Cover Image

Not for Sale



View/Hide Left Panel
Click for next page ( 80


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 79
79 CHAPTER 13 Conclusions Archived AVL-APC indeed holds great potential for improv- and departure times at stops, identify holding, and deal ing transit management and performance. This report has with multiple apparent stops and starts at bus stops and in reviewed the history and state of the practice in AVL-APC data terminal areas. Having data from both door sensors and collection and analysis. System design affects in many ways the odometers is particularly valuable in this respect. type and quality of data captured, which in turn affects the Integrating AVL with the fare collection system offers a types of analyses that the data can support. To develop guid- potentially powerful means of measuring ridership pat- ance for system design, existing and potential analyses and tools terns, because matching fare media serial numbers offers a that use AVL-APC data to improve management and per- means of observing linked trips. formance are reviewed and their data needs analyzed. Analy- sis tools for running time, waiting time, and crowding were Automatic data collection can revolutionize schedule plan- developed in the course of this project. ning and operations quality monitoring as agencies shift from AVL systems have traditionally been designed primarily for methods constrained by data scarcity to methods that take real-time applications. For investigating specific incidents, advantage of data abundance. The large sample sizes afforded archives of almost any AVL data stream can be used in play- by automatic data collection allow analyses that focus on back mode. However, for analysis of historical data combining extreme values, which matter for schedule planning (e.g., how multiple days of observations, data requirements substantially much running time and recovery time are needed, what head- exceed what will suffice for real-time monitoring. The follow- way is needed to prevent overloads) and service quality mon- ing are the main conclusions regarding how AVL-APC systems itoring (e.g., how long must passengers budget for waiting, should be designed to provide a valuable data archive: how often do they experience overcrowding). Stop-level data recording provides a basis for stop-level scheduling, a practice Storing data on board frees the system from the capacity with potential for improved customer information and better restrictions of transmitting data records over the air. operational control. With AVL-APC data, trends can be found Time-at-location data (i.e., stop and timepoint records) is that might otherwise be hidden, such as operator-specific ten- needed for analyses that aggregate over multiple days of dencies and sources of delay en route. Regularly analyzing AVL observation, such as running time and schedule adherence data gives a transit agency a tool for taking greater control of analysis. Location-at-time (i.e., polling) data is not suitable. its running times by offering a means of detecting causes of Stop-level records permit running time analysis and sched- delay and evaluating the effectiveness of countermeasures. ule making at greater geographic detail than timepoint Two sets of analysis tools were developed as part of this records, better serving passenger information needs and project. One uses running time data to suggest periods of supporting better operational control. homogeneous running times, analyze user-selected running Integrating on-board devices adds information to the data time periods and scheduled running times, and create stop- stream that can be valuable in its own right as well as aid in or timepoint-level schedules. It includes a valuable "what-if " matching captured data to the base map and schedule. The tool that allows schedule planners to propose a scheduled most valuable devices to integrate are door sensors, odome- running time period and running time, and immediately ter (transmission), and radio control head. see how that running time would have performed based Designers should pay attention to a data collection and on the historical data. The tool offered for segment-level run- processing system's ability to accurately determine arrival ning times uses a statistical approach that, if combined with

OCR for page 79
80 operational control in the form of holding early trips, has presented that prevents not only negative departing load, but great potential to improve on-time performance. These run- also negative through load, a stronger feasibility criterion. ning time tools are part of the software package TriTAPT, Approaches and data structures are described for dealing developed at the Delft University of Technology, and are avail- with end-of-line passenger attribution, especially on routes able with no license fee to U.S. and Canadian transit agencies ending in loops and on interlined routes in which passengers through the end of 2009. can be inherited from one trip to the next. On routes ending New tools were also developed on a spreadsheet platform with short loops (short enough that it can be assumed that no to evaluate waiting time and crowding from the customer's passenger trips both begin and end in the loop), a parsing and perspective using AVL and APC data. Unlike traditional meth- balancing method was developed that makes the loop effec- ods, they focus on the extreme events (e.g., very early and late tively serve as a zero-load point. On routes lacking a natural buses, very long headways, very crowded buses) that most zero-load point, such as downtown circulators or routes with affect customer satisfaction. A whole new framework was loops on both ends, APC systems may need operator input to developed for evaluating passenger waiting time, one that fix the load at a key point on each round trip. gives attention to the time that passengers have to budget for The typical fleet penetration rate (10% to 15%) for APCs waiting, not just the time they actually spend waiting. Three is shown to be adequate for all passenger count applications new measures of waiting time are proposed: budgeted waiting except for monitoring extreme crowding. The sample sizes time, potential waiting time, and equivalent waiting time, the typically afforded by APCs are shown to be sufficient to sat- latter being a comprehensive summary of passengers' waiting isfy NTD requirements for passenger-miles reporting. Statisti- cost. This framework is superior to traditional measures of cal requirements on systematic error in load or passenger-miles waiting time because it accounts for the impact of service measurements to meet NTD reporting requirements are unreliability on passenger waiting time. also given. The new crowding measures developed are defined from Analysis procedures using AVL-APC data should account the passenger rather than vehicle perspective. With APC data, for variable sampling rates by aggregating and weighting the percentage of passengers at the maximum load point who observations based on the schedule. Simply aggregating over sit and who stand can be inferred, and they can be further all the stop or timepoint records for a chosen route, period, divided into those who do and do not sit next to an unoccu- and date range can bias results in favor of trips that were pied seat, and those standing at various levels of crowding. measured more frequently. The result is a distribution of passengers by crowding experi- While some analyses involve simple aggregation over a ence, something that should correlate well with passenger selected set of records, others require that analysis follow a complaints and with service objectives. particular sequence of stops. For such an analysis to involve Routine use of automatic passenger counts poses a special data from multiple patterns operating on a common trunk, a challenge because imperfect counting accuracy requires trip- special data structure is required to align the patterns. Such a level parsing and balancing both for consistency and to avoid data structure, called the "virtual route," was developed in Tri- drift errors. Accuracy measures are described and analyzed. TAPT; it permits analysts to see, for example, headway analy- The researchers show, both from empirical data and from log- sis and passenger load profiles on trunks shared by multiple ical considerations, that accuracy in measured load (and, by routes and patterns. extension, in passenger-miles) can be significantly worse than The development of AVL-APC data collection and analysis accuracy in on-off counts; therefore, load accuracy should be capability poses numerous organization challenges. Perhaps an important consideration in system specification, design, the greatest is raising awareness within the organization of the and testing. value of archived AVL-APC data in order to ensure that AVL Shortcomings in several existing methods of load balancing systems, which are often procured to serve real-time applica- are pointed out; some of them can bias load and passenger- tions, have a design and the database support needed to miles estimates upwards. A method of trip-level balancing is achieve their potential for archived data analysis.