT has fewer stops and higher passenger boardings. 2 With off-vehicle fare collection and multi-door boarding

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-16 Bus Rapid Transit Practitioner’s Guide EXHIBIT 5-14 Analysis - Scenario 1 Item Analysis Existing Condition:
1 mi at 6 mph (10 min/mi) = 10 min (CBD mixed traffic)
14 mi at 10 mph (6 min/mi) = 84 min (mixed traffic)
Total travel time = 94 min Proposed Condition 1 - Local:
1 mi (CBD) at 8 min/mi = 8 min  Start with 6-mph speed  1.5-mph gain due to bus lane speeds  Net speed = 7.5 mph = 8 min/mi
14 mi at 10 mph (6 min/mi) = 84 min (on local streets)
Total travel time = 92 min Proposed Condition 2 - BRT:
1 mi (CBD) at 7 min/mi = 7 min  Start with 6-mph speed  1.5-mph (2 min/mi) gain from CBD bus lane  1.1-mph (1 min/mi) gain from fewer stops (only 3 in CBD)  Net speed = 8.6 mph = 7 min/mi
14 mi at 38 mph (1.58 min/mi) = 22 min (busway)  50-mph busway running speed  10 stops at 1.4 mi/stop  For 1.5 mi between stops, 15-sec dwell = 38 mph*  Net speed = 38 mph = 1.58 min/mi
Total travel time = 29 min (overall average speed 31 mph) Travel times If the BRT scenario uses on-vehicle fare collection, assume a dwell of 20 sec/stop, or 37 mph. This translates to 30 min for the total trip. The average speed for the total trip is then 35 mph. Assumed initial allocation of base riders:
BRT: 10,000 riders and 10-min maximum headway
Local: 10,000 riders and 10-min minimum headway The preferred method of estimating BRT and local bus ridership is using the four-step model (trip generation, trip distribution, mode split, and trip assignment) and applying incremental logit models. An alternative method using elasticities is described below. Ridership estimates 1. Apply travel time elasticity factors to BRT: 12 1211 2 )1()1( )1()1( TETE RTERTER +−− +−− = where R1 = initial ridership, R2 = anticipated ridership, T1 = initial travel times, T2 = travel times with BRT, and E = travel time elasticity factor = -0.4. 361,15 )946.0()294.1( )000,10296.0()000,10944.1( 2 = ×−×− ××−××− =R

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-17 System Packaging, Integration, and Assessment 2. Estimate additional ridership generated by BRT features. Weights applied to up to 25% increase in base ridership are obtained from elasticity computations. (See Exhibit 3-22 for details.) Component Percentage Running Way 20% Stations 12% Vehicles (normal floor boarding) 10% Service pattern 15% ITS (assumed) 10% Branding 10% Subtotal 77% BRT component synergy 15% Total 92% The 92% applies to an increase in base ridership of 25% beyond that obtained by elasticities. 0.92 x 25% = 23%. Anticipated additional BRT ridership = 10,000 x 23% = 2,300. Total anticipated BRT ridership in the base year = 15,361 + 2,300 = 17,661. Peak-hour peak-direction riders in the maximum load section (P): DirectionPeakinHourPeakin Turnover RidersDaily P %% ××= Turnover is assumed to be 1.20. (The BRT line extends on only one side of the CBD.) 883%)60(%)10( 20.1 661,17 ==P At 60 passengers per bus, 15 buses are needed in the maximum load section (with 4-minute headways). Fleet requirements BRT vehicles needed: 17 4 1058 = + = + Headway TimeLayoverTimeRunningTripRound Add 4 spares to get 21 buses. Estimated costs See Exhibit 5-15. * From Transit Capacity and Quality of Service Manual (8), Exhibit 4-47

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-18 Bus Rapid Transit Practitioner’s Guide EXHIBIT 5-15 Estimated Development Costs - Scenario 1 Item Units Unit Cost Total Cost Busway 14 miles $13 million/mile $182 million Bus lane 1 mile $100,000/mile $100,000 Stations - busway 10 stations $2.5 million/station $25 million Stations - CBD 3 stations $60,000/station $180,000 Passing lane 2 lane-miles1 $2.7 million/lane-mile $5.4 million Specialized, articulated BRT vehicles 21 vehicles $1.3 million/vehicle $27.3 million Off-board fare collection 28 ticket vending machines2 $65,000/machine $1.8 million Station information 26 locations3 $6,000/location $156,000 Vehicle information 21 vehicles $4,000/vehicle $84,000 Total (2004 dollars) $242.0 million 1 (10 stations x 2 directions) at 0.1 mile each 2 Two per station plus two additional machines 3 13 stations x 2 locations/station NOTE: Excludes park-and-ride lot costs Scenario 2: At-Grade Busway The description and assumptions for this scenario are shown in Exhibit 5-16 and Exhibit 5-17. Exhibit 5-18 describes the detailed analysis, and Exhibit 5-19 gives the estimated costs by BRT feature. An at-grade busway extends for 7 miles on each side of the city center. The two busways are connected by bus lanes on one mile of downtown streets. Specialized, articulated BRT vehicles would be used. Fare collection would occur off of the vehicle. The local bus route would remain on city streets. The BRT service would have the following effects: • Reduce the number of stops from 90 to 17 • Reduce the one-way running time from 94 to 43 minutes • Increase the daily BRT ridership from 10,000 to 15,699 Development costs for Scenario 2 were estimated at $109.4 million. CBD LOCAL BUS ROUTE AT-GRADE BUSWAY (7 MILES, 7 STATIONS) AT-GRADE BUSWAY (7 MILES, 7 STATIONS) BUS LANE (1 MILE, 3 STATIONS) EXHIBIT 5-16 BRT Development Scenario 2

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-19 System Packaging, Integration, and Assessment EXHIBIT 5-17 Key Assumptions of BRT Development Scenario 2 Computed Results Feature Existing Service (Base) Local BRT Daily Ridership 20,000 10,000 15,699 Stops 90 90 17 Dwell/Stop 15 sec 15 sec 20 sec1 10-15 sec2 Frequency 8 min 10-min minimum 6 min Speed 6 mph CBD; 10 mph elsewhere 9.8 mph 20.9 mph2 One-Way Travel Time 94 min 92 min 43 min Fare Collection On board On board On board Vehicle Conventional Conventional Specialized TSP No No Along busway Passenger Information Stations Vehicles No No No No Yes Yes Branding No No Yes Development Costs N/A N/A $109.4 million 1 The 20-second dwell time for BRT assumes no change in fare collection or door channels available, in comparison to the existing service. The 20-second dwell time for BRT is higher than the dwell time for local bus because BRT has fewer stops and higher passenger boardings. 2 With off-vehicle fare collection and multi-door boarding EXHIBIT 5-18 Analysis - Scenario 2 Item Analysis Existing Condition:
1 mi at 6 mph (10 min/mi) = 10 min (CBD mixed traffic)
14 mi at 10 mph (6 min/mi) = 84 min (mixed traffic)
Total travel time = 94 min Proposed Condition 1 - Local:
1 mi (CBD) at 8 min/mi = 8 min  Start with 6-mph speed  1.5-mph gain due to bus lane speeds  Net speed = 7.5 mph = 8 min/mi
14 mi at 10 mph (6 min/mi) = 84 min (on local streets)
Total travel time = 92 min Travel times Proposed Condition 2 - BRT:
1 mi (CBD) at 7 min/mi = 7 min  Start with 6-mph speed  1.5-mph (2 min/mi) gain from CBD bus lane  1.1-mph (1 min/mi) gain from fewer stops (only 3 in CBD)  Net speed = 8.6 mph = 7 min/mi
14 mi at 22 mph (2.7 min/mi) = 38 min (busway)  2.7 min/mi is the initial speed for L.A.’s Orange Line busway  TSP time saving, 28 locations at 5 sec/intersection = 140 sec, or 2 min (rounded)
Total travel time = 43 min (overall average speed 20.9 mph) Ridership estimates Assumed initial allocation of base riders:
BRT: 10,000 riders and 10-min maximum headway
Local: 10,000 riders and 10-min minimum headway The preferred method of estimating BRT and local bus ridership is using the four-step model and applying incremental logit models. An alternative method using elasticities is described below.

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-20 Bus Rapid Transit Practitioner’s Guide 1. Apply travel time elasticity factors to BRT: 12 1211 2 )1()1( )1()1( TETE RTERTER +−− +−− = where R1 = initial ridership, R2 = anticipated ridership, T1 = initial travel times, T2 = travel times with BRT, and E = travel time elasticity factor = -0.4. 499,13 )946.0()434.1( )000,10436.0()000,10944.1( 2 = ×−×− ××−××− =R Since the BRT service is on a new alignment, there would be no increase in frequency that would allow application of service frequency elasticities. 2. Estimate additional ridership generated by BRT features. Weights applied to up to 25% increase in base ridership are obtained from elasticity computations. (See Exhibit 3-22 for details.) Component Percentage Running Way 15% Stations 12% Vehicles (normal floor boarding) 10% Service pattern 15% ITS (assumed) 10% Branding 10% Subtotal 72% BRT component synergy 15% Total 87% The 87% applies to an increase in base ridership of 25% beyond that obtained by elasticities. 0.87 x 25% = 22%. Anticipated additional BRT ridership = 10,000 x 22% = 2,200. Total anticipated BRT ridership in the base year = 13,499 + 2,200 = 15,699. Peak-hour peak-direction riders in the maximum load section (P): DirectionPeakinHourPeakin Turnover RidersDaily P %% ××= Turnover is assumed to be 1.80. (The BRT line extends on both sides of the CBD.) 523%)60(%)10( 80.1 699,15 ==P At 60 passengers per bus, 9 to 10 buses are needed in the maximum load section (with approximate 6-minute headways). Fleet requirements BRT vehicles needed: 16 6 1086 = + = + Headway TimeLayoverTimeRunningTripRound Add 4 spares to get 20 buses. Estimated costs See Exhibit 5-19.

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-21 System Packaging, Integration, and Assessment EXHIBIT 5-19 Estimated Development Costs - Scenario 2 Item Units Unit Cost Total Cost Busway 14 miles $5 million/mile $70.0 million Bus lane 1 mile $100,000/mile $100,000 Stations - busway 14 stations $150,000/station $2.1 million Stations - CBD 3 stations $60,000/station $180,000 Passing lanes 2.8 lane-miles1 $2.7 million/lane-mile $7.6 million Specialized, articulated BRT vehicles 20 vehicles $1.3 million/vehicle $26.0 million Off-board fare collection 36 ticket vending machines2 $65,000/machine $2.3 million Station information 34 locations3 $6,000/location $204,000 Vehicle information 20 vehicles $4,000/vehicle $80,000 TSP 28 intersections $30,000/intersection $840,000 Total (2004 dollars) $109.4 million 1 (14 stations x 2 directions) at 0.1 mile each 2 Two per station plus two additional machines 3 17 stations x 2 locations/station Scenario 3: At-Grade Busway and Median Arterial Busway The description and assumptions for this scenario are shown in Exhibit 5-20 and Exhibit 5-21. Exhibit 5-22 describes the detailed analysis, and Exhibit 5-23 gives the estimated development costs for various BRT features. A one-mile pair of downtown bus lanes connects with a 5-mile median arterial busway, a 5-mile at-grade busway, and 4 miles of mixed-traffic operations. Stylized, articulated buses would be operated, and fare collection would be off the vehicle. TSP would be provided at signalized intersections outside the CBD. The BRT service would replace the local bus service, which has a base daily ridership of 20,000. The BRT service would have the following effects: • Reduce the number of stops from 90 to 22 • Reduce the one-way running time from 94 to 47.9 minutes • Increase the daily ridership from 20,000 to 33,022 If the base ridership were 10,000, the BRT ridership would be 16,511. Development costs for Scenario 3 were estimated at $84.3 million. CBD MEDIAN ARTERIAL BUSWAY (5 MILES, 10 STATIONS) MIXED TRAFFIC (4 MILES, 4 STATIONS) BUS LANE (1 MILE, 3 STATIONS) AT-GRADE BUSWAY (5 MILES, 5 STATIONS) EXHIBIT 5-20 BRT Development Scenario 3

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-22 Bus Rapid Transit Practitioner’s Guide EXHIBIT 5-21 Key Assumptions of BRT Development Scenario 3 Feature Existing Service (Base) Computed Results (BRT) Daily Ridership 20,000 33,022 Stops 90 22 Dwell/Stop 15 sec 20 sec1 10-15 sec2 Frequency 8 min 3 min Speed 6 mph CBD; 10 mph elsewhere 18.8 mph One-Way Travel Time 94 min 47.9 min Fare Collection On board Off board Vehicle Conventional Stylized, articulated TSP No Yes Passenger Information Stations Vehicles No No Yes Yes Branding No Yes Development Costs N/A $84.3 million 1 The 20-second dwell time for BRT assumes no change in fare collection or door channels available, in comparison to the existing service. The 20-second dwell time for BRT is higher than the dwell time for local bus because BRT has fewer stops and higher passenger boardings. 2 With off-vehicle fare collection and multi-door boarding EXHIBIT 5-22 Analysis - Scenario 3 Item Analysis Travel times Existing Condition:
1 mi at 6 mph (10 min/mi) = 10 min (CBD mixed traffic)
14 mi at 10 mph (6 min/mi) = 84 min (mixed traffic)
Total travel time = 94 min

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-23 System Packaging, Integration, and Assessment Proposed Condition:
1 mi (CBD) at 7 min/mi = 7 min  Start with 6-mph speed  1.5-mph (2 min/mi) gain from CBD bus lane  1.1-mph (1 min/mi) gain from fewer stops (only 3 in CBD)  Net speed = 8.6 mph = 7 min/mi
5 mi at 19 mph (3.13 min/mi) = 15.6 min (median arterial busway)  For 2 stations/mi and 20-sec dwell, using Exhibit 4-56 of TC&QSM (8) results in a base speed of 2.73 min/mi  Add 0.7 min/mi for traffic delay from Exhibit 4-57 of TC&QSM (8)  Subtract 0.3 min/mi for TSP (5 sec/signal for 4 signals/mi)  Net speed = 19 mph = 3.13 min/mi
5 mi at 22.6 mph (2.66 min/mi) = 13.3 min (at-grade busway)  For 2 stations/mi and 20-sec dwell, using Exhibit 4-56 and Exhibit 4-57 of TC&QSM (8) results in a base speed of 2.73 min/mi  Busway has 1 station/mi, so subtract 0.6 min/mi for one less stop, acceleration, and deceleration based on Exhibit 5-6 of this Guide  Add 0.7 min/mi for traffic delay from Exhibit 4-57 of TC&QSM (8)  Subtract 0.17 min/mi (5 sec/signal for 2 signals/mi) for TSP  Net speed = 2.66 min/mi = 22.6 mph
4 mi at 20 mph (3.0 min/mi) = 12 min (mixed traffic)  With 2 stations/mi and 20-sec dwell, using Exhibit 4-56 of TC&QSM (8) results in a base speed of 2.73 min/mi  Busway has 1 station/mi, so subtract 0.6 min/mi for one less stop, acceleration, and deceleration based on Exhibit 5-6 of this Guide  Add 1.2 min/mi for traffic delay from Exhibit 4-57 of TC&QSM (8)  Subtract 0.33 min/mi (5 sec/signal for 4 signals/mi) for TSP  Net speed = 3.0 min/mi = 20 mph
Total travel time = 47.9 min
Total speed = 18.8 mph The headway using existing (non-articulated) buses is 8 minutes. The BRT headway using articulated buses is 6 minutes. It is assumed that all bus service will be BRT service. The preferred method of estimating BRT ridership is using the four-step model and applying incremental logit models. An alternative method using elasticities is described below. 1. Apply travel time elasticity factors to BRT: 12 1211 2 )1()1( )1()1( TETE RTERTER +−− +−− = where R1 = initial ridership, R2 = anticipated ridership, T1 = initial travel times, T2 = travel times with BRT, and E = travel time elasticity factor = -0.4. 974,25 )946.0()9.474.1( )000,209.476.0()000,20944.1( 2 = ×−×− ××−××− =R Ridership estimates 2. Estimate increased ridership due to increase in service frequency (R3). Elasticity (frequency) = +0.4 Existing 8-min headway = 7.5 buses/hour Proposed 6-min headway = 10 buses/hour 122,29 )5.74.1()106.0( )974,25104.1()974,255.76.0( 3 = ×−×− ××−××− =R

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-24 Bus Rapid Transit Practitioner’s Guide 3. Estimate additional ridership generated by BRT features. Weights applied to up to 25% increase in base ridership are obtained from elasticity computations. (See Exhibit 3-22 for details.) Component Percentage Running Way Median arterial busway (5 mi) (5 mi/15 mi) x 10% = 3.3% At-grade busway (5 mi) (5 mi/15 mi) x 15% = 5% Mixed traffic (4 mi) (4 mi/15 mi) x 0% = 0% CBD bus lane (1 mi) (1 mi/15 mi) x 0% = 0% Total 8% (weighted average) Stations 10% Vehicles (normal floor boarding) 10% Service pattern 15% ITS (assumed) 10% Branding 10% Subtotal 63% BRT component synergy 15% Total 78% The 78% applies to an increase in base ridership of 25% beyond that obtained by elasticities. 0.78 x 25% = 19.5%. Anticipated additional BRT ridership = 20,000 x 19.5% = 3,900. Total anticipated BRT ridership in the base year = 29,122 + 3,900 = 33,022. Peak-hour peak-direction riders in the maximum load section (P): DirectionPeakinHourPeakin Turnover RidersDaily P %% ××= Turnover is assumed to be 1.50. 101,1%60%10 8.1 022,33 =××=P At 60 passengers per bus, approximately 20 buses are needed in the maximum load section (with 3-minute headways). Vehicle requirements BRT vehicles needed: 36 3 108.95 = + = + Headway TimeLayoverTimeRunningTripRound Add 4 spares to get 40 buses. Estimated costs See Exhibit 5-23.

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-25 System Packaging, Integration, and Assessment EXHIBIT 5-23 Estimated Development Costs - Scenario 3 Item Units Unit Cost Total Cost Bus lane 1 mile $0.5 million/mile $500,000 Median arterial busway 5 miles $4 million/mile $20 million At-grade busway 5 miles $5 million/mile $25 million Stations - CBD 3 stations $60,000/station $180,000 Stations - median arterial busway 10 stations $150,000/station $1.5 million Stations - at-grade busway 5 stations $150,000/station $750,000 Stations - mixed traffic 4 stations $60,000/station $240,000 Passing lanes (busway) 1.0 lane-mile1 $2.7 million/lane-mile $2.7 million Stylized articulated BRT vehicles 40 vehicles $780,000/vehicle $31.2 million Off-board fare collection 6 ticket vending machines2 $65,000/machine $390,000 Station information 44 locations3 $6,000/location $264,000 Vehicle information 40 vehicles $4,000/vehicle $160,000 TSP 46 intersections $30,000/intersection $1.4 million Total (2004 dollars) $84.3 million 1 (Five stations x two directions) at 0.1 mile each 2 Two per station at three CBD locations 3 22 stations x 2 locations/station Scenario 4: Bus Lanes and Transit Signal Priority The description and assumptions for this scenario are shown in Exhibit 5-24 and Exhibit 5-25. Exhibit 5-26 describes the detailed analysis, and Exhibit 5-27 gives the estimated development costs for various BRT features. This scenario includes 11 miles of bus lane, of which 10 miles would have specially delineated pavement. The outlying two miles on each side of the route that extends through the CBD includes operation in mixed traffic. TSP would be provided at 12 non-CBD locations. The BRT service would use stylized articulated buses with on-vehicle fare collection. This scenario would have the following effects: • Reduce the number of stops from 90 to 31 • Reduce one-way travel times for the BRT service from 94 to 49.7 minutes • Daily BRT ridership would increase from 8,000 to 11,600 Local buses using the bus lanes and TSP would have their one-way running time reduced from 94 to 81 minutes. Development costs for Scenario 4 were estimated at $40.3 million.

