Click for next page ( 26

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 25
25 In some applications, data from more than one ITS tech- the system, APC and/or fare card data prevent researchers from nology can be used to address a service delivery question, entering the field "blind," or with incomplete or outdated in- although different levels of customer or system resolution formation. For example, passenger counts by time and location may call for different data or analysis techniques. For exam- provide a contemporary census of the riding population, from ple, the market segmentation question "What is the ridership which sampling plans can be designed and personnel and other to special events?" can be addressed using APC, EFB, mag- resources can be assigned (TriMet Case Study). After survey netic stripe, or smart card data, but the details captured will data are collected, ITS data on actual service attributes can be differ by technology. For this particular question, APCs pro- compared with surveyed perceptions and attitudes to obtain a vide only passenger counts; EFBs further segment boardings more complete picture of customer preferences (CTA Case by fare type; magnetic stripe or unregistered smart cards pro- Study; TriMet Case Study). Finally, completed surveys may vide additional detail on linked trip paths; and registered provide inputs into subsequent ITS data analysis by validating, smart cards provide customers' identities and place of resi- for example, O-D models estimated from fare transactions dence. All of the listed technologies provide data suitable for data (CTA Case Study). ITS data can thus enter at any point in addressing the general research question; however, the depth the cycle of an on-board survey project: from design through of analysis is limited by the level of customer resolution that administration, analysis, and evaluation. each ITS technology is capable of achieving. Thus, even if an Telephone surveys are commonly used to gather attitudi- agency currently has only a partial complement of the tech- nal data from both riders and non-riders. Surveyed percep- nologies just listed, the data still may be suitable for analyzing tions of specific service attributes can be compared with a considerable range of market research questions. actual system performance, as measured by ITS data. Joint Particularly when the object of analysis is a specific area or analysis of survey responses and service delivery data can then set of areas, integration of ITS data with a GIS is a convenient be used to target service improvements with the greatest way to incorporate spatial boundaries. Because ITS data are potential for improving satisfaction and increasing ridership usually available for an entire transit system, the potential (TriMet Case Study). for spatial analysis of user-defined subareas is great. Unlike a Although they do not constitute a scientific sample, cus- traditional survey in which geographic sampling boundaries tomer complaints provide important snapshots of customer must be set in advance, analysis area boundaries can be drawn perceptions of service quality. TriMet and Madison Metro and redrawn as necessary with locationally referenced ITS compare customer complaints with ITS data to "validate" the data. GIS applications using ITS data also allow the incor- complaint (e.g., was a given bus actually speeding, or did it poration of census demographic data or tax-lot data for a actually leave the stop early?), to explain to a customer why defined service area. Finally, the mapping capability of a GIS the event associated with their complaint occurred (e.g., does provides an effective means of communicating analysis and event data indicate that a pass-up occurred because the vehi- information to decisionmakers and stakeholders. cle was overloaded or directed by dispatch to skip stops?), or to better understand how actual operating conditions relate to rider perceptions (e.g., are routes with more complaints ITS Data Applications: Leveraging per thousand boardings also less reliable or subject to more Traditional Market Research crowding?). In addition to applications of ITS data in monitoring service On-street intercept surveys benefit from ITS data in many delivery, the case studies identified examples in which ITS data of the same ways as on-board surveys. Data on passenger are combined with traditional methods and data. Figure 3-2 movements from APCs and electronic fare cards aid in iden- presents a range of applications in which ITS data combines tifying intercept locations (TriMet Case Study). Other ITS with traditional market research methods and data. The appli- data contribute to the analysis of intercept data. ITS data on cations are organized by traditional market research data reliability and passenger volumes can be used to identify sim- collection methods: on-board surveys, telephone surveys, on- ilar locations to use as a "control" when surveying a "treat- street surveys, mail surveys, and focus groups. Following the ment" group's perceptions of service improvements (CTA organization of applications in monitoring service delivery, Case Study). Intercept surveys can also inform or validate the presentation in Figure 3-2 identifies the specific technolo- models estimated from ITS data. For instance, CTA intends gies and data elements or reports necessary to perform the to use intercept surveys to validate its electronic fare card- analysis, and refers to the case study where such analysis has based O-D model. By providing service delivery data cover- been performed. ing an entire transit system, ITS data provide useful inputs for On-board surveys benefit from ITS data in the research de- intercept surveys; at the same time, intercept surveys also sign, survey administration, and response evaluation phases. provide key data inputs to support models estimated from By providing documentation of actual customer flows through ITS data.

