Developing a Fuzzy Logic Based Sampling Protocol for Transit Route Choice Modeling

Abstract

In the route choice modeling, a popular way of route set generation is to generate only a sample of routes, not the full set. Consequently, this approach can not use the valuable information regarding the characteristics of the routes. This study developed a sampling protocol by using fuzzy logic to sample from a full set of alternatives. The sampling protocol was compared with the simple random sampling protocol and found to be promising especially if the full choice set size is huge compared to the desired choice set size. Moreover, few explanatory variables related to the passenger’s strategic choice was extracted by generating the full route choice set. Later, these variables were used a..