Click for next page ( 167


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 166
Appendix 3 Samuel Lau, "Truck Travel Surveys: A Review of the Literature and State-of-the-Art," MTC Oakland, CA, January 1995 (Excerpts)

OCR for page 166
Jack Faucett Associates, Inc. Final Report March 1997 TRUCK TRAVEL SURVEYS: A REVIEW OF THE LITERATURE AND STATE-OF-T~-ART-MTC OAKLAND (EXCERPTS) This literature and state-of-the-art review reveals that few urban areas in the country have had extensive experience in conducting truck surveys and truck travel demand forecasting. Most metropolitan planning organizations (MPOS) or regional transportation planning agencies continue to generate their truck trip estimates based on orig~n-destination surveys conducted in Me 1960s and 70s. In the last ten years, only a few metropolitan areas, namely Chicago (1970 and 1986), Ontario (1978, 1983, and 1988), Vancouver (1988), Phoenix (1991), Alameda County, California (1991), New York-New Jersey (1991-94), E1 Paso, Texas (1994), and Houston-Galveston (1994) have undertaken significant efforts to collect truck travel data or develop new techniques in forecasting truck traffic. Out of the eight urban areas, only Chicago and Phoenix have had Weir truck model development and forecasting methodologies documented in detail, and only Ontario and the Port Authority of New York & New Jersey (PANYNJ) have systematically collected truck travel data. This report documents the experiences of different urban areas in Me U.S. and Canada. The following is a summary of Me results. Types of Data Collected The eight most recent truck travel surveys all collected orig~n-dest~nation information (see Exhibit A3-1). With the exception of roadside surveys conducted In New York and New Jersey, most truck surveys requested land use at destination and truck odometer readings from respondents. Most surveys classified trucks by weight, number of axles, or by truck type. With the exception of roadside surveys, all other survey types included trip diaries (Chicago, Phoenix' E1 Paso' Houston-Galveston, and Alameda County). NCHRP Multimodal Transportation A3-1 Planning Data Project 8-32(5)

OCR for page 166
Jack ~aucett Associates, lac. Final Report March 1097 I ...... .. ~'m~ o.~ ~,ru9,~ ,l't ve' ~.v ~ ~ ~I: : : Chicago Ontario Phoenix N.Y. & NJ. Alameda County, Calif. N.Y. & NJ. - El Paso 1986 1988 1991 1991 1991 1992-94 1994 Houston- | 1994 Galveston Mailout- Mailback Roadside Interview Combined Telephone Mailout Mailback Roadside Interview Combined Telephone Mailout Mailback & Roadside Interview Roadside Interview Telephone Interview Combined Tdephone Mailout Mailback 19,225 270 4,500 2,200 over 8,000 14,671 188 900 (1) Cost induded data collection, data coding, and model development. (2) This was the sampling rate. No response rate was given. (3) This was a multi-agency effort, with partnership from the New Jersey Department of Transportation (NJDOT), the New York Metropolitan Transportation Council (NYMTC), and the Port Authority of New York and New Jersey. The survey was conducted at 18 locations with 3 interviewers per toll plaza for 24 hours. (4) Cost induded sample design, survey design, data collection, coding, reporting, survey analysis, and model development. (5) The higher cost was due to a high number of incomplete surveys. Truck travel model development CorridorlRoute analysis Effects of toll on trucks Truck speed simulation mode Truck activity mapping 96.5% 30.0% NA 79% NA 37.8%(2) 42.6% 35%-40% Time series comparison Evaluate & design road geometries Pavement management planning Truck accident analysis Dangerous goods regulation & enforcement analysis Truck driver characteristics Driver education program Truck travel model development Evaluate dedicated route/corndor proposal Traffic management for highway reconstruction Time series freight analysis Frdght-economic analysis I-880 corridor analysis Create truck travel submodel for corridor analysis Generate 24-hour & PM peak volumes by axle NA Truck travel model development Part of regional travel study Truck emissions analysis Truck travel model development $200,000 NA S90,000(1) NA NA $312,000(3) $65,000(4) $150,000 S57/survey NA S125tsurvey(1) NA NA $21/survey $345tsurvey(5) S167/sun~ey NCHRP Multimodal Transportation Planning Data A3-2 Project 8-32(5)

OCR for page 166
Jack Faucett Associates, lace Final Report March 1997 The commodity data collected ranged from a simple classification of commodity by type to detailed description of the actual commodities being carried. The 1988 Ontario survey is the only commercial vehicle survey that gathered information on truck driver characteristics. The 1994 E1 Paso commercial truck survey was the only survey route choice information for the surveyed trip. Uses of Truck Survey Data The most common uses of truck data are for regional truck travel model development and corridor/route analysis. Chicago, Phoenrx, E1 Paso, and Vancouver have used their truck survey data to develop regional truck travel demand models. Ontario has seen the most use of its truck survey information. The truck data have been used for time series comparisons, evaluation of road design and geometries, pavement management planning, truck-related accident analysis, dangerous goods movement regulation and enforcement, understanding truck driver characteristics and for planning truck driver education programs. E1 Paso has mainly used its truck data for regional travel and truck emissions modeling. The Port Authority of New York and New Jersey has used its truck data for traffic management purposes during highway and bridge/tunnel reconstructions and freight- economic analysis. Chicago has used its truck data to generate truck activity maps of the Greater Chicago region; truck speed simulation; and modeling the effects of toll facilities on truck route choices within the context of the Chicago regional travel model. The Southern California Association of Governments (SCAG) has used its truck travel data to estimate heavy truck VMT and model truck emissions in the South Coast Air Basin (SCAB) for the Los Angeles area. It has also used truck data to conduct analysis of truck traffic to the Port of Hueneme in Ventura County. Caltrans and Alameda County has used its truck survey data to estimate truck traffic entering and leaving the County, as well as seaport planning for the Port of Oakland. NCNRP Multimodal Trar~sport~ion A3-3 Planning Data Project 8-32(5)

OCR for page 166
Jack Faucett Associates, Inc. Final Report March 1997 Truck Travel Survey Methods Used . The most common survey method for conducting truck travel surveys in urban areas was the combined telephone-mailout-mailback method. Three urban areas in the country Phoenix, Arizona; Alameda County, California; and Houston-Galveston, Texas - have recently conducted truck travel surveys using the combined telephone-mailout-mailback method. The combined telephone-mailout-mailback survey method is more cost-effective and yields a reasonably high response rate. The second most used survey method was the roadside interview method. The Province of Ontario, Canada and the Port Authority of New York & New Jersey have conducted numerous roadside interviews. Roadside interviews produce very high response rates with complete information. They are ideal for cordon surveys or surveying trucks traveling in from outside the survey area. The most common source for drawing the survey sample is the Department of Motor Vehicle (DMV) registration files. Other sample sources include lists of truck registration files available from commercial vendors (R.~. Polk, Texas Vehicle Information and Computer Services, Inc., etch. A summary of different survey characteristics for eight urban truck travel surveys is found in Table A-l, and a summary of different truck travel survey methods (typical response rate, advantages, and disadvantages, etc.) is found in Table A-2. Number of Completed Surveys and Response Rate The approximate number of completed surveys from the eight urban truck surveys varied from ISS to 19,225. Roadside surveys produced the highest number of completed surveys and the best response rate (nearly 100 percent). NCHRP Multimodal Transportation AM Project 8-32~5) Planning Data

OCR for page 166
Jack Faucett Associates, Inc. Final Report March 1997 Mailout-mailback surveys produced the lowest overall and item response rates. Combined- telephone-mailout-mailback surveys produced improved response rates over mailout- mailback or telephone surveys alone (See Exhibit A3-l and Exhibit A3-2~. Survey Cost Telephone interviews are the m. ost costly to conduct. They require a large number of staff and time for data collection. . Combined telephone-mailout-mailback surveys are the most cost-effective to conduct. They yield reasonably high response rates over mailout-mailbacks alone. The phone contact portion of the survey can help assess non-response biases when analyzing and weighting the mailback survey samples. Comparison of Survey Findings A summary of the general findings from various truck travel surveys is provided below. Characteristics of Commercial Vehicles Average Vehicle Weight: The only survey that reported average vehicle weight was the 1991 Phoenix Commercial Vehicle Survey. The average vehicle weight per commercial trip was ll,870 lbs. Truck Size: The share of different truck sizes used varied from urban area to urban area. Characteristics of Commercuz! Vehicle Trips . . Average Trip per Commercial Vehicle: Light trucks have a higher average trip frequency than for heavy trucks. Regional vs. Through Trips: Most truck trips serve local regional needs. Of the few through trips (usually less than 10 percent), most are made by heavy trucks. NCHRP Multimodal Transportation A3-5 Project8-32~5) Planning Data

OCR for page 166
Jack Faucet' Associates, lace Pinal Report March 1997 s mm f r Or Telephone Interview Mailout Mailback Combined Telephone -Mailout Mailhn~k _ _ Roadside Intercept/ Interview N.Y.(1964) Calgary(71) El Paso(94) Chicago(86) Phoenix(91) Houston(94) Alameda, CA(91) Calgary(71) Ontario(78, 83, 88) N.Y. & N.J. (74, 82, 85, 91-94) Alameda County, CA(91) 4%-15% 1%-5% 3%-10% 8%-35%(3) 40%-50% 10% 45%(1) 30%-V0%(2) 95%-1 00% High response rate Easy to follow-up Less costly Good response rate "/certified mail Only follow-up Of nonresponses is necessary Improved response rate over mailout-mailback alone Early identification of owners who agree to participate & potential nonresponses through phone contact Phone contact may help adjust sample size for mailout-mailback Complete information High response rate BeKer sampling control Good representative sample of trucks entering or leaving a cordon line Easy comparison with mainstream traffic through field counts at survey location (l) The higher response rate was due to better survey participation front large truck Beet operators. l (2) The higher response rate was due to an employer survey conducted in California (1991 Caltrans-Alameda County Survey). (3) The higher percentage is from the 1988 Ontario survey which curve-Is For;- Al ^ 0~= ~ Can only call during business hours "Phone-tagging" problem Limited time on phone if respondent is busy Requires access to vehicle registration file Low overall & item response rate Possible bias due to better response from some drivers/owners Low response from small truck owners Low response from out-of-state trucks Need to follow-up on nonrcsponses Difficult to ensure that the driver will fill out the form, instead of the owner or Beet manager who receives the survey forms Requires access to registration file Same disadvantages as telephone survey method above High cost of telephone follow-ups Need phone reminders for trip diary More costly than above methods Potential disruption to traff c Quality and conduct of survey affected by weather, lighting Hazardous to survey crew Time constraint No follow-up possible Enforcement problems Drivers avoiding the survey station Only represent trucks traveling on road along survey station, not entire region vets ",C. ~ ~,oa~nour period throughout the Ontario Province. NCHRP Multimodal Transportation Planning Data A3-6 Project 8-32~5)

OCR for page 166
Jack Faucett Associates, Inc. Final Report Average Trip Length: Heavy trucks make longer trips than lighter trucks. Vehicle Miles Traveled: Heavy trucks log a higher VMT per day than light trucks. March 1997 Time of First Commercial Vehicle Trip: Most "first" truck trips occur early in the morning (between 6:00 a.m. to 9:00 a.m.). This pattern, however, varies by weight category. Light trucks were more likely to start their first Hip between 6:00 a.m. and 9:00 a.m. Heavy trucks, however, started their first trip before 6:00 a.m. Time-of-Day Distribution: Most truck trips seem to occur during the midday period between 9:00 a.m. and 3:00 p.m. Truck "through" traffic seems to avoid peak periods and tend to travel at night. Truck Travel During Peak Periods: The results vary by urban area and by individual locations. In New York and New Jersey, over 35 percent of trucks made trips during the morning peak period (6:00 a.m. to 10:00 am.). In comparison with AM and PM peaks for private vehicle travel, the results found that the AM peak period travel was as important for commercial vehicles as for private vehicles. Truck Travel During Peak Periods as Percent of Total Vehicular Volume: Truck traffic range from less than 9 percent to as high as 17 percent of the total vehicular volume during peak periods. Day-of-Week Distribution: Truck traffic typically occurs on weekdays and decreases significantly on the weekends. . Average Trip Duration: Trip time generally increases with vehicle weight. The 1991 Phoenix survey recorded that the overall average trip tune for truck travel was 28.1 minutes. Truck Travel by Facility Type: Few surveys or studies have attempted to analyze truck trips based on facility types used. Only the Canadians used facility types to classify their truck trips. A 1991 Barton Aschman Study of Alameda County truck trips found that many of the approximately 5,000 daily truck trips in the Port of Oakland area are local trips that never access a freeway. Route Choice for Return Trips: The only survey that analyzed route choice for return trips was the 1991 New York and New Jersey Truck Commodity Survey. It found that 73 percent of the truck drivers interviewed in the toll direction indicated that they would use the same route for the reverse trip. NCHRP Multimodal Transportatwn A3-7 Planning Data Project 8-32~5J

