May-June 2016: The Bus Rennaissance - Intercity Travel, Bus Rapid Transit, Technology Advances, Rural Services (2016)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

TR N EW S 30 3 M AY –J UN E 20 16 28 The author is Associate Professor, Geography and Planning Department, State University of New York at Albany. Data and information technology are not onlytransforming the way that people use busesbut also the way that bus systems are being planned to serve riders more effectively. Operational data and other information routinely collected by bus systems are being put to work in new ways to assist bus riders and system planners. Bus operators have a long history of collecting data to answer questions from customers and staff. For example, riders want to know when a bus is scheduled to arrive at a stop and what stops the bus will make. In some urban areas, a bus rider can open a mobile device, enter a desired destination, and immediately receive information about what bus to take, when it will arrive, and where. The data and information sys- tems providing this service rely on open data resources—and on bus operators who are willing to provide and maintain reliable, machine-readable data. Although these data applications can provide con- siderable benefits to riders, the effective use of data also can aid bus system planning. When planning their systems, bus operators want to know the fol- lowing: u How many riders typically are on a particular bus, what is the ridership on a particular route, or what is the ridership on the overall system? u Where do the majority of riders get on and off a bus? u How much revenue do the bus services gener- ate, overall and by specific routes and route seg- ments? u What set of stop locations and routes would maximize service levels for customers and boost rev- enues? u Where are the buses during service hours in real time? Transformative Trends in Bus Data A Bright Future Ahead C A T H E R I N E T. L A W S O N The Bus Renaissance A Washington, D.C., Metro employee collects bus arrival and departure data at the Silver Spring Transit Center in Maryland. TRN_303.e$S_TRN_303 7/1/16 11:46 AM Page 28

TR N EW S 303 M AY–JUN E 2016 29 Types of Data To answer these questions, operators traditionally have relied on “paper and pencil” data collection programs. For example, traffic checkers can collect three types of data: u Ride checks: the number of passengers getting on and getting off and the time for each stop; u Point checks: a record of the time, the number of passengers getting on and off, and of passing buses at a particular point; and u Fare checks: an onboard record of each pas- senger’s fare category when boarding at a bus stop. Onboard surveys also are used to query passen- gers about their trip origins and destinations, trip purposes, fare types, and sociodemographics. In addition, supervisors may monitor adherence to schedules by observing bus traffic and may consult the fare payment records for ridership information. When computers were introduced into the bus agency work environment, the staff converted the manually collected data into machine-readable for- mats, making it possible to store and retrieve digital records with ease and efficiency. Manually entering the data, however, was time-consuming. The solu- tion was to “harvest” and archive machine-readable data automatically from the equipment installed on buses for operational purposes. Harvesting Valuable Data A variety of equipment installed on modern buses provides a range of data for analyzing bus operations and planning. Dispatching and Tracking Data Computer-assisted dispatch and automatic vehicle location equipment track buses in real time. The tools improve operational efficiency and support incident management, security response, and restoration of service. The captured and archived data contain a rich set of information that is geocoded—for example, recording the exact position of the bus—and time- stamped. These data serve many purposes, such as comparisons of actual stops with scheduled stop locations to monitor on-time performance, calcula- tions of estimated arrival times, and the dissemina- tion of the estimated times to bus shelter monitors, web displays, and mobile phone services. Other onboard systems generate data from the openings and closings of the doors and from elec- tronic event logs. These data can be useful in ana- lyzing bus services—for instance, by examining data on the deployment of lifts, bus operators can gain a better understanding of how the disabled community uses buses. Similarly, the locational data on door activity can inform the placement of stops on a bus route. Automated Passenger Count Data Automatic passenger counting equipment generates data that can be archived and linked to the vehicle’s location and time-stamp data. Having actual data on how many passengers enter a bus at a particular stop and at a particular time of day can guide the deploy- ment of on-board surveys so that the results can be generalized to a larger population of riders. Addi- tionally, market researchers can use these data to understand passenger behavior at the stop, route, and system levels. The data can serve other planning and adminis- trative purposes as well. For example, the Federal Transit Administration requires transit agencies to submit operations data to the National Transit Data- base. Buses equipped with automatic passenger counters can capture many of the required data ele- ments in much larger quantities than the traditional manual counting methods, improving the quality of the data. Automated Fare Collection Data Automated fare collection techniques, such as smart- cards, not only provide convenience for bus cus- tomers, but also create an opportunity to harvest valuable operations data. Transit agencies often “roll up” the fare collection data to produce monthly or annual revenue figures for financial reporting, but the original transactions data can assist in system planning. Generated at the stop level, fare data can yield passenger flow information, including origin and Minneapolis–St. Paul, Minnesota, Metro Transit police officers check a customer’s fare. Onboard fare checks also function as data collection. A solar-powered kiosk at Metro-North’s Woodlawn station in New York allows customers to access real-time train, subway, and bus arrival information. P H O TO : M ETR O T R A N SIT PH O TO : A A R O N D O N O V A N , M ET R O PO LI TA N TR A N SP O R TA TI O N A U TH O R IT Y TRN_303.e$S_TRN_303 7/1/16 11:46 AM Page 29

TR N EW S 30 3 M AY –J UN E 20 16 30 destination matrices. These data also can inform fare policy strategies and facilitate revenue modeling. For example, bus systems with fare zones can examine patterns in the archived data to inform decisions about fare adjustments. Data Challenges Using these data encounters many challenges—some technical and others related to policy. Ensuring that the data are in a useful geocoded format is an example of a technical challenge. Real- time bus routing and operations require a dynamic spatial approach, not a static map. Geographic infor- mation systems (GIS), traditionally used for making maps and for scoping service-level planning, are not well suited for projecting day-to-day bus operations. The GIS data, however, can be useful in creating and maintaining bus asset inventories, including the loca- tion of bus stops, routes, and time points. Analyzing the rich source of data from smartcards raises concerns about privacy, a policy-related chal- lenge. Bus agencies must balance the need to gather detailed trip information against the need to protect rider privacy. This may require adopting privacy principles to communicate to customers how the data will be used and protected. Ownership of the archived data can also raise issues. In most cases, the equipment and software in buses are purchased from commercial vendors. Bus agencies therefore need an agreement with each ven- dor to spell out the ownership and uses of the archived data. Data Sciences Approach Many bus operators are unable to take advantage of these data sources because of a lack of skills in the work force and a lack of the software to handle large quantities of data, or “big data.” To deal with these resource challenges, data scientists increasingly are using open-source software, and whenever possible, open data. Open-source software has source code that is available for modification or enhancement by any- one. This openness provides opportunities for addi- tional progress by teams with new ideas, as well as feedback on these features and improvements to the original software creators. Similar to open-source software, open data can be used and distributed freely. Data science notably came to the assistance of bus planning in 2005, when the Tri-County Metro- politan Transportation District of Oregon, or Tri- Met, partnered with Google to develop an open data scheduling strategy. The efforts led to the creation of the General Transit Feed Specifications (GTFS), which enable the generation of static schedule infor- mation—such as stop location, route geometrics, and stop times—in a standard format.1 GTFS, com- bined with automatic vehicle location data, produces Open-Source Resources for Bus Systems A Sampling Open Transit Indicators: a web-based, user-friendly application for generating transit service, accessibility, and performance indicators from GTFS, GTFS-R, and GIS census data; https://github.com/WorldBank-Trans- port/open-transit-indicators. TransitWand: a mobile application for collecting GTFS data in the field https://github.com/conveyal/transit-wand. GTFS Editor: a web-based, visually based application for creating and editing a GTFS feed; https://github.com/conveyal/gtfs-editor. Open Trip Planner: a web-based multimodal trip planning tool; https://github.com/openplans/OpenTripPlanner. Open-Trip Planner Analyst: a web-based platform for using GTFS to analyze transit service data; https://github.com/openplans/open tripplanner-analyst. GTFS-RT Admin Tool: a web-based platform for tracking transit service delays; https://github.com/conveyal/gtfs-rt-admin. Sakay: a web and SMS—or short message service—platform for trip planning that supports informal transit routes; https://github.com/thats mydoing/sakay-gateway and https://github.com/ahelpingchip/sakayph. SimpleGTFSCreator: a Python programming script that generates GTFS feeds from GIS files; https://github.com/PatSunter/SimpleGTFS Creator. TransitFeeds: a worldwide directory of GTFS and GTFS-R feeds; http://transitfeeds.com. Source: World Bank, https://github.com/WorldBank-Transport/GTFS-Training- Materials/wiki/Link-repository-for-international-GTFS-training-materials. FIGURE 1 Screen display from app for accessing bus schedule information in Portland, Oregon. (Source: https://trimet.org) 1 See https://developers.google.com/transit/gtfs/. (continued on page 32) TRN_303.e$S_TRN_303 7/1/16 11:46 AM Page 30

TR N EW S 303 M AY–JUN E 2016 31 Microtransit is not an on-demand car and not a scheduledbus but fills a gray area in between. The Kansas City Area Transportation Authority (KCATA) is the first major transit agency to partner with a technology company to employ microtransit as a public transportation option. KCATA is piloting on-demand bus service through a partner- ship with Bridj, a company based in Boston, Massachusetts. For a traditional bus service, riders go to a bus stop located on a prescheduled route, but with the RideKC Bridj microtransit ser- vice, the stop comes to the rider, on the rider’s schedule. Like the ride-hailing apps, the Bridj soft- ware sorts through billions of data points to determine where riders want to go and how to get them there at the times they want to depart and arrive. The Bridj routes are formed with algorithms that sift through transit data, social media, and requests for service on the mobile Bridj app. Riders requesting the service head to convenient “pop-up” stations indicated on the app; these pop-up stops usually are within a 5-minute walk from where the rider lives or works. Getting Started KCATA and Bridj selected Ford Transit vehicles for the project (see photo, below). Ten leased vehicles, built at the automaker’s Kansas City assembly plant, were specially outfitted for the ser- vice. The vehicles have a 14-passenger seating layout and a cus- tom-installed large running board for passenger convenience and safe boarding. Bright logos posted on the vehicles’ sides make for easy visibility. The service operates during weekday rush hours, offering an introductory fare of $1.50, the same as the local bus fare. Bridj rid- ers enjoy fewer stops and do not need to make transfers. The shuttle-like bus is equipped with free Wi-Fi and mobile device chargers. KCATA expects that the smaller vehicles with amenities and flexible scheduling will give riders a feeling of choice and comfort, more like traveling by car than by transit bus. During planning, KCATA and Bridj identified several metro- politan areas with large clusters of jobs, as well as other areas that could benefit from transit service. The planners focused on areas in downtown Kansas City, the neighborhoods east and south of downtown, three regional hospitals, and a slice of the Kansas sub- urbs that had infrequent or inconvenient bus service. The so- called Hospital Hill area, for example, hosts 22,000 jobs, and parts of the Kansas side of the metropolitan area contain more than 10,000 jobs. Making Adjustments By studying the travel request data and ascer- taining where the demand for Bridj service is the highest, planners identified potential new markets and routes. After an influx of travel requests from Kansas City’s River Mar- ket area, for example, the service area was adjusted to include that part of the city. As a result, Bridj is providing reverse com- mutes from downtown Kansas City back to the Kansas side of the state line and to the University of Kansas Medical Center, which has many employees but few parking spots. The dynamic nature of the service has allowed the agency to test and match demand to hours of operations and to areas of the city in ways that were not possible with the more traditional fixed-route bus service. KCATA has used a mix of grassroots marketing and geotar- geted, mobile advertising to reach potential customers. The agency has fielded myriad questions from customers who are unfamiliar with microtransit service. Making Connections Because the microtransit operators must interact more closely with the riders, Bridj has recruited staff from customer service industries. The operator training covers the traditional areas of driving, safety, and customer service but also includes instruction on using mobile devices to connect with passengers booking rides. After the successful launch, KCATA President and CEO Robbie Makinen said that “Bridj is an empowering tool that gives peo- ple one more alternative to driving their personal vehicle. There is no one-size-fits-all transit solution. That’s why KCATA is com- mitted to enhancing its portfolio of services to help people con- nect with all sorts of opportunities.” Information for this article was supplied by Kansas City Transportation Authority, Kansas City, Missouri; for more information, visit www.kcata.org. Kansas City Pilots Microtransit Partnership with Bridj Delivers On-Demand Bus Service The Bus Renaissance Bridj customers can hail buses directly to pop-up locations within a 5-minute walk from their location. The areas served by Bridj pop-up buses in Kansas City, Missouri. TRN_303.e$S_TRN_303 7/1/16 11:46 AM Page 31

TR N EW S 30 3 M AY –J UN E 20 16 32 GTFS-Realtime, or GTFS-R, for an on-screen display of buses traveling in real time. Universal access to GTFS, including the instruc- tions, is available via the Internet. A community of GTFS producers and consumers has demonstrated the value of this approach in developing mobile apps that deliver schedule information to bus riders on their smartphones, tablets, and computers. Figure 1 (page 30) shows the screen display; entering a cur- rent location—which can be detected automatically in some instances—and the desired destination can produce instant instructions for the nearest bus ser- vice. The World Bank provides developing countries with open-source tools and instructions for creating GTFS. Tests show that editing and using this pro- gram does not require training or experience (for additional resources, see sidebar, page 30). Pulling It All Together More advanced bus system analysis and planning tools are under development, including tools for internal bus system planning that leverage the inter- operability of the open-source approach. The new generation of web-based, open-source tools for bus planning and forecasting can combine archived bus operations data with new open data resources, par- ticularly GTFS and the U.S. census. For example, Bus Transit Market Analyst seam- lessly predicts bus ridership by following a sequence of steps using open-source tools; GTFS routes are used to define the market areas. The Census Bureau’s American Community Survey and the Census Trans- portation Planning Products can extract the demo- graphic information to predict bus ridership from census tracts adjacent to bus routes. A mobile app, Open Trip Planner, serves as a A s the testing of driverless cars gains headlines and sparksthe public’s imagination, automation is making inroads into transit bus operations. For example, bus rapid transit (BRT) is a good candidate for applications of lane assist tech- nology. Already widely deployed in passenger cars, lane assist systems can help prevent crashes caused by a driver drifting out of the lane—the device alerts the driver and provides a countersteering force. When the right-of-way for building new lanes for BRT is lim- ited, lane assist technology can allow buses to operate safely in narrow lanes, such as a freeway shoulder. These systems also can help buses pull to within centimeters of the curb, allowing faster loading and unloading of passengers, including riders with special needs. The Minnesota Valley Transit Authority and Lane Transit District in Eugene, Oregon, oper- ate buses equipped with lane assist technology, and the Federal Transit Administration is sup- porting additional research. Crash avoidance systems also promise benefits for transit bus operations. Several transit agencies are working with the Washington State Transit Insurance Pool to test a collision avoidance system. The system uses front- and side-mounted sensors to detect pedestrians, cyclists, or vehicles close by and to alert the driver for timely, corrective action. Although the prospect of fully autonomous transit bus oper- ation is more futuristic, some researchers and developers are offering glimpses into early applications. For example, in National Harbor, Maryland, Local Motors is working with stake- holders to develop and deploy self-driving shuttles for trans- porting passengers at low speeds, to enhance mobility and make connections to other transportation modes. The author is Director of Mobility Systems, Local Motors, Inc., National Harbor, Maryland, and Vice Chair of the TRB Automated Transit Systems Committee. Automation Advancing into Transit Bus Operations Lane Assist, Crash Avoidance—and More M A T T H E W L E S H Olli, the self-driving, low-speed, community shuttle at the Local Motors microfactory in Arizona. Minnesota Valley Transit Authority buses are equipped with lane assist technology. TRN_303_TRN_303 7/7/16 6:37 PM Page 32

TR N EW S 303 M AY–JUN E 2016 33 microsimulation routing engine. Bus riders take microsimulated trips in Open Trip Planner based on the GTFS schedule, as if they were using their smart- phones to navigate their way to work via the bus. Open-source tools that tap automated farebox data validate the outcome of the microsimulation. These demand modeling tools can include cus- tomizable inputs for forecasting future conditions at the regional level, such as increases or decreases in population and employment; at the route level, such as increases or decreases at the census tract level; and at the stop level, such as land use changes within walking distance of the stop. Next Steps Many bus operators are making their services more appealing by providing GTFS for customer-focused apps, but the real leap forward will occur when the bus operators themselves can take full advantage of all the emerging open-source data and programming capabilities available for bus system planning. Several efforts would assist that leap: u Encouraging cooperation by vendors to adopt data format standards—for example, for farebox data—to facilitate data integration; bus operators could specify standardized data and data ownership policies in their procurements; u Training bus agency staff to use the available open-source tools with their own archived data; u Developing new courses and webinars in trans- portation planning to allow staff to gain acquain- tance and expertise with web-based deployments, including the ability to customize code; and u Encouraging the use of data resources by small and medium-sized bus agencies, so that bus opera- tors everywhere can take advantage of the open- source resources to improve their systems, as the World Bank effort to assist bus services in develop- ing countries has demonstrated. Leveraging the Tools The upward trajectory for the adoption of open source and open data is promising, given the success of GTFS. Even if onboard equipment remains pri- marily an environment of retrofitted advancements, and vendors cannot participate in the production of standardized data elements, data science can help to address these issues. As more bus systems join the initiative to provide GTFS for mobile apps that assist bus riders, they will contribute to a rich data resource for their own plan- ning needs. The ability to leverage tools for bus sys- tem planning and forecasting, using GTFS as a foundational data resource, promises a bright future. Additional Reading Acumen Building Enterprise, Inc., and Booz Allen Hamilton, Inc. TCRP Report 115: Smartcard Interoperability Issues for the Transit Industry. Transportation Research Board of the National Academies, Washington, D.C., 2006. Fleishman, D. L., N. Shaw, A. Joshi, R. Freeze, and R. Oram. TCRP Report 10: Fare Policies, Structures, and Technologies. Transportation Research Board, National Research Coun- cil, Washington, D.C., 1996. Furth, P. TCRP Synthesis of Transit Practice 34: Data Analysis for Bus Planning and Monitoring. Transportation Research Board, National Research Council, Washington, D.C. 2000. Hemily, B. The Use of Transit ITS Data for Planning and Man- agement, and Its Challenges; A Discussion Paper. Prepared for ITS America. July 28, 2015. Khani, A., G. Abram, N. R. Juir, and J. Duthie. A Web-based Tool for Visualization of Transit Modeling Results and Data. Presented at the 15th TRB Applications Conference, Atlantic City, N.J., May 20, 2015. Krambeck, H., and L. Qu. Toward an Open Transit Service Data Standard in Developing Asian Countries. Presented at the Transportation Research Board 94th Annual Meeting, Washington, D.C., January 2015. Lawson, C. T. and A. Muro. The Application Programming Interface Advantage: Utilizing Cloud Data Sources for Transit Modeling. Presented at the 2013 GIS in Transit Conference, Washington, D.C., October 16–17, 2013. Lawson C. T., A. Muro, and E. Krans. Integration of Bus Stop Count Data with Census Data for Improving Bus Service and Efficiency. Presented at New Jersey Department of Transportation, Trenton, N.J., June 22, 2015. Schiavone, J. J. TCRP Report 43: Understanding and Applying Advanced On-Board Bus Electronics. Transportation Re - search Board, National Research Council, Washington, D.C., 1999. Strathman, J. G., T. J. Kimpel, J. Broach, P. Wachana, K. Coffel, S. Callas, B. Elliot, and R. Elmore-Yalch. TCRP Report 126: Leveraging ITS Data for Transit Market Research: A Practi- tioner’s Guidebook. Transportation Research Board of the National Academies, Washington, D.C., 2008. Sutton, J. C. TCRP Synthesis of Transit Practice 55: Geographic Information Systems Applications in Transit. Transportation Research Board of the National Academies, Washington, D.C., 2004. P H O TO : S TU A R T S EEG ER. F LIC K R Tri-Met light rail in Portland, Oregon. GTFS developed out of collaboration between Tri-Met and Google. TRN_303.e$S_TRN_303 7/1/16 11:46 AM Page 33