Municipality of Anchorage

Transcription

1 Municipality of Anchorage Public Transportation Department High Priority Transportation Corridor Plan October 28, 2010

2 Table of Contents Chapter 1 Introduction Chapter 2 Context -- Anchorage and Public Transit Chapter 3 High Performance Transportation Corridors Chapter 4 Examining Transit Corridors Performance Chapter 5 Findings and Recommendations

3 Chapter 1 Introduction In 1972 the population of Anchorage was 144,200 persons. That year voters approved a ballot issue to inaugurate Municipal public transportation service. A companion funding measure was narrowly defeated so startup of People Mover transit service was delayed two years. In July 1974 People Mover began bus service and has operated continuously ever since. People Mover grew rapidly over its first several years. By 1982, its buses provided 156,000 hours of service and attracted 4.01 million passenger boardings as Anchorage growth mushroomed with military base expansion and Prudhoe Bay oil development and the construction boom that followed. But as the booming environment slowed, over 50,000 hours of annual bus service was sliced while transit ridership declined, then remained essentially stable through the 1990s. Fast forward to 2008, 34 years later. Anchorage was home to nearly 284,000 residents. People Mover carried 4.22 million passengers, its highest level in its entire history. Figure 1 illustrates People Mover annual bus ridership over the twenty-six year period from its early peak in 1982 through 2008; bus service hours per capita are also displayed. Figure 1. People Mover Riders and Service Hours, ,500, ANNUAL RIDERSHIP HOURS PER CAPITA 4,000, ,500,000 RIDERSHIP 3,000, HOURS PER CAPITA 2,500,000 2,000, ,500, The turn-around in annual bus ridership from levels to 2008 is remarkable. Even more impressive is that the percentage gain in passengers exceeded the corresponding increase in bus revenue hours of service as well as bus hours per capita. The stimulus for People Mover s ridership recovery was a 5-year People Mover Blueprint Plan for service - 1 -

4 restructuring, completed in Implementation steps for the Blueprint Plan were programmed over four-phases and several years. Initial restructuring changes began in 2002 and continued through Impressive results were achieved in those years as the following changes were put in place: Introduction of Route 1 Crosstown weekday service and suspension of services with low productivity, most notably in the Hillside area. Revision of peak period schedules to 30-minute headways on primary routes and institution of memory headways (consistent schedule times from hour to hour of the day) on most routes; airport service added on Route 7; Saturday and Sunday service additions including expanded hours of operation. Memory headways implemented on additional routes and revamping of the route structure for Route 102, Routes 13, 15, and 77 replacing Routes 11, 12, 74 and 76 and changes in Eagle River plus demand scheduled service there and Hillside. Establishment of a Muldoon transfer center. But in 2005, efforts for the Blueprint Plan s fourth phase to provide all-day weekday 30- minutes service frequency were curtailed due to budgetary restrictions. Full completion of the Blueprint Plan has not been possible with available operating budget resources. The People Mover rider gains between 2002 and 2007 were largely due to the Blueprint route restructuring program made possible by a federal Congestion Mitigation for Air Quality grant. Modest service improvements were introduced in 2006 and in A fare increase was an offsetting factor. Municipal population and economic growth also was a contributing factor as was a sharp increase in gasoline prices in mid Genesis for this Study Initiative People Mover has sought to sustain the significant ridership gains and momentum achieved over the past several years. In 2006, People Mover management began to consider ways to maintain vigorous ridership gains experienced with the initial implementation phases of the Blueprint Plan. Management efforts focused on implementing advanced technology hardware and software systems to aid transit management and improve service delivery. Contractor services were retained for the following technology elements: bus schedules and driver assignment rostering optimization driver payroll records incident management module on-board data communication and display terminals automated vehicle location tracking automated passenger counters supporting database management, analysis and reporting capabilities - 2 -

5 People Mover management recognizes that increased transit ridership is critical to attain funding support and enable further People Mover service enhancements. Bus passenger boardings between 2002 and 2005 jumped 28% due to the Blueprint Plan service changes, and then flattened due to funding limits. As noted earlier, bus ridership increased in mid as gas prices spiraled to record levels. Other related Municipal programs and planning complemented the People Mover Blueprint Plan work. A Comprehensive Plan for the Anchorage Bowl was adopted by the Municipal Assembly in The Comp Plan identified four priority corridors for transitsupportive development. It also articulated a number of policies to encourage transit use - denser urban development, a more walkable community, and provision of more travel options to reduce reliance on private vehicle transportation. About the same time, the Municipality s Long Range Transportation Plan update was launched with a household travel survey of Anchorage residents in spring Subsequently, the Anchorage Metropolitan Area Transportation Solution (AMATS) Long Range Transportation Plan (LRTP) was completed and adopted by the Anchorage Assembly in Recommendations in the LRTP include increased transit service frequency and improvements to attract more riders. All three planning efforts the People Mover s Blueprint Plan, the MOA Comprehensive Plan, and AMATS Long Range Transportation Plan envision more extensive and robust public transportation services in future years. Prototype Transit Corridor Study Sustaining People Mover s recent five-year achievements is the objective for the research reported in this study. The study examines strategies and initiatives that may be most effective to further improve service, productivity and ridership. Elements encompassed in the investigations include route and service structure, service reliability, service scheduling/timing, bus stop locations, application of advanced technology, transit signal priority, transit queue jumping, marketing, and passenger information systems. Part of the analyses reported here addresses the People Mover system as a whole; other parts focus on a particular prototype corridor as a microcosm of the entire system. Route 7 serving the Spenard Jewel Lake corridor is used as the Prototype Corridor case. Report Organization Four chapters follow this introductory chapter. Chapter 2 presents an overview of the Anchorage transit system, its level of service, and related land use and transportation planning. Chapter 3 discusses different perspectives on what defines high performance transportation and identifies the Glenn Highway as the major regional, multimodal transportation corridor and People Mover Route 45 as the dominant high performance transit corridor respectively. Chapter 4 analyzes the service changes made in People Mover s Blueprint Plan that led to the significant ridership gains between 2002 and It reviews - 3 -

6 systemwide features and operations and evaluates Route 7 in detail. Chapter 5 presents findings and recommendations for higher performance transit services

7 Chapter 2 Context -- Anchorage and Public Transit Anchorage s population in 2008 was 284,000; the Municipality added over 62,000 residents in the past two decades. Even so, the Municipality ranked 140 th of the 147 U.S. metropolitan areas with over 250,000 population in the 2000 US Census. Thus, among US metropolitan areas, the Municipality remains a relatively small urban area. Development patterns in Anchorage are generally medium to lower density. Major commercial and employment centers are dispersed. There are six areas with significant employment Downtown, Midtown, two military bases, U-Med, airport, and the Dimond Center area. The dispersed geographic distribution of major activity centers means that travel patterns are also broadly dispersed, rather than concentrated on a few primary destinations. The multiple activity centers and dispersed travel patterns make it more difficult for transit to attract a large share of traveler destinations. Current Transit System Configuration 1 The People Mover system in 2010 consists of fourteen (14) routes serving Anchorage, and the Eagle River and Chugiak areas. The system focus is downtown Anchorage; all but one of the fourteen routes serve the downtown Anchorage Transit Center hub. Transit service also converges on three smaller satellite hubs within the Anchorage Bowl one in the U-Med area, another at Dimond Cener in South Anchorage and a third at the Muldoon Town Center in East Anchorage. Another hub is found in the Eagle River commercial district (See map on following page). Bus schedules are time-coordinated to enable easier transfers at the Downtown Transit Center, and to a lesser extent, at the other hubs. Table 1 summarizes People Mover service hours and average headways in 2010 for each route. Bus service generally begins operation on weekdays between 5:30 am and 6:30 am and ends between 10:00 pm and 11:30 pm. Transit service runs from 7:30 am to 9:00 pm on Saturdays, and from about 9:30 am to 7:00 pm on Sundays. During weekdays Route 45 service operates most frequently (approximately 20 minute headways) and carries the highest route ridership. Four routes (3, 7, 9 and 15) operate every 30 minutes during peak and midday hours. Six routes (2, 13, 36, 60 and 75) have an additional trip(s) during the weekday peak to provide 30-minute headways. Three routes (1, 8 and 14) operate at 60-minute frequencies all day on weekdays. 1 This section is abstracted from the 2009 People Mover Short Range Transit Plan Update and updated for 2010 service. -5-

8 Table 1 People Mover Bus Routes and Headways, 2010 Route Service Hours Average Headway (Weekdays) Peak Midday Evening Saturday Sunday 1 6:12a 10:14p :30a 10:23p :55a 11:29p :155a 11:23p :17a 10:09p :51a 10:26p :45a 11:09p :45a 10:08p :15a 10:36p :00a 11:21p :50a 11:22p / :12a 10:05p :26a 9:56p :44a -8:09a 2:45-p - 7:01p

9 In summary, weekday bus service during peak periods is at 20 minute headway on two routes, 30 minute on 9 routes and at 60 minute on three others. For the majority of time, many bus routes have 60 minute frequencies. The 2020 Anchorage Comprehensive Plan Vision The 2020 Anchorage Bowl Comprehensive Plan identifies four transit-supportive development corridors which generally connect town centers with the three major employment centers - Downtown, Midtown and U-Med. The Comprehensive Plan intent for the transit-supportive development corridors is that these corridors represent optimal locations for more intensive commercial and residential land use patterns which will encourage and support higher levels of transit service. The overall intent is to create a city in which there will be more opportunities to live a less automobile-dependent lifestyle. The plan promotes transitsupportive development corridors, pedestrian accessible developments and multimodal roadway and trail networks. Residents of town centers could use the high frequency bus service provided along the transit-supportive development corridors to reach their job sites in major employment centers. 2 These corridors are not intended to represent a transit route map, but illustrate where new medium to high-density housing development will occur. Anchorage 2020 stresses that travel choice is an important goal so that residents are not, by necessity, required to drive everywhere. The foundation for travel choice is frequent transit service to all mid-density and higher-density areas. A typical transit-supportive development corridor includes the following: medium to high-density housing (over 8 dwelling units per acre) within one-fourth mile of the major street at the center of the corridor small scale commercial sites oriented to the street multi-modal facilities, emphasizing bus, pedestrian, and bicycle transportation expanded sidewalks and crosswalks, street furniture, bus shelters, and landscape improvements Transit-supportive development corridors tie major elements of the Land Use Policy Map together. Higher residential density is a key to increasing transit ridership along these corridors. Bus routes serving transit corridors should achieve a 15-minute headway during peak periods and a 30-minute headway during non-peak periods. Most of the town centers are linked to one or more employment centers by transitsupportive development corridors. For example the town center located near the intersection of Jewel Lake Road and Dimond Boulevard is connected to the major employment 2 Transit-supportive development corridors are intended to be multimodal with the primary emphasis on bus, pedestrian and bicycle transportation. Bus routes serving transit corridors should achieve a 15-minute headway during peak hours and 30-minute headway during non-peak periods. (This reflects nationally accepted standards.) See pages 54-55, Chapter 4, Land Use Concept Plan. Anchorage 2020 Anchorage Bowl Comprehensive Plan -7-

10 centers in Midtown and Downtown by the transit-supportive development corridor along Jewel Lake Road and Spenard Road. This corridor is examined for this prototype High Performance Transit Corridor Project Anchorage Bowl Comprehensive Plan Map -8-

11 Public Transit in the AMATS 2027 Long Range Transportation Plan The AMATS Long Range Transportation Plan (LRTP) notes that funding is a fundamental determinant of the level of transit service that can be provided, and that public policy and public perceptions of transit service value define willingness to support public funding. The LRTP analyses of future growth and transportation scenarios finds that improved transit service operations and service delivery can increase riders. The LRTP concludes that public transportation patronage can likely be doubled (from 2002 levels), perhaps tripled. But to get the higher number of riders, public funding will need to expand. The LRTP recommends high frequency bus rapid transit service from Chugiak/Eagle River along the Glenn Highway corridor. Increasing transit service frequency to 15 minutes during weekday peak periods by 2013 is also recommended on seven corridors Routes 2, 3, 7, 9, 15, 36 and

12 Chapter 3 High Performance Transportation Corridors This chapter examines the most important transportation corridors in the Anchorage region. Initially the focus is on regional multimodal transportation corridors because these corridors serve high travel demand and therefore impact many travelers as well as the communities and neighborhoods they traverse. Coordinated, multi-modal, and multi-faceted transportation systems are addressed as well as demand management strategies and policies to yield significant benefits to travelers, transportation providers and the general public. Next, public transportation corridors are reviewed to determine the highest performing corridor. Potential ways to enhance performance are discussed. Interpreting Transportation Performance Travel for work, shopping, recreation, school and other needs is a basic element of urban life. That is why transportation facilities and services are important parts of the core infrastructure in urbanized communities and regions. Where travel demand is intense, higher performance transportation facilities and services are required. Public transit works best in high demand travel corridors because more travelers have similar origin and destination patterns that can support better, more frequent, direct transit service. What constitutes a high performance transportation corridor and how does one define or measure high performance? The term high performance transit corridor may convey different meaning or interpretations to various audiences. Some may interpret the phrase in hardware or equipment terms, such as rail or guideway transit service operated on a separate right-of-way (i.e., commuter rail, light rail transit), or perhaps what has been coined bus rapid transit in which buses operate frequently and are given preferential space and/or road priority to improve travel speed. Others may use a different perspective thinking in terms of how well the transportation service satisfies travel needs (i.e. customer service) with respect to easy access, frequency of service, directness to desired destinations, comfort, and safety/personal security considerations. So, defining performance may depend upon one s perspective. Stakeholders with an interest in transportation performance include service providers, transportation users, and the general public. The perspectives of each stakeholder group are discussed in the following paragraphs. Service Provider Perspective Almost all public transportation service in the United States is operated by public agencies. Organizations providing public transit service operate within the constraints of finite financial, equipment, facilities, and human resources. Of these, financial resources -11-