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-26 Bus Rapid Transit Practitioner’s Guide CBD BUS LANE (5 MILES, 10 STATIONS) 0 of 6 SIGNALS WITH PRIORITY MIXED TRAFFIC (2 MILES, 4 STATIONS) BUS LANE (5 MILES, 10 STATIONS) MIXED TRAFFIC (2 MILES, 4 STATIONS) 2 OF 8 SIGNALS WITH PRIORITY 4 OF 20 SIGNALS WITH PRIORITY 4 OF 20 SIGNALS WITH PRIORITY 2 OF 8 SIGNALS WITH PRIORITY BUS LANE (1 MILE, 3 STATIONS) EXHIBIT 5-24 BRT Development Scenario 4 EXHIBIT 5-25 Key Assumptions of BRT Development Scenario 4 Computed Results Feature Existing Service (Base) Local BRT Daily Ridership 16,000 8,490 11,600 Stops 90 90 31 Dwell/Stop 15 sec 15 sec 20 sec1 Frequency 8 min 10 min 10 min Speed 6 mph CBD; 10 mph elsewhere 11.2 mph 18.1 mph One-Way Travel Time 94 min 81 min 49.7 min Fare Collection On board On board On board Vehicle Conventional Conventional Stylized, articulated Passenger Information Stations Vehicles No No No No Yes Yes TSP No No 12 signals Development Costs N/A N/A $40.3 million 1 The 20-second dwell time for BRT assumes no change in fare collection or door channels available, in comparison to the existing service. The 20-second dwell time for BRT is higher than the dwell time for local bus because BRT has fewer stops and higher passenger boardings. EXHIBIT 5-26 Analysis - Scenario 4 Item Analysis Existing Condition:
1 mi at 6 mph (10 min/mi) = 10 min (CBD mixed traffic)
14 mi at 10 mph (6 min/mi) = 84 min (mixed traffic)
Total travel time = 94 min Travel times Proposed Condition - Local:
Local buses would also benefit from a bus-only lane.
1 mi (CBD) at 7 min/mi = 7 min  Start with 6-mph speed  1.5-mph (2 min/mi) gain from CBD bus lane  1.1-mph (1 min/mi) gain from fewer stops (only 3 in CBD)  Net speed = 8.6 mph = 7 min/mi
10 mi at 12 mph (5 min/mi, non-CBD bus lanes) = 50 min  Start with 10 mph = 6 min/mi  Subtract 1 min/mi (see Exhibit 5-8 of this Guide)  No TSP  Net speed = 5 min/mi
4 mi at 10 mph (6 min/mi, mixed traffic) = 24 min
Total travel time = 81 min

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-27 System Packaging, Integration, and Assessment Proposed Condition - BRT:
1 mi (CBD) at 7 min/mi = 7 min  Start with 6-mph speed  1.5-mph (2 min/mi) gain from CBD bus lane  1.1-mph (1 min/mi) gain from fewer stops (only 3 in CBD)  Net speed = 8.6 mph = 7 min/mi
10 mi at 22 mph (2.73 min/mi) = 27.3 min (non-CBD bus lanes)  With 2 stations/mi and 20-sec dwell, using Exhibit 4-56 of TC&QSM (8) results in a base speed of 2.73 min/mi  Add 0.7 min/mi for traffic delay from Exhibit 4-57 of TC&QSM (8)  Subtract 0.7 min/mi (5 sec/signal for 8 signals) for TSP  Net speed = 2.73 min/mi = 22 mph
4 mi at 15.6 mph (3.85 min/mi) = 15.4 min (mixed traffic)  With 2 stations/mi and 20-sec dwell, using Exhibit 4-56 of TC&QSM (8) results in a base speed of 2.73 min/mi  Add 1.2 min/mi for traffic delay from Exhibit 4-57 of TC&QSM (8)  Subtract 0.08 min/mi (5 sec/signal for 4 signals) for TSP  Net speed = 3.85 min/mi = 15.6 mph
Total travel time = 49.7 min
Total speed = 18.1 mph Assumed initial allocation of base riders:
BRT: 8,000 riders and 10-min maximum headway
Local: 8,000 riders and 10-min minimum headway The method of estimating BRT and local bus ridership using the four-step model and applying incremental logit models may be applicable for bus lanes on city streets. A method using elasticities is described below. Ridership estimates 1a. Apply travel time elasticity factors to BRT for improved travel time: 12 1211 2 )1()1( )1()1( TETE RTERTER +−− +−− = where R1 = initial ridership, R2 = anticipated ridership, T1 = initial travel times, T2 = travel times with BRT, and E = travel time elasticity factor = -0.4. 251,10 )946.0()7.494.1( )000,87.496.0()000,8944.1( 2 = ×−×− ××−××− =R 1b. Apply travel time elasticity factors to local bus service for improved travel time: 12 1211 2 )1()1( )1()1( TETE RTERTER +−− +−− = where R1 = initial ridership, R2 = anticipated ridership, T1 = initial travel times, T2 = travel times with BRT, and E = travel time elasticity factor = -0.4. 490,8 )946.0()814.1( )000,8816.0()000,8944.1( 2 = ×−×− ××−××− =R

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-28 Bus Rapid Transit Practitioner’s Guide 2a. Estimate the proportion of BRT riders who would save time by taking the first bus (BRT or local) that arrives. An approximate estimate can be obtained based upon the following relationship: 2 1 2 t ha ×= where a = approximate length of route where riders would take first bus (assumed equal to % of riders), h = BRT headway (min), and t2 = time saved over local bus (total length of route). For Scenario 4, h = 10 min and maximum time savings is 94 – 49.7 = 44.3 min. %11 3.44 1 2 10 =×=a The proportion of BRT riders who would save time by taking the first bus that arrives is 11% x 10,251 = 1,128. 2b. Estimate increased ridership due to increase in service frequency (R3). Elasticity (frequency) = +0.4 Initial 8-min headway = 7.5 buses/hour Proposed 5-min headway (combined routes) = 12 buses/hour 357,1 )5.74.1()126.0( )128,1124.1()128,15.76.0( 3 = ×−×− ××−××− =R Since 1,128 BRT riders were already included in the ridership estimate, the net increase is 1,357 - 1,128 = 229 BRT riders. 3. Estimate additional ridership generated by BRT features. Weights applied to up to 25% increase in base ridership are obtained from elasticity computations. (See Exhibit 3-22 for details). Component Percentage Running Way Bus lanes, special pavement (10 mi) (10 mi/15 mi) x 5% = 3.3% Mixed traffic (4 mi) (4 mi/15 mi) x 0% = 0% CBD bus lane (1 mi) (1 mi/15 mi) x 0% = 0% Total 3% (weighted average) Stations (enhanced) 10% Vehicles (stylized, articulated) 10% Service pattern 12% ITS 10% Branding 10% Subtotal 55% BRT component synergy 0% Total 55% The 55% applies to an increase in base ridership of 25% beyond that obtained by elasticities. 0.55 x 25% = 14%. Anticipated additional BRT ridership = 8,000 x 14% = 1,120. Total anticipated BRT ridership in the base year considering only travel time savings = 10,251 + 1,120 = 11,371. Total anticipated BRT ridership in the base year considering both travel time and service frequency changes = 10,251 + 229 + 1,120 = 11,600. If 10,000 riders were initially allocated to BRT, the preceding ridership values would be increased 10/8 or 25% to 14,214 and 14,500, respectively. Fleet requirements Peak-hour peak-direction riders in the maximum load section (P): DirectionPeakinHourPeakin Turnover RidersDaily P %% ××=

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-29 System Packaging, Integration, and Assessment Turnover is assumed to be 1.80. 379%)60(%)10( 80.1 371,11 ==P to 387%)60(%)10( 80.1 600,11 ==P At 60 passengers per bus, seven buses are needed in the maximum load section (with approximate 10-minute headways). BRT vehicles needed: 2.12 10 10112 = + = + Headway TimeLayoverTimeRunningTripRound Round up to 13. Add 3 spares to get 16 buses. Estimated costs See Exhibit 5-27. EXHIBIT 5-27 Estimated Development Costs - Scenario 4 Item Units Unit Cost Total Cost Bus lane - CBD 1 mile $100,000/mile $100,000 Bus lane - outlying (special pavement) 10 miles $2.5 million/mile $25 million Mixed traffic 4 miles — — Stations - CBD (enhanced) 3 stations $60,000/station $180,000 Stations - bus lanes (enhanced) 20 stations $60,000/station $1.2 million Stations - mixed traffic (enhanced) 8 stations $60,000/station $480,000 Stylized, articulated BRT vehicles 16 vehicles $780,000/vehicle $12.5 million Station information 62 locations* $6,000/location $372,000 Vehicle information 16 vehicles $4,000/vehicle $64,000 TSP 12 intersections $30,000/intersection $360,000 Total (2004 dollars) $40.3 million * Two per station (for 31 stations) Scenario 5: Bus Lanes Only (No Transit Signal Priority) The description and assumptions for this scenario are shown in Exhibit 5-28 and Exhibit 5-29. Exhibit 5-30 describes the detailed analysis, and Exhibit 5-31 gives the estimated development costs for various BRT features. This scenario includes 11 miles of bus lanes and 4 miles of mixed-traffic operation (on each side of the CBD). Conventional articulated buses with on- vehicle fare collection would be used. The bus lanes would be delineated by pavement markings and signage. This scenario would have the following effects: • Reduce the number of stops from 90 to 31 • Reduce BRT one-way running times from 94 to 57 minutes • Daily BRT ridership would likely increase from 8,000 to 10,886 The one-way running time for local buses using the bus lanes would be reduced from 94 to 81 minutes. Development costs for Scenario 5 were estimated

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-30 Bus Rapid Transit Practitioner’s Guide at $12.5 million. The cost savings over Scenario 4 results from not rebuilding the bus lane. CBD BUS LANE (5 MILES, 10 STATIONS) 6 TRAFFIC SIGNALS MIXED TRAFFIC (2 MILES, 4 STATIONS) BUS LANE (5 MILES, 10 STATIONS) MIXED TRAFFIC (2 MILES, 4 STATIONS) 8 TRAFFIC SIGNALS 20 TRAFFIC SIGNALS 20TRAFFIC SIGNALS 8 TRAFFIC SIGNALS BUS LANE (1 MILE, 3 STATIONS) EXHIBIT 5-28 BRT Development Scenario 5 EXHIBIT 5-29 Key Assumptions of BRT Development Scenario 5 Computed Results Feature Existing Service (Base) Local BRT Daily Ridership 16,000 8,490 10,886 Stops 90 90 31 Dwell/Stop 15 sec 15 sec 20 sec1 Frequency 8 min 10-min minimum 10-min maximum Speed 6 mph CBD; 10 mph elsewhere 11.2 mph 15.8 mph One-Way Travel Time 94 min 81 min 57 min Fare Collection On board On board On-board smart card Vehicle Conventional Conventional Conventional, articulated Passenger Information Stations Vehicles No No No No Yes Yes Development Costs N/A N/A $12.5 million 1 The 20-second dwell time for BRT assumes no change in fare collection or door channels available, in comparison to the existing service. The 20-second dwell time for BRT is higher than the dwell time for local bus because BRT has fewer stops and higher passenger boardings. EXHIBIT 5-30 Analysis - Scenario 5 Item Analysis Travel times Existing Condition:
6 mph over 1 mi = 10 min (CBD mixed traffic)
10 mph over 14 mi = 84 min (mixed traffic)
Total = 94 min

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-31 System Packaging, Integration, and Assessment Proposed Condition - Local:
Local buses would also benefit from a bus-only lane.
1 mi (CBD) at 7 min/mi = 7 min  Start with 6-mph speed  1.5-mph (2 min/mi) gain from CBD bus lane  1.1-mph (1 min/mi) gain from fewer stops (only 3 in CBD)  Net speed = 8.6 mph = 7 min/mi
10 mi at 12 mph (5 min/mi, non-CBD bus lanes) = 50 min  Start with 10 mph = 6 min/mi  Subtract 1 min/mi (see Exhibit 5-8 in this Guide)  No TSP  Net speed = 5 min/mi
4 mi at 10 mph (6 min/mi, mixed traffic) = 24 min
Total travel time = 81 min Proposed Condition - BRT:
1 mi (CBD) at 7 min/mi = 7 min  Start with 6-mph speed  1.5-mph (2 min/mi) gain from CBD bus lane  1.1-mph (1 min/mi) gain from fewer stops (only 3 in CBD)  Net speed = 8.6 mph = 7 min/mi
10 mi at 17.5 mph (3.43 min/mi) = 34.3 min (non-CBD bus lanes)  With 2 stations/mi and 20-sec dwell, using Exhibit 4-56 of TC&QSM (8) results in a base speed of 2.73 min/mi  Add 0.7 min/mi for traffic delay from Exhibit 4-57 of TC&QSM (8)  Net speed = 3.43 min/mi = 17.5 mph
4 mi at 15.3 mph (3.93 min/mi) = 15.7 min (mixed traffic)  With 2 stations/mi and 20-sec dwell, using Exhibit 4-56 of TC&QSM (8) results in a base speed of 2.73 min/mi  Add 1.2 min/mi for traffic delay from Exhibit 4-57 of TC&QSM (8)  Net speed = 3.93 min/mi = 15.3 mph
Total travel time = 57 min
Average Speed = 15.8 mph The 3.43-min/mi speed in the bus lanes reflects the initial speed for Los Angeles’s Orange Line busway adjusted +0.7 min/mi for traffic delay. The 3.93-min/mi speed is the initial speed for Los Angeles’s Orange Line busway adjusted +1.20 min/mi for traffic delay. Assumed initial allocation of base riders:
BRT: 8,000 riders and 10-min maximum headway
Local: 8,000 riders and 10-min minimum headway The method of estimating BRT and local bus ridership using the four-step model and applying incremental logit models may be applicable. A method using elasticities is described below. Ridership estimates 1a. Apply travel time elasticity factors to BRT: 12 1211 2 )1()1( )1()1( TETE RTERTER +−− +−− = where R1 = initial ridership, R2 = anticipated ridership, T1 = initial travel times, T2 = travel times with BRT, and E = travel time elasticity factor = -0.4. 739,9 )946.0()574.1( )000,8576.0()000,8944.1( 2 = ×−×− ××−××− =R 1b. Apply travel time elasticity factors to local bus service for improved travel time: 12 1211 2 )1()1( )1()1( TETE RTERTER +−− +−− = where R1 = initial ridership, R2 = anticipated ridership, T1 = initial travel times, T2 = travel times with BRT, and E = travel time elasticity factor = -0.4. 490,8 )946.0()814.1( )000,8816.0()000,8944.1( 2 = ×−×− ××−××− =R

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-32 Bus Rapid Transit Practitioner’s Guide 2a. Estimate the proportion of BRT riders who would save time by taking the first bus (BRT or local) that arrives. An approximate estimate can be obtained based upon the following relationship: 2 1 2 t ha ×= where a = approximate length of route where riders would take first bus (assumed equal to % of riders), h = BRT headway (min), and t2 = time saved over local bus (total length of route). For Scenario 5, h = 10 min and maximum time savings is 94 - 57 = 37 min. %5.13 37 1 2 10 =×=a The proportion of BRT riders who would save time by taking the first bus that arrives is 13.5% x 9,739 = 1,315. 2b. Estimate increased ridership due to increase in service frequency (R3). Elasticity (frequency) = +0.4 Initial 8-min headway = 7.5 buses/hour Proposed 5-min headway (combined routes) = 12 buses/hour 582,1 )5.74.1()126.0( )315,1124.1()315,15.76.0( 3 = ×−×− ××−××− =R Since 1,315 riders were already included in the ridership estimate, the net increase is 1,582 - 1,315 = 267 riders. 3. Estimate additional ridership generated by BRT features. Weights applied to up to 25% increase in base ridership are obtained from elasticity computations. (See Exhibit 3-22 for details). Component Percentage Running Way 0% Stations (enhanced) 7% Vehicles (articulated) 5% Service pattern (regular) 12% ITS 10% Branding 10% Subtotal 44% BRT component synergy 0% Total 44% The 44% applies to an increase in base ridership of 25% beyond that obtained by elasticities. 0.44 x 25% = 11%. Anticipated additional BRT ridership = 8,000 x 11% = 880. Total anticipated BRT ridership in the base year considering only travel time savings = 9,739 + 880 = 10,619. Total anticipated BRT ridership in the base year considering both travel time and service frequency changes = 9,739 + 267 + 880 = 10,886. If 10,000 riders were initially allocated to BRT, the preceding ridership values would be increased 10/8 or 25% to 13,274 and 13,608, respectively. Fleet requirements Peak-hour peak-direction riders in the maximum load section (P): DirectionPeakinHourPeakin Turnover RidersDaily P %% ××=

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-33 System Packaging, Integration, and Assessment Turnover is assumed to be 1.80. 354%60%10 80.1 619,10 =××=P to 363%60%10 80.1 886,10 =××=P At 60 passengers per bus, six are needed in the maximum load section (with 10-minute headways). BRT vehicles needed: 134.12 10 10114 == + = + Headway TimeLayoverTimeRunningTripRound Add 3 spares to get 16 buses. Estimated costs See Exhibit 5-31. EXHIBIT 5-31 Estimated Development Costs - Scenario 5 Item Units Unit Cost Total Cost Bus lane - CBD 1 mile $100,000/mile $100,000 Bus lane - outlying 10 miles $100,000/mile $1 million Mixed traffic 4 miles — — Stations - CBD (enhanced) 3 stations $60,000/station $180,000 Stations - bus lanes (enhanced) 20 stations $60,000/station $1.2 million Stations - mixed traffic (enhanced) 8 stations $60,000/station $480,000 Conventional articulated BRT vehicles 16 vehicles $570,000/vehicle $9.1 million Station information 62 locations* $6,000/location $372,000 Vehicle information 16 vehicles $4,000/vehicle $64,000 Total (2004 dollars) $12.5 million * Two per station (for 31 stations) Scenario 6: Transit Signal Priority Only The description and assumptions for this scenario are shown in Exhibit 5-32 and Exhibit 5-33. Exhibit 5-34 describes the detailed analysis, and Exhibit 5-35 gives the estimated development costs for various BRT features. This scenario includes TSP for BRT at all non-CBD signalized intersections. Bus-only lanes would operate within the CBD. Conventional articulated buses would operate with on-board fare collection. This scenario would have the following effects: • Reduce the number of BRT stops from 90 to 31 • Reduce BRT one-way running times from 94 to 58.2 minutes • Increase daily BRT ridership from 8,000 to 10,817 Development costs for Scenario 6 were estimated at $11.4 million.

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-34 Bus Rapid Transit Practitioner’s Guide CBD 0 OF 6 SIGNALS WITH PRIORITY MIXED TRAFFIC (7 MILES, 14 STATIONS) 28 OF 28 SIGNALS WITH PRIORITY MIXED TRAFFIC (1 MILE, 3 MIXED TRAFFIC (7 MILES, 14 STATIONS) 28 OF 28 SIGNALS WITH PRIORITY EXHIBIT 5-32 BRT Development Scenario 6 EXHIBIT 5-33 Key Assumptions of BRT Development Scenario 6 Computed Results Feature Existing Service (Base) Local BRT Daily Ridership 16,000 8,000 10,817 Stops 90 90 31 Dwell/Stop 15 sec 15 sec 20 sec1 Frequency 8 min 10 min 10 min Speed 6 mph CBD; 10 mph elsewhere 9.8 mph 15.5 mph One-Way Travel Time 94 min 92 min 58.2 min Fare Collection On board On board On-board smart card Vehicle Conventional Conventional Conventional, articulated Passenger Information Stations Vehicles No No No No Yes Yes Development Costs N/A N/A $11.4 million 1 The 20-second dwell time for BRT assumes no change in fare collection or door channels available, in comparison to the existing service. The 20-second dwell time for BRT is higher than the dwell time for local bus because BRT has fewer stops and higher passenger boardings. EXHIBIT 5-34 Analysis - Scenario 6 Item Analysis Existing Condition:
6 mph over 1 mi = 10 min (CBD mixed traffic)
10 mph over 14 mi = 84 min (mixed traffic)
Total = 94 min Travel times Proposed Condition 1 - Local:
1 mi (CBD) at 8 min/mi = 8 min  Start with 6-mph speed  1.5-mph gain due to bus lane speeds  Net speed = 7.5 mph = 8 min/mi
14 mi at 10 mph (6 min/mi) = 84 min (on local streets) Total travel time = 92 min

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-35 System Packaging, Integration, and Assessment Proposed Condition 2 - BRT:
1 mi at 7.8 min/mi = 7.8 min (CBD)  The 7.8-min/mi speed reflects a base travel time of 3.3 min/mi and +4.5 min/mi for traffic delays (because traffic signals are more frequent than bus stops).
14 mi at 3.6 min/mi = 50.4 min (non-CBD)  The 3.6-min/mi speed is the initial speed for Los Angeles’s Orange Line busway adjusted +1.20-min/mi for traffic delay and -5 sec/signal for TSP.
Total = 58.2 min Assumed initial allocation of base riders:
BRT: 8,000 riders and 10-min maximum headway
Local: 8,000 riders and 10-min minimum headway The method of estimating BRT and local bus ridership using the four-step model and applying incremental models may be applicable. A method applying elasticities is described below. 1. Apply travel time elasticity factors to BRT: 12 1211 2 )1()1( )1()1( TETE RTERTER +−− +−− = where R1 = initial ridership, R2 = anticipated ridership, T1 = initial travel times, T2 = travel times with BRT, and E = travel time elasticity factor = -0.4. 662,9 )946.0()2.584.1( )000,82.586.0()000,8944.1( 2 = ×−×− ××−××− =R Ridership estimates 2a. Estimate the proportion of BRT riders who would save time by taking the first bus (BRT or local) that arrives. An approximate estimate can be obtained based upon the following relationship: 2 1 2 t ha ×= where a = approximate length of route where riders would take first bus (assumed equal to % of riders), h = BRT headway (min), and t2 = time saved over local bus (total length of route). For Scenario 5, h = 10 min and maximum time savings is 94 - 58 = 36 min. %14 36 1 2 10 =×=a The proportion of BRT riders who would save time by taking the first bus that arrives is 14% x 9,662 = 1,353. 2b. Estimate increased ridership due to increase in service frequency (R3). Elasticity (frequency) = +0.4 Initial 8-min headway = 7.5 buses/hour Proposed 5-min headway (combined routes) = 12 buses/hour 628,1 )5.74.1()126.0( )353,1124.1()353,15.76.0( 3 = ×−×− ××−××− =R Since 1,353 riders were already included in the ridership estimate, the net increase is 1,628 - 1,353 = 275 riders.

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-36 Bus Rapid Transit Practitioner’s Guide 3. Estimate additional ridership generated by BRT features. Weights applied to up to 25% increase in base ridership are obtained from elasticity computations. (See Exhibit 3-22 for details). Component Percentage Running Way 0% Stations (enhanced) 7% Vehicles (conventional, articulated) 5% Service pattern 12% ITS 10% Branding 10% Subtotal 44% BRT component synergy 0% Total 44% The 44% applies to an increase in base ridership of 25% beyond that obtained by elasticities. 0.44 x 25% = 11%. Anticipated additional BRT ridership = 8,000 x 11% = 880. Total anticipated BRT ridership in the base year considering only travel time savings = 9,662 + 880 = 10,542. Total anticipated BRT ridership in the base year considering both travel time and service frequency changes = 9,662 + 275 + 880 = 10,817. Peak-hour peak-direction riders in the maximum load section (P): DirectionPeakinHourPeakin Turnover RidersDaily P %% ××= Turnover is assumed to be 1.80. 351%60%10 80.1 542,10 =××=P to 361%60%10 80.1 817,10 =××=P At 60 passengers per bus, six buses are needed in the maximum load section (with 10-minute headways). Vehicle requirements BRT vehicles needed: 136.12 10 10116 == + = + Headway TimeLayoverTimeRunningTripRound Add 3 spares to get 16 buses. Estimated costs See Exhibit 5-35.

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-37 System Packaging, Integration, and Assessment EXHIBIT 5-35 Estimated Development Costs - Scenario 6 Item Units Unit Cost Total Cost Stations - CBD (enhanced) 3 stations $60,000/station $180,000 Stations - other (enhanced) 28 stations $60,000/station $1.7 million Conventional articulated BRT vehicles 16 vehicles $570,000/vehicle $9.1 million Station information 62 locations* $6,000/location $372,000 Vehicle information 16 vehicles $4,000/vehicle $64,000 Total (2004 dollars) $11.4 million * Two per station Summary and Comparison of BRT Development Scenarios Exhibit 5-36 compares the following for the six scenarios analyzed: • Existing and anticipated BRT travel and the likely percentage reduction • Existing and anticipated BRT ridership and the likely percentage increase • Anticipated development costs EXHIBIT 5-36 Summary of Anticipated BRT Travel Times, Ridership, and Costs Scenario 1 Scenario 2 Scenario 3 Scenario 4 Scenario 5 Scenario 6 Item Grade- Separated Busway At-Grade Busway At-Grade Busway & Median Arterial Busway Bus Lanes (Rebuilt) & TSP Bus Lanes Only TSP Only Existing (base) one-way travel time 94 min 94 min 94 min 94 min 94 min 94 min BRT in-vehicle travel time 29 min 43 min 48 min 50 min 57 min 58 min % reduction 69% 54% 49% 47% 39% 38% Assumed BRT base ridership 10,000 10,000 20,000 8,000 8,000 8,000 Anticipated BRT ridership 17,660 15,700 33,020 11,600 10,885 10,815 % increase 77% 57% 65% 45% 36% 35% Existing local bus ridership 20,000 20,000 20,000 16,000 16,000 16,000 Anticipated local bus ridership 10,000 10,000 - 8,490 8,490 8,000 Estimated development costs* $242.0 million $109.4 million $84.3 million $40.3 million $12.5 million $11.4 million * In 2004 dollars NOTE: Numbers have been rounded. SOURCE: Computed The six BRT scenarios studied in this chapter reduce transit running times 38% to 69%.

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-38 Bus Rapid Transit Practitioner’s Guide Exhibit 5-37 shows the likely sources of the anticipated increases in BRT ridership for each of the six BRT scenarios. The anticipated future BRT ridership is also “normalized” to a “base ridership” of 10,000 daily riders. The BRT features accounted for about 8% to 10% of the bus lane/TSP riders and 12% to 14% of the busway ridership. EXHIBIT 5-37 Source of Anticipated BRT Ridership Scenario 1 Scenario 2 Scenario 3 Scenario 4 Scenario 5 Scenario 6 Item Grade- Separated Busway At-Grade Busway At-Grade Busway & Median Arterial Busway Bus Lanes (Rebuilt) & TSP Bus Lanes Only TSP Only Base condition 10,000 10,000 20,000 8,000 8,000 8,000 Reduced travel time 5,360 3,500 5,974 2,250 1,740 1,660 Increased Frequency — — 3,148 230 265 275 Subtotal 15,360 13,500 29,120 10,480 10,005 9,935 BRT features* 2,300 2,200 3,900 1,120 880 880 Total 17,660 15,700 33,020 11,600 10,885 10,815 Normalized to 10,000 base riders 17,760 15,700 16,510 14,500 13,605 13,520 * Percentage applied to subtotal depending on extent of BRT features. Maximum is 25%. NOTE: Numbers have been rounded. SOURCE: Ridership estimates for each scenario The effect of travel time savings for the 15-mile BRT route on daily BRT ridership (assuming a base BRT ridership of 10,000) is shown in Exhibit 5-38. There is a consistent linear trend, showing steady ridership growth as the time savings (in absolute time or minutes per mile) increases. 0 5,000 10,000 15,000 20,000 25,000 0 10 20 30 40 50 60 70 80 One-Way Time Savings (minutes) D ai ly B R T R id er s (n or m al iz ed to 1 0, 00 0 ba se ri de rs ) EXHIBIT 5-38 Daily BRT Ridership v. Travel Time Savings Daily ridership increases consistently with the savings in one-way travel times.

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-39 System Packaging, Integration, and Assessment The anticipated time savings are related to estimated development costs in Exhibit 5-39. As development costs increase, there is a corresponding gain in the travel time saved. Exhibit 5-40 shows how BRT ridership would grow as development costs increase. Again, there is a consistent linear trend. 0 10 20 30 40 50 60 70 0 50 100 150 200 250 300 Estimated Development Costs ($ millions, 2004) M in ut es S av ed (o ne -w ay , e nt ire ro ut e) EXHIBIT 5-39 Time Savings v. Estimated Development Costs 0 5,000 10,000 15,000 20,000 25,000 0 50 100 150 200 250 300 Estimated Development Costs ($ millions, 2004) D ai ly B R T R id er s (n or m al iz ed to 1 0, 00 0 ba se ri de rs ) EXHIBIT 5-40 Daily BRT Ridership vs. Estimated Development Costs Investment in busways translates into greater time savings and higher ridership.

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-40 Bus Rapid Transit Practitioner’s Guide The information contained in these exhibits can be developed for any series of BRT proposals for a given corridor. While the numbers and relationships shown are specific to the six scenarios analyzed, several patterns emerge: • As development costs increase, there is a consistent reduction in travel times and a growth in BRT ridership. • Faster travel times reduce operating costs for any given bus volume. • The busway scenarios, because of their exclusive right-of-way and wider station spacing, have the greatest gains in speeds and ridership. • The low-cost scenarios (i.e., bus lanes and TSP) have the smallest time savings and ridership gains. • Travel time savings appear to be the greatest contributor to BRT ridership gains, followed in turn by the provision of special BRT features. While BRT may run at short intervals, the splitting of corridor service between BRT and local bus operations may limit the computed ridership gains from combined bus frequencies. Any city-specific analyses should reflect local conditions in terms of land and construction costs, population and employment growth, and land development impacts. Current experience suggests that major investments such as busways or reconstructed arterial streets may encourage new investments. These effects are discussed in Chapter 6. Assessment of BRT Development Scenarios The preceding ridership estimates are largely influenced by the initial allocation of street (or corridor) ridership between BRT and local transit service. The proportions of riders that would use BRT in the six scenarios are as follows: • Scenario 1 - 64% • Scenario 2 - 61% • Scenario 3 - 100% • Scenario 4 - 58% • Scenario 5 - 56% • Scenario 6 - 57% A review of ridership along busways in Miami, Pittsburgh, and Ottawa shows a wide range in usage depending on geography, street system, and assigned routes. In Miami, all parallel local service now operates on the busway. In Ottawa, about two-thirds of all transit riders use the busway for at least a portion of their trip. The mix between local and limited (or BRT) service on city streets ranges from about 35% to 65%. (In Vancouver, however, all local services in the corridor were replaced by the 98-B BRT service.) Specific values reported were as shown in Exhibit 5-41. These comparisons suggest that the 56% to 58% allocations anticipated for Scenarios 4, 5, and 6 may be optimistic. More conservative BRT ridership estimates, based on allocating 40% of the initial base ridership to BRT, results in BRT accounting for about 47% of total riders. (See Exhibit 5-42.) This percentage is close to current experience. In any real-world situation, review of detailed ridership patterns and station usage, along with possible changes in routes and services, will permit more refined estimates.

Bus Rapid Transit Practitioner’s Guide Bus Rapid Transit Practitioner’s Guide Page 5-41 System Packaging, Integration, and Assessment EXHIBIT 5-41 Allocation of Ridership between Local Bus and BRT (or Limited-Stop) Service Street % of Riders Using BRT Wilshire Boulevard (Los Angeles) 34% Grand Concourse (New York City)* 35% 1st and 2nd Avenues (New York City)* 46% Flatbush Avenue Southbound (New York City)* 54% Fordham Road (New York City) 54% Ventura Boulevard (Los Angeles) 66% Average 48% *Average of 6:00 a.m. to 10:00 a.m. SOURCE: New York City Transit Authority EXHIBIT 5-42 Anticipated Route Ridership with 40% of Base Ridership Allocated to BRT - Selected BRT Development Scenarios Service Scenario 4 Scenario 5 Scenario 6 BRT 9,280 8,710 8,650 Local 10,190 10,190 9,600 Total 19,470 18,900 18,250 % BRT 48% 46% 47% REFERENCES 1. Zimmerman, S. BRT: A Primer Paper. Prepared for the 2001 ITE Annual Meeting. Chicago, IL, 2001. 2. Levinson, H., S. Zimmerman, J. Clinger, S. Rutherford, R. Smith, J. Cracknell, and R. Soberman. TCRP Report 90: Bus Rapid Transit: Vol. 1, Case Studies in Bus Rapid Transit, and Vol. 2, Implementation Guidelines. Transportation Research Board of the National Academies, Washington, D.C., 2003. 3. Kittelson & Associates, Inc., Herbert S. Levinson Transportation Consultants, DMJM+Harris, and OURCO, Inc. TCRP Project A-23A Interim Report. Unpublished. 2004. 4. Diaz, R.B., M. Chang, G. Darido, E. Kim, D. Schneck, M. Hardy, J. Bunch, M. Baltes, D. Hinebaugh, L. Wnuk, F. Silver, and S. Zimmerman. Characteristics of Bus Rapid Transit for Decision-Making. FTA, Washington, D.C., 2004. 5. Vehicle Catalog 2005 Update. A Compendium of Vehicles and Hybrid Drive Systems for Bus Rapid Transit Service. WestStart-CALSTART, 2005. 6. Moffat, G.K., A.H. Ashton, and D.R. Blackburn. TCRP Synthesis 40: A Challenged Employment System: Hiring, Training, Performance Evaluation, and Retention of Bus Operators. Transportation Research Board, National Research Council, Washington D.C., 2001. 7. Kittelson & Associates, Inc., H.S. Levinson, and DMJM+Harris. TCRP A-23A April-June 2005 Quarterly Progress Report. Unpublished. Jul. 2005. 8. Kittelson & Associates, Inc., KFH Group, Inc., Parsons Brinckerhoff Quade and Douglass, Inc., and K. Hunter-Zaworski. TCRP Report 100: Transit Capacity and Quality of Service Manual. Transportation Research Board of the National Academies, Washington, D.C., 2003.

Bus Rapid Transit Practitioner’s Guide System Packaging, Integration, and Assessment Page 5-42 Bus Rapid Transit Practitioner’s Guide This page is intentionally blank.