OCR for page 25
26 Research ITS Method Technology* ITS Data Application Reference On-board APC Load Summaries From load summaries, determine sampling rates, how many [TriMet Case Study] Surveys by Stop, Route surveys to print, and weighting factors for expanding sample survey responses to population totals APC Load Summaries Survey responses on satisfaction with "overcrowding" are [TriMet Case Study] by Route, Trip compared with passenger load data to identify specific circumstances where high passenger loads are affecting customer satisfaction APC Load Summaries Survey vendors draw on stop and route passenger data to gain a [TriMet Case Study] by Route, Trip, better understand of the dynamics of the system's operating Stop environment APC Load summaries Compare downtown trip estimates from O-D survey with the [TriMet Case Study] by Stop number of APC boardings/alightings downtown AVL Performance Compare route-level reliability indicators (on-time [TriMet Case Study] Reports by Route performance, headway maintenance, excess waiting time) with surveyed satisfaction with "reliability" to assess correspondence APC, MAG, SC Boarding and Inform sampling plan and provide expansion factors for O-D [CTA Case Study] Fare Card Use survey Summaries by Trip, Stop MAG, SC Transaction Calibrate stated preference-based Fare Change Model using [CTA Case Study] Summaries actual card usage data GPS/GIS Geocoded Survey Use geocoding digital pens to record where surveys were [CTA Case Study] Locations administered MAG, SC Transaction University student (Upass Card) ridership related to surveyed [CTA Case Study Summaries by perceptions of safety and security (considering)] Pass Group SC, MAG Card ID, Date & Use fare card data to continuously update survey-based O-D [CTA Case Study] Time, tables (after initial validation using on-board O-D survey totals) Vehicle/Station ID Telephone APC, AVL Load Summaries, Compare surveyed customer satisfaction on specific service [TriMet Case Study] Surveys Performance attributes with actual system performance on specific attributes Reports to identify improvement areas with greatest potential to improve satisfaction APC, AVL Load Summaries, Compare changes in customer perceptions of service attributes [TriMet Case Study] Performance over time with actual performance trends to determine whether Reports actual performance trends correspond to changes in customer satisfaction APC, AVL Load Summaries Given attitudinal market segments determined by surveys, [TriMet Case Study] (+GIS) by Route, Trip, evaluate effectiveness of targeted marketing/service Stop; Service improvement programs by comparing ridership response in Frequency & areas where targeted segments are prevalent versus other areas Coverage MAG Card Usage Employ card usage data to inform sampling plan for perceptions [Madison Case Study Summaries by survey (planned)] Stop, Route SC Card ID Compare surveyed demographics and perceptions of registered [CTA Case Study] and non-registered smart card users with those of other riders/non-riders On-street APC Boarding & Identify stops with similar passenger volumes to use as [CTA Case Study] Surveys Alighting treatment and control groups for survey of perceived reliability Summaries by before and after installation of real time arrival displays Stop AVL Performance Compare surveyed customer perceptions of waiting time and [CTA Case Study] Reports on Stop- reliability with changes in actual reliability indicators before Schweiger (2003) level Reliability and after a change (e.g., installation of real time arrival display) Indicators APC Boarding & Determine the best times to survey and when or where more [TriMet Case Study] Alighting than one surveyor is needed Summaries by Stop by Period MAG, SC Card ID, Date & Use surveys to validate cross-platform transfer rates estimated [CTA Case Study] Time, Station ID from card data Mail APC Load Summaries Use ridership data for household travel survey sampling and [TriMet Case Study Surveys by Route, Trip, expansion factors (planned)] Stop Figure 3-2. Applications of ITS data: leveraging traditional market research methods.

OCR for page 25
27 Research ITS Method Technology* ITS Data Application Reference MAG Card Transaction Replace pass program user surveys with card transaction [Madison Case Study] Summaries by summaries to document pass program ridership in negotiating Pass Group pass program contracts SC Card ID, Validate household travel survey responses with actual recorded [CTA Case Study Transaction Date transit trips from smart cards issued to survey respondents (planned)] & Time, Vehicle/Station ID Focus MAG, SC Card ID and Card Structure fare policy focus groups by card type and use to [CTA Case Study] Groups Usage History; represent different perspectives Customer Contact Information APC, AVL Load Summaries Recruit "rider experience" focus group based on actual service [TriMet Case Study] and Performance delivery data Reports Note: Italicized entries in the ITS data column indicate a need for the ITS data to be processed or analyzed to address a given market research question. APC = Automatic passenger counter AVL = Automatic vehicle location GIS = Geographic information system GPS = Global positioning system MAG = Magnetic stripe card SC = Smart card Figure 3-2. (Continued). The three case study properties had not yet directly com- capture a desired range of customer perspectives. For exam- bined ITS data with mail surveys. However, both CTA and ple, fare policy focus groups may be structured by fare card TriMet plan to incorporate ITS data in the future. ITS data type and use data. could potentially provide sampling and expansion factors for As in the case of service monitoring applications presented travel diary surveys. In addition, if survey participants are in Figure 3-1, the leveraged applications presented in Figure given registered smart cards, trips recorded by the cards could 3-2 do not exhaust the potential of ITS data to support tran- serve as a check on transit trips recorded in travel diaries. The sit market research. As ITS data become more familiar and correspondence between participants' "smart card trips" and accessible to market researchers, additional applications will "diary trips" could be assessed to determine underreporting be developed. Even so, the applications reported by the case rates and provide prompting information to correct diaries study properties provide a framework for pursuing leverag- for missing or misreported trips. ing opportunities. First, ITS data provide inputs about For transit properties with websites, web logs of site visits system activity that facilitate the use of traditional market can provide information on how well information is reach- research methods. Second, since ITS technologies gather ing customers. TriMet tracks how many times important data continuously and comprehensively, they allow surveyed service announcements are viewed. If views are less frequent perceptions and preferences to be compared with actual than expected, additional communication may be needed. service performance. Third, leveraging may also work in the ITS data is useful for focus group research, particularly for reverse direction, with traditional survey techniques provid- selecting participants. Both CTA and TriMet have used fare ing benchmark information for validating models based on card and APC data to target specific rider groups in efforts to ITS data.