Developing a fuzzy-neuro travel demand model (trip distribution and mode choice)

Abstract

Various methods are currently used in travel demand modelling (TDM), for example, the Four-Step model, which is widely used and is perhaps the most famous one, Discrete Choice Models, Fuzzy Set Theory and the Neural Network Approach. The emergence of these different methods is due to, for instance, different areas having different problems. Hