Title: CAPTURING INCOME EFFECTS ON TRANSIT MODE CHOICE AND ASSIGNMENT
Abstract: Many travel demand modeling platforms ignore or distort the income effects in the modeling of transit mode choice and assignment. The fundamental reason for this is that the maximum utility transit path for a particular origin, destination, and income class may not in fact be the shortest time path. This problem can be serious when transit options with higher fares and lower travel times (e.g., commuter rail and some high-end express bus services) are modeled in the same alternative with lower fare/higher travel time options (e.g., a standard, urban fixed route service). Working around this problem is difficult, especially with travel demand modeling software which is limited to single shortest path, all-or-nothing transit assignment, or where the basic mode choice model does not explicitly account for these differences in transit service. This paper describes a straightforward approach to solving this problem. The approach includes three basic steps, using a standard multinomial logit transit mode choice model including an income class variable, and a MINUTP model platform. The first step is splitting the transit network into two pieces, based on fares and level of service distinctions (e.g., standard fixed-route transit and commuter rail). The second step is tallying the source network for the maximum transit utility for each origin, destination and income class. The third step is splitting the final transit trip tables, and assigning trips to the correct (i.e., maximum utility) transit network. This approach would be of interest to travel demand modelers, transportation analysts and planners evaluating higher fare/lower travel time transit options such as commuter rail and deluxe express bus service. The paper presents the results of using this approach for a test network in the Sacramento metropolitan area. The results include: a comparison of a test analysis using the approach, with a standard approach not accounting for income effects; an accounting of the extra network development, computation, and analysis time in using the approach; and a qualitative assessment of the value of the approach. The approach was developed for use in a corridor study, which will include analysis of commuter rail options.
Publication Year: 2002
Publication Date: 2002-01-01
Language: en
Type: article
Access and Citation
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot