Research Threads

This page contains abstracts of several published articles in the areas of Network Editing, Transit Ridereship Forecasting, Land Use Forecasting, Modeling in General, Benefits Assessment, Traffic Assignment and Delay, and Intermodal Systems.  All of these articles should be readily attainable through major university libraries.

Network Editing

APPLICATIONS OF THE GENERAL NETWORK EDITOR

Horowitz, Alan J.

Microcomputer Applications in Transportation II, Proceedings of the North American Conference on Microcomputers in  Transportation. 1987 Sponsored by : ASCE, Urban Transportation Div, New York, NY, USA ASCE New York, NY, USA p 262-269

 

Abstract: The General Network Editor is a graphical, data base manager for computer-aided design of transportation networks. Its principal use is for entering network data into the computer: drawing a network on the CRT screen; entering numerical data; editing of strings and numbers; calculating attribute values from other pieces of data; and finding network elements that meet certain criteria. Through a variety of menus, GNE can be configured for nearly any type of transportation system. GNE produces and ASCII text data file that may be read by user-supplied analysis routines or interfaced to other existing software. GNE was implemented under MS-DOS, and it is in the public domain.  This paper describes how GNE was attached to the Quick Response System II, developed for the Federal Highway Administration (FHWA) as

 a major upgrade of an existing program for highway planning.

 

 

TOWARD A DATA STRUCTURE FOR COMPUTER-AIDED DESIGN OF TRANSPORTATION NETWORKS

Horowitz, Alan J.; Pithavadian, Anand R.

Transportation Research, Part B: Methodological v 21B n 4 Aug 1987 p 317-321

 

Abstract: Although there are a great variety of transportation network design applications, there are important similarities across all transportation networks. These similarities permit the construction of a single, relatively simple, data structure that satisfies the needs of both graphics and algorithms. This paper describes how this data structure was implemented as the basis of the General Network Editor - a program for graphical preparation of network data.  10 Refs.

 

 

GENERALIZED COMPUTER AIDED DESIGN OF TRANSPORTATION NETWORKS

Horowitz, Alan J.; Pithavadian, Anand R.

Transportation Quarterly v 41 n 3 Jul 1987 p 397-409

 

Abstract: Computer graphics is widely recognized as a highly useful feature for many different transportation design applications. Reported applications run the gamut from physical design of transportation facilities to solving mathematical models of travel demand. This article is particularly concerned with the most abstract end of this application spectrum - those that require data in the form of a network description. If a single computer program were available to perform all network computer-aided design (CAD) functions, then software development time could be drastically cut, less ambitious applications could receive network graphics capability, and standard network coding procedures could be adopted by all transportation specialties. This article reports on efforts to construct the General Network Editor (GNE), a standard program for microcomputers.  3 Refs.

 

 

INTEGRATING GIS CONCEPTS INTO TRANSPORTATION NETWORK DATA STRUCTURES

Horowitz, Alan J.

Transportation Planning and Technology v 21 n 1-2 1997 Gordon & Breach Science Publ Inc Newark NJ USA p 139-153

 

Abstract: Transportation network data structures must be designed to meet the requirements of the analyses being conducted and must be compatible with the selected graphical user interface. Increasing interest in geographic information systems (GIS) and intelligent transportation systems (ITS) have further burdened the network data structure. It is possible to implement object oriented programming (OOP) technology to satisfy these needs, without making the data structure excessively complicated. This paper shows how a well-developed network data structure can incorporate major capabilities normally associated with stand-alone GIS's. The design of a network data structure derives from both theoretical and practical considerations. A design of a network data structure, composed entirely of objects, is presented. Examples of its implementation, limitations, advantages, and possible extensions are drawn from experience with the General Network Editor (GNE). 9 Refs.

Transit Ridership Forecasting

EXTENSIONS OF STOCHASTIC MULTIPATH TRIP ASSIGNMENT TO TRANSIT NETWORKS

Horowitz, Alan J.

Transportation Research Record 1987 p 66-72

 

Abstract: The procedure extends an existing traffic assignment algorithm by (a) establishing strict criteria for transit zone definition; (b) using a comprehensive measure of disutility of transit trips; and (c) reconstructing the transit network so that all passenger movements are explicitly represented. The assignment procedure was tested on a large section of the Milwaukee County Transit System, which was specifically chosen to reveal any undesirable properties in the procedure. The assignment procedure was found to be free of those problems previously associated with applications of stochastic multipath traffic assignment in automobile networks.  14 Refs.

 

 

IMPLEMENTATION OF SERVICE-AREA CONCEPTS IN SINGLE-ROUTE RIDERSHIP FORECASTING

Horowitz, Alan J.; Metzger, David N.

Transportation Research Record n 1037 1985 p 31-39

 

Abstract: How the service-area concept was incorporated into the Transit Ridership Forecasting Model is explained and the application of the concept on routes selected from the Milwaukee urban area is illustrated. It is shown that use of the service-area concept removes a serious tendency for four-step models to underestimate ridership on marginal routes. It is also shown that proper application of the service-area concept can reduce both computation time and data requirements.  16 Refs.

 

 

DESIGN OF A SINGLE-ROUTE RIDERSHIP FORECASTING MODEL

Horowitz, Alan J.

Transportation Research Record 1984 p 112-117

 

Abstract: Travel demand estimation is considered an integral part of transportation planning. The transit ridership forecasting model (TRFM) has been designed to overcome many of the serious obstacles to implementation presented by previous methods of forecasting ridership on a single route. TRFM simplifies, optimizes, and repackages conventional ridership forecasting techniques to make the job of the planner as easy as possible. The model exploits the advantages of a popular, modest-sized microcomputer (e. g. , animated color graphics). 12 Refs.

 

 

SUBJECTIVE VALUE OF TIME IN BUS TRANSIT TRAVEL

Horowitz, Alan J.

Transportation v 10 n 2 Jun 1981 p 149-164

 

A scaling technique, magnitude estimation, is used to rate time spent on various elements of bus transit trips. Relative values of time are found for in-vehicle portions of trips, walking, waiting and transferring. Because magnitude estimation produces a ratio scale, results can be directly incorporated into modal choice analyses, route planning and evaluation procedures where monetary values of time are not necessarily required. 13 Refs.

 

 

TRANSFER PENALTIES: ANOTHER LOOK AT TRANSIT RIDERS' RELUCTANCE TO TRANSFER

Horowitz, Alan J.; Zlosel, Dennis J.

Transportation v 10 n 3 Sep 1981 p 279-282

 

Abstract: Two on-board surveys were conducted to determine how transit riders perceive transfers. The surveys were conducted before and after the imposition of a transfer in the middle of an existing bus route. Results of the surveys showed that riders perceive bus transit trips as significantly worse when the trip requires a transfer, even if transfer time is neglibible. 2 Refs.

 

 

TRANSIT RIDERSHIP FORECASTING - A MICROCOMPUTER APPLICATION

Horowitz, Alan J.

Transportation Quarterly v 38 n 2 Apr 1984 p 297-310

 

Abstract: This article describes the advantages and disadvantages of implementing transit ridership forecasting techniques on microcomputers, taking into account necessary and desirable simplifications and modifications. One model, which has been successfully implemented on a microcomputer, will be presented as an indication of the considerations involved. Forecasting transit ridership can be accomplished by a variety of methods of varying degrees of sophistication. For example, elasticities have been estimated between ridership and a large number of travel factors including fare and headway. With these elasticities, the effects of fare and headway changes on an existing route can be quickly and roughly estimated. Other techniques, which assume ridership is proportional to the number of dwelling units in the service area, enable calculation of impacts of route extensions.  5 Refs.

 

 

SIMPLIFICATIONS FOR SINGLE-ROUTE TRANSIT-RIDERSHIP FORECASTING MODELS  Horowitz, Alan J.

Transportation v 12 n 3 May 1984 p 261-275

 

Abstract: The growth in popularity of microcomputers has reemphasized the need for simplified transit-planning techniques. This paper describes and evaluates a single-route ridership forecasting model which is designed to fit within a modest-sized microcomputer. The model is based upon the traditional four-step urban transportation modeling process, but it is simplified by removing the possibility of multiple transfers and by eliminating the highway network. An analysis of model error shows that these simplifications do not appreciably affect the accuracy of the forecasts. A particular advantage of implementing the model on a microcomputer is the user-friendliness that can be achieved by employing interactive color graphics for data input.

Land Use Forecasting

EXPERT PANEL METHOD OF FORECASTING LAND USE IMPACTS OF HIGHWAY PROJECTS

Mulligan, Patricia M.; Horowitz, Alan J.

Transportation Research Record 1986 p 9-15

 

Abstract: The validity of expert panel forecasts of land use impacts of highway projects in small urban areas was evaluated. A panel was assembled consisting of individuals with backgrounds in different aspects of land use and forecasting. This panel of experts was asked to predict the changes that have occurred over the past 20 years from a 1965 perspective. The panel received information on each of the two case study cities, as well as brief descriptions of the projects. The forecasting instrument consisted of a map for each city and a questionnaire to elicit evaluations of 31 features of community development. Each feature was rated as to whether an impact would occur, whether the impact was negative or positive, and the magnitude of the impact and its importance. On the map the panelists predicted the areas in which residential, retail, service, and industrial impact would occur.  10 Refs.

Modeling in General

PURE PROBABILISTIC MODEL OF TRIP TOURS BY AUTOMOBILE

Horowitz, Alan J.

Transportation Planning and Technology v 7 n 2 1982 p 99-108

 

Abstract: A mathematical model of automobile trip tours is presented. Within a framework of eight restrictions on automobile trip making, all travel behavior is assumed random and all of the ways in which tours can be arranged are assumed equally likely. Three probability distributions are derived from the model: the probability that a household makes a given number of tours in a day; the probability that a household makes a given number of trips in a day; and the probability that a tour reaches a given number of destinations. 9 Refs.

Benefits Assessment

ASSESSING TRANSPORTATION USER BENEFITS WITH MAXIMUM TRIP LENGTHS

Horowitz, Alan J

Transportation Planning and Technology v 6 n 3 1980 p 175-182

 

Abstract: The author shows how maximum time lengths of urban trips can be used to calculate major user benefits of transportation improvements. These benefits include those associated with reduced cost of travel, those associated with travel to additional destinations, and those associated with improved quality of destinations. Maximum trip lengths can be measured using techniques of psychological scaling. 12 Refs.

 

 

METHODS AND STRATEGIES FOR TRANSIT BENEFIT MEASUREMENT

Horowitz, Alan J.; Beimborn, Edward

Transportation Research Record n1496 Jul 1995 National Research Council Washington DC USA p 9-16

 

Abstract: Benefit assessment is done to make decisions, and a general discussion is given of how to view benefits for that purpose. Benefit assessment practices from many agencies in the United States are described. Agencies' reported benefits and their use of benefit measures in actual practice are compared. The political environment surrounding transit decisions was found to have a major effect on procedures that are adopted for benefit analysis. The paper also shows how consequences of transit can be illustrated through the use of a benefit tree, which allows planners to show how transit service provides an alternative means of travel, results in changes of trip making by automobile and transit, affects land-use activity, and leads to direct and indirect employment. Approaches are described for quantifying benefits. As an example, a method is presented for calculating the enhanced consumer surplus as a broad measure of user benefits of a project alternative. Recommendations are made on how to effectively use benefit measures for selection of project alternatives within a political decision-making environment. 3 Refs.

 

 

ASSESSING USER BENEFITS OF TRANSIT SYSTEM IMPROVEMENTS WITH SPATIALLY VARYING DEMANDS

Horowitz, Alan J.

Transportation Research Record n1452 Dec 1994 National Research Council Washington DC USA p 1-9

 

Abstract: Transit planners recognize that spatially varying demands affect the assessment of transit system alternatives. However, they do not yet possess the tools necessary to properly determine the effects of the variation on estimates of user benefits. An extended measure of user benefits that is consistent with net consumer surplus from classical economic theory is presented. Also presented is the structure of a travel forecasting model that can show the effects of activity allocation, trip distribution, and route choice on net consumer surplus. Individual components of the model have already been extensively tested in practice and are described in the academic literature, but the transit ridership properties of the model, as a whole, have not been established. The model is capable of finding a joint equilibrium solution between activity allocation, mode split, trip distribution, and traffic assignment. Tests of the model on real networks indicate that spatial redistribution of activities resulting from a transit service improvement can be large enough to determine whether the improvement should be implemented. 15 Refs.

 

 

ESTABLISHING RELATIONSHIPS BETWEEN PAVEMENT ROUGHNESS AND PERCEPTIONS OF ACCEPTABILITY

Garg, Arun; Horowitz, Alan; Ross, Fred

Transportation Research Record n 1196 1988 p 276-285

 

Abstract: A psychological scaling experiment was conducted in Wisconsin to establish relationships between pavement roughness and users' perceived need to improve the road. A total of 32 road segments were selected for user evaluation. Except for their surface, they had very similar characteristics (speed limit, length, terrian, traffic volumes, scenery, etc.). Physical roughness was measured with both a response-type instrument (roadmeter) and a profilometer. Fifty paid subjects were selected randomly from the general population. They were asked to rate ride quality on both the traditional Weaver/AASHO categorical scale and on a newly designed magnitude estimation scale. In addition, subjects were asked, using a Likert scale, about their willingness to resurface and were asked to estimate the amount of extra time they would be willing to spend to avoid a particular segment, considering its roughness. The experiment yielded several useful mathematical relations between physical roughness and users' willingness to resurface. It was found that the magnitude estimation scale was preferable to the Weaver/AASHO scale for measuring subjective roughness. Surprisingly the roadmeter was better than the profilometer for measuring physical roughness.  8 Refs

Traffic Assignment and Delay

INTERSECTION DELAY IN REGIONWIDE TRAFFIC ASSIGNMENT: IMPLICATIONS OF 1994 UPDATE OF THE HIGHWAY CAPACITY MANUAL

Horowitz, Alan J.

Transportation Research Record n 1572 Nov 1996 National Research Council Washington DC USA p 1-8

 

Abstract: The original 1985 Highway Capacity Manual (HCM85) described widely recognized relationships for traffic delay that could be incorporated into travel forecasts. Applications of the HCM85 procedures demonstrated that such delay relationships were both technically feasible and beneficial. In early 1995, the Transportation Research Board released the 1994 update to the HCM (HCM94), heavily revising the signalized and two-way stop intersection procedures and introducing a detailed all-way stop intersection procedure. These new procedures have the potential to improve the accuracy of forecasts and to make forecasts consistent with other design practices. Implementation of the HCM94 procedures into travel forecasts reveals that fewer adjustments are required to make them work within equilibrium traffic assignments. The two-way stop procedure can be used nearly intact. The signalized intersection procedure, although still requiring some adjustments, allows a greater range of traffic conditions and phasing options. The all-way stop procedure cannot be incorporated into travel forecasts because of its restrictions on allowable volumes and turning movements. Tests of the HCM94 procedures in traffic assignments indicate that they produce noticeably different results (both volumes and link delays) than the original HCM85 procedures. Multiple equilibrium solutions are possible, but the differences between these solutions are small and manageable. 12 Refs.

 

 

REVISED QUEUEING MODEL OF DELAY AT ALL-WAY STOP-CONTROLLED INTERSECTIONS

Horowitz, Alan J.

Transportation Research Record n 1398 1993 p 49-53

 

Abstract: An existing M/G/I queueing model of delay at all-way stop-controlled intersections is updated to reflect recent empirical evidence of driver behavior. In particular, the queueing model has been expanded to account for turning and coordination between drivers who get priority at the same time. An exact expression for an important variable in the model, the variance of service time, could not be found. However, an approximate expression could be verified using a Monte Carlo simulation. A queueing model can produce values of delay that are similar to strictly empirical models, but that apply to a much greater range of intersection volumes. 4 Refs

 

 

CONVERGENCE PROPERTIES OF SOME ITERATIVE TRAFFIC ASSIGNMENT ALGORITHMS

Horowitz, Alan J.

Transportation Research Record n 1220 1989 p 21-27

 

Abstract: This paper examines the convergence properties of four popular traffic assignment algorithms: Frank-Wolfe decomposition for fixed-demand equilibrium assignment, and ad hoc variation of the Evans algorithm for elastic-demand equilibrium assignment, fixed-demand incremental assignment, and elastic-demand incremental assignment. The algorithms were evaluated according to errors associated with insufficient iterations, arbitrary selection of starting point, inexact theory, and small variations in data. Each of the four algorithms reached its intended solution, but did so very slowly. Elastic-demand incremental assignment emerged as the preferred technique, principally because of its more accurate response to small variations in data and its adaptability to various models of travel demand. 9 Refs.

Intermodal Systems

GENERIC OBJECTIVES FOR EVALUATION OF INTERMODAL PASSENGER TRANSFER FACILITIES

Horowitz, Alan J.; Thompson, Nick A.

Transportation Research Record n1503 Jul 1995 National Research Council Washington DC USA p 104-110

 

Abstract: A list of generic objectives is a tool for initiating the evaluation process for project alternatives for an intermodal passenger transfer facility. Such a list should contain all objectives that might be important to any project. This paper presents a list of 70 objectives developed through a literature review and through interviews with users. Each objective on the list was rated by a panel of experts on transportation planning and station design. An analysis of the ratings revealed that most important were objectives for ensuring safety and security and objectives for improving transfers and transfer opportunities. Less important were objectives relating to the environment and to finance. Architectural, building, and site considerations were rated as least important. 5 Refs.

 

 

PERFORMANCE MEASURES FOR INTERMODAL PASSENGER FACILITIES

Horowitz, Alan J.; Thompson, Nick A.

Transportation Congress, Proceedings Proceedings of the Transportation Congress. Part 2 (of 2) Oct 22-26 1995 v2 1995 San Diego, CA, , USA, Sponsored by : ASCE ASCE New York NY USA p 1529-1542

 

Abstract: An expert panel was formed to rate 70 generic objectives for the design and operation of intermodal passenger transfer facilities. The panel indicated that a good facility should pay particular attention to issues of transportation efficiency. Four methods of measuring transportation efficiency are presented, which address the issues of facility location, access priorities, modal interaction and user benefits.