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 78
79 APPENDIX C Literature on Statewide and Intercity Passenger Travel Forecasting This appendix was principally written by David Farmer, with trip makers at an individual or household level and are typically contributions from Alan Horowitz. The material has been ex- used to generate the probability that a particular trip is taken or cerpted from the Guidebook on Statewide Travel Forecasting mode is used. In terms of model structure, a direct-demand Models. model is one that calculates all of the desired travel information in one, singly calibrated step. (Direct-demand models are sometimes called econometric models because of their resem- INTERCITY PASSENGER LITERATURE blance to statistical models of economic demand.) A sequential model, on the other hand, divides the modeling process into Intercity travel is a broad heading that includes statewide several individually calibrated steps. The urban "four-step" travel. As used here, the term "intercity" forecasting involves modeling process, which many departments of transportation the prediction and assignment of traffic volumes between (DOTs) have adopted for the statewide modeling purposes, pre- cities or other points of interest that are separated by some sents the quintessential example of a sequential model. significant distance. The term intercity is also used to distin- guish these models from "urban" models, which typically in- volve travel between more closely spaced points of interest Aggregate Direct-Demand Models within a localized area. Intercity models include corridor, statewide, regional, and national models. Statewide models The earliest intercity models were of the direct-demand type are therefore a subset of intercity models. The main point, and were developed in the 1960s as part of an examination first expressed as early as 1960 (C1,C2), is that the charac- of the Northeast Corridor (C6). The most famous of these teristics of intercity travel are inherently different from those was Quandt and Baumol's abstract mode model (C8). The of travel within an urban area. It is assumed that people travel reader is referred to the reviews referenced in the previous according to a somewhat different set of rules over longer section [especially Koppelman et al. (C6)] for a more com- distances and between metropolitan areas. The intercity plete historical perspective of significant intercity modeling models encountered in the literature are often associated with efforts. The following direct-demand models--some of an academic exercise, and therefore make use of fewer, more which are not mentioned in those references--are noted here carefully chosen origindestination (OD) pairs than would because they possess features that might prove useful to normally be included in a meaningful statewide model. Con- modeling at the statewide level. sequently, they generally present situations that are a little more abstract in nature. The similarities to statewide models A notable early innovation was attempted by Yu (C9). Yu are many. took the standard direct-demand formulation--regressed from cross-sectional data--and recognized that the elastici- Types of Intercity Passenger Models ties present in the cross-sectional data would not necessarily remain constant over time. His paper presents two single- A number of reviews have been made of the early history of purpose (one for business travel and one for personal travel) intercity modeling (C3C7) and most include some discus- direct-demand models in which the regression coefficients sion of the taxonomy of intercity models. Intercity models can each include a timeseries component. It is a novel idea that essentially be divided into four types on the basis of two cat- does not appear to have been picked up by succeeding au- egories: data and structure. The models can make use of either thors. Another innovative idea is found in Cohen et al. (C10). aggregate or disaggregate data, and can be of a direct-demand Here, as part of two single-purpose (business and nonbusi- or sequential structure. The four resulting combinations are: ness) direct-demand models, the authors propose to use a (1) aggregate direct-demand models, (2) aggregate sequential pivot-point procedure. The procedure is intended to elimi- models, (3) disaggregate direct-demand models, and (4) dis- nate the effects (on the traffic volumes to be forecasted) that aggregate sequential models. Intercity travel demand models result from variables that have been excluded from the mod- can be further classified by whether they encompass only a els. Description of the pivot-point procedure is brief, how- single mode (mode-specific) or multiple modes (total de- ever, and use of this procedure does not seem to have been mand), and by which trip purposes they include. adopted by other researchers. Aggregate data make use of the socioeconomic data for the By the late 1970s, direct-demand models were being con- OD pairs in the model and can also include the service charac- structed to include an increasingly wider range of variables teristics of the modes of travel between them. Disaggregate to account for the enormous variety of factors that influence data go further to examine the motives and characteristics of the travel behavior. Models presented by Peers and Bevilacqua
OCR for page 79
80 (C11) and Kaplan et al. (C12) give some sense of this trend. Make Trip Peers and Bevilacqua describe a model that includes a long list of policy-sensitive variables, arranged into three groups: (1) extensive variables, including population and employ- Destination 1 Destination 2 Destination i ment; (2) intensive variables, including persons per house- hold, income per household, and employment per acre; and (3) system variables, including travel speeds and costs. Meanwhile, Kaplan et al. describe their Passenger Oriented Mode 1 Mode 2 Mode j Intercity Network Travel Simulation (POINTS) model, a multimodal model that explicitly includes consideration of accessibility to the transportation system. Both of these mod- els provide a bridge from an earlier emphasis on aggregate Rent a car Don't rent a car modeling to the growth in disaggregate modeling research by the early 1980s. FIGURE C1 Structure of Morrison and Winston's model. Disaggregate Sequential Models One Disaggregate Direct-Demand Model One of the first applications of disaggregate (or behavioral) modeling was for the mode-choice step of sequential models. Another model of interest is the disaggregate direct-demand It is possible to develop a mode-choice model without disag- model developed in the 1980s by the Egypt National Trans- gregate data, as DiRenzo and Rossi did, using a "reasoned portation Study (C23C25). The Egyptian Intercity Trans- diversion model" (C13). Disaggregate models, however, typ- portation Planning Model estimates travel on seven modes ically use a logit formulation to provide a convenient way of for travelers in three income levels. It is unusual in its use of including a number of mode-abstract, transportation accessi- disaggregate data in a single equation (direct-demand) for- bility, policy-related, and behaviorally based variables in the mat. Also, unlike many intercity passenger models, it in- modeling process. Owing to parallel research in urban area cludes capacity restraints on the network, most notably for forecasting in the early 1980s, these models became more at- the shortage of passenger rail cars. Because it deals with a tractive. They were thought to be especially useful in the ef- very practical situation, the Egyptian model could reasonably fort to estimate the shifts in mode share that were expected be noted in the section of this appendix describing statewide from deregulation in the air and intercity bus industries, and forecasting techniques; however, because the transportation from the anticipated implementation of high-speed rail trans- situation in Egypt is sufficiently an abstraction of the situa- portation (C14,C15). Again, Koppelman et al. (C6) provides tion in the United States, it seems fitting to include it with the a review of many of the earlier disaggregate mode choice intercity models. It might also be noted that, in its treatment models. In addition, Miller (C16), Forinash (C17), and of rail car capacity restraints, it resembles some freight mod- Forinash and Koppelman (C18) provide studies of the various els, as well. structures (binomial, multinomial, and nested-multinomial) available to more realistically represent the cross-elasticities between modes and to eliminate irrelevant alternatives in the Trip Frequency logit mode-split formulation. Armed with an increasing understanding about the imple- One None mentation of disaggregate modeling techniques and fueled by the increasing availability of disaggregate data, several re- searchers have developed complete travel-demand models based on the analysis of disaggregate data in a number of dis- Destination 1 Destination 2 Destination i crete, nested steps. Morrison and Winston, for example, pres- ent multimodal models (one for vacation travel and one for business) with the hierarchical structure shown in Figure C1 (C19). Similarly, Koppelman (C20) and Koppelman and Hirsh Mode 1 Mode 2 Mode j (C21,C22) present a multimodal model with a structure shown in Figure C2. Morrison and Winston make use of the 1977 National Travel Survey data, whereas Koppelman and Hirsh use both the National Travel Survey and the 1977 National Personal Transportation Survey (NPTS) data. Both pairs of Service Class 1 Service Class 2 Service Class k researchers sought to use this disaggregate data in a model structure that mimics the behavioral logic of trip making. FIGURE C2 Structure of Koppelman and Hirsh's model.
OCR for page 80
81 Single-Mode and Single-Purpose Models models. Until further data are available, their use will remain limited. Besides the ubiquitous single-mode automobile models, there are two other types of single-mode models of interest: It should also be noted that there is a place for aggregate bus and air. (Most passenger rail models are a part of a mul- direct-demand models at a statewide scale. This econometric timodal model.) Modeling of intercity bus travel has proven type of model can be especially useful in tying the forecast to be difficult (C26) and examples of intercity bus models are of single quantity (e.g., annual vehicle-miles traveled or rare. One interesting bus model is presented by Neumann and emissions) to forecasts of socioeconomic data. Byrne (C27). His model describes a probabilistic (disaggre- gate) model based on a Poisson distribution of ridership, as opposed to a regression model. He concludes that this for- STATEWIDE PASSENGER FORECASTING mulation provides a simpler and more reasonable estimate of LITERATURE ridership on rural bus routes. Despite the amount of research involving the characteristics of intercity travel and its concentrations on econometric Several air travel models are also of interest. As early as models and probability-based models, passenger travel fore- the 1960s it was recognized that the year-to-year growth in casting, as practiced by the various state DOTs, has remained air travel makes the use of timeseries techniques valuable, much more basic. In most of the states contacted as part of and a 1968 paper by Brown and Watkins (C28) addresses this the research for this appendix no travel modeling is done on issue with simple linear regression techniques. A later paper by Oberhausen and Koppelman (C29) also looks at time a statewide level. At the majority of state DOTs, forecasting series analysis of air traffic patterns using a BoxJenkins is done for specific projects only, and forecasts are made procedure to account for cyclical (seasonal and yearly) vari- based on historic trends, rather than on some formal model. ations in travel behavior. In another study, Pickrell (C30) uses a combination of techniques to assess future trends in For the states that are engaged in some type of modeling intercity air travel. Pickrell uses a single-mode direct- process, the models used are all "four-step" models, with a demand model to estimate the total demand for air travel. At modeling procedure borrowed almost entirely from the ur- the same time, he uses an aggregate mode-choice model to ban transportation planning (UTP) process. This is likely a predict the percentage of market share that the air mode function of the ready availability of urban modeling soft- could generate under several alternative futures. Other air ware and personnel trained to use it. As early as 1967, Ari- travel models of interest include a regression analysis of zona and Illinois had developed UTP-style models (C34), travel between small cities in Iowa by Thorson and Brewer and by 1972 at least 19 different states were using or prepar- (C31), and an elaborate direct-demand model of intercity air ing statewide models (C35). Modeling activities were evi- travel based on quality-of-service measures by Ghobrial and dently so popular that in 1973 FHWA perceived the need to Kanafani (C32). standardize the thinking about statewide modeling, and issued a guidebook on the subject (C36)--effectively insti- Finally, the one other single-purpose intercity model tutionalizing the UTP-style model for statewide use. The worth noting is the disaggregate model of recreational enthusiasm for developing statewide models that was pres- travel presented by Gilbert (C33). Gilbert's model is suffi- ent in the late 1960s and early 1970s soon waned, however, ciently abstract to be included here with the other intercity whether owing to funding cuts or to frustration with the models, but more will be said about recreational travel model results, and little activity seems to have taken place models in Section 2. It should be sufficient to state here [studies in Florida and Kansas (C37C39) were an excep- that Gilbert's paper, published in 1974, is one of the latest tion until very recently]. Apparently, only Connecticut, papers found to specifically address the recreational trip Kentucky, and Michigan have been continuously develop- purpose. ing models from the earlier period. By the early 1990s, prompted by new federal legislation Discussion (Clean Air Act Amendment of 1990 and Intermodal Surface Transportation Efficiency Act of 1991), several states were As will be seen in the following sections, the intercity fore- rethinking their strategies. New Mexico (C40) and Texas casting techniques employed in most existing statewide (C41) produced interesting reports that outline this renewed models are principally those of the aggregate sequential type. focus on statewide modeling. The New Mexico report ad- This is partly owing to the strong traditions of and training in dresses both passenger and goods movement models within the four-step modeling process, but it is also the result of the broader context of statewide transportation planning. The the general failure of disaggregate techniques at a statewide Texas report, which includes reviews of circa-1990 models scale. Although disaggregate models are attractive because from Florida, Kentucky, and Michigan, concentrates more on of their ability to include the behavioral aspects of travel, the details of statewide modeling, especially the difficulties their principal drawback is the lack of sufficient disaggregate in isolating interzonal trips and the proliferation of data for calibration of statistically meaningful statewide "K-factors" in recent models. Despite this promising trend,
OCR for page 81
82 neither New Mexico nor Texas is currently involved in Data Collection for Passenger Travel statewide modeling. (Texas is, however, scheduled to issue a request for proposal for a model development contract in the Ideally, travel forecasts are based on some sort of travel data. fall of 1997.] A list of states contacted that sent information One obvious source of travel data is the survey. Surveys have about their current passenger modeling efforts is presented in been conducted at the statewide level since the earliest days Table C1, and these are discussed below. of highway modeling (C42), and continue to be conducted at TABLE C1 CURRENT STATEWIDE PASSENGER MODELS State TAZs Modes Purposes Comments Connecticut 1,300 total 1. SOV 1. HBW · Mode split based on LOS information 2. HOV 2. HBNW · Iterative-equilibrium assignment for 3. Bus 3. NHB highways 4. Rail Florida 440 Highway 1. HBW · All trips are modeled to maximize use internal vehicles 2. HB shop of MPO models only 3. HB soc./rec. · Gravity friction factors based on MPO 32 external 4. HB misc. urban models 5. NHB · Mode split is auto occupancy only 6. Truck/taxi based on production zone · Extensive use of K-factors Indiana 500 1. Auto 1. HBW · Under development internal 2. Truck 2. Other business · Internal TAZs at the township level 5060 3. Transit 3. HBO · Aggregate mode choice external 4. NHB 5. Recreational 6. Truck Kentucky 756 Auto only 1. HBW · Model includes a large portion of internal 2. HBO surrounding states 706 3. NHB · NPTS national average data used for external trip generation Michigan 2,392 total Auto only 1. HB work/biz. · All trips modeled--previous models 2. HB soc./rec./vac. did not consider local trips 3. HBO · Two possible mode split models: (1) 4. NHB work/biz. simple cross-classification and (2) 5. NHB other LOS-based · LOS-based mode split model still under development · NTPS data used for calibration; CTPP data used for validation · Extensive use of K-factors New 1 per 5,000 1. SOV 1. HBW · Under development Hampshire pop. 2. HOV2 2. Business related · Logit trip generation and distribution 3. HOV3+ 3. Personal · Time of day and seasonal factors 4. Bus 4. Shopping 5. Rail 5. Recreational 6. Other New Jersey 2,762 -- -- · Model created by merging five MPO internal models 51 external Vermont 622 Highway 1. HBW · Based on extensive statewide survey internal vehicles 2. HB shop only 3. HB school 70 external 4. HBO 5. NHB 7.Truck Wisconsin 112 1. Auto 1. Business · Under development internal 2. Air 2. Other · No external trips considered 3. Rail · Network used only to develop impe- 45 external 4. Bus dances for mode share calculations Wyoming 5 internal 1. Auto -- · Model created mostly to demonstrate 2. Truck techniques 5 external · Summer weekend travel is modeled · Full trip tables estimated using entropy maximization technique Notes: TAZ = transportation analysis zone; SOV = single-occupancy vehicle; HOV = high-occupancy vehicle; HBW = home-based work; HBNW = home-based nonwork; NHB = nonhome-based; MPO = metropolitan planning organization; LOS = level of service; CTPP = Census Transportation Planning Package; HBO = home-based other; NPTS = National Personal Transportation Survey.
OCR for page 82
83 the statewide level (C43,C44). However, they are relatively for their state trunk highways (C63); however, such docu- expensive to conduct and must be supplemented by other mentation seems to be the exception. Some indication of the data. Two other options make use of data that are already possibilities of trendline analysis is given in a paper by available: federal survey data and statewide traffic counts. Harmatuck (C64) for the Wisconsin DOT. In it he provides U.S. Census data have always been valuable as inputs to further insight into the particular ways of dealing with traffic travel modeling. The 1990 Census improved on this by in- data as a timeseries. In addition, at least one state contacted cluding a journey-to-work survey, and by introducing for this appendix indicated that a growth factor method, sim- the Census Transportation Planning Package (C45). The ilar to the method outlined for updating coverage counts in journey-to-work has proven especially useful in estimating FHWA's 1992 Traffic Monitoring Guide (C65), is used for home-based work trips on a statewide level, but has been forecasting purposes. Otherwise little information is avail- criticized for its lack of information about other purposes able on travel forecasting techniques in the absence of a (C46). The Census Transportation Planning Package pro- statewide model. vides transportation-related information at a transportation analysis zone level, which can be readily aggregated into Statewide Models of Passenger Travel township- or county-level data for statewide modeling. Another federal data source is provided by the U.S.DOT, Of the states contacted as part of the research for this appen- which conducted its most recent NPTS in 1995. The NPTS dix, those having ongoing modeling efforts sent documenta- data, which measure some intercity travel, have been used in tion of their progress. A summary of the passenger models in the development of a number of statewide models. In addi- existence or under development is presented in Table C1. tion to the aforementioned federal government sources, it This includes work done in Connecticut (C66,C67), Florida should also be noted that estimated and forecasted data are (C68,C69), Indiana (C70), Kentucky (C71), Michigan (C72), also available from a wide variety of state, academic, and New Hampshire (C73), New Jersey (C74,C75), Vermont commercial sources. (C76), Wisconsin (C77), and Wyoming (C60,C61). In addi- tion to the states cited in Table C1, California has a statewide Of course, for many years state DOTs have had in place model, but it is being redesigned; therefore, documentation systems of traffic counting equipment operating at a is currently unavailable. Oregon is also in the early stages of statewide scale. Research in the early 1980s (C47C49) developing a comprehensive forecasting model that will in- developed statistical methods of clustering together traffic clude a land use element (C78). Several other states are counts on different roads based on their similar functional currently in the initial stages of modeling projects--issuing and geographical characteristics. In association with the in- requests for proposals to interested consultants. troduction of FHWA's Traffic Monitoring Guide in 1985 (C50), Pennsylvania (C51), Washington State (C52C54), As can be seen from Table C1, most of the models con- and New Mexico (C55,C56) began to reevaluate their traffic sider a large number of trip types (as many as five or six), but monitoring systems to take advantage of clustering. The re- only a few modes. All of the models are of the four-step style. sult is a larger and more statistically valid collection of traf- All use fairly standard UTP procedures, except for the model fic count data available for use in travel forecasting. under development for New Hampshire. New Hampshire proposes to use logit formulations for trip generation and Data Synthesis for Passenger Travel distribution. The Wisconsin model is unique in that it is es- sentially an intercounty model, with comparatively few Even with advanced systems for traffic data collection, it is transportation analysis zones. The Florida and New Jersey difficult for a state DOT to collect enough data to account for models are also interesting in the degree to which they have all of the likely paths between OD pairs being examined. To attempted to incorporate existing metropolitan planning or- get around this difficulty, optimization methods have been de- ganization models into the statewide modeling effort. The veloped to synthesize trip tables from available traffic count Kentucky and Michigan models are two of the more recent information (C57C59). These methods have subsequently useable models from states with long histories of model been applied to statewide analyses in Wyoming (C60,C61). development and are representative of the current state of the Attempts have also been made to synthesize trip tables from practice. census data at a sub-state level in New Jersey (C62). Recreational Travel Models Trend Analyses of Passenger Travel As early as 1963, recreational trips were considered an im- As noted earlier, many of the DOT officials contacted for this portant enough purpose to warrant separate study (C79). appendix indicated that the only forecasts they make are not Indeed, in the late 1960s and early 1970s NCHRP (C80), based on models, but are instead based on the extrapolation Indiana (C81,C82), Kentucky (C83,C84), and other states of trends observed in historical data. The Minnesota DOT (C85,C86) conducted studies of the special characteristics of has formalized this process as it applies to forecasting traffic recreational travel. However, although Americans seem to
OCR for page 83
84 have dedicated an increasing amount of time to pursuing C11. Peers, J.B. and M. Bevilacqua, "Structural Travel De- recreational activities, the last of these studies was published mand Models: An Intercity Application," Transporta- more than 20 years ago. Because many state economies de- tion Research Record 569, Transportation Research pend heavily on recreational activities, it would seem that Board, National Research Council, Washington, D.C., this trip type might be important enough to require a closer 1976, pp. 124135. examination than it has received in the past two decades. C12. Kaplan, M.P., A.D. Vyas, M. Millar, and Y. Gur, "Forecasts of Intercity Passenger Demand and En- ergy Use Through 2000," Transportation Research Discussion Record 870, Transportation Research Board, Na- tional Research Council, Washington, D.C., 1982, Using trendline procedures in statewide forecasting is prob- pp. 8390. ably better than not forecasting at all, especially for short- C13. DiRenzo, J.F. and L.P. Rossi, "Diversion Model for term planning horizons where large variations from recent Estimating High-Speed Rail Use," Highway Research trends are less likely. The use of travel forecasting models, Record 369, Highway Research Board, National Re- however, grounds the forecast in the underlying statewide search Council, Washington, D.C., 1971, pp. 1525. and national socioeconomic trends. Although these socio- C14. Brand, D., T.E. Parody, P.S. Hsu, and K.F. Tierney, economic trends are themselves forecasts, it is hoped that "Forecasting High-Speed Rail Ridership," Transporta- they broaden the basis of the transportation model suffi- tion Research Record 1342, Transportation Research ciently to provide a more reasonable forecast of future travel. Board, National Research Council, Washington, D.C., 1992, pp. 1218. REFERENCES C15. Buckeye, K.R., "Implications of High-Speed Rail on Air Traffic," Transportation Research Record 1341, C1. Church, D.E., "Outlook for Better Regional and Na- Transportation Research Board, National Research tional Forecasts of Highway Traffic and Finance," Council, Washington, D.C., 1992, pp. 1927. HRB Bulletin 257, Highway Research Board, National C16. Miller, E.J., "Intercity Passenger Travel Demand Mod- Research Council, Washington, D.C., 1960, pp. 3638. elling: Present State and Future Possibilities," Pro- C2. Burch, J.S., "Traffic Interactance Between Cities," ceedings of the 27th Annual Meeting of the Canadian HRB Bulletin 297, Highway Research Board, National Transportation Research Forum, University of Research Council, Washington, D.C., 1961, pp. 1417. Saskatchewan, Saskatoon, 1992, pp. 378389. C3. Vogt, Ivers and Associates, NCHRP Report 70: Social C17. Forinash, C.V., A Comparison of Model Structures for and Economic Factors Affecting Intercity Travel, Intercity Travel Mode Choice, M.S. thesis, Northwest- Highway Research Board, National Research Council, ern University, Evanston, Ill., 1992. Washington, D.C., 1969. C18. Forinash, C.V. and F.S. Koppelman, "Application and C4. Watson, P.L., "Comparison of the Model Structure and Interpretation of Nested Logit Models of Intercity Predictive Power of Aggregate and Disaggregate Mod- Mode Choice," Transportation Research Record 1413, els of Intercity Mode Choice," Transportation Research Transportation Research Board, National Research Record 527, Transportation Research Board, National Council, Washington, D.C., 1993, pp. 98106. Research Council, Washington, D.C., 1974, pp. 5965. C19. Morrison, S.A. and C. Winston, "An Econometric C5. Rallis, T., Intercity Transport: Engineering and Plan- Analysis of the Demand for Passenger Transporta- ning, Wiley, New York, N.Y., 1977. tion," Research in Transportation Economics, Vol. 2, C6. Koppelman, F.S., G.-K. Kuah, and M. Hirsh, Review 1985, pp. 213237. of Intercity Passenger Demand Modeling: Mid-60's to C20. Koppelman, F.S., "Multidimensional Model System the Mid-80's, The Transportation Center, Northwest- for Intercity Travel Choice Behavior," Transportation ern University, Evanston, Ill., 1984. Research Record 1241, Transportation Research C7. Khan, A.M., "Towards the Development of Innovative Board, National Research Council, Washington, D.C., Models of Intercity Travel Demand," Transportation 1989, pp. 18. Quarterly, Vol. 39, No. 2, Apr. 1985, pp. 297316. C21. Koppelman, F.S. and M. Hirsh, "Intercity Passenger C8. Quandt, R.E. and W.J. Baumol, "The Abstract Mode Decision Making: Conceptual Structure Data," Trans- Model: Theory and Measurement," Journal of Re- portation Research Record 1085, Transportation gional Science, Vol. 6, No. 2, 1966, pp. 1326. Research Board, National Research Council, Washing- C9. Yu, J.C., "Demand Model for Intercity Multimode ton, D.C., 1986, pp. 7075. Travel," Journal of Transportation Engineering, Vol. C22. Koppelman, F.S. and M. Hirsh, Intercity Passenger 96, No. 2, May 1970, pp. 203218. Travel Demand Analysis and Forecasting, The Trans- C10. Cohen, G.S., N.S. Erlbaum, and D.T. Hartgen, "Inter- portation Center, Northwestern University, Evanston, city Rail Travel Models," Transportation Research Ill., 1987. Record 673, Transportation Research Board, National C23. Moavenzadeh, F., M.J. Markow, B.D. Brademeyer, Research Council, Washington, D.C., 1978, pp. 2125. and K.N.A. Safwat, "A Methodology for Intercity
OCR for page 84
85 Transportation Planning in Egypt," Transportation C36. Statewide Travel Demand Forecasting, U.S. Depart- Research-A, Vol. 17A, No. 6, 1983, pp. 481491. ment of Transportation, Federal Highway Administra- C24. El-Hawary, M. and A. Gadallah, "Egypt Intercity tion, Washington, D.C., 1973. Transportation Model," In Research for Tomorrow's C37. Assessment of Modeling Techniques and Data Sources Transportation Requirements, The Centre for Trans- for Multi-modal Statewide Transportation Planning, portation Studies, University of British Columbia, Peat, Marwick, Mitchell & Co., Washington, D.C., Vancouver, BC, Canada, 1986, pp. 9971009. Oct. 1977. C25. Safwat, K.N.A., "Application of Simultaneous Trans- C38. Person Travel Forecasting Procedures for Multi- portation Equilibrium Model to Intercity Passenger modal Statewide Transportation Planning, Peat, Mar- Travel in Egypt," Transportation Research Record wick, Mitchell & Co., Washington, D.C., Feb. 1979. 1120, Transportation Research Board, National Re- C39. Russell, E.R., J.H. Jatko, and M. Ahsan, Investigation search Council, Washington, D.C., 1987, pp. 5259. and Development of a Statewide Traffic Assignment C26. Dean, D.L., "Modeling Dilemma of Intercity Bus Model for the State of Kansas, Kansas Department of Transportation," Transportation Research Record Transportation, Topeka, 1977. 887, Transportation Research Board, National Re- C40. Barton-Aschman Associates, A Research Process for search Council, Washington, D.C., 1982, pp. 3537. Developing a Statewide Multimodal Transport Fore- C27. Neumann, E.S. and B.F. Byrne, "A Poisson Model of casting Model, New Mexico State Highway and Trans- Rural Transit Ridership," Transportation Research portation Department, Santa Fe, 1991. Record 661, Transportation Research Board, National C41. Benson, J.D., J.A. Mullins, and G.B. Dresser, Feasi- Research Council, Washington, D.C., 1978, pp. 2127. bility of Developing a Statewide Modeling System for C28. Brown, S.L. and W.S. Watkins, "The Demand for Air Forecasting Intercity Highway Volumes in Texas, Travel: A Regression Study of Time-Series and Cross Texas Transportation Institute, College Station, 1991. Sectional Data in the US Domestic Market," Highway C42. Green, M.K., "Multiple Screenline Study to Determine Research Record 213, Highway Research Board, Na- Statewide Traffic Patterns," HRB Bulletin 253, High- tional Research Council, Washington, D.C., 1968, pp. way Research Board, National Research Council, 2134. Washington, D.C., 1960, pp. 139144. C29. Oberhausen, P.J. and F.S. Koppelman, "Time-Series C43. Virkud, U. and C.S. Keyes, "Design and Implementa- Analysis of Intercity Air Travel Volume," Transporta- tion of a Statewide Roadside Origin Destination Sur- tion Research Record 840, Transportation Research vey in Vermont," Transportation Research Record Board, National Research Council, Washington, D.C., 1477, Transportation Research Board, National Re- 1982, pp. 1521. search Council, Washington, D.C., 1995, pp. 1525. C30. Pickrell, D.H., "Model of Intercity Travel Demand," In C44. Crevo, C.C., R.S. Niedowski, and D.J. Scott, "Design Deregulation and the Future of Intercity Travel, J.R. and Conduct of a Statewide Household Travel Survey Meyer and C.V. Oster, Eds., MIT Press, Cambridge, in Vermont," Transportation Research