National Academies Press: OpenBook
« Previous: 1. Introduction
Page 5
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 5
Page 6
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 6
Page 7
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 7
Page 8
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 8
Page 9
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 9
Page 10
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 10
Page 11
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 11
Page 12
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 12
Page 13
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 13
Page 14
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 14
Page 15
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 15
Page 16
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 16
Page 17
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 17
Page 18
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 18
Page 19
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 19
Page 20
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 20
Page 21
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 21
Page 22
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 22
Page 23
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 23
Page 24
Suggested Citation:"2. Data Collection Methods." Transportation Research Board. 1997. Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources. Washington, DC: The National Academies Press. doi: 10.17226/6341.
×
Page 24

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.

2.0 Data Collection Methods 2.1 Surveys Within the traditional four step travel forecasting framework the major impetus for conducting surveys has been to supply the data needs of four basic models: 1) trip generation, 2) trip distribution, 3) Nodal split, and 4) network assignment. The passage of the 1990 CAAA and ISTEA initiated the process of altering these models to accommodate the new regulatory requirements. Changes and enhancements to the models precipitated an obvious need For reevaluating the surveys which support these models. - The basic types of traditional travel surveys includes: household travel surreys, workplace surveys, stated preference surveys, longitudinal and panel surveys, transit on-board ridership Evens, commercial vehicle (truck) surveys, and extemal station surveys. Household Travel Surveys-Household surveys provide data on the characteristics of families and individuals, as well as their travel movements, mode choice, and time of travel. Due to the extent of infonnaticn Cat can tee gathered Tom a househo;~.survey, !nany of-~-infomadon reseeds prompted by the 1990 CAAA and ISTEA win result.in an enhancement and/or alteration of the tra&tior~ household survey. For this reason, the household survey demands additional attention when examining He changes in travel related Sundays. The characteristics and uses of household surveys are numerous. Both telephone collection and mail- out/mail-back surveys can be used. Small sample sizes are generally acceptable for calibrating regional trip generation and trip distnbudon models. Larger sample sizes are needed to adequately calibrate "Short-Term Travel Model Improvements. Travel Mode! Improvement Program (DOT, FlIWA, FTA, EPA), October 1994. - Publ.# DOT-T-95~5. 5.

mode choice models. Household survey results have also been traditionally used to calibrate trip attraction models. The evolution to multimodal planning has created specific information needs that can be addressed by new or revised survey processes. For instance, ISTEA requires analysis of ad modes so the collection of travel data by aD modes, including non-motorized modes, is necessary. The collection of information on specific vehicle-type trips (e.g., auto, van, or pick-up) by trip type is needed for air quality planning, as is intonation on the physical characteristics of vehicles such as make, model, fuel type, odometer-reading. Likewise, to meet the level of detail regarding emissions and dispersion of pollutants, air quality models need to be supplied with travel estimates by facility, by vehicle type, by hour of the day, and by vehicle operating mode (e.g. cold start). This type of information could possibly show a correlation between the type of hip and the type of vehicle typically used. Furthermore, household travel surveys have typically focused on trips made by household members. Modelers have advocated changing the focus of household surveys from- surveys of trips to surveys of activities of household members. The Ho pr~napal reasons for the shift to activity-based modeling are described below. A Cinp,'is an abstract term describing movement Dom one point to another. It is not always wed understood by He population 'Gregg surveyed and can lead to unreported tnps. People recall their daily activities much better than the actual trips ~emseives. vity-based surveys allow more ~nfor~r~ion to be gathered on the reasons for trip making. order to properly understand and mode] the ejects of the changing transportation supply and socioeconomic pressures on travel, we need to understand the activities being performed and the decision process a household uses in determining He activities that are performed during a day. Work Place Surveys-Work place surveys have been used to gather detailed information, including the attraction purposes (e.g., visitor, customer, employee) of trips at the location attracting those trips. The surveys provide disaggregate data that can be used to estimate trip attraction rates. Their primary use has been to support the calibration of trip attraction models. Information such as parking cost and 6

walk distance can also be use in other models such as the mode choice model. Although woric place surveys potentially can provide detailed trip attraction data, they are expensive and difficult to peon. S=ed-Preference Surveys-In a stated-preference survey, each respondent is asked to make a travel decision for a scenario which describes the available aItematives and their characteristics. They represent an attempted increase in the volume and variety of data that has been traditionally taken from household and other surveys (revealed preference data). Stated-preference surveys allow respondents to respond to alternative scenarios, resulting in more useable data for estimating travel behavior and charactenshcs. They have mditionady been used In long~istance Eves demand modeling. Data collected through stated-preference Coveys, however, do not reflect actual travel behavior, rather, people respond in a manner that characterizes how they would prefer to behave. In addition, respondents are provided with more information than is available to a typical traveler. Therefore, models operating off of these data frequently underestimate He level of uncertainty present under actual conditions. Stated-preference surveys have recently been proposed for urban area travel modeling (e.g., by He . . . Oregon and New Hampshire DOTs). There.has also been discussion on how to best combine revealed and stated-preference data for model development. l~ongitudir~al or Panel Surveys-Longitudinal and pane! surveys are characterized by a sample of . . . . households Hat are surveyed over time (2-3 year intervals) to determine changes In travel behavior of the same individual households under different socioeconomic and transportation supply con&lions. Aspects of travel behavior that are not realized under typical snapshot household surveys, but are potential outputs of a panel study, include understanding the process of information acquisition, emergence and learning, and behavioral turnover. Trait On Board MOBS Surreys-Traditionally used by transit operators to gain an understanding of ridership profiles. TOB surveys have also been used by travel demand modelers to develop trip tables for travel model validation and to enhance household survey data for development of mode 7

. . . choice models. TOB surveys are typically self-administered and short enough to complete during the transit run. Results of on-board surveys have been combined with the results of household travel surveys to develop 'thoice-based"calibration data files for mode choice model estimation. Eternal Station Surveys-External station surveys are used to provide information for trips traveling into, out of, and through a region. Survey techniques include roadside interviews, postcard handout/mailback surveys, and license plate recording/survey mailing. The recording of license plates, matching numbers with vehicle registration, and mailing a survey fomm has uncovered a host of privacy issues that need to be addressed ~ ture survey administration. . ~ Recent changes and developments associated with the design and execution of external surveys are described below. . The method and technology of roadside collection has been enhanced by some regions via microcomputer-based data entry procedures. Through the use of dBASE {V programming and Rem Tune Data Entry, automatic geocod~ng can be achieved. Surveys conducted in large transportation centers (e.g., bus stations, airports, and railroad stations) can provide infommion similar to that found In an extemal station survey, but Bom an individual, rather than vehicle, standpoint. Information on passenger movements, Rome, mode choice, etc. can be obtained via these surveys. Experimentation at these large transportation centers with computer-administered (CA) surveys completed by self-selected participants have revealed some of the advantages and disadvantages of such a system.2 The surveys are initiated via touchscreen computers placed in a common area kiosk One advantage of the CA survey is that, if properly designed, it can miIii~riize the Bequengy of respondent errors. For example, only certain questions should be answered by particular respondents, depending on their purpose at Me location In He paper- and-penci} survey, the respondent has to skip several questions, whereas in die CA survey He proper questions are automatically displayed. Another advantage is that the data from He CA survey can automatically be input into useable fees, while the tractional surveys must be transcribed or scanned into computer files. Additionally' the use of touch screen technology enables maps to be &splayed and allows respondents to choose specific locations from the maps for their responses to origin/destination questions. This significantly advances He use and precision of geocoding (discussed below) as part of the surrey process. Finally, CA surveys can be continuous and provide real time data and larger data sets than the papers d 2Papacostas. C.S. et aL "Computer-Administered Surveys at Honolulu International Airport', Innovations in Travel Su~vev Methods. Transportation Research Record No. 1412, Transportation Research Board, 1993. 8

. -: ~, ·:-:-: . : . :-;~;!;:;:;:;.;~';!;~;,;. ~ pencil surveys. This allows data to be collected during time periods when supervised surveyors may be costly or difficult to administer and avoids the problems associated with one time surveys (e.g., abnormal conditions). Computer-administered surveys do not eliminate the need for facilitator-supervised surveys, however, because studies have shown that the there is a statistical difference between the type of responses received Mom CA surveys and supervised surveys. To provide control data, as wed as survey those that do not typically respond to a touch-screen system, traditional surveys would be needed in some capacity. A hyena surrey technique for determining Hip characteristics, which was tested at a shopping center in 1993, was able to increase the questionnaire response rate Rom ~ ~ percent to nearly 30 percents The-~vey utilized techniques Dom both roadside mterv~ews and postcard handout/mailback~ irvey postcards were distributed- on parked cars throughout the shopping center. As We Divers eaten Me center Me surveys were collected by members of the survey . team. Additionally, monetary incentive was provide by promising respondents a $1.00 payment for returning a complied surveyor. Commercial Ve~dcie Surveys Traditionally, commercial vehicle surveys were used to collect information on Duck trips made In a region. Because of the confidendaDy associated with some commercial bucking information; the difficulty in detennining the population to be surveyed, and We incomplete data provided by truck registrations (due to out-of-state tops), few comprehensive truck surveys have been conducted In recent years. Recent changes and developments assoc~a ted Wig! We design and execution of commercial vehicle surreys are described below. : Similar to the movement of the household survey towards acEvity-based modeling, work has been conducted which focuses on conunodity movements, rawer Man truck trips. Studies have recently been conducted using hand held computers, caped Persor~ Digital Assistants (PDA's), as truck survey instruments rather than the standard paper-and-pencil ~vey.4 The Street Smart Company (Zulus, GA) was contracted by the Federal Highway Administration to conduct ong~n~estination and commodity surveys using the PDA technology. Several advantages were identified over the course of Tree case studies: I) 3Landis, Bruce W. Improved Sampling Techniques to Determine Inp Characteristics for TraBic Impact Analyses", Transportation Research Record 1400, !993. 4Lau, Samuel. Truck Travel Su~vevs: A Review of the Literature and State-of-the-Art, Metropolitan Transportation Commission, Oakland, CA, January 1995 9, i

b;~ improved accuracy of data because of the ease with which data could be input and quickly reviewed, 2) collection time is reduced primarily due to the elimination of recentering data before processing, and 3) cost of data collection is reduced by removing data entry and transfer, as well as eliminating erasers, pencils, clipboards, and the printing of new forms. Some other types of surveys that may not be applicable to all transportation planning agencies include visitor surveys and parking surveys5. Visitor surveys are sometimes utilized in areas where visitors contribute significantly to the amount of travel. Many large metropolitan areas and even relatively small areas may host thousands of visitors on a daily basis due to attractions such as businesses, convention centers' sporting complexes, amusement parks, etc. The surveys are typically designed to obtain information about the characteristics of the non-residents who are staying in the area (e.g., hotels, motels, bed-and-breakfast), as well as the number and type of trips being taken. The data collected can be used in several ways. Trip generation estimates such as Hips generated per occupied hotel room can be estimated. The data can also be used to estimate the visitor demand for possible new travel modes or services (e.g., added bus service, people movers). In addition, estimates of increased trip generation by visitors due to new development or We addition of a major visitor attraction can also be made. The two primly methods of conducting visitor surveys involve distributing self~ompletion surveys and in-person intercept interviews, typically in the hotel lobby. Recent advances in hand-held computers have allowed some interviewers to conduct computer-assisted personal interviews (CAP0. The advantages of CAPI surreys are: The hand-held computers can be programmed to accept valid entrees which would reduce field data entry errors. The storage capacitor allows previous entrees to be checked to avoid inconsistencies wad over entries. The computer automatically guides the interviewer Trough the questions so none are missed or asked out of order. 5Travel Survey Manual (Draft). Prepared by Cambridge Systematics for the Travel Model Improvement Program 0)OT, FlIWA, FTA, EPA), Track D, 1995. 10

; ~; ~; ~ ~ ; -; -; 2 ~; ~ The interviewer can use visual information on the computer screen to better communicate With respondents. Disadvantages of the CAPI system include: The computer program for the survey requires a significant amount of time and effort to ensure precision because program corrections are usually not feasible once the interviewers have gone into the field. Interviewers must be skilled in using the CAPI system. · The lack of hardcopy records places greater reliance on Me info~Tnation entered. Parking Surveys have evolved from being used exclusively to project parking supply, demand, utilization, and turnover to using the data to provide trip generation figures associated with a particular parking area. Pricing information collected during a parking survey can also assist wig understanding and predicting price elasticity of paring costs and the effects on travel behavior. Three methods of collecting parking ~rey.data incillde: · Interviewing duvers as they enter or exit a parldng facility. · Placing mail-back questionnaires on the windshields of parked cars. . Matching parked car license plates with addresses Tom DMV files and mailing out surveys. Payback surveys tend to be less costly, but the response rate is generally much lower than the interview method. In addition, Me accuracy of the cost and dumbon of the paring stay are more accurate with the interview method. Possible References The foDo~nng references may assist transportation planners in designing and evaluating various survey methods. Lau, Samuel. Truck Travel Surveys: A Review of the Literature and State-of-the-Art., Metropolitan Transportation Commission, Oakland, CA, January 1995. Transportation Research Board. Innovations in Travel Survey Methods. Transportation Research Record No. 1412,Washington,D.C. 1993. Travel Survey Manual (Draft). Prepared by Cambridge Systematics for the Travel Mode] Improvement Program (DOT, FHWA, FTA, EPA), Track D, 1995. 11

: :~: ~:. :; .;~;~;. ~.;. ~. .;- a;;; ::;:;; ';:;:; :~:: 2.2 Primary Collection Methods - Travel Monitoring Aside from surveys, another way to support the existing and new generation of travel models and their associated data needs is via traditional traffic mon~tonng programs and the application of advanced technologies. The following section describes some of the new and emerging technologies associated with the collection of travel monitoring data (e.g., speed, travel time, vehicle occupancy, and vehicle classification and counts). Where possible, associated case studies have been provided wad contact names and phone numbers. Traditional Traffic Count Techniques Traffic count data needs encompass coverage counts, Long-Tem~ Pavement Performance KNAPP) counts, project related counts, special count requests, and data obsolescence counts. Three basic types of traffic counting equipment used in collecting the traffic data include traffic. volume, vehicle classifiers, and we~gh-'n-modon.6 Traditio'2al framed volume counters - As with all three types of equipment, traffic volume counters may be either portable or permanent in design. They utilize a single awe sensor and may include time - period or cumulative counts recorded or. punch tape, printed paper, or electrc;~irvally. The perl~nanent traffic counters use ~ single inductive loop that recognizes passing vehicles.and records the data on vehicle lengths and speed for a given time period. Data can be retrieved Trough periodic collection of paper tape, downloading the data to computer, or, in Me newer systems' telecommunication technology. Vehicle clasQfication record - The portable recorders use t~ro-a~e sensors, such as road tubes, tape switches, tape-down piezo~ectric film or piezo-cable to sort vehicles into thirteen FEWA established categories. The permanent designs utilize two ~n-pavement inductive loops, two piezo awe 6Guidelines for Traffic Data Programs. American Association of State Highway and Transportation Officials (AASHTO), Washington, DC, 1992. 12 r ;

sensors, or combinations of two loops and one age sensors, or two axle sensors and one loop. Data recorded can include date and time, axle spacing, number of axles per vehicle, and speeds. Weigh-in-mofion ~ - Both portable and permanent WIM equipment can collect data on date and time, vehicle lengths, speed, and age weights and spacing. The portable WIM uses a combination of two loops and a capacitance weigh pad to collect data. Units are available that require no personnel on-site. The permanent model uses a combination of one or two inductive loops and one or more age weight sensors for data collection. As with volume counters, the data can potentially be retrieved through manual coDechon or telecommunications, depending on the sophistication of the equipment. ITS Rotated Technology Intelligent Transportation Systems gTS) refers to a broad range of systems that win use sophisticated microcomputer and communications technology to monitor, guide, or control operation of vehicles and provide travelers with information about highway and travel condition~s.7 8 Due to the data focus of this analysis, the concentration win be on the monitoring capabilities of fRs technology. ITS can potentially provide real time travel data to both customers and traffic managers. Supplying current tragic infonnation to travelers could gready improve their transportation decision-making, aDowing travelers to alter Coheir routes or modes accordingly Additionally, it allows traffic n`~agemer~t decisions to be made immediately following Me identification of problems. It is important to note that the technologies described below do not encompass We universe of ad ITS- related technologies. There are numerous ITS technologies that relate to traffic management, improved communications, data management, etc. The technologies described below are Rose that, among other possible functions, have the ability to collect pr~rnary data usefid to transportation planners. 7 'Transit and IRAQIS in the New Jersey/New York Region", ITE Journal. December 1992. 8Pisarski, Alan E. "Appendix B - New Technologies for Transportation Data Collection and Analysis: Opportunities and Applications, Data for Decisions: Requirements for National Transportation Police Making. Transportation Research Board, 1992.

· .~: i.:-. :~:~-~:~-:-:~: ~ : Electron~c ticketing and automated trip payment- This technology allows patrons to use ticket machines to select and purchase (e.g., credit card, cash) tickets, while automated trip payment machines allow patrons to quickly access a transportation vehicle or system (e.g., Metro Card access to subway) without engaging fare collecting personnel. These machines can provide a host of information on readership characteristics such as: total number of riders by hip and line; total revenue collected by trip and line; fare medium used-~~ash, monthly pass, transfer, etc.; and zonal ridership by patrons. This information, which in most cases is collected off-line and downloaded daily to a primary database, could eventually be collected on-line and provide real time data directly to the traffic management center. Case Study - The city of Phoenix is allowing bus passengers to pay by major credit card and a special BusCard Plus Posit System credit card. The system has been able to reduce bottlenecks when boarding, as wed as collect readership data, including passenger count. : . . Vehicle operations and: communications - The prirriary purpose of this technology is to provide constant communication between the transit operators in the field and the supervisors and management staff in charge of dispatching, routing, and providing information to riders. In 1992, New Jersey Transit, for example, installed an 800 megahertz bus radio system, supported by a statewide microwave network, which allows virtually uninterrupted contact between the entire 1800+ bus Beet, 220+ supervisory personnel, and 11 transit police units. Up-to-date information on delays can be relayed to the control centers where routing and scheduling can be adjusted. The New Jersey Transit communication system is also equipped with automatic vehicle location (AVL) capabilities. Using signpost monitors and on-board odometers, AVL allows the position of buses to be determined approximately even ten minutes. 14 . . . . . . .

:; ~: .-::: : :~- : -. . :: : ~ :~ :~:~- Case Study - Minnesota Guidestar's Travlink project goes a step further than the work done by the new Jersey Transit by using a global positioning system to pinpoint bus locations for dispatch and then providing that information to commuters. Electronic signs at park-and-rides, as well as touch screen kiosks provide real time information on the buses status. The kiosks can also be used to pant schedules, plan routes, and receive information on other travel related items. Automatic toR collection - An automatic toll collection system's main function is not primary data collection, but rather serves as a congestion management tool. However, the technology, which allows vehicles to be electronically recorded as they pass through toD plazas, can be utilized for .pr~mary data collection as wed. The New Jersey Transit's automatic tog collection system, caged Electronic ToD and Traffic Management (ETTM), equips aD New Jersey Transit buses with transponders that interface with equipment at the ton plaza and record the tog. The Mass Transit Authority (MTA) in New York/New Jersey has been trying to implement 'electronic ton tags" on public and private vehicles for automatic ton -collection. This technology, in coordination with roadside antennas, can eventually be utilized as a vehicle locator system (similar to New Jersey Transit's system) to relay real time data to the control center. This same type of technology can also be applied to train systems as wed, to locate individual trains, relay congestion information to control centers, and lessen delays. Case Study - Electronic ton collection projects in Oklahoma have almost eliminated vehicle accidents in die toil lanes that have been automated and reduced emissions by 25- to 70 percent. The resulting savings due to the new toll system has been estimated at $160,000 per lane.9 Video Imlzging Technolo'tyl°-Since the early 1990's, video imaging technology has made some serious strides towards becoming a widely accepted traffic mon~tonug tool. These 'smart" cameras 9Traveling with Success: How Local Governments Use Intelligent Transportation Systems. Public Technology, Inc., Washington, DC, p. 21, 1995. Funded under a cooperative agreement with the U.S. Department of T~an~ortation, Federal Highu ay Administration. '°Kyte, Michael et al. 'Using Machine Vision (Video Imaging) Technology To Collect Transportation Data". Transportation Research Record 1412: Innovations in Travel Survey Methods. Transportation Research Board, 1993. 15 ..- me.

use a video-imaging detection system which can distinguish between normal rush-hour congestion and unusual incidents, as well as record traffic volumes, speeds, and vehicle classifications. Several video imaging technologies have been developed in recent years. For instance, the California Department of Transportation has identified eight systems which have the potential to analyze traffic data. These include the following: 1) Aspex Traffic Analysis System (ATAS); 2) Camera and Computer Aided Traffic Sensor (CCATS) developed by Devlonics in Belgium; 3) Sirgru, developed by Eliop in Spain; 4) Traffic Analysis System (TAS); 5) Titan developed by the "trench Institut National de Recherche sur tee Transports et tour Securite; 63 Traffic Tracker, 7) Tulip; and 8) the Autoscope which was developed by the University of Minnesota and hnage Sensing Systems, Inc. The Autoscope has been utilized in a recent study In Boise, Idaho and is further detailed below as an illustration of this technology. The Autoscope system is actually a combination of video cameras and computer processing equipment. hnages are transmitted from a series of cameras to a personal computer which digitizes and stores each image at a rate of 30 frames per second. The information is then sent to a second computer, the Autoscope, which actuary processes the information. The Autoscope recognizes the video content of the stationary background image and identifies any changes in the luminance as potential vehicle movements. The Autoscope can also measure vehicle lengths and heights at ports of entry. It is the eventual aim that the Autoscope's information be used for Dings like automatically adjusting the rep meter traffic signals associated with the monitored roadway o. for immediately contacting emergency personnel when an incident is detected. Case Study - The Idaho study, referenced above, concluded that the racy of the Autoscope system can come very close to that of manual field collection if proper placement of the cameras can be achieved and if the cameras have an unobstructed view of the vehicles. The study also concluded that the presence of pedestrians can significantly reduce the quality of the results, making the present technology more suitable for highway analysis. Although much of the same data can be provided by currently used inductive loop systems, the video imaging technology provides some advantages. For one, the delays caused by the maintenance of in 16

: By: pavement loop detectors can be avoided because of the roadside nature of the video equipment. In addition, the visual infonnation can allow tragic managers to assess more expediently the traffic or emergency situation and can improve both the quality and speed of the mitigating action(s). Closed circuit Gzmeras~1 - Closed circuit cameras relay pictures back to a control center where accidents and congestion can be identified and corrective measures can be implemented much more readily than waiting for calls from the scene from emergency personnel, passing motorists, or traffic surveillance (e.g:' traffic helicopter). Case Studio - As off 1992, the Virginia Department of Transportation ~30T3 liati 48 c~sed-circu~t cameras strategically placed along the busiest roadway as part of its traffic management system (I=;). VDOT has also experimented with cameras mounted on Fairfax County aircraft to get an overall picture of traffic flow, incidents, and congestion. Future VDOT plans include the testing of the Autoscope camera license Plate Mashing ~`te@)12 - TICS technology involves the use of at least two video cameras aimed and/or placed at two separate sections along a road segment. The images of the license plates of the downstream vehicles are matched with those of the upstream images to determine the travel time and, therefore, speed of the vehicles passing that segment. One camera is usually required per lane. The plate numbers on the video unages can De matched either manually, which is labor intensive, or through a machune vision system which can automatically read the plate images. The machine vision technology is capable of capturing 50 percent of vehicles passing at speeds of up to 1 00 MPH One of He mayor benefits of He technology is that there is no interference with the moving Panic during He data coDecdon and traffic is not disrupted to set up He equipment because noting has to be placed on or below the pavement surface. ~'~irginia's Traffic Mi~nagemene System"ITE Journal. July 1999. Travel Time Data Collection: Field Tests-Lessons Learned. Prepared for the Office of Highway Information, Federal Highway Administration by Volpe National Transportation Systems Center, DOT-VNTS- FHWA-96-1, January 1996. 17

:i: ;:;~:::~:~ :;;;;-::; One of the advantages of collecting this type of data on video images is that it is possible to review the tape to extract other critical statistics such as volume, lane occupancy, vehicle classification, headway, etc. Case Study - The University of Flonda's Center for Urban Transportation Research conducted a field demonstration sway in 1995 to evaluate the feasibility of using v~deo-based technology for data collections. The study was conducted in Hillsborough County, Florida as a data collection tool for the count:y~s Congestion Management System (CMS). The technology was focused on the collection of Gavel time, on~-destination, and average vehicle occupancy data. 'Over the three~ay ~M. pealc-per~od evaluation (six total hours of traffic perfonnance mon'=r~ng), this field demonstration found that v~deo-based Marc data collection compared to manuaDy~odected traffic data resulted In a total of 2,746 (almost &50 percent) more useable observations, each requiring about seven minutes less time to collect and process, at a cost of only 50 cents per unit more. The more expanded, real-6me sampling capabilities of video- basest collection also facilitated the creation of more mearungfill traffic performance measures (e.g., 15-minute volume versus average vehicle occupancy, 15-minute volume versus average speed, and total person trips) at specific points or by movement within Me transportation system" Automated Vehicle ~&en6fication (A~4~\S _ AVl technology refers to technology which enables vehicles to be uniquely identified as they pass a specific point without any action from the driver and without any second party observer. The technology basically incorporates three components: 1) a transponder or electronic tag mounted on the vehicle, 2) a roadside reader unit, and 3) a computer system for storage and data processing. Information can be collected regardless of lighting con&lions and Ravel speeds, which adversely affects some other new technology (e.g., video ~ ng). The i3Demonstration of Video Based Technology for Automation of Traffic Data Collection. Prepared for the Hilldborough County Metropolitan Planning Organization by He Center for Urban Transportation Research, Universitr of South Florida, January 1996. ~4"AV! Monitors Traffic and Reduces Congestion in Houston". Congestion Management News Volume 2), i 1 b Journal, September, 1995. 1SAssessment of Advanced Technologies for Relieving Urban Traffic Congestion. National Cooperative Highway Research Program Report #340, TRB, December 1991. 18

technology provides current travel time information which can improve response time and relief strategies. Most AV! systems also avoid any in-pavement maintenance problems typically associated with other sensory equipment (e.g., inductive loops). The four main categories of AVI systems include: optical systems, infrared systems, inductive loop AVl systems, and radio frequency and microwave systems. Case Study - In 1994, the Texas Department of Transportation (TxDOT) completed the first phase of installing AVI technology over 50 miles of an eventual 220 miles of freeway and HOV lanes. TxDOT has equipped over 40,000 vehicles with transponders, including all buses using designated HOV lanes. The standard transponders and tags can also be used for automated ton collection (as described above), either deducting toss from a prepaid account or initiating a biding process. More advanced forms of transponders, tags, and readers could also be used to provide two-way communication on traffic conditions. Vehicles equipped with two-way communication devices could potentially receive current information on traffic conditions or route change information which could speed travel time and reduce congestion. Other possible applications of AVI technology beyond current travel time information include HOV lane enforcement, congestion pricing (e.g., charging users according ~ frequency and time of use), transit for special events, traffic signal preemptions, bus terminal information, and' commercial vehicle identification. Computer~DataAcquisition~n (CDA5716-Acomputenzed data acquisition system utilizes tape switch sensors on the roadway to quicldy and accurately collect traffic characteristics. Characteristics collected include headway, flow rate, speed, acceleration, wheelbase, and lane position. Traditional manual collection of this data involves significant amounts of labor and time, particularly because only one or two characteristics can be collected at one time by any individual in the field. Because ofthe new emphasis placed ' on the temporal component of traffic analysis, the CDAS's ability ~6Bnan Moen et al., 'Traffic Data Collection Using a Computerized Data Acquisition System", Innovations in Travel Survey Methods, Transportation Research.Record No. 1412, TRB, 1993. 19

·: ~ to measure events to the thousandth of a second proves its super~on~cy over traditional manual field collection. A typical CDAS consists of a portable computer, digital timer, tape switches, photocells, and a video camera The video camera selves as a backup for the tape switch sensors and a visual guide to iden~ing circumstances involved in data ar~omalies. Although computer data acquisition is typically more accurate, precise, and less time consuming than manual collection, the data collected through a CDAS is in a very basic form that requires some da a reduction before proper analysis, whereas manual data can be collected In the desired format. Cellular Phones (probe vehiclesJi'- The increasing popularity of cellular phones, as well as their ability to provide real time traffic information, has caused many traffic management centers to examine their possible use in daily operations as a means of communication between users and traffic management. Case Study - The city of Houston recently completed an 18-month demonstration project which involved using 200 volunteers, called Probes" who reported haflic conditions at pre-deterrnined intervals. The Diver information uras analyzed and processed at a central location where Ravel times and average speeds were calculated. The information was In turn relayed to the public via message boards and add y services Using the Floating car Methodt with Distance Measuring Instruments (DA!I)I' This methodology is deemed as an improvement over the manual collection and calculadon of travel times and speeds using a coating car. The equipment involves a distance measuring instrument, similar to a tachometer, which is attached to the transmission or an anti-lock brake and a laptop computer. ~7"CelWar Phone Demo Project: Congestion Management News (Volume l, No. 5), 1= Journal, September, 1995. t8Benz, Robert "The Development, Benefits, and Use of Computer Aided Travel Time Data Collection", Presented at the National Traffic Data Acquisition Conference ~ATDAC), May 1996. 20 I · ; . .; ·; ~-; .

: ~;:~.';~: ~:;:;'- .:-~-: ::.:: . ;;''::-:: : ~:;i;: Case Study - In the spring on 1995, the Texas Transportation Institute collected data on 195 miles of arsenal routes and 98 miles of freeway routes. `The DMI provided 0.5 second interval speeds Dom which speed profiles average speed range, and acceleration characteristics such as noise (standard deviation of the acceleration) were calculated.~,The technique only required one technician and proved to be a cost-effective, safe, and accurate method to collect speed time and distance information. Acceleration and filet consumption estimates were also calculated using the collected data. Global Posz~ioning Systems (GPSJ~9- GPS is a "space-based radio positioning, navigation, and time transfer system"which utilizes satellites to locate vehicles equipped with the GPS emitting technology. The systems could be used on buses, trucks, rail cars, and marine Tic to relay data on vehicle locations at various points In time. In fact, many of the recent studies, as detailed below, involve using GPS for collecting travel time data Aggregated GPS data can provide an overall procure of traffic flows and travel patterns. In addition, a constant Dow of GPS data can indicate when a vehicle's position is not changing, indicating either a congestion problem or maintenance situation ~ either event, Me infonnation could assist In providing a more immediate response. Some of the weaknesses ofthe GPS include cost and the signal disruption that can occur when vehicles are shielded by bridges, tunnels, and All buildings. The discontinuous signal could adversely effect We usefulness of the data and require the GPS to be augmented by art additional ground-based system. Case Study - A recent study conducted by Louisiana State University used GPS to collect travel time data on 3 10 miles of Interstate and Principal Arterial routes: The equipment for the study included a 'floating ear,' equipped with a GPS unit, a di~er.ential correction unit (not mandatory), and a low-end notebook for data storage. The total cost of the equipment was about $2,000 ($1,000 without the notebook). A spatial model of the study corridor was developed using GIS, as well as a procedure to link the GPS data to the GIS corridor data. The GPS data, which is collected every ~ second, was reduced- in volume by using only the data found at every 0.32-km (0.2mi). The routes were color- coded according to travel times to visually represent the areas of highest congestion. I9Assessment of Advanced Technologies for Relieving Urban Traffic Congestion. National Cooperative Highway Research Program Report #340 TRB December 1991. 21 i

Intelligent Parking MetersI° - Intelligent parking meters recognize when a vehicle departs a space and automatically resets itself to zero. Beyond increasing revenue, the meters perform some basic primary data collection by storing information on usage, revenue, and violations by location which can be useful to transportation planners in estimating parldng demand and trips Case Study - New Hope, Pennsylvania is currently utilizing these new meters. Infrared Sensors - Infrared technology is fairly simple and straight fort arc. It consists of a light source and a light receiver mounted together which emits a beam of infiared light at a reflector. When a vehicle passes over or in front of a rejector, thus interrupting the signal, an electronic switch is closed to indicate the presence of a vehicle. Case Study - In December of l9g4, the Center for Transportation Research (CTR) at The University of Texas installed infiared sensors on a sign bridge spanning a 15-lane Deeway near Houston The insulation occurred during two weekends versus the minimum of five weekends of labor needed to inset loop detectors in He pavement. The equipment costs were approximately $130 per transm~tter/receiver unit and $30 (maximum) for the rejectors. The technology proved to be cost- effective and reliable In perfonn~ng vehicle counts. The ~nstadation oftwo sensors would allow speed, length, and headway to also be calculated. 20 Traveling with Success: How Local Governments Use Intelligent Transportation Systems. p. 17. 22 . /

Acoustic .Ve`tsors21 - Acoustic sensors have been tested recently to determine if various vehicle classes, vehicle spread, traffic volume, and lane occupancy can be identified based on the difference in sound levels made by each type of vehicle. This type of passive sensor is becoming increasingly more desirable as the cost of large closures and inbound installations and repairs become more expensive. Case Study - The Center for Transportation Research at Virginia Polytechnic Institute conducted a test (approx. 1994) for classifying vehicles using an AT&T SmartSonic acoustic sensor and neural networks. The sensor was installed over a lane and detected acoustic signals as the vehicles passed the sensor. The goal of the test was to separate the vehicles into four categories: passenger cars, small trucks, heavy duty trucks, arid tractor trailers. The best results of the test show that the acoustic data was able to correctly distinguish between cars and trucks 96 percent of the time. In addition' using a Two Stage Multilayer Perceptron ~P) network there was 95 percent accuracy in classifying the vehicles into the four desired categories. The belief is that with enough data and analysis the acoustic data could eventually be used to identify eight classes of vehicles. Other Study - The Arizona Department of Transportation (ADOT) conducted a test between temporary in-ground loops and passive acoustic detectors (PAD). The test lasted for IS0 days and data was processed to determine vehicle spread, traffic volume, and lane occupancy. The results ofthe test indicate that there was only a three to five percent difference bet:,areen the data collected via the in-ground loop and those collected using the PAD. Based on the results of the test, PADs are being installed at over 30 locations instead of ~r~-ground loops. [inking Specific Feld Data Collection Sites with God -'The widespread use and lowering costs of GIS systems has enabled the technology to be of assistance with relatively simple tasks such as scheduling and siting of data collection in the field. 2iJames, Pcbert and Sornkiat Sampan (Virginia Polytechnic Institute and State University). "Vehicle Classification of Acoustic Signals Using Neural Networks". Accepted for presentation and publication at the Intelligent Transportation Society of America Fifth Annual Meeting, March 1995. 22Blewett et al., 'The Albuquerque Traffic Monitoring System: Beyond Traffic Data Collection Standards", presented at the Fourth National Conference on Transportation Planning Methods Applications, 1993. 23

Case Study - The Albuquerque traffic monitoring system uses GIS to limit data collection mistakes and redundancy. Each location selected as a traffic monitoring site is randomly selected each month. The locations are linked to a GIS system and each field personae! is given a detailed map of where the collection material is supposed to be placed. These map schedules have been so e~ec~ve that a counter has not been placed in the wrong location once in two years. 24 ;

Next: 3. Secondary Data Sources »
Multimodal Transportation Planning Data: Compendium of Data Collection Practices and Sources Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!