13 dominate 1. Allocating available resources to achieve maximum effectiveness is a primary goal. Effectiveness can be measured by service productivity how many riders/users are attracted per mile or hour of bus service offered. A priority goal for People Mover is to increase transit ridership while maintaining accepted standards for riders per hour of bus service. Travelers and Transit Riders Perspective Travelers seek to minimize the time and effort required to accomplish their travel needs. They make time of departure, mode, route and other decisions based on the options available to them. Many travelers choose only auto travel. For others, several transit service characteristics are important factors in deciding whether to use transit: Service frequency. Travelers have an aversion to scheduling constraints, waiting, and especially unreliable service. Time is the dominant factor influencing travelers choice of travel mode. In choosing a travel mode, people weigh time for walking and waiting as two-to-three times more important than the same amount of time spent riding in a vehicle. Extensive transit research and experience indicate that travelers who have a choice between travel by automobile or by transit view 15- minute transit service frequency as a threshold. Less frequent service is judged unfavorably. 2 Direct service to destination. A transfer from one bus to another to reach a destination has a highly negative effect. Few travelers are willing to accept the inconvenience, added waiting delay, uncertainty, and longer journey time. Transit versus automobile travel time. Bus travel times are often about two and a half to three times longer than times for the same journeys by automobile. Similar comparative travel time conditions are observed in Anchorage. That disparity needs to be significantly reduced to attract more riders. Longer spacing between bus stops and transit priority at signalized intersections can help improve transit speed. Accessibility. A reasonable walking distance acceptable to most people is typically one-quarter mile. Bus stops both where trips originate and where they are destined need to be accessible and safe, ideally with sidewalk connectivity. The walking environment matters too, especially during winter conditions. Dependability, security, comfort, value. On-time reliability (dependability), sense of personal safety and security, creature comforts, and cost-value perceptions 1 The significance of funding is aptly reflected in the severe impact of the current deep global recession on transit agencies. An American Public Transportation Association membership survey in March 2010, found nearly 70% of the nation s transit operators project budget shortfalls this year, 84% have made or are considering service reductions and/or fare increases, while more than half have eliminated positions and one-third laid off employees. 2 The elasticity of transit use with respect to service frequency (called a headway elasticity) averages about 0.5, meaning that each 1 percent increase in transit service frequency increases ridership by 0.5 percent. But considerable variation is observed. See Chapter 9 Transit Scheduling and Frequency, TCRP Web Document 12 (Project B-12) [ and TCRP Web Document 32 (Project H-32) Elements Needed to Attract High Ridership Transit Systems: Interim Guidebook [ -12-

14 contribute to positive traveler assessments of transit service. These factors generally are subordinate to time and direct routing, but nonetheless come into play when a traveler decides to use transit instead of a personal vehicle. General Public Perspective The majority of personal travel in small and mid-sized urban areas is made by personal vehicle. Public transit generally accounts for less than five percent of all trips in these communities. Naturally, then the general public tend to focus on traffic and road conditions when they think about transportation corridors and service. The public perceives transportation primarily in terms of travel time to desired destinations, safety and pavements that are not full of potholes or deeply rutted. Citizens accept some degree of congestion during commute periods, but they expect roads to be relatively congestion-free most of the day. Since travel is predominantly in autos, adequate parking is yet another dimension of public perception of transportation service. While many people do not use public transportation frequently, or at all, nonetheless there is generally broad appreciation and support for public transit service 3. The dominant deterrents to broader transit ridership are twofold the relatively sparse service frequency that restricts flexibility and limited direct-ride destination coverage that makes service convenient. The public understands that not all citizens can drive or own vehicles and that access, participation and contribution to the community is important for all. They support specialized transportation for seniors, people with disabilities, low income, and other needy individuals. They support school bus services to get students to and from schools. Further, the public comprehends the impacts (positive and negative) of auto dependency on the community, on the environment and air quality, and on neighborhood cohesion and intrusion. The public wants a transportation system that meets travel needs, that is affordable, cost-effective and efficient, that enables accessibility for all members of the community, that is safe, that is consistent with community development plans, and that satisfies existing and future growth conditions. Glenn Highway Regional High Performance Transportation Corridor The Glenn Highway is arguably the most important high performance transportation corridor in the Anchorage metropolitan region. 4 It serves over 51,000 vehicles per day (which translates to about 70,000 daily person trips). The Glenn Highway is part of the National Highway System linking the Anchorage Bowl to Chugiak/Eagle River and the 3 Surveys by RLS & Associates Inc. (People Mover Rider Survey, 2001; 2009 People Mover Route Restructure Plan Update Community Survey, 2008); Craciun Research Group Inc.(Municipality of Anchorage Bus Riders and Non-Bus Riders, October 2008; LRTP Public Opinion Survey Preliminary Results, September 2005); NuStats (2002 Anchorage Household Travel Survey). An April 2010 transit bonds ballot proposition garnered 48% in favor 52% against. 4 The Seward Highway also ranks as an important regional corridor, carrying over 58,000 vehicles per day south of Dowling Road. But trip lengths on the Seward Highway are much shorter and its trafficshed area is not growing nearly as rapidly as is the Glenn tributary areas in the Mat Su Borough and Chugiak/Eagle River. -13-

15 Mat Su Borough to the north. It is a primary motor freight backbone for freight and goods movement flowing from the Port of Anchorage and Ted Stevens Anchorage International Airport. The Anchorage Long Range Transportation Plan (LRTP) prepared and adopted by AMATS in 2007 examined primary transportation corridors throughout the Municipality. The LRTP designates the Glenn Highway corridor as a priority corridor and proposes a number of multimodal projects to address future transportation needs. LRTP projects include roadway widening, bridge and interchange improvements, more frequent transit service, upgrading express bus or bus rapid transit, as well as ridesharing, telecommuting and other travel demand management initiatives. There are four compelling reasons why the Glenn Highway is a priority High Performance Transportation Corridor for Anchorage: High and growing travel demand between central Anchorage (the terminus of the Glenn Highway) and Eagle River/ Chugiak and the Mat Su Borough. The State of Alaska, the Municipality, the Mat Su Borough and the federal government each have interest and a stake in sustaining efficient transportation service in the corridor. Agency collaboration in funding, operating and coordinating transportation services is a cost-effective, win-win approach for all parties and the public. Fully integrated, multi-modal system management and demand management strategies will provide traveler choice options that sustain level of service while reducing capital investment requirements. Multi-modal management strategies for the corridor reinforce the Anchorage Comprehensive Plan and goals by reducing traffic-generated environmental and air quality impacts and by lessening vehicle traffic volumes and parking requirements in the Bowl. Rapid Growth & Long Trips The Mat Su Borough is the fastest growing area in the state. Mat Su Borough population has exploded five-fold over the past 30 years, far exceeding Anchorage s growth rate. By 2008, the Borough s population was 82,500. Recent economic projections by the Institute of Social and Economic Research at the University of Alaska, Anchorage estimate Mat Su population at 95,400 in 2015 and 170,800 in Scott Goldsmith, Economic and Demographic Projections for Alaska and Greater Anchorage , Institute of Social and Economic Research, UAA, December,

16 90,000 80,000 70,000 60,000 50,000 40,000 30,000 20,000 10, Mat Su Borough Population Growth Source: Alaska Department of Labor. Over 12,000 Mat Su residents commute to workplaces in Anchorage where there are many more employment opportunities and jobs are generally better paying. Forty-mile journey to work commutes are common from Mat Su to Downtown and Midtown Anchorage; these trips are much longer than average trips within Anchorage. 6 14,000 12,000 Mat Su to Anchorage Anchorage to Mat Su No. of Commuters 10,000 8,000 6,000 4,000 2, Commuters from Mat Su Borough and Anchorage Source: Alaska Department of Labor and Workforce Development, Research and Analysis Section. (Note: chart excludes federal and self-employed workers) Mat Su commuters use the Glenn Highway corridor. Thus, Mat Su Borough growth, combined with Mat Su commuting trends to Anchorage workplaces, has significantly increased traffic on the Glenn Highway. Daily traffic at Mirror Lake has grown at a compounded annual rate of nearly 5% over the past thirty years. In 1978 daily traffic was 6,950; by 2008 daily traffic volume had reached 28, Within Anchorage, most work trips are less than 8 miles long, using less than 8 vehicle-miles of capacity. Each long Mat Su commute trip consumes more vehicle miles of roadway capacity. For example, a 40-mile solo driver trip provides 40 passenger miles and uses 40 vehicle-miles of capacity. A 40-mile bus trip with 30 passengers uses the same 40 vehicle-miles of roadway capacity but serves 1,200 passenger miles. -15-

17 Chugiak/Eagle River also has grown more rapidly than the Anchorage Bowl, with many workers commuting to Anchorage. Daily Glenn Highway traffic flow east of Muldoon Road was 51,400 in 2008, about 23,000 more than north of Chugiak/Eagle River. + 35,000 30,000 Glenn Highway at Mirror Lake 25,000 20,000 15,000 10,000 5, Average Daily Traffic Glenn Highway daily traffic volume trends at Mirror Lake, Source: ADOT&PF Central Region Traffic Data Glenn Highway LRTP Capital Investments Projects over $750 Million The AMATS LRTP anticipates the Glenn Highway corridor will require over $ 750M for road improvement projects by Major improvements include adding lanes, rebuilding interchanges, and widening bridges. In addition, the LRTP recommends $21 million for new buses, more frequent express bus service, shelters, park n ride lots and improvements to the Eagle River transit hub. Collaborative Framework for Glenn Highway The State of Alaska Department of Transportation and Public Facilities owns and operates the Glenn Highway. Transit service in the Glenn Highway corridor is operated mostly by the Municipality of Anchorage Public Transportation Department, but limited transit service also is provided from the Mat Su Borough by MASCOT, a non-profit service, and Valley Mover (a private operator). The Anchorage Rideshare Program manages vanpools operating 7 Table 8-1 Recommended Road Improvement Projects, AMATS Long Range Transportation Plan, This cost estimate includes the Glenn Highway to Seward Highway Connection for which an environmental impact study is currently underway. Actual costs may significantly exceed $750 million because the LRTP project cost estimates were done in 2004; more recent H2H engineering and cost updates are higher than previously estimated. -16-

18 in the corridor. In addition there is a large number of commuter carpools on the Glenn Highway. State, Municipality, Mat Su Borough and federal governments have a vested interest in maintaining the Glenn Highway transportation level of service. A collaborative advocacy for coordinated, multimodal, integrated transportation service strategies should be nurtured. Coordination and funding support for transportation improvements and system management by the state, the MOA, and the Mat Su Borough should emphasize balanced transportation investments to maintain service levels, enhance mode options, provide mobility for all citizens, and mitigate auto dependency and adverse community impacts. Multi Modal Strategies for the Glenn Highway Transit service, vanpool, and carpool choices reduce vehicle demand in the Glenn Highway corridor. Long commute trips to Downtown and Midtown Anchorage that are made in higher occupancy vehicles instead of solo driver vehicles will reduce roadway capacity expansion requirements. Leveraging good transit services, ridesharing, telecommuting and intelligent system management measures could effectively reduce peak vehicle demand as much as an equivalent freeway lane. About one-third of Mat Su Borough employed persons work outside the Borough. US Census data indicate 23% of those who work outside the Borough travel to work in carpools. 8 Most of these carpoolers commute on the Glenn Highway to worksites in Anchorage. Approximately 900 Mat Su commuters are active participants in vanpools traveling to and from jobs in Anchorage. In 2007 People Mover s Route 102 operated 16 weekday trips between Chugiak/Eagle River and Downtown Anchorage with most trips scheduled during peak commute hours. Route 102 passenger trips average 12.3 miles, nearly three times the 4.5 mile systemwide average for People Mover. Passenger Miles per Bus Hour Rte 102 People Mover Systemwide Rider counts and bus hours data for People Mover reveal that Route 102 carried 232 passenger miles per bus hour of service in 2007 compared to a systemwide average of 100 passenger miles per bus hour. Route 102 is a high performance transit corridor; it s highspeed Glenn Highway express bus service can provide competitive peak period travel times for Downtown and Midtown workers. Service frequency for Route 102 needs to be increased to give travelers greater flexibility and more departure time choice. 8 US Census, American Community Survey 3-Year Estimates, Table B Means Of Transportation To Work By Place Of Work--State And County -17-

19 The cost for significantly more frequent Glenn Highway express bus service is modest relative to the anticipated LRTP $750 million for Glenn Highway roadway projects. Assume People Mover service were improved to 10-minute frequency over the next three years, then 30-minute service initiated to Wasilla over the next four years and both services were continued at that level for another eight years (i.e. a total of 15 years). The net cost increase to provide that significantly improved transit service would be under $18 million. Improved transit service costs likely could be offset by avoided capital costs to expand road capacity. 9 Table 2. Estimated Cost for Improved Glenn Highway Commuter Express Bus Service Service Frequency Scheduled Weekday Bus Trips Est. Annual Revenue Miles Est. Annual Revenue Hours Annual Cost (1000 s in 2007$) Existing 16 96,640 3,690 $ minute ,000 6,190 $ minute ,200 7,180 $ minute ,800 10,650 $1,360 Collectively, transit, vanpool and carpool, and other options including telecommute can dampen the vehicle demand on the Glenn Highway. Reducing vehicle demand during commute periods can limit or delay the need for freeway capacity expansion. Transportation System Management Intelligent transportation systems management has demonstrated considerable success across the nation in achieving better transportation system performance. 10 Example applications include incident detection/response, traffic management through construction work zones, and ramp traffic control to improve operations. These management best practices and techniques could be deployed in the Glenn Highway corridor to reduce travel delay and improve traffic flow and reliability. Travelers in all modes -- autos, buses, van and car pools -- would benefit. People Mover Route 45 High Performance Transit Route Route 45 is by far the best performing route in the People Mover system. It carries more passengers, generates more rider revenue, and delivers the best operating results per unit of bus service of all routes in the entire system. Route 45 has the most weekday riders, Saturday riders and Sunday riders. It also has the highest number of boarding riders per timetable revenue hour of service and the highest riders per revenue mile of service (3.6 vs. system average of 1.9). Route 45 is clearly a high performance transit route it sports the best route operating metrics for the entire People Mover bus system. 9 Estimated annual cost to increase Route 102 to 10-minute commute period express bus service is about $850,000 per year in 2007 dollars. Another $570,000 per year could provide Mat Su Valley commuters with 30-minute commute period bus headways to Wasilla. 10 Notable examples include the State of Washington traffic incident management program and the State of California statewide Performance Evaluation Measurement System (PEMS). -18-

20 Passengers per Weekday 3,000 2,500 2,000 1,500 1, Weekday Service Route B/G Passenger per Revenue Bus Hour Weekday Service Route B/G Route 45 travels between the Downtown Transit Center and the Alaska Native Medical Center via 3 rd Avenue, Commercial Drive, Bragaw Road, UAA Drive, Providence and Elmore Road. In 2008 a sub-route, 45G, was introduced to address high ridership and overcrowding. It operates between Downtown and Mountain View and provides service to the Glenn Square Mall; 45G does not travel south of the Glenn Highway. Route 45 has consistently performed much better than all other People Mover peer routes. Its history of riders per bus hour of revenue service is markedly better than the People Mover systemwide average. Route 45 serves an area which clearly needs and uses public transit service. Moreover, Route 45 is effective because it links denser residential neighborhoods with jobs and commercial establishments in downtown, with community and social services along the corridor, with hospitals and retail centers, and with the university community. People Mover began to implement its Blueprint Plan for route restructuring in Weekday revenue service hours increased from 381 bus hours daily in 2002 to 439 bus hours in 2007 (roughly 15%). See Table

21 60.0 Passengers per Revenue Bus Hour Route 45 Systemwide Average Route 45 has continuously outperformed other routes since 2001 by a wide margin in passengers per revenue bus hour of service. Its productivity (riders per bus hour) is 55% better than the People Mover weekday average and 38% better than the next best route. Table and 2007 Weekday Revenue Hours and Riders 2007 Weekday Route 2002 Timetable Revenue Hours 2007 Timetable Revenue Hours 2002 Average Weekday Riders 2007 Average Weekday Riders Timetable Revenue Hours % Timetable Revenue Hours Average Weekday Riders % Average Weekday Riders % % , % % 3/ ,298 1, % % ,118 1, % % % % % % 13/15* ,237 1, % % % -8-4% % -30-5% ,007 2, % % % 34 5% % % % 77-B/G % 8 2% % % Total ,729 13, % 2,835 * Routes 13 and 15 in 2007 are compared to Routes 11 and 12 in % The figure below displays the increase in weekday bus service revenue hours by route versus the change in daily route ridership between 2002 and Route 45 timetable revenue hours barely changed, hours vs hours per weekday; nonetheless, Route 45 gained more weekday riders (561) than any other route over that time span. Route 1, which started operation in July 2002, was next most productive; it gained 438 additional riders per average weekday with 6.95 additional timetable revenue hours of -20-

22 Weekday Rider Gain Route 45 Route 1 Route 7 Route 2 Route 102 Route 8 Route 75 Route Weekday Bus Hours Increase Year : Change in People Mover weekday bus service hours by route vs. change in weekday average riders by route. Route 45 increased riders by 561 per weekday with only 0.25 weekday revenue bus hours added. Route 1 showed 438 weekday passengers for 6.95 more revenue bus hours; Route 7 had 420 riders and added weekday revenue bus hours service. Route 7 gained 420 weekday riders with added hours of service. Initiatives to Enhance High Performance Some actions to enhance higher performance on Route 45 have already been taken. The schedule frequency has been increased so time intervals between successive buses are shorter. Additional service has been added to reduce overcrowding on selected trips and time-of-day. More service has been introduced by short-turning some bus schedules to serve North Mountain View and the Glenn Square Mall. Additional high performance initiatives may include the following for consideration: Reduce headway further as long as boarding passengers per revenue bus hour exceed other routes in the People Mover system. Route 45 in 2009 had 38% more passengers per weekday revenue hour than the next best major route and 55% more than systemwide. This means each additional bus service hour is far more effective on Route 45 than anywhere else. Institute transit signal priority to attain higher bus speed and reduce transit travel times. Interline Route 45 with other routes to enable 1-seat bus rides to more regional major destinations. For example, interlining Route 45 with Route 7 or Route 3 would enable access to additional major trip attractions in the Bowl (public service agencies, Dimond Center, Midtown commercial areas and jobs, Airport, etc.). Extend Route 45 service to the Tikahtnu Commons Mall to provide access to jobs and major commercial establishments. -21-

23 Chapter 4 Examining Transit Corridors Performance Chapter 3 addressed multi-modal coordinated strategies for the Glenn Highway, the primary regional multimodal transportation corridor. It also identified Route 45 as the best performing transit route in the People Mover bus system and described continuing operational initiatives for that route. In this chapter the focus shifts to what actions and measures might be appropriate for the other 12 bus routes within the Anchorage Bowl. One of those 12 routes Route 7 is used as a prototypical case for analysis of higher performance opportunities. Lessons from the People Mover Blueprint Plan, People Mover began the first phase of its Blueprint Plan service restructuring in July Three phases were carried out through July A fourth phase, to extend 30-minute all day headways to most routes, could not be fully executed within budget resources and service remained generally static from 2005 through The Route 65 circulator in the lower hillside area was eliminated in mid In July 2008, 30- minute midday service on Route 9 and 15 was introduced and added trips were scheduled on Route 45 to address overcrowding. Impressive passenger gains and operations productivity have been realized since People Mover carried over 4,222,600 annual passengers in 2008, compared to 3,120,500 in The breakdown of that 1.1 million rider gain is shown below for weekdays, Saturdays and Sundays. Table 4. Increase in People Mover Annual Bus Passengers, 2002 to 2008 Weekdays Saturdays Sundays Total 921, ,160 78,700 1,100,620 Source: MOA Public Transportation Department Ridership Reports It is instructive to examine the individual Blueprint Plan service restructuring actions and how each contributed to the passenger gains that People Mover achieved between 2002 and The service changes that People Mover began in 2002 fall into six primary categories: Eliminate poor performing routes/services Increase service on existing routes, adding more frequent schedules Institute memory schedules with bus arrivals at the same time each hour 1 Not long before the first restructuring phase, People Mover bus service to Elmendorf Air Base and Fort Richardson was eliminated along with bus services to military bases across the nation due to heightened national security in the aftermath of the September 11, 2001 terrorist attacks on the World Trade Center in New York and the Pentagon. -23-

24 Coordinate arrival and departure times Downtown and at other hubs for easy transfers Start new routes Increase Saturday and Sunday service span and frequency of service Each of these six service change actions contributed to the 1.1 million passenger growth realized by Other factors such as gasoline price and population growth also played a role. 2 An evaluation of the effect of each service change action and external factors on ridership was conducted to interpret their respective importance. The results are s presented in the following paragraphs. Eliminate Poor Performers In 2001 People Mover averaged 28.9 weekday riders per timetable revenue hour (TTRH) of weekday bus service. But four routes serving the lower Hillside area had single digit riders per revenue hour. People Mover eliminated those poor performers and redirected the saved revenue hours to increase service on better performing routes. That action yielded a gain of 153,000 annual riders by Increase Service Frequency Between 2002 and 2008 about 70 additional weekday timetable revenue hours of serviced were added. That enabled 103 weekday bus trips to be added. Nearly every routes gained revenue hours and trips to increase bus frequency (Route 36 had a decrease of two trips). The added service (i.e., more TTRH s) resulted in about 359,000 additional weekday riders annually by Institute Memory Schedules In 2003 and again in 2004 bus schedules were adjusted to memory headways with bus arrival times consistent from hour to hour. Users now could remember bus times more easily without constant referral to printed schedules. The easy-to-remember weekday schedules also contributed additional weekday transit passengers an estimated 84,000 annually by Coordinated Arrival & Departure Schedules Downtown Each People Mover bus route serves a limited geographic corridor from a beginning termini to its ending termini. Nearly all routes end at the Downtown Transit Center. The Blueprint Plan revamped route schedules to come together and leave Downtown at common times. This pulse scheduling practice enables passengers to more easily transfer from one route to another without long delays. Coordinated scheduling was also introduced (to the extent practical) at transit hubs serving U-Med, the Dimond Center and the Muldoon Center. Coordinated scheduling contributed about 44,000 additional annual riders by Gasoline prices spiked at over $4.40 per gallon in July

25 New Route Startup People Mover initiated Route 1 in July 2002, a new crosstown route from DeBarr & Muldoon roads to the Dimond Center. Route 1 operates from Muldoon via DeBarr Road, Baxter Road, Tudor Road & Elmore to U-Med, then Providence to Lake Otis Parkway and Abbot Road to the Dimond Center. Starting with weekday service only, Route 1 service expanded from 14.5 TTRH in 2002 to 22.7 weekday TTRH in 2008 plus 14.8 hours on Saturdays and 11.1 hours on Sundays. Weekday ridership grew from 24,600 in 2002 to 170,400 in Some of the weekday passenger growth is attributed to the gasoline price surge in 2008; the net effect of new route service TTRHs is estimated at 136,200 weekday riders. 3 Deploying transit service on a new route has both upside and downside effects. The upside is new riders will be attracted to the service; the downside is a startup delay in rider response to a new route. A similar but lesser effect occurs when service is added on a pre-existing route. Two examples are examined below the first for the Route 1 introduction and the second for added service on Route 9 in Route 1. The figure below shows weekday riders and riders per timetable revenue hour of bus service for the first 30-months operation of Route 1, beginning in July Route 1 underperformed the systemwide average each of its first 30 months of operation. People Mover s systemwide average was 28 to 29 passengers per revenue bus hour on weekdays in 2002 through The new Route 1 did not achieve that systemwide passenger productivity level until well after 30 months of operation. Route 1 Weekday Passengers & Passengers per Revenue Bus Hour, Startup to 30 months Weekday Passengers Passengers Passengers per Revenue Bus Hour System Avg Pass/Bus Hr Service increased 40% r Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Passengers per Revenue Bus Hou In addition, weekend service on Route 1 carried 30,200 riders in The weekend riders are considered within the added weekend service category. -25-

26 In the first 9 months, Route 1 passengers per revenue bus hour were less than one-half the system average. Ridership grew quite steadily and began to level off at about 525 weekday passengers after month 25. Had the Route 1 bus revenue hours been deployed elsewhere instead, at the systemwide average riders per revenue bus hour, 4,860 more passengers would have been carried. So, new routes carry a startup surcharge they are unlikely to be as productive in attracting riders as the system average for the first two to three years.. Route 9 (added service to existing route). A similar, but lesser, startup lag generally should be expected when service is increased on an existing route. For an established route the scale and duration of the lag is less because riders are already familiar and adaptation to improved service occurs more quickly. The figure below shows Route 9 performance for 18 months before and 21 months after weekday service frequency was upgraded from hourly to 30-minute midday in July 2008 (just as gasoline prices were peaking at $4.40 per gallon). Passengers per weekday trended upward immediately, then fell back for several months, eventually holding at about 960 approximately 200 more passengers per weekday than were carried before the service improvement. Route 9 Before and After Added Weekday Revenue Bus Hours, July Passengers Passengers/Revenue Bus Hour 48 Passengers per Weekday Min Midday Service Begins Passengers per Revenue Bus Hour 400 Jan 2007 Mar May Jul Sep Nov Jan 2008 Mar May Jul Sep Nov Jan 2009 Mar May Jul Sep Nov Jan 2010 Mar 20 Passengers per weekday TTRH of bus service averaged 40.2 for January through June 2008 prior to the added midday service; it fell immediately but then rebounded to average 36 passengers per bus hour through July Route 9 continues to attain higher passengers per revenue hour than the weekday systemwide average. [Note: Route 15 also was increased in 2008 from hourly to 30 minute frequency. Its passenger volume increased from 700 weekday riders to 950 while passenger per revenue bus hour initially declined sharply but then rebounded to about the same level as in 2007.] -26-

27 Added Weekend Service Saturday riders on People Mover number a little less than one-half as many as on weekdays. Sunday riders number about one-quarter as many as on weekdays. But for both weekend days, the trend in ridership has been upward since implementation of the Blueprint Plan began in Weekend rider gains were due to more revenue service scheduled as well as earlier and later hours of operation. Most of the weekend service increases and rider gains took place in In 2008 People Mover carried 179,800 more weekend passengers than in Annual Saturday and Sunday Riders, , ,000 Saturday Sunday Annual Passengers 300, , , , ,000 50, Other Factors Gasoline Price Spike Gasoline prices in Anchorage began to rise sharply in late February By May gasoline was over $4.00 per gallon and spiked at over $4.40 per gallon in July Travelers responded to the sharply higher gas prices by shifting to transit. From May through August 2008, People Mover carried nearly 95,000 more passengers than for the same months in The gas price surge is estimated to account for nearly 148,000 added bus trips in Other Factors Population Growth -27-

28 Population in Anchorage grew by about 16,000 persons from 267,860 in July 2002 to 283,900 as of July These new residents account for an estimated 650 annual transit trips; thus they represent a very small fraction of the ridership gain that People Mover achieved. Summary of Findings The six Blueprint Plan restructuring actions together with the effect of gasoline price spiking collectively grew People Mover ridership by 1.1 million between 2002 and Population growth in the Anchorage region had a negligible ridership impact. The effect of the various service change actions are illustrated in the pie chart below. By far the largest ridership effect came directly from adding weekday timetable revenue hours on existing routes; that contributed 33% of the ridership gain. Added timetable hours for weekend service accounted for 16% of total passenger growth. Cancelling poor performing route and redirecting those TTRH hours to other routes provided 1.1 Million Riders Gained, 2002 to % of the rider gain. Establishing a new route (Route 1) delivered 12% of the new riders; the gasoline price surge in 2008 contributed 13%. Memory headways and coordinated schedules downtown reinforced and strengthen the effect of other service change actions; together they contributed 12% of the total system gain. Add Weekend TTRH Higher Gas Price New Route Drop Poor Rtes Add Weekday TTRH Over 61% of the total rider gain is attributed to added timetable revenue hours of service on weekdays, weekends, and a new route. Coord Schedules Memory Hdwys Re-assigning TTRH from poorly performing routes and instituting memory headways and coordinating arrival and departure route schedules downtown yielded about 25% of the ridership growth. Added Service vs. Added Riders The strong relationship between timetable revenue hours of service and ridership also is apparent for individual routes. The composite effect of added TTRHs of service, memory headways and coordinated pulse scheduling Downtown is examined here for weekdays. The chart below displays the percent change for each route in TTRH 4 Sources: US Census 2000, Alaska Department of Labor and Workforce Development, Research and Analysis Section, Demographics Unit. Table Estimates for Alaska Places,

29 from 2002 to 2007 vs. the corresponding percent increase in passengers. Year 2007 is used in this analysis to exclude the effect of the gasoline price spike in Route 1 and Route 102 had large percentage changes in TTRH and riders; they stand out at the right side of the graph. All other routes exhibit a collinear pattern but are clustered around the lower left of the chart. Route 102 exhibited a percentage increase in riders very close to the percentage increase in TTRH of service. This service elasticity the percentage change in riders for a one percent change in service hours -- is nearly 1:1; that is unusually high. Route 1 had even higher rider elasticity, showing 227% increase in riders compared to 49% increase in TTRHs of service over the fiveyear period. Increase in TTRH vs. Increase in Riders by Route, % Increase in Revenue Bus Hours 300% 250% Route % 150% 100% 50% Route 1 0% -50% 0% -50% 50% 100% 150% 200% 250% 300% % Increase in Riders A larger scale display of the same information for selected routes (with Routes 1 and 102 excluded) is shown in the next chart. Again, it is apparent that a number of routes achieved higher percentage rider gains than their corresponding percentage increase in timetable revenue bus hours. Routes 60, 3, 9, 2, 8, and 45 performed well. Route 45 s performance at 28% rider increase and only 0.5% increase in timetable hours was particularly noteworthy. System Route Structure and Coverage A public transit system has to focus services where there is greatest demand and opportunity to attract riders. People Mover orients nearly all of its routes to serve Downtown because travel and activity is concentrated, downtown is walkable, parking spaces are limited, and there are parking fees. Over ninety percent of transit riders are traveling either from or to their home. Therefore bus routes need to be near as many homes as possible and connect to a variety of destinations. Most people choose not to walk more than one-quarter mile to get to or from bus stops., Accordingly, each bus route effectively serves a relatively -29-

30 Increase in TTRH vs. Increase in Riders for Selected Routes, % Route 7 40% % Increase in Revenue Bus Hours 30% Route 77 Route 2 20% Route 14 Route 75 Route 9 10% Route 8 Route 60 Route 3 0% Route 45-20% -10% 0% 10% 20% 30% 40% 50% -10% Route 36-20% % Increase in Riders narrow 1/2 mile geographic band, ¼ mile each side of its path. As a consequence there are many potential traveler destinations which are not directly accessible on a particular bus route; transfer to a second bus route is required for such destinations. Parts of the Anchorage Bowl, notably the Hillside area and south Anchorage, are simply not hospitable to transit service due to low density development and terrain. People Mover s route system generally blankets the remainder of the Bowl. Transit service is accessible to approximately three-quarters of the population within the transit focus area. Where Transit Riders Are Going Boarding and alighting counts at People Mover bus stops give some clue about where riders are going. Most bus stops have low counts because they serve adjacent residential areas. Busier bus stops or route segments with large numbers of boarding passengers are generally around major destinations. The most active bus stops, displayed below, are highly concentrated Downtown and at U-Med, the Dimond Center and Muldoon transit hubs 5. Downtown is the dominant destination. Others include the Northway Mall and Mountain View. Transit passenger destinations are highly focused in areas that are served by multiple bus routes. Bus Stops with High Passenger On and Off Activity, March Many passengers transfer to reach their destination. Transferring passengers are included in boarding counts within the destination concentrations, especially in the Downtown area. -30-

31 The Burden of Transferring Majority of 2001 and 2005 Transit Riders Transferred to Reach Destinations two, 18% three or more, 7% one, 27% 2001 none, 48% two, 21% three or more, 18% one, 22% none, 38% 2005 Note: Values show number of transfers, percentage of riders. Direct service from origin to destination is important to most travelers. Changing buses or modes en route is a significant deterrent to attractive service and typically not favorably viewed. A 2001 People Mover rider survey queried passengers about transfers. A second People Mover rider survey in 2005 also asked passengers if they had to transfer to complete their bus trip. The 2001 survey found that 52% of People Mover riders transfer buses to reach their destinations; while the 2005 survey reported 62% of riders reached their destinations by transferring. -31-

32 TBEST Route and Service Planning Tools Travel within cities is diverse and complex. In Anchorage over one million individual trips are made each day by residents for a variety of purposes to a broad array of destinations. Estimating how many of those daily journeys will use transit is complicated. This project implemented a transit ridership estimation model to help People Mover s planning staff evaluate People Mover transit routes and potential service changes. The Florida Department of Transportation sponsored research leading to development of the ridership model, TBEST. TBEST is a computer simulation model that uses descriptions of the transit system routes, bus stop locations, service frequency by time of day and day of week, and detailed socio-economic characteristics of the population along each route to compute ridership estimates at individual stops. TBEST is useful for evaluating ridership changes due to routing changes, headway improvements, hours of operation, and other modifications. The TBEST model was calibrated for People Mover. It can be used by planning staff for continuing analysis and assessments of possible service changes. Schedule Interlining Selectivity May Help People Mover currently assigns buses and drivers to bus schedules without limiting the itinerary that an individual driver/bus may be assigned. For example, a driver may start his workshift on a Route 45 trip, then make a Route 7 trip, next complete a Route 60 trip schedule and so on throughout the day. Unconstrained interlining is used to find the least costly way to operate all of the scheduled bus trips. But this practice has two other consequences. First, People Mover bus driver work runs commonly span multiple routes each day, so drivers must know every route and its bus stop locations. Drivers assigned to a limited number of routes would be more familiar with those routes and therefore may provide superior service. Also improved passenger convenience may be realized if a conscious alignment of bus routings with passenger transfer patterns could increase the likelihood that passengers transferring between routes might obtain a one-seat ride. Passenger surveys indicate strong transfer patterns between certain People Mover routes. Routes 45, 75, 3 and 7 experience the highest passenger transfer interchange. Selective interlining of these high transfer routes may serve riders better at very minor cost to People Mover. Additionally, interlining between selective routes to enlarge access to major employment, medical, services, and shopping attractions could provides riders more destination accessibility opportunities. People Mover should explore and evaluate these options. -32-

33 Route 7 Prototype Analysis for High Performance Transit Corridor Route 7, serving the Spenard Road and Jewel Lake corridor is analyzed as a Prototype Corridor case. The analysis examines a number of service attributes to achieve better service quality and higher ridership response. The prototype corridor analysis is a pilot study on a single People Mover route. Study considerations include route coverage, service headways, ridership, bus running times, on-time performance, stop spacing, coordination among routes, origin-destination patterns, transit signal priority, and household socio-economic profiles along the corridor. In the future, the analysis methodology may be replicated on other People Mover routes. Service Schedule & Frequency Route 7 Service Profile Route 7 provides transit service between the Downtown Transit Center and Dimond Center via L and I streets, Spenard Road, Jewel Lake Road, Arlene Street, and Dimond Boulevard. The route is split into two separate travel patterns, or sub-routes : 7 and 7A. Every 7A goes to the Airport while Route 7 does not except evenings and weekends. South of International Airport Road, 7 and 7A operate different route patterns between Jewel Lake/Collins Way and between Raspberry/Arlene, as shown here. Route 7 transfers can be made to other People Mover routes (3, 9 and 36) in addition to transfers at the transit centers at Downtown and the Dimond Center. Coverage -- Split Routing Makes Sense The development density along Route 7 is markedly different north of the Airport than south of the Airport where much lower density prevails. In the latter environment, added route coverage is achieved by spitting service among the Route 7 and Route 7A branches so that transit service is accessible to more households in the Jewel Lake community Weekdays: Before 6 PM the overall frequency of weekday transit service along the majority of the Route 7 corridor is every half-hour, alternating between the respective hourly service of 7 and 7A. After 6 PM all buses (both 7 and 7A) go into the airport. South of International Airport Road, 7 and 7A maintain their respective (different) routing. -33-

34 Table 5. Weekday Scheduled Bus Travel Times (end-to-end), Effective July 2009 Direction and Route AM peak Mid-day PM peak Inbound , Inbound 7A 50 50, Outbound , Outbound 7A 53 53, Saturday and Sunday: Weekend frequency is hourly, with all buses (both 7 and 7A) going into the airport. South of International Airport Road, 7 and 7A maintain their different route patterns. Service Time Span Service on weekdays starts at approximately 6 AM and continues through 11:30 PM. Weekend service is more abbreviated, spanning approximately 13 hours on Saturdays from 8 AM through 9 PM while Sunday service operates approximately 10AM to 7 PM. Table 6. Weekday, Saturday & Sunday Service Start & End Times, Effective July 2009 Route 7 Direction Weekday Saturday Sunday Inbound First bus leaves Dimond Center 6:02 AM 7:49 AM 9:51 AM Last bus arrives Downtown 10:49 PM 8:41PM 6:40 PM Outbound First bus leaves Downtown 6:15 AM 8:35 AM 10:35 AM Last bus arrives Dimond Center 11:36 PM 9:26 PM 6:56 PM Passenger Boardings Systemwide service changes were made in July 2003 and again in July 2004, generally increasing frequency on routes and providing better connections between routes. Service improvements on Route 7 made in July 2003 continued through 2006; they added 15 more weekday scheduled trips, 7 more Saturday trips and 8 more Sunday trips. In 2007 service was reduced slightly when intra-airport service on Postmark Drive was discontinued. Ridership on Route 7 increased by nearly 38% between 2002 and 2007, gaining 420 riders per weekday. The corresponding systemwide rider increase was 27.1%. Route 7 Average Weekday Riders, 2000 to 2007 Route 7 Average Weekday Ridership Passenger Boardings Year -34-

35 Passengers per timetable revenue hour (productivity) on Route 7 fell in 2003 and 2004, even as the number of riders rose steadily Productivity declined somewhat in 2003 and 2004 because passenger gains lagged the revenue hours increase in Productivity has edged upward since Route 7 Passengers per Timetable Revenue Hour Route 7 Passengers per Timetable Revenue Hour # of Passengers per Rev Hour (YTD 2nd Qtr.) Weekday Saturday Sunday Year Passengers per scheduled bus trip inbound and outbound are shown below for weekdays and Saturdays in mid Over 20 riders per trip are carried except for the early morning and late evening schedules. On weekdays, the Route 7 trips scheduled at 58 minutes after the hour consistently carry more than the Route 7A trips (at 12 minutes after each hour) because the Route 7A weekday trips take an additional eight to ten minutes time to serve the airport. Route 7 Weekday Average Inbound Riders by Time of Day (2 nd Quarter 2007) 7 Inbound Weekday, 2nd Quarter 2007 (YTD Average Ridership by Time of Day) # of Riders :58 AM 6:12 AM 6:58 AM 7:12 AM 7:58 AM 8:12 AM 8:58 AM 9:12 AM 9:58 AM 10:12 AM 10:58 AM 11:12 AM 11:58 AM 12:12 PM 12:58 PM 1:12 PM 1:54 PM 2:10 PM 2:54 PM 3:10 PM 3:54 PM 4:10 PM 4:55 PM 5:10 PM 5:54 PM 6:10 PM 7:12 PM 8:12 PM 9:12 PM 10:00 PM Start of Inbound Run -35-

36 Route Route 7 Weekday 7 Average Average Outbound Riders Riders by Time by Time of Day, of Day (2 nd (2Quarter nd 2007) 2007) 7 Outbound Weekday, 2nd Quarter 2007 (YTD Average Ridership by Time of Day) # of Riders :15 AM 6:45 AM 7:15 AM 7:45 AM 8:15 AM 8:45 AM 9:15 AM 9:45 AM 10:15 AM 10:45 AM 11:15 AM 11:45 AM 12:15 PM 12:45 PM 1:15 PM 1:45 PM 2:15 PM 2:45 PM 3:15 PM 3:45 PM 4:15 PM 4:45 PM 5:15 PM 5:45 PM 6:15 PM 6:45 PM 7:45 PM 8:45 PM 9:45 PM 10:45 PM Start of Outbound Run Route 7 Saturday 7 Average Inbound Inbound Saturday, Riders 2nd by Quarter Time of 2007 Day (2 nd Quarter 2007) (YTD Average Ridership by Time of Day) # of Riders :46 AM 8:46 AM 9:46 AM 10:46 AM 11:46 AM 12:46 PM 1:46 PM 2:46 PM 3:46 PM 4:46 PM 5:46 PM 6:46 PM 7:16 PM Start of Inbound Run Route 7 Saturday Average Outbound Riders by Time of Day (2nd Quarter 2007) 7 Outbound Saturday, 2nd Quarter 2007 (YTD Average Ridership by Time of Day) # of Riders :26 AM 9:26 AM 10:26 AM 11:26 AM 12:26 PM 1:26 PM 2:26 PM 3:26 PM 4:26 PM 5:26 PM 6:26 PM 7:26 PM 8:26 PM Start of Outbound Run -36-

37 Route 7 Passengers On and Off by Segment Route 7 has strong destination attractions at its northern Downtown termini and at its southern Dimond Center termini. Most Route 7 passengers either board or alight at the Downtown Transit Center or the Dimond Center. Between 75% and 80% of passengers get on or off at those two centers. Elsewhere along the route passenger boarding and alighting activity is significantly lower with only a few bus stops accommodating more than 10 passengers per day. Passenger on and off counts are higher north of the Airport than are counts south of the Airport. This reflects both higher density and greater transit usage in the Spenard community than in the Jewel Lake neighborhoods. Route 7 Inbound Boarding Riders by Route Segment, March 2005 Route 7 Inbound Riders by Segment (March 2005) 400 # of Riders Boardings Alightings Chester Creek to DowntownTC 2. Chester Creek to Spenard/Minnesota 3. Spenard/Minnesota To Int. Airport Rd. 4. Airport Area 5. Int. Airport Rd. to Jew el Lake/Straw berry 6. Jew el Lake/Straw berry to Minnesota/Dimond 7. Minnesota/Dimond to Dimond TC Segment Route 7 Outbound Boarding Riders by Route Segment, March 2005 Route 7 Outbound Riders by Segment (March 2005) # of Riders Chester Creek to DowntownTC 2. Chester Creek to Spenard/Minnesota 3. Spenard/Minnesota To Int. Airport Rd. 4. Airport Area 5. Int. Airport Rd. to Jew el Lake/Straw berry 6. Jew el Lake/Straw berry to Minnesota/Dimond 7. Minnesota/Dimond to Dimond TC Boardings Alightings Segment Traffic Signal Impacts on Travel Times Route 7 has over 30 traffic signals along its path between the Downtown Transit Center and the Dimond Center. A series of auto travel time runs were made with a GPS receiver that recorded time and location readings every five seconds. A total of 28 runs were made in both directions. The auto travel route followed the #7 bus route between the Downtown Transit Center (TC) and International Airport Road, then continued south (skipping the Airport terminals) along Jewel Lake Road to Dimond Boulevard, then east to the Dimond -37-

38 Center. Data was collected on weekdays between May 16 and June 14, Typical travel time by auto varies from 25 to 35 minutes depending on traffic conditions. Waiting at traffic signals accounts for approximately one-third of the total travel time. Total Auto Travel Time and Signal Delay by Time of Day, Spring Travel Minutes Travel Time AM PEAK MIDDAY PM PEAK EVENING Signal Delay Multiple travel time runs between the Downtown Transit Center and the Dimond Center measured auto travel times, averaging 25 to 30 minutes for the inbound direction and 30 to 34 minutes for the outbound direction. Time stopped at traffic signals comprises about one-third of the total travel time. Bus Travel Time Bus travel time vis-à-vis auto travel time is a key factor in judging transit attractiveness. In general, travel by bus in Anchorage can take two to three times the journey time by auto for equivalent itineraries. 1 A detailed analysis of bus running times on Route 7 was conduced in 2007 using on-board GIS recorders to log bus travel time along the route. Data collected on 20 scheduled bus trips provides a picture of how much time is devoted to passenger pick up and drop off, traffic congestion, and delay at traffic signals. The Route 7 scheduled bus times between Downtown and the Dimond Center in effect during the travel time field runs are summarized in Table 7 (published in the July 10, 2006 People Mover Timetable Ride Guide). Route 7A schedule times were longer than Route 7 by 14 to 17 minutes per trip due to service into Ted Stevens Anchorage International Airport. 2 Passengers alighting and/or boarding for 400 individual events were observed. The bus Table 7. Route 7 and 7A Weekday Scheduled Bus Travel Times, Spring The 2008 US Census American Community Survey reports average travel time to work for transit passengers in Anchorage was 41.6 minutes; commute time for auto drivers averaged 19.4 minutes. 2 The airport service on 7A included service on Postmark Drive to and from the US Post Office and freight carriers. People Mover service on that loop was discontinued in July

39 Direction and Route AM peak Mid-day PM peak Inbound , Inbound 7A 54 54, Outbound , Outbound 7A 57 57, dwell time for one passenger boarding (with no bicycle or ramp use) at a stop is approximately 15 seconds; a single passenger alighting takes 9 seconds. Most stops have only one or two passengers boarding or alighting so bus dwell times per stop are relatively low. 3 Somewhat over three percent of the 400 stop events involved passengers mounting a bicycle on the bus bike rack before boarding. Stops where a passenger with a bicycle is getting on or off have longer dwell times, largely dictated by the passenger s time to get the bicycle on or off the bus s bicycle rack. These dwell times ranged between 17 and 82 seconds, averaging 37 seconds. A very few stop events involve boarding or exiting passengers in wheelchairs for which the bus loading ramp is extended. Ramp use necessarily extends the bus dwell time as do passengers using a walker or crutches. Observed dwell times ranged from 45 seconds to 164 seconds for passengers using walkers or wheelchairs, including time to properly secure the wheelchairs. The breakdown of time for Route 7 bus trips is displayed in the chart below. Dwell time at passenger bus stops varies from 15% to 20% of total travel time while traffic signal delay accounts for 22% to 36% of total time. The actual time a bus is in motion typically accounts for about one-half to two-thirds of the bus total travel time. Route 7 Bus Time Spent In Motion, Signal Delay, and Stop Dwells 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% AM Peak Mid-Day PM Peak Average Bus Travel Signal Delay Passenger Stops Bus schedule time on Route 7 typically took about 40 minutes in 2007 (56 to 63 minutes for trips that serve the Airport). The total travel time is divided between time spent stopping to pick up and 3 Dwell time is the time the bus was stopped; it does not include either deceleration before the stop or acceleration time after the stop. -39-

40 drop off passengers (16%), time waiting at traffic signals (27%) and time in-motion along the route (46%). Signal delay puts more stress on on-time bus schedule performance. Other cities and regions have implemented traffic signal schemes termed transit signal priority -- that extend or initiate signal green times for buses to improve bus speeds. Traffic signals delay tends to increase over time with congestion. Signal delay lengthens bus schedule times which in turn increases transit operating costs. While the impact of delay at individual signals may seem slight, that is not the case for all weekday bus trips aggregated over a full year. Route 7 has 60 trips per weekday or 15,300 per year. Assume implementation of transit signal priority could reduce bus running times by 3% to 7%; that equates to roughly $30,000 to $98,000 annual savings for Route 7. Extrapolation of the Route 7 transit signal priority savings over the entire People Mover system might yield annual savings of $280,000 to $900,000 per year. These preliminary estimates suggest it is worthwhile to conduct and evaluate a Transit Signal Priority Pilot Test; the Public Transportation Department proposed such a pilot test for Assembly approval and is currently conducting a limited pilot test on Routes 7 and 9. On Time Performance Monitoring and Measurements On-time schedule performance was measured for 20 bus runs on Route 7 in June Eleven trips arrived on-time at their Downtown or Dimond Center route termini. Nine (9) trips or 45% were three or more minutes late; five trips (25%) arrived five or more minutes late. On-time performance and reliability of bus service is important to riders. The People Mover 2001 Blueprint Plan specifies 85% or better on-time performance as satisfactory; it notes that schedule adherence is a critical success factor. 4 Collection of on-time schedule performance data for People Mover has been limited by availability of staff resources to conduct field studies and summarize results. As of 2008, on-time performance data is becoming available from the Automatic Vehicle Locator (AVL) system that People Mover installed on its buses. Reports from the AVL system show the on-time schedule adherence for bus stops along a chosen route for a specified date range. Other reports display the trip-by-trip elapsed time between successive time checkpoints along a route for a specified date span. The AVL system and associated reporting features are not fully operational as of June Still, interim Route 7 schedule adherence reports for January, February and early March 2009 show a disturbing disparity between scheduled and actual time departures for Route 7 bus stops. Similarly, AVL-generated running time reports for Route 7 in October 2009 indicate recurring late arrivals. 4 Nationally, rail services typically set 95% or better as their on-time performance standard; bus on-time standards tend to be lower, generally above 80% or 85%. -40-

41 Bus Operations Information System The AVL system referenced above is one of several advanced technology components of People Mover s emerging information management system for operations, scheduling and management. Other components include mobile data computers, data and voice communications between drivers and dispatch, computerized scheduling and driver assignments, automated passenger counters, and passenger information signs. Evaluation for final acceptance of the installed information system and reporting features is in progress at People Mover as of June The capabilities provided by the technologies and associated information system are important to People Mover. They enable real-time monitoring for dispatch and operations management and report alerts for unexpected events. Automated data collection supports analysis of on-time performance, scheduled vs. actual running times between timepoints, passenger boardings and alightings by location and passenger miles of travel, and driver duty logs. The passenger boarding, on-time performance, and running time reports will prove particularly powerful in evaluating and refining schedules for the entire system. Reporting system performance metrics provides a benchmark for on-going analysis and improvements. Bus Stop Spacing People Mover Route 7 has over 50 inbound bus stops and over 60 outbound stops. The stop locations determine how accessible bus service is for nearby residents and businesses. Stops also affect bus travel speed; the more stops the slower the bus speed. Bus drivers pull up to a bus stop if there is someone waiting or if a passenger wishes to get off the bus. As a consequence, a typical Route 7 trip stops at only one-third to one-half of all bus stops. Table 8 shows the number and percentage of bus stops utilized per bus trip for Route 7 and 7A by direction and time-of-day for a sample of bus trips. Buses on Route 7 generally carry more passengers per trip and make more stops than do Route 7A buses. Table 8. Route 7 Number of Stops Utilized and Percent of All Stops, Sampled Bus Trips Direction and Route AM peak mid-day PM peak Inbound 7 (52 total stops) 21 (40.4%) 24 (46.2%) 28 (53.8%) Inbound 7A (55 total stops) 20 (36.4%) 14 (25.5%) 16 (29.1%) 15 (27.3%) 11 (20%) 21 (38.2%) 16 (29.1%) Outbound 7 (64 total stops) 22 (34.4%) 24 (37.5%) 33 (51.6%) 22 (34.4%) 23 (35.9%) 16 (25%) Outbound 7A (68 total stops) 20 (29.4%) 22 (32.4%) 21 (30.9%) 20 (29.4%) Spenard Town Center Analysis: Spenard Road between Hillcrest Drive and 30 th Avenue has the highest population density in the Route 7 corridor as well as the highest transit stop usage. A planned Spenard Road reconstruction project anticipates reconfiguring the -41-

42 roadway. A detailed analysis of Route 7 bus stop locations and spacing was conducted. Population and jobs within ¼ mile of each intersection were computed to determine which potential bus stop location(s) provide accessibility to the largest number of residents and job opportunities. Existing bus stop locations were then compared to these most accessible locations. Proposed changes to bus stop locations are shown in the figure below. Two southbound bus stops (north of 26 th and south of 27 th ) are proposed to be replaced by a single mid-block stop. The northbound bus stop north of 23 rd is proposed to be moved closer to Hillcrest Drive. These stop location adjustments are intended to improve bus speed and enhance access and pedestrian safety. Spenard Road Existing and Proposed Bus Stop Locations, Hillcrest to 29 th Avenue Existing stops in purple (left) and proposed new locations (right) for stops in the Spenard Town Center area, Red numbers show the population within ¼ mile of points designated by the red triangles. Overlapping Routes - Route 7, 3 and 36 Routes 3, 7 and 36 traverse common paths between the Downtown Transit Center and Chester Creek. South of Chester Creek, Route 7 (inbound and outbound) and Route 3 (outbound) run on Hillcrest and Spenard to Benson; Route 3 inbound operates on Northern Lights Blvd and Minnesota. Route 36 operates inbound and outbound on Minnesota to Northern Lights. The scheduled departure time from the Downtown Transit Center for these three routes is staggered by five minutes: Route 36 leaves at 10 minutes past the hour, Route 7 at 15 and

43 minutes past and Route 3 at 20 and 50 minutes after the hour. The result is uneven service along L and I streets with buses separated by 5 minutes, then 25 minutes, and again 5 minutes followed by 20 minutes. Alternatively, the five scheduled bus trips each hour could be timed every 12 minutes and provide consistently-timed, superior service. There is a tradeoff if coordinated 12 minute service were to be considered. Current scheduling practice is structured around coordinated arrival/departure times for multiple routes into and out of the Downtown Transit Center. Bus scheduling is intended to enable easy transfer opportunities. In fact, route schedules arriving and departing are somewhat staggered. This is partially due to limited curb space which cannot accommodate buses from all routes simultaneously at the Transit Center, but also because of individual route operational schedules and running times. Would putting routes 3, 7 and 36 on a consistent 12 minute service interval along K and L streets disrupt easy transfers between routes downtown? Passenger transfer activity at the Downtown Transit Center was analyzed using data collected for six days in April A total of 477 transferring passengers were observed. Key findings from the transfer observations are: Route 45 had the highest number of transferring passengers. Forty-six percent (46%) of all observed Downtown transfers were to or from Route 45 Route 7 had the second highest number of transfers; 20% of total observed transfers were to or from Route 7. The highest number of passengers transferring from Route 3, Route 7 and Route 36 in each case was to Route passengers were observed transferring between Routes 3 and passengers were observed transferring between Routes 7 and passengers were observed transferring between Routes 36 and 45 Bus schedules for Routes 3, 7 and 36 arriving and departing the Downtown Transit Center were examined to determine the wait time for transfers to and from Route 45. Table 9 shows the wait time for passengers arriving Downtown on Routes 3, 7 and 36 who transfer to outbound trips on Route 45. Bus schedules from early morning through mid-day are displayed. The elapsed time from scheduled arrival on each of the three routes to scheduled Route 45 departure times is commonly over 20 minutes. Only four of the scheduled trips shown have transfer wait time of less than 10 minutes. -43-

44 Table 9. Downtown Transit Center Transfer Times from Routes 3, 7 and 36 to Route 45, July 2009 Rte 3 Arr Rte 45 Dep Trsfr Wait (minutes) Rte 7 Arr Rte 45 Dep Trsfr Wait (minutes) Rte 36 Arr Rte 45 Dep Trsfr Wait (minutes) 6:49 7: :42 7: :42 7: : : : : : : : : : : : : : : : : :42 10: : : : : :42 11: : : : : :42 12: : : : : :46 13: :22 13: :12 13: : :00 18 Table 10 shows transfer wait times for passengers arriving at the Downtown Transit Center on Route 45 who transfer to Routes 3, 7 and 36. Table 10. Downtown Transit Center Transfer Times from Route 45 to Routes 3, 7 and 36, July 2009 Rte 45 Arr Rte 3 Dep Trsfr Wait (minutes) Rte 45 Arr Rte 7 Dep Trsfr Wait (minutes) Rte 45 Arr Rte 36 Dep Trsfr Wait (minutes) 6:34 6: :34 6: :34 7: : : : : : :26 8: : : : :23 8: : :23 9: : : : :07 9: :07 9:15 8 9:07 9: : : : : :09 10: :09 10: :09 10: : :13 11: :13 11: :13 12: : :03 12: :03 12: :03 12: :03 13:22 18 The transfer wait times for passengers going from Route 45 to Route 3, 7 and 36 spans a range from one minute to 56 minutes; in general, the transfer passengers from Route 45 have longer transfer times than those transferring from the three routes to Route 45. For -44-

45 many cases the transfer time is over 20 minutes. Transfer wait times after 9 AM generally are shorter than during the morning peak period except for Route 36. Further examination of shifting Route 3, 7 and 36 schedules to more uniformly spaced trips is suggested to provide better service between Downtown and Benson Boulevard. Route 7 has the most transfers with other routes Downtown. Therefore Route 7 may be the logical time-coordinated connection to other routes at the Transit Center. Route 7 also has the largest number of boarding and alighting passengers between Downtown and Northern Lights Boulevard so it could be the service backbone through this area. Route 7 Bus Service vs. O-D Patterns The destinations of Route 7 bus riders are dominantly focused in the Downtown area and the Dimond Mall area. In excess of three-quarters of all passengers go to those locations while only a modest share of riders has intermediate destinations along the route. Data from AMATS 2007 origin destination travel patterns were summarized for the Sand Lake, Jewell Lake and Spenard communities served by Route 7. Residents of these areas make about 140,000 daily person trips by all modes. Approximately 10% of trips by corridor residents are to Midtown and six percent to Downtown. Over 20% of resident trips have destinations within the corridor area. Individuals from elsewhere in the Anchorage Bowl make about 70% (65,000 person trips) of all destinations in the Route 7 corridor, compared to 30% or 28,500 trip destinations by corridor residents. 5 Route 7 directly serves trips made within the corridor; it also enables transfer connections at the Dimond Center (Routes 1, 2, 9 and 60 for destinations along Old Seward, Lake Otis Parkway and Arctic) and to east-west Routes 3 and 36 across Midtown to the U-Med area. The travel patterns clearly show that travel destinations of corridor residents are much more broadly dispersed than are transit trips by corridor bus riders. Transit in the Route 7 corridor attracts most of its riders from persons who are going downtown, but does not compete well for residents going elsewhere in the Anchorage Bowl. Route 7 Household Socio-Economic Profiles The demographic and socio-economic profile of households is a strong indicator of propensity for transit usage. The TBEST transit rider estimation computer model uses these characteristics and transit service measures to estimate bus riders. Data from the AMATS 2007 travel modeling database for each traffic analysis zone (TAZ) within 1/4 mile each side of Route 7 were compiled and mapped. Statistical summaries also were tabulated for five geographic subareas along the Route 7 corridor: 6 5 For more detail on travel to and from individual geographic subareas in the corridor and trip purposes see Working Paper Travel Patterns in the Route 7 Corridor. 6 The Airport area is designated as a sixth subarea but is not relevant to the socio-economic analysis because it has no residential component. -45-

46 The five subareas are: 1. Chester Creek to Downtown 2. Chester Creek to Spenard/Minnesota 3. Spenard/Minnesota to International Airport Road 4. International Airport Road to Minnesota/Dimond 5. Minnesota/Dimond to Dimond TC Population Density People are not willing to walk very far to a bus stop or a destination. Higher density close to bus service means people have convenient access to bus stops. Conversely single family, large lot neighborhoods are much more spread-out, often requiring longer walking distances to reach a bus route. Population density for the five Route 7 subarea segments is displayed below. Highest density is in subarea 2 - Chester Creek to Minnesota at Spenard Road, south of 36 th. The Jewel Lake area south of Strawberry and the south Spenard Road to International Airport Road subarea are the next highest in population density along the corridor. Route 7 Corridor Population per Square Mile by Subarea Segments Population Density Persons / Sq. Mile ChesterCrk- Dow ntow n 2. ChesterCrk to Spenard/Minn. 3. Spenard/Minn. to Int. Airport Rd. Airport Area 4. Int. Airport Rd. to Jew el Lake/Straw berry 5. Jew el Lake/Straw berry to Minnesota/Dimond 6. Minnesota/Dimond to Dimond TC Segment Population density by traffic analysis zones is mapped below. The Spenard Road subarea from Chester Creek to Minnesota has the highest density. Higher density is found further south and west of Spenard Road to Wisconsin as well. A smaller density cluster can be seen from Strawberry to 88 th west of Northwood. -46-

47 Population per Square Mile by TAZ, 2007 Multifamily Housing The proportion of dwelling units in multifamily housing is summarized in Table 11. In the two northerly subareas between Downtown and Minnesota at Spenard, multifamily housing comprises nearly two-thirds of all housing units. The more southerly subareas have from 44% to 32% in multifamily units. Multifamily and Total Residential Units by Subarea, Corridor Subarea Residential Units in Multifamily Housing Total Residential Units Percent Units in Multifamily Housing 1. Chester Creek to Downtown % 2. Chester Creek to Spenard/Minnesota % 3. Spenard/Minnesota to Int. Airport Rd % 4. Int. Airport Rd. to Minnesota/Dimond % 5. Minnesota/Dimond to Dimond TC % Household Income The percent of households with income less than $40,000 per year along the Route 7 corridor is displayed in the chart below. The corridor segments from Chester Creek to -47-

48 Minnesota has 57% of households in that income group while Spenard to International Airport Road stands just under 50% of all households. Subareas south of the Airport have notably lower percentage of households with incomes less than $40,000. Households with Annual Income Less than $40,000 Households with an Annual Income of Less than $40,000 60% Percentage of Households 50% 40% 30% 20% 10% 0% 1. ChesterCrk- Downtown 2. ChesterCrk to Spenard/Minn. 3. Spenard/Minn. to Int. Airport Rd. Airport Area 4. Int. Airport Rd. to Jew el Lake/Straw berry 5. Jew el Lake/Straw berry to Minnesota/Dimond 6. Minnesota/Dimond to Dimond TC Segment Household Vehicle Availability The same pattern prevails with respect to household vehicle availability. The northern corridor subareas have much higher percentage of households with no vehicle while those south of the Airport have notably fewer households with no vehicle. The percentage of households with one vehicle is more consistently similar along the whole corridor. Percent Households with No or One Vehicle Available Percentage of Households with No or One Automobile Percentage of Households 60% 50% 40% 30% 20% 10% No Auto Only One Auto 0% 1. ChesterCrk- Dow ntow n 2. ChesterCrk to Spenard/Minn. 3. Spenard/Minn. to Int. Airport Rd. Airport Area 4. Int. Airport Rd. to Jew el Lake/Straw berry 5. Jew el Lake/Strawberry to Minnesota/Dimond 6. Minnesota/Dimond to Dimond TC Segment -48-

49 Chapter 5 Findings and Recommendations People Mover has provided public transportation service in Anchorage since the 1970 s. Management over the past seven years achieved a resurgence in transit ridership, hitting a historic high of over 4 million annual passengers in This high performance transit corridor project was initiated in search of strategies and actions to maintain ridership growth momentum. The Anchorage Long Range Transportation Plan and the People Mover Blueprint Plan present evidence that more frequent service and reduction in travel time have a major impact. Those findings provide a platform for interim and longer term strategies and actions for transit in Anchorage. Funding Funding resources are pivotal for transit services. Funding dictates the level of service that can be operated and controls what service improvements are possible. Indeed, in 1972 the funding measure to support creation of People Mover was defeated and startup of transit service was delayed two years. Again in 2003 and 2004, additional funding from the federal Congestion Management and Air Quality Program enabled the People Mover Blueprint Plan improvements to be initiated. And later stages of the Blueprint Plan were derailed by funding limits. People Mover s objective of maintaining ridership growth momentum is not an easy one. A gain of 1.1 million annual passengers (35%) was realized between 2002 and 2008 as People Mover added over 30,000 annual timetable revenue hours of service. But annual systemwide transit costs concurrently escalated $9.0 million over those same years 1. Continued service and rider increase clearly depends on sustained funding. The prevailing global economic climate poses extreme fiscal challenges for federal, state and local governments. Resources are stretched to a degree not experienced since the Great Depression. Transit operators across the nation face restricted funding regimens. It is difficult to anticipate more funding to increase transit service levels until the economy regains its footing from this recession environment. Higher-cost service improvements likely have to take a back seat in the near term. But that does not mean there are no avenues available to improve transit performance. Two strategies guide the recommendations presented in this chapter. The first strategy focuses on steps to improve service delivery. These actions are relatively low cost. The 1 Bus operating expense was $12,681,744 in 2002 and $21,719,335 in Source: Federal Transit Administration National Transit Database The Anchorage Consumer Price Index increased 19.78% between 2002 and 2008, suggesting $1.8 million of People Mover s operating expense increase could have been driven by inflation

50 second underlying strategy is to continuously direct resources to the most productive services when implementing candidate transit performance improvements. In order to boost the overall system performance, in the absence of added funding, underperforming services may need to be trimmed. Service reductions, if necessary, should focus on the lowest performing elements. The recommendations laid out below emphasize low cost initiatives for the near-term and address more costly options over the mid to longer-term. Exploiting Advanced Technology People Mover s emerging information management system for operations, scheduling and management encompasses multiple components, including GPS vehicle location logging, mobile data computers, data and voice communications between drivers and dispatch, computerized scheduling and driver assignments, automated passenger counters, and realtime passenger information signs. The capabilities provided by the technologies and associated data reporting are important to People Mover. They enable real-time monitoring for dispatch and operations management and report alerts for unexpected events. Automated data collection supports analysis of on-time performance, scheduled vs. actual running times between time checkpoints, passenger boardings and alightings by location, passenger miles of travel, and driver duty logs. Refining Running Times and Schedules Bus schedules are based on estimated travel time between route time checkpoints. Travel times vary with traffic conditions by time-of-day. The entire bus and driver scheduling system, as well as on-time performance, depends upon reliable travel time (running time) estimates. People Mover s evolving technology and information systems provide new capabilities to refine schedule running times that were never before available. The automated vehicle location system captures and reports actual bus running time between successive time checkpoints, segment by segment along each route. Statistical measures based on large data samples of actual time data can now be used to derive more accurate bus running times for all routes. This means that bus scheduling information is more accurate. In turn, on-time performance should be more reliable because schedules are data-grounded. People Mover scheduling staff should construct a database of field-measured, actual bus route segment running times for each route, direction and time-of-day. Analysis on this database should be performed to derive statistically-valid segment running times. This new process should then drive construction of bus schedules and driver run assignments for each scheduling cycle. Monitoring On-Time Performance The new transit management system technology helps drivers maintain on-time schedules by alerting them when they are running ahead or behind schedule. It also can log scheduled vs

51 actual arrival times by bus stop and report the percent of bus trips which were on schedule, behind and ahead of schedule for a designated calendar period. On-time schedule reliability is important to transit riders. People Mover s schedule adherence service standard is 85% or higher for satisfactory performance. Observations of actual schedule adherence on Route 7 indicate that on-time performance is not meeting the agency service standard. People Mover can improve its customer service by tracking on-time performance trends and continually focusing attention on problem situations. An on-time performance monthly benchmark report should be developed. On-time performance should be monitored on a continuing basis. Corrective actions should be devoted to routes, hot spot locations, time periods, and drivers who deviate from the schedule and performance standard. Transit Signal Priority to Reduce Bus Delay Bus travel times exhibit a considerable disadvantage compared to auto travel times for many trips. Bus speeds are slowed by signal light delays, traffic congestion, and dwell times at bus stops. Field measurements of traffic signal delays for Route 7 reveal that they account for as much as one -third of bus travel times 2. Transit signal priority techniques offer a way to help make transit service more reliable, faster, and more cost effective. Signal priority schemes increase the likelihood of buses getting signal green time through intersections, thus reducing traffic light delay for buses. The result is improved bus speed and shorter travel time. Signal priority systems consist of four basic components: 1. A priority request emitter on the vehicle that transmits a priority request to the traffic signal control system 2. A detection system that lets the traffic signal controller know where a vehicle requesting signal priority is located 3. Software that is programmed to manage the designated priority control strategy and determine whether and how to grant priority 4. Management reporting that collects data and produces reports of priority events Transit signal priority is implemented in many urban areas across the country. A handbook for planning and implementing transit signal priority projects has been published by the US Department of Transportation. 3 Case studies of a number of bus signal priority implementations report bus travel time savings in excess of 10 percent. In Chapter 4, a more modest estimate of the impact of implementing transit signal priority in Anchorage led to projected potential bus operating savings approaching $1 million per year. The following figure illustrates how the primary technology elements interact in a signal priority system. 2 See Chapter 4 for details 3 Gannett Fleming Inc., Transit Signal Priority: A Planning and Implementation Handbook, May

52 Source: An Overview of Transit Signal Priority (ITS America, 2004) The MOA Traffic Department has installed the OptiCom traffic control system at its signalized intersections within the Municipality. The system is already in use for emergency fire and police unit response. OptiCom can provide Early Green (Red Truncation) and Green Extension which are the two most commonly used TSP strategies. People Mover is currently conducting a Transit Signal Priority pilot test on Routes 7 and 9 in cooperation with MOA Traffic Engineering. The pilot test involves transit signal priority only at selected intersections and uses an unconditional priority scheme; this priority strategy is the simplest approach and provides priority to all transit buses approaching the selected intersections. An evaluation of the pilot test will be completed in If the results are favorable, transit signal priority implementation should be programmed for additional routes. Consideration should be given to more sophisticated signal priority control strategies and integration with the People Mover automated vehicle location system. Other Strategies to Reduce Transit Travel Time Reducing transit travel time can be achieved by increasing transit operating speed, by decreasing delays, and by shortening the access or waiting time for transit service. Queue jumping: Enabling buses to bypass traffic congestion queues is an oft-cited technique to decrease transit delay. But queue bypassing is practical only under very limited circumstances. The physical roadway space required to permit buses to maneuver around stalled traffic queues is seldom readily available. A large number of buses must use a queue

53 bypass road segment to justify it. On most routes in Anchorage there are generally 1 or 2 buses per hour; the limited number renders queue jumping impractical. One location where a reserved lane/space for buses may have merit is at the Dimond Center. Multiple routes serve the Dimond Center and buses often experience delay getting into and out of the Center. The Municipality of Anchorage Assembly provided local match funding in to support a Dimond Center Intermodal Facility Feasibility Study. People Mover could find it advantageous to advocate a bus circulation proposal to address bus delay within the shopping center as part of the study, and explore the proposal with Dimond Center management. Reduce Headway by Coordinating Route 3, 7 and 36 Schedules: Three routes 3, 7 and 36 - traverse a common path along L and I streets between the Downtown Transit Center and Chester Creek, then along Minnesota or Spenard Road. Scheduled departure times from the Downtown Transit Center for these three routes are staggered by five minutes; in 2009 Route 36 left at 10 minutes past the hour, Route 7 at 15 and 45 minutes past and Route 3 at 20 and 50 minutes after the hour. The result is irregular service frequency along L and I streets with buses separated by 5 minutes, then 25 minutes, and again 5 minutes followed by 20 minutes. The five scheduled bus trips for Routes 3, 7 and 36 each hour could be timed every 12 minutes to provide consistent headways and superior service. Current bus schedules are designed to enable easy timed-transfers among routes Downtown. By far the highest number of Downtown transfers for Routes 3, 7 and 36 are to or from Route 45; but these transfers often involve waits over 20 minutes. Route 7 has the highest volume of transferees and its schedule should be timed to facilitate connections with other routes Downtown, particularly Route 45. Route 3 and 36 trip schedules could be adjusted to uniform, more frequent service. People Mover Planning and Marketing should conduct a market research analysis of the Spenard to Downtown corridor, determine passenger boarding and alighting counts, and identify the numbers of transfer passengers to and from these three routes Downtown and their destinations. Planning should examine bus scheduling with uniform headways and minimum Downtown transfer times with Route 45. More complete facts then could help define the best operating plan. See further discussion below under Cost to Increase Service Frequency. Spenard Road Bus Stop Locations: Bus stops spaced too closely increase transit travel time. However analysis of bus stops along Route 7 revealed that stop spacing is not an issue because passengers are getting on or off at about one-third to one-half of the available bus stops. A planned Spenard Road reconstruction project anticipates reconfiguring the roadway. Concurrent changes to the bus stops along Spenard Road between Hillcrest Drive and 29 th Avenue are proposed in conjunction with the reconstruction project. This road segment has the highest population density in the Route 7 corridor and the highest transit stop usage. The nine proposed bus stops will serve the community well and improve bus flow. 4 Assembly Resolution No approved May 12,

54 Traveler Information: People Mover gained additional riders when they introduced memory schedules that made it easier for people to remember bus schedule times. The internet has become a common conduit for transit information. User-friendly transit websites such as Portland s Tri-Met have proven to be powerful and valuable assets in meeting rider information needs. Website layouts and design are important - they should be visually attractive, easy to navigate and understand, and provide information quickly. People Mover should explore best practices elsewhere and use that insight to streamline and simplify their current website offering. People Mover s website currently does not include a tool for trip planning. Other transit agencies have website aids to help travelers find a transit route to their desired destination, determine bus schedule times and where they can board. The Public Transportation Department intends to implement Google Transit for trip planning and is currently developing data required for installation setup of the Google application. Using Google Transit for customer information provides real-time information and mapping services via the internet, phone, cell phone, or handheld/pda. Transit Marketing: People Mover has traditionally used print advertising and radio spots for marketing both transit and rideshare services. Budget resources are split between print and radio in roughly equal proportions. Marketing strategies seek to promote transit to selected market segments such as youth, seniors, university students and staffs, visitors, etc. Measures of effectiveness to assess the outcomes achieved by marketing efforts and expenditures are needed. As part of this High Performance Transit Corridor analysis, a direct mail marketing effort along Route 7 was conducted in May A direct mail transit information packet was distributed to nearly 10,000 households. The packet contained a Route 7 map, bus time schedules, and a coupon for a day pass or a monthly pass discount. The mailer campaign proved successful with nearly 7% of households north of Airport Road using the transit coupon and 4% coupon users south of Airport Road. The mailer campaign is estimated to have added 900-1,400 transit trips over a one-moth time span on Route 7. The 2010 national census will soon provide a rich resource of socio-economic and travel data that can help identify transit opportunities. People Mover should collaborate with AMATS staff to exploit these market research data resources. Additionally, People Mover should continue to refine and sharpen its marketing strategies for positioning and branding, targeted marketing, customer information, community partnering initiatives, and market research to understand customers and potential customers. Costs to Increase Service Frequency The Blueprint Plan implementation experience demonstrates that added service weekdays and weekends accounted for the majority of added riders. In numerous interviews and surveys, many Anchorage residents have responded that transit is not viable because existing bus service frequency is too limiting

55 Hourly service is currently operated on the two Route 7 branches traversing the Jewel Lake community and 30-minute frequency is offered from Downtown to International Airport Road and on Dimond Boulevard between Arlene Street and the Dimond Center. North of International Airport Road, the route is much more heavily used than south of that location. A cost analysis was conducted for Route 7 to estimate the funding needed to increase service frequency. Table 12 summarizes operating cost estimates for six weekday service scenarios with differing peak period, midday and evening service frequencies; the route branch segments within the Jewel Lake area would have only one-half the service shown in the table for each scenario. Transit service existing in 2009 was used as a benchmark. Increasing the current 30 minute service during morning and afternoon peak periods to 15 minute service would increase operating costs by about 40%; reducing peak period headways further to 12 minutes would add 57% while 10 minute peak headways would cost 72% more. Table 12. Cost Estimates for Route 7 Increased Service Scenarios Weekday Schedule (headways in minutes) No of Daily Trips Est. Annual Cost (2008$) % increase over Actual am & pm peak 30 midday 30 after 6:30pm 60 am & pm peak 15 midday 30 after 6:30pm 60 am & pm peak 12 midday 30 after 6:30pm 60 am & pm peak 10 midday 30 after 6:30pm 60 am & pm peak 15 midday 15 after 6:30pm 60 am & pm peak 15 midday 15 after 6:30pm $1,976, $2,766, % 94 $3,096, % 103 $3,392, % 109 $3,590, % 116 $3,820, % Reducing midday and peak period headways to 15 minutes adds 82% more operating cost; if evening service is upgraded to 30 minute headways as well the cost expands to 93%. Route 7 accounts for about 11% of People Mover s annual operating hours and cost. The table above shows that costs rise rather dramatically with transit headway improvements 40% to double peak period service and 93% for all day service improvements. So, why is increasing transit service frequency desirable? Service frequency is a primary determinant of transit ridership. Over one-million individual person trips are made every weekday in Anchorage. If People Mover service were able to attract only one percent more of those daily trips it would gain about 70% more riders. 5 5 For additional discussion and empirical data, see Chapter 9 Transit Scheduling and Frequency, TCRP Web Document 12 (Project B-12), Traveler Response to Transportation System Changes Interim Handbook, Richard H. Pratt et al, March [

56 Estimated Ridership Impact of Increased Route 7 Service Frequency: Route 7 ridership increased about 70% from 285,200 annual weekday riders in 2002 to 405, 400 in During that period timetable revenue hours of service increased 47%. The impressive ridership gains were fueled by four primary factors increased bus service, transition to memory headways, coordinated route schedules Downtown and at the Dimond Center, and sharply higher gasoline prices. Future service increases may not result in comparable rider gains because the effect of memory headways and coordinated route schedules will be muted. Future gasoline price changes are not known but they too may not be as pronounced. The ridership gain due to the three service options shown above that increase only peak period service to 15 minute, 12 minute or 10 minutes is estimated to range from 78,000 to 140,000 more annual riders. For the two options that also increase midday service, ridership gain is estimated at 160,000 to 183,000 annual riders. The estimated added cost to provide these improved levels of service range from $790,000 to $1.84 million per year in 2008 dollars. Coordinated Routes 3, 7 and 36 Headways for Low Cost Frequent Service As described earlier, coordinating Route 3, 7 and 36 schedule times on the inner portion of Route 7 from Downtown to Northern Lights Boulevard/Spenard could offer a 12-minute headway. Increasing ridership is a primary goal for People Mover. More frequent service is the largest contributor to increased ridership. However adding timetable revenue hours of service requires additional budget allocation to cover higher operating costs. But coordinating service schedules for Routes 3, 7 and 36 could implement 12 minute headways with nominal if any additional cost. This is a unique opportunity to test the ridership impact of much more frequent service in Anchorage. Coupling more frequent service with development of more accurate running times, diligence in achieving high on-time performance, and testing transit signal priority on this corridor is proposed as an effective near-term action. Route 45 High Performance Transit Route Route 45 carries the most riders of all routes and has the highest riders per revenue bus hour of all routes. It is clearly the high performance transit route in Anchorage. In 2008 Route 45G was introduced as a shortened segment serving the Glenn Square Mall and north Mountain View. The chart below shows average passengers per weekday on each inbound bus trip for the first quarter of Route 45G s performance may not look impressive compared to the full Route 45. However, that perception is deceptive; 45G actually carried about 50 passengers per weekday timetable revenue bus hour compared to 52 for Route 45. Both 45 and 45G are well above the system weekday average of 33.3 passengers per revenue bus hour

57 2010 1st Quarter Avg Riders :50 AM 6:42 AM 7:35 AM 8:30 AM 8:55 AM 9:55 AM 10:55 AM 11:34 AM 12:05 PM 12:35 PM 1:05 PM 1:35 PM 2:30 PM 3:25 PM 3:55 PM 4:55 PM 6:09 PM 6:54 PM 7:56 PM 9:02 PM Route 45 and 45G (in dark tone above) average riders for each weekday inbound scheduled bus trip in the first quarter of Fewer riders use 45G trips. However, riders per bus hour of service on 45 and 45G are considerably higher than the People Mover system average. Additional Route 45 initiatives for consideration include the following: Continue to add service and reduce headways as long as boarding passengers per revenue bus hour exceed other routes in the People Mover system. Route 45 in 2009 had 38% more passengers per weekday revenue hour than the next best major route and 55% more than the systemwide average. This means each additional bus service hour is far more effective on Route 45 than anywhere else. Following evaluation of a transit signal priority pilot test on Route 7, institute signal priority on Route 45 to attain higher bus speed and reduce transit travel times. Selectively interline Route 45 with other routes to enable 1-seat bus rides to more regional major destinations. For example, interlining Route 45 with Route 7 or Route 3 would enable access to additional major trip attractions in the Bowl (public service agencies, Dimond Center, Midtown commercial areas, Airport, etc.). Glenn Highway High Performance Strategy The Glenn Highway is a primary transportation corridor in Anchorage as well as the sole roadway connection between Anchorage and the Matanuska Susitna Borough. The Mat Su is the fastest growing area in the state. The 2027 Anchorage Long Range Plan incorporates over $750 million in capital improvements for the corridor to accommodate future travel growth. A multimodal corridor development strategy with roadway, transit, vanpooling, carpooling, travel demand management, and advanced technology systems management elements would provide superior transportation choice and service as well as mitigate capital costs and single occupant vehicle environmental impacts. The MOA Public Transportation Department and AMATS should advocate crafting a collaborative, integrated, multimodal Glenn Highway strategy and funding consensus among

58 regional transportation providers. The pending 2011 AMATS LRTP update should be used as the vehicle to formulate Glenn Highway operational strategies and funding for multiple traveler mode options to meet transportation demand, exploit technology deployment to manage congestion, relieve parking requirements and traffic impacts in the Anchorage Bowl, minimize system costs, and improve air quality. AMATS can build from the recent work for the Eagle River LRTP and analyses performed for the Highway to Highway Connection project. The pending Regional Transportation Authority study is another opportunity to frame integrated multimodal solution strategies for the corridor. Significantly more frequent transit service to Chugiak/Eagle River and new transit service extended to Wasilla could be implemented for less than three percent of the anticipated capital cost for Glenn Highway roadway improvements

59 Technical Memorandum Travel Origin and Destination Patterns In the People Mover Route 7 Corridor Municipality of Anchorage People Mover High Performance Transit Corridor Study October 30, 2007 Ch7-Rte7CorridorTravelPatterns, October 19,

60 Travel Patterns in the Route 7 Corridor This memorandum presents an analysis of travel patterns travel patterns to and from the west side of Anchorage, generally the area served by People Mover Route 7. The travel analysis is part of an assessment of potential opportunities for improving transit service within the Route 7 corridor. The travel patterns data is based on the AMATS travel model for the year Travel is measured in person-trips. Travel origins and destinations are summarized into districts which are clustered groups of traffic analysis zones (TAZs). Four designated district groups were chosen to represent the Route 7 corridor. The district groups are composed of several adjacent TAZ districts, and each TAZ district is in turn composed of a number of TAZs. The path of Route 7, along with the location of each district group in the Anchorage Bowl, is displayed on the map below: Ch7-Rte7CorridorTravelPatterns, October 19,

61 Route 7 Origin & Destinations Four district groups (5, 12, 14 and 15) were selected to represent the corridor along Route 7, with person-trip production-attraction data assembled from the 2007 AMATS TAZbased travel model matrix. Person-trip attractions were tabulated at the district group level, while the productions from the corridor were tabulated at the TAZ district level. These TAZ districts selected for trip productions along the Route 7 corridor correspond directly to the district groups 5, 12, and 15, though make up only a selected portion of district group 14, as highlighted in blue on the map below: Ch7-Rte7CorridorTravelPatterns, October 19,

62 Total person-trips produced in the Route 7 corridor: Ch7-Rte7CorridorTravelPatterns, October 19,

63 Summary Major Destinations Of the total universe of trips predicted by the model for the entire Municipality Of Anchorage, the selected TAZ districts along the Route 7 corridor produced 154,240 daily person-trips. Of this total number of trip productions within the designated Route 7 corridor district groups, 77% (119,164) were home-based travel, and 23% (35,076) of the trips produced were non-home-based travel (defined as trips that have neither origin or destination at the home of the trip-maker). About 40,100 (26%) of the 154,240 daily person-trips produced in the Route 7 corridor s selected TAZ districts are going to attractions (i.e., destinations) within the four Route 7 corridor district groups (5, 12, 14, and 15). Table 1: Breakdown of the total attractions by district group, for all the 154,240 daily person-trips produced within the selected TAZ districts along the Route 7 corridor: Dist Grp Grp Description Total Daily Person-Trips 13 West Dowling/Raspberry % 12 Sand-Jewel Lake % 16 Midtown % 17 Airport % 15 Spenard % 2 Downtown % 9 Dimond/Abbott % 14 Turnagin % 6 Rogers Pk & U-Med % 4 NE Bowl % 10 Hillside % 11 Southpark % 1 Mil Bases & N of Glenn % 8 Dowling Central % 3 Fairview % 7 Scenic Hills % 5 Inner city s of CBD % 20 Girdwood % 18 Eagle River % 19 Chugiak/Birchwood MatSu % Yellow highlights are district groups in the Route 7 corridor. Ch7-Rte7CorridorTravelPatterns, October 19,

64 Trip Productions - Sand Lake/Jewel Lake Daily person-trips produced within district group 12 (Sand Lake/Jewel Lake) total 75,047; 28% of them (20,863) are to attractions within the four Route 7 corridor district groups (5, 12, 14, and 15). Table 2: Breakdown of the total attractions by district group, for the 75,047 daily person-trips produced within the district group 12 (Sand Lake Jewel Lake): Dist Grp Grp Description Total Daily Person-Trips 12 Sand-Jewel Lake % 13 West Dowling/Raspberry % 17 Airport % 9 Dimond/Abbott % 16 Midtown % 10 Hillside % 11 Southpark % 2 Downtown % 6 Rogers Pk & U-Med % 14 Turnagin % 15 Spenard % 8 Dowling Central % 4 NE Bowl % 1 Mil Bases & N of Glenn % 7 Scenic Hills % 3 Fairview % 20 Girdwood % 5 Inner city s of CBD % 18 Eagle River % 19 Chugiak/Birchwood MatSu % Trip Productions - Outer Spenard Of the 33,597 daily person-trips produced within the selected TAZ districts of district group 14 (Outer Spenard), 26% of them (8870) are to attractions within the four Route 7 corridor district groups (5, 12, 14, and 15). Table 3: Breakdown of the total attractions by district group, for the 33,597 daily person-trips produced within the selected TAZ districts of district group 14 (Outer Spenard): Dist Grp Grp Description Total Daily Person-Trips 17 Airport % 13 West Dowling/Raspberry % 14 Turnagin % 16 Midtown % 15 Spenard % 2 Downtown % 6 Rogers Pk & U-Med % 12 Sand-Jewel Lake % Ch7-Rte7CorridorTravelPatterns, October 19,