OCR for page 166
I 1 Jack Faucett Associates, Inc. FinalReport March 1997 On-Street Stops: The 1991 Phoenix survey was the only to report the number of on-street stops made by trucks. The results found that over one-third of all commercial vehicles stops were made on-street. Light vehicles made half of their stops on-street. Commercial Vehicle Trips and [and Use Trips by Land Use: Light trucks make more residential scrips than anv other trek r~tr'~r~rv Retail attracted many more light and medium truck trips. terminal/warehouse land uses. 1 A ~ ^_ __ ~ Heavy trucks dominated Activities at Trip Ends: Light trucks are heavily used for service delivery and personal business. Heavy trucks are most used for loading and unloading cargo at their trip ends. Other Truck-Related Findings . . Truck Travel and Dangerous Goods Movement: The Ontario survey (1988) was the only survey that obtained information on dangerous goods movement. It found that a total of about 5 to 6 percent of all truck trips surveyed involved the carrying of dangerous goods. Flammable liquids (47 percent) were the most frequently transported dangerous goods, followed by compressed gases (24 percent)' and corrosive substances (20 percent). Truck-Related Accidents: The 1988 Caltrans Urban Freeway Gridlock Study found that 5 to 10 percent of all truck-related incidents were found to cause major incidents which closed two or more freeway lanes for at least two hours. Recommendations This report recommends the following for conducting a regional truck travel survey and truck travel demand forecasting model if MTC should be interested in developing new truck data and tools: NCHRP Multimodal Transportation A3-8 Project8-32~5) Planning Data

OCR for page 166
Jack FauceuAssocia:tes, Inc. Final Report March 1997 Survey Conduct For internal-to-internal or internal-to-external truck trips, draw the survey sample from the DMV registration file or regional truck registration files ~PUC, or private truck registration databases). Conduct either a telephone or mailnut-mnilhn~k Rev an combination of both to obtain a better response rate. . . ~ ~^ ~ _} , ~ ~ For external-to-internal or external-to-external truck trips, conduct roadside intercept surveys at various roadway facilities and links in the network. The best places to conduct them are "weigh-in-motion" stations. This would minimize tragic delay for the mainline ~ ~ 1 ~ _ 1 ~ 1~ ~ _ _ ^_ ~ i_ .( _ Ally WUU1U O~ sarer tar me survey crews compared to conducting the survey at the roadside. Consider conducting intercept surveys at bridge toll plazas. For a better explanation of how to conduct roadside surveys at toll plazas, review We experiences in New York and New Jersey. For roadside interviews or cordon surveys, conduct vehicle classification counts at the same time and at the same location where the actual survey/interview is conducted. This will provide the basic information for sample expansion and analysis. For obtaining trip diaries, using a combination of fleet-employer samples and truck unit samples is desirable. Sub-sampling fleet employers will provide better sample control and reduce the problem of over sampling large fleet operators. Over sample smaller or individual truck operators. The 1986 CATS survey has shown that large fleet operators tend to respond better (more manpower, time, or incentive to reply to surveys) and smaller operators tend to yield higher nonresponses. To reduce the cost of conducting a full-scale truck survey, consider making the survey a multi-agency effort. Consider soliciting the help of private freight/trucking agencies or organizations. Open a dialog with interested parties to facilitate cooperation and to request assistance, especially in the design of the survey. Truck Travel Analysis NCNRP-Multimoda/ Transportation A3-9 Project 8-325) Planning Data

OCR for page 166
Jack Faucett Associates, inc. Final Report March 1997 Time-of-day (24-hour), day-of-week, and seasonal variations in truck travel should be examined. Analyze trips by facility types used (include questions Hat obtain facility type information for each trip). Conduct further analysis on the impact trucks have on peak period congestion. Several surveys (New York, New Jersey, and Ontario) have found Nat in comparison with AM and PM peaks for private vehicle travel, AM peak period travel was as important for commercial vehicles as for private vehicles. Estimate total truck hours of delay by facility to help reduce truck operating cost. Conduct further analysis on Be impact of truck traffic on pavement, especially Me impact of waste-refuse trucks and buses (considered as "passenger~arrying trucks") on residential arterials and streets. The origins and destinations of trips that begin and end within the study area should be geocoded to the transportation analysis zones (TAZs) rather than at the city or zip code level. This would improve the accuracy of truck trip generation models based on zonal socioeconomic attributes. . Exercise extreme caution when using or applying vehicle ' equivalency factors (VEQs) in truck travel analysis. NCHRP-Multimode Transportation A3-10 Project 8-32(5) Planning Data