Record 1477, Mass., 1987, pp. 249260. Transportation Research Board, National Research C31. Thorson, B.A. and K.A. Brewer, "Model to Estimate Council, Washington, D.C., 1995, pp. 2630. Commuter Airline Demand in Small Cities," Trans- C45. Glass, D., R.W. Tweedie, and R.J. Zabinski, "Trans- portation Research Record 673, Transportation Re- portation Planning Data for the 1990s," Transportation search Board, National Research Council, Washington, Research Record 1283, Transportation Research D.C., 1978, pp. 187193. Board, National Research Council, Washington, D.C., C32. Ghobrial, A. and A. Kanafani, "Quality-of-Service 1990, pp. 4550. Model of Intercity Air-Travel Demand," Journal of C46. Meyer, M.D. and G. Mazur, "Census Data Use for Transportation Engineering, Vol. 121, No. 2, Mar./Apr. Statewide Transportation Planning and Transportation 1995, pp. 135140. Planning in Smaller Metropolitan Areas," In Conference C33. Gilbert, G., "Recreational Travel Behavior: The Case Proceedings 4: Decennial Census Data for Transporta- for Disaggregate, Probabilistic Models," In Special tion Planning, Transportation Research Board, National Report 149, Transportation Research Board, National Research Council, Washington, D.C., 1995, pp. 7582. Research Council, Washington, D.C., 1974, pp. C47. Sharma, S.C. and A. Werner, "Improved Method of 223230. Grouping Provincewide Permanent Traffic Counters," C34. Whiteside, R.E., C.L. Cothran, and W.M. Kean, "In- Transportation Research Record 815, Transportation tercity Traffic Projections," Highway Research Record Research Board, National Research Council, Wash- 205, Highway Research Board, National Research ington, D.C., 1981, pp. 1217. Council, Washington, D.C., 1967, pp. 110135. C48. Bester, C.J. and J.D. De Joubert, "Classifying a Rural C35. Hazen, P.I., "A Comparative Analysis of Statewide Road Network for Traffic Counting," Transportation Transportation Studies," Highway Research Record Research Record 1090, Transportation Research 401, Highway Research Board, National Research Board, National Research Council, Washington, D.C., Council, Washington, D.C., 1972, pp. 3954. 1986, pp. 2226.
OCR for page 85
86 C49. Albright, D., "A Quick Cluster Control Method: Per- Research Record 1499, Transportation Research Board, manent Control Station Cluster Analysis in Average National Research Council, Washington, D.C., 1995, Daily Traffic Calculations," Transportation Research pp. 16. Record 1134, Transportation Research Board, Na- C62. Walker, W.T., "Method to Synthesize a Full Matrix tional Research Council, Washington, D.C., 1987, pp. of Interdistrict Highway Travel Times from Census 5764. Journey-to-Work Data," Transportation Research C50. Traffic Monitoring Guide, Federal Highway Adminis- Record 1236, Transportation Research Board, National tration, U.S. Department of Transportation, Washing- Research Council, Washington, D.C., 1989, pp. 5058. ton, D.C., 1985. C63. Procedures Manual for Forecasting Traffic on the C51. DiRenzo, J., N. Serianni, and L. King, "Improving the Rural Trunk Highway System, Traffic Forecasting Productivity of the Pennsylvania Transportation De- Unit, Minnesota Department of Transportation, Min- partment's Traffic Count Program," Transportation neapolis, Apr. 1985. Research Record 1050, Transportation Research C64. Harmatuck, D.J., Improving Traffic Forecasting with Board, National Research Council, Washington, D.C., Time Series Models, Wisconsin Department of Trans- 1985, pp. 1319. portation, Madison, Jan. 1997. C52. Hallenbeck, M.E., "Development of an Integrated C65. Traffic Monitoring Guide, Federal Highway Adminis- Statewide Traffic-Monitoring System," Transporta- tration, U.S. Department of Transportation, Washing- tion Research Record 1050, Transportation Research ton, D.C., 1992. Board, National Research Council, Washington, D.C., C66. ConnDOT Modeling Procedure, Connecticut Depart- 1985, pp. 512. ment of Transportation, Hartford, Feb. 1997. C53. Ritchie, S.G., "A Statistical Approach to Statewide C67. PERFORM Person Forecasting Model: Trip Genera- Traffic Counting," Transportation Research Record tion, Staff Paper 93-10, Connecticut Department of 1090, Transportation Research Board, National Re- Transportation, Hartford, 1993. search Council, Washington, D.C., 1986, pp. 1421. C68. Shimpeler, Corridino Associates, Florida Highway C54. Ritchie, S.G. and M.E. Hallenbeck, "Evaluation of a Traffic Forecasting Models Technical Report 1: Statewide Highway Data Collection Program," Trans- Statewide Model Development, Florida Department of portation Research Record 1090, Transportation Re- Transportation, Tallahassee, Sep. 1990. search Board, National Research Council, Washing- C69. Shimpeler, Corridino Associates, Statewide Highway ton, D.C., 1986, pp. 2735. Traffic Forecasting Model