65 9 Dimond/Abbott % 4 NE Bowl % 1 Mil Bases & N of Glenn % 8 Dowling Central % 3 Fairview % 11 Southpark % 7 Scenic Hills % 10 Hillside % 5 Inner city s of CBD % 20 Girdwood % 18 Eagle River % 19 Chugiak/Birchwood MatSu % Trip Productions - Inner Spenard Of the 32,097 daily person-trips produced within district group 15 (Inner Spenard), 23% of them (7254) are to attractions within the four Route 7 corridor district groups (5, 12, 14, and 15). Table 4: Breakdown of the total attractions by district group, for the 32,097 daily person-trips produced within district group 15 (Inner Spenard): Dist Grp Grp Description Total Daily Person-Trips 16 Midtown % 13 West Dowling/Raspberry % 15 Spenard % 2 Downtown % 17 Airport % 6 Rogers Pk & U-Med % 4 NE Bowl % 14 Turnagin % 1 Mil Bases & N of Glenn % 9 Dimond/Abbott % 12 Sand-Jewel Lake % 3 Fairview % 8 Dowling Central % 7 Scenic Hills % 11 Southpark % 10 Hillside % 5 Inner city s of CBD % 20 Girdwood % 19 Chugiak/Birchwood MatSu % 18 Eagle River % Ch7-Rte7CorridorTravelPatterns, October 19,

66 Trip Productions - North of Chester Creek Of the 13,499 daily person-trips produced within district group 5 (Inner City West), 22% of them (3030) are to attractions within the four Route 7 corridor district groups (5, 12, 14, and 15). Table 5: Breakdown of the total attractions by district group, for the 13,499 daily person-trips produced within district group 5 (Inner City West): Dist Grp Grp Description Total Daily Person-Trips 2 Downtown % 16 Midtown % 13 West Dowling/Raspberry % 5 Inner city s of CBD % 15 Spenard % 1 Mil Bases & N of Glenn % 17 Airport % 4 NE Bowl % 6 Rogers Pk & U-Med % 3 Fairview % 14 Turnagin % 9 Dimond/Abbott % 12 Sand-Jewel Lake % 8 Dowling Central % 11 Southpark % 7 Scenic Hills % 10 Hillside % 20 Girdwood % 18 Eagle River % 19 Chugiak/Birchwood MatSu % Ch7-Rte7CorridorTravelPatterns, October 19,

67 Travel between Home and Worksites Trips which have origin or destination at the home of the trip-maker are termed homebased. For home-based trips, the home is considered the trip production and the non-home end is defined as the trip attraction. So, for example, the home-based work (HBW) trips produced within the Route 7 corridor are trips that start or end at the traveler s home in the corridor and are attracted to worksites. HBW data from district groups 12, 14 and 15 were tabulated. District group 5, north of Chester Creek, was excluded from the HBW analysis. Summary HBW Major Destinations (Attractions): Of the 25,320 daily HBW trips produced from the designated parts of the Route 7 corridor (district groups 12, 14 and 15), only 10.8% of them (2728) are to attractions (i.e. worksites) within the same three district groups. Total HBW attractions by district group: Dist Grp Grp Description Total Daily HBW Person-Trips 17 Airport % 16 Midtown % 13 West Dowling/Raspberry % 2 Downtown % 6 Rogers Pk & U-Med % 1 Mil Bases & N of Glenn % 9 Dimond/Abbott % 4 NE Bowl % 15 Spenard % 7 Scenic Hills % 14 Turnagin % 12 Sand-Jewel Lake % 3 Fairview % 11 Southpark % 8 Dowling Central % 10 Hillside % 18 Eagle River % 20 Girdwood % 5 Inner city s of CBD % 19 Chugiak/Birchwood MatSu % Ch7-Rte7CorridorTravelPatterns, October 19,

68 Technical Memorandum Transit Signal Priority Municipality of Anchorage People Mover High Performance Transit Corridor Study October 4, 2007 Transit Signal Priority Review Oct 4, 2007 Page 1 of 12

69 Introduction Transit Signal Priority (TSP) is a traffic signal control strategy to provide preferential treatment to transit vehicles at signalized intersections. TSP is a type of Intelligent Transportation System (ITS) which helps buses get through intersections more quickly. TSP operational strategy typically involves giving transit vehicles at traffic signals a small amount of extra green light time or a small amount of shorter red light time. The net effect of TSP, also referred to as bus priority, is to reduce the waiting time of transit vehicles at intersections. TSP has been implemented successfully in many cities worldwide, particularly in Europe, and its use is rapidly growing in North America. Dozens of U.S. cities have implemented TSP, including large cites such as Los Angeles, Chicago, Houston, Seattle and Portland, as well as in smaller communities such as St. Cloud, Minnesota and Bremerton, Washington. The purpose of this memo is to make a case, based on evidence elsewhere, that TSP could work for at least some Anchorage intersections where transit buses experience significant delays. ITS America reports (2004) that on average, transit buses spend about 15% of their trip time waiting for traffic signals. Field data observations in Anchorage indicate a higher proportion of time for People Mover buses is absorbed by traffic signal delay. Benefits of Transit Signal Priority (TSP): Reduced bus travel time Improve schedule adherence/reliability Reduced stops which leads to reduced vehicle wear and tear Reduced bus idling and resulting emission/fuel use Improve customer ride quality Consistent on-time performance helps build ridership. Reduction in bus travel time makes transit more competitive with the private automobile as well as potentially reducing operating bus-hours and costs needed to service a particular route. Bus travel time savings on the order of 15% are reported due to the implementation of TSP. However, substantial variability exists in the nature of particular TSP deployments and magnitude of impacts. Potential negative impacts of TSP mostly consist of delays to non-priority traffic, although such side-street delays have proven to be minimal in cities where TSP has been implemented (Smith et. al., 2005). In dozens of cities across the country which have already implemented TSP, only very minor impacts on overall intersection operations have been reported. In Anchorage, relatively few buses traverse a given intersection per hour, so it seems likely that TSP would improve bus service with minimal impact to other traffic. However, that thesis needs to be demonstrated. Improvements in transit travel time with a TSP must not unfairly disrupt the normal traffic operations at the affected intersections. Transit Signal Priority Review Oct 4, 2007 Page 2 of 12

70 Pedestrian activity should be analyzed for each of the intersections along transit routes where TSP is to be deployed, especially the length of time it takes for a pedestrian to safely cross the signalized intersection. Pedestrian safety should not be compromised by the implementation of TSP. As described in the document An Overview of Transit Signal Priority (ITS America, 2004): In most instances, the time required for a pedestrian to safely cross the street at a signalized intersection limits the time available to provide TSP, and can decrease the responsiveness of certain priority treatments (e.g., early green), due to the necessity of waiting for the pedestrian phase to terminate before the priority phase can be activated. Importantly, the pedestrians are often transit customers- hence they require service at the same time as the transit vehicles. Delay for pedestrians should always be considered, as [transit signal] priority can potentially lengthen the maximum waiting time for pedestrians waiting to cross the priority corridor. The impacts of shortening pedestrian walk times should be evaluated on an intersection-by-intersection basis to investigate these impacts and institute appropriate mitigation measures (e.g. establish a maximum red time for pedestrians). A potential problem may arise if the TSP system enables a bus to get ahead of its schedule. Bus trip schedules need to recognize the time-saving effect of TSP and be set to somewhat higher speed, yet maintain on-time performance. This problem could be avoided by driver training and dispatch oversight. A general reduction in bus travel time resulting from a successful TSP implementation along a route should lead to slightly shorter scheduled running time for the route, as well as less schedule variability caused by signal delay. While the traffic signal terms priority and preemption are often used synonymously, in practice they are different processes. Traffic signal priority modifies the normal signal operation process to better accommodate certain (i.e., transit) vehicles. Traffic signal preemption interrupts the normal signal process, and is usually reserved for emergency vehicles (Smith et. al., 2005). As described in the document An Overview of Transit Signal Priority (ITS America, 2004): When a traffic signal is preempted there is no consideration for maintaining the existing timing signal plan such that coordination can be maintained between adjacent traffic signals. Preemption uses a special timing plan, requiring the traffic signal controller to transition out of and back into the coordinated operation of the normal signal timing plan. Transit Signal Priority attempts to provide some priority service opportunities within the coordinated operation of the traffic signal. This allows the objectives of priority to be considered without significantly impacting other traffic. Traffic signal preemption is generally reserved for emergency vehicles to reduce response times. Types of TSP Control Strategies: Traffic signal priority at an intersection can be implemented in a variety of ways. Technical aspects and more detailed descriptions of the various TSP options are described in two reports by the Intelligent Transportation Society of America: Transit Signal Priority Review Oct 4, 2007 Page 3 of 12

71 An Overview of Transit Signal Priority (ITS America, 2004) Priority Treatments Transit Signal Priority (TSP): A Planning and Implementation Handbook (Smith et. al., 2005). Several different types of priority treatments for TSP are available. The most common type is active priority, which selectively detects transit vehicles and communicates this information to the traffic controller. In this treatment, signal priority is only given when the transit vehicle is very near the intersection approach. The main active priority strategies are listed below: Green extension: extends the green light when a TSP-equipped bus is approaching. Early green: expedites the return to a green light when a TSP-equipped bus approaches a red signal, also called red truncation. Actuated transit phases are only displayed when a transit vehicle is detected at the intersection. Such strategies, for example, might insert a left-turn green or be used in conjunctions with a queue-jump. This application may be more advanced than would be practical in Anchorage for the near future. A simplified schematic of the process in executing a signal priority treatment is illustrated in the figure on the following page. Four sequential stages of the priority treatment control are described. The source of the schematic representation is Transit Signal Priority (TSP): A Planning and Implementation Handbook (Smith et. al., 2005) Transit Signal Priority Review Oct 4, 2007 Page 4 of 12

72 Unconditional and Conditional Active Priority There are generally two different approaches to providing active TSP: Unconditional Priority is the simplest approach, providing priority to all transit buses approaching equipped intersections. Conditional Priority is a more sophisticated approach, usually employing GPSbased Automatic Vehicle Location (AVL) systems, which selectively provides priority to transit vehicles based on whether or not the bus is behind schedule. Unconditional priority is the more commonly used type of TSP and is simpler and less expensive to implement than conditional priority systems. Components of a TSP system: Transit Signal Priority Review Oct 4, 2007 Page 5 of 12

73 Diagram copied from An Overview of Transit Signal Priority (ITS America, 2004) As described in the report Transit Signal Priority (TSP): A Planning and Implementation Handbook (Smith et. al., 2005), a TSP system consists of four basic components: 1. Detection: system to deliver data indicating that a bus is approaching a certain distance/time away from the signalized intersection. 2. Priority Request Generator/Server: system that requests priority from the traffic control system and manages multiple requests if needed. 3. Priority Control Strategies: Software enhancement for the traffic control system to implement the signal priority strategies. 4. TSP System Management: integration of both transit and traffic management to configure settings, record events, and provide reporting capabilities. To make the case to municipal decision-makers for funding a TSP system installation for the Municipality Of Anchorage, realistic computer simulations of traffic patterns can help demonstrate its viability. Traffic simulation models provide an opportunity to assess the impact of TSP on transit and non-transit traffic. An estimated cost for implementing TSP on all or some People Mover bus routes in Anchorage is not yet known; however a major cost component is covered in the intersection control device hardware updates which Traffic Engineering is currently installing to handle signal preemption for emergency vehicles. People Mover and the MOA Traffic Department must work closely together to develop implement and evaluation a TSP project. A systems engineering approach is necessary to integrate traffic signal operation and transit priority. For the People Mover system, the MOA Opticom signal controls (described below) can provide for Early Green (Red Transit Signal Priority Review Oct 4, 2007 Page 6 of 12

74 Truncation) and Green Extension, which are the two most commonly used TSP strategies. The Opticom Priority Control System: In Anchorage, implementation of a system-wide traffic signal preemption system for emergency vehicles is in progress. Signal pre-emption for emergency vehicles (fire trucks and ambulances) has been installed at 20% of MOA intersections as of the present time [September 2007], and is expected to be installed on all traffic signals within the MOA by fall of The MOA Traffic Department has chosen the Opticom signal priority/preemption system, which is capable of two modes of operation: High priority: preemption for emergency vehicles (fire department and ambulances, not police). Low priority: signal priority for buses, consisting of the early green (or red truncation), and green extension strategies described above. This low priority mode is subject to a more constrained set of rules, in order to minimize impact Once the Opticom Priority Control System is installed at all MOA signalized intersections for emergency vehicle preemption, the cost of implementing TSP would be much less than if People Mover would be installing the system from scratch. The amount of priority to be granted for buses must be determined by an intersection-byintersection basis. Intersections where there is no bus route would be excluded from TSP implementation. The Opticom Priority Control System was developed by 3M Traffic Management Solutions, a division which was recently sold by 3M to a Toronto-based private investment firm. This group is forming a new Minnesota-based company called Global Traffic Technologies, LLC. Opticom website: ticom/ Opticom system components: Emitter: broadcasts a secure, encoded priority request (using infrared light) to the detector on the intersection signal Detector: installed at the intersection signal, it receives the infrared transmission and relays the request to the phase selector Phase Selector: validates the priority request and provides input to the traffic signal control system. The Opticom TSP provides three options for controlling the emitter on-board each bus: Transit Signal Priority Review Oct 4, 2007 Page 7 of 12

75 Conditional basis (using buses equipped with an AVL system) Unconditional basis (also called continuous operation) Driver discretion (so the individual driver can manually request bus priority when needed) (Diagram copied from Opticom brochure, downloaded from 3M website) People Mover will have installed GPS-based automatic vehicle location (AVL) systems on all buses by March The AVL systems by Mentor Engineering ( ) sends a signal to People Mover central dispatch, showing the location and on-time performance status of all transit buses in the city. The AVL system also displays to the driver whether or not they are running on time. Automatic passenger counters (APC) are installed item on People Mover buses. The AVL and APC units combined provide passenger boarding/alighting data and location referencing for all stops and all bus runs. The Opticom TSP could be coordinated with the Mentor AVL system to provide conditional priority for buses. However, unconditional TSP appears to be the simplest and most practical approach for an initial Anchorage pilot TSP project. Integrating TSP with AVL is more complex, and there appears no immediate requirement for an on-board Opticom TSP emitter to interface with the Mentor AVLs. Spenard Road Corridor The Spenard Road corridor could serve as a pilot project for implementing TSP in Anchorage. The Spenard corridor s high traffic volume and high transit potential (based on census data) make it a favorable location for a pilot project for an initial pilot implementation of TSP technology in Anchorage. A TSP pilot study could also test ontime schedule adherence and possible ridership growth in this high-priority transit corridor. Route 7 would be the main beneficiary of TSP along the Spenard corridor, although parts of the routes 3 and 36 also operate along Spenard Road. The Spenard Road Reconstruction project, proposed by the state and municipal governments, is planned to occur along the busiest section of Spenard, or the same section where TSP would make the most sense ( ). The key intersections for potential TSP deployment along this part of the Route 7 Spenard Road corridor are listed below: Transit Signal Priority Review Oct 4, 2007 Page 8 of 12