Technical Report 2: Col- C55. Albright, D., C. Blewett, and R. Goldstine, "Develop- lection and Assembly of Data, Florida Department of ment of State Traffic Monitoring Standards: Imple- Transportation, Tallahassee, Oct. 1990. menting Traffic Monitoring Guide in New Mexico," C70. Cambridge Systematics, Major Corridor Investment- Transportation Research Record 1236, Transportation Benefit Analysis System, Appendix A: Scope of Work, Research Board, National Research Council, Wash- Indiana Department of Transportation, Indianapolis, ington, D.C., 1989, pp. 913. undated. C56. Albright, D.P., "Revision of Statewide Traffic Data C71. Wilbur Smith Associates, Kentucky Statewide Traffic Standards Indicated During Implementation of a Traffic Model Final Calibration Report, Kentucky Trans- Monitoring System," Transportation Research Record portation Cabinet, Frankfurt, Apr. 1997. 1271, Transportation Research Board, National Re- C72. KJS Associates, Inc., Statewide Travel Demand Model search Council, Washington, D.C., 1990, pp. 4854. Update and Calibration: Phase II, Michigan Depart- C57. Van Zuylen, H.J. and L.G. Willumsen, "The Most Likely ment of Transportation, Lansing, Apr. 1996. Trip Matrix Estimated from Traffic Counts," Trans- C73. Cambridge Systematics, New Hampshire Statewide portation Research-B, Vol. 14B, 1980, pp. 281293. and Subarea Travel Models Plan, New Hampshire De- C58. Bell, M.G.H., "The Estimation of an Origin-Destination partment of Transportation, Concord, Mar. 1995. Matrix from Traffic Counts," Transportation Science, C74. URS Consultants, Inc., Effects of Interstate Comple- Vol. 17, No. 2, May 1983, pp. 198217. tion and Other Major Highway Improvements on Re- C59. Hendrickson, C. and S. McNeil, "Estimation of Origin- gional Trip Making and Goods Movement: Network Destination Matrices with Constrained Regression," Development, New Jersey Department of Transporta- Transportation Research Record 976, Transportation tion, Trenton, June 1995. Research Board, National Research Council, Wash- C75. URS Consultants, Inc., Effects of Interstate Comple- ington, D.C., 1984, pp. 2532. tion and Other Major Highway Improvements on Re- C60. Wilson, E.M. and J. Wang, Statewide Transportation gional Trip Making and Goods Movement: Auto Trip Planning--An Interactive Modeling Process, Wy- Table, New Jersey Department of Transportation, oming Department of Transportation, Laramie, May Trenton, Mar. 1995. 1995. C76. Vanasse, Hangen Brustlin, Inc., Statewide Travel De- C61. Wilson, E.M. and J. Wang, "Interactive Statewide Trans- mand Model Development, Vermont Agency of Trans- portation Planning Modeling Process," Transportation portation, Montpelier, Apr. 1996.
OCR for page 86
87 C77. KPMG Peat Marwick, Translinks 21: Multi-Modal In- C82. Robinson, D.C. and W.L. Grecco, "Stability of Recre- tercity Passenger Analysis, Wisconsin Department of ational Demand Model," Highway Research Record Transportation, Madison, June 1995. 392, Highway Research Board, National Research C78. Parsons Brinckerhoff Quade & Douglas, Inc., Consul- Council, Washington, D.C., 1972, pp. 147156. tant Recommendations for the Development of Phase C83 Deacon, J.A., J.G. Pigman, and R.C. Deen, "Travel to II Databases, Models, and Forecasting Methods: Outdoor Recreation Areas in Kentucky," Highway Re- Transportation and Land Use Model Integration Pro- search Record 392, Highway Research Board, Na- gram Phase I, Task 1.6, Oregon Department of Trans- tional Research Council, Washington, D.C., 1972, pp. portation, Salem, Sep. 13, 1996. 134135. C79. Crevo, C.C., "Characteristics of Summer Weekend C84. Deacon, J.A., J.C. Pigman, K.D. Kaltenbach, and R.C. Recreational Travel," Highway Research Record 41, Deen, "Models of Recreational Travel," Highway Re- Highway Research Board, National Research Council, search Record 472, Highway Research Board, National Washington, D.C., 1963, pp. 5160. Research Council, Washington, D.C., 1973, pp. 4562. C80. Ungar, A., NCHRP Report 44: Traffic Attraction of C85. Gyamfi, P., "A Model for Allocating Recreational Rural Outdoor Recreational Areas, Highway Research Travel Demand to National Forests," Highway Re- Board, National Research Council, Washington, D.C., search Record 408, Highway Research Board, National 1967. Research Council, Washington, D.C., 1972, pp. 5061. C81. Matthias, J.S. and W.L. Grecco, "Simplified Procedure C86. Berg, W.D., P.A. Koushki, C.L. Krueger, and W.L. for Estimating Recreational Travel to Multi-Purpose Bittner, "Development of a Simulation Model for Re- Reservoirs," Highway Research Record 250, Highway gional Travel," Transportation Research Record 569, Research Board, National Research Council, Wash- Transportation Research Board, National Research ington, D.C., 1968, pp. 5469. Council, Washington, D.C., 1976, pp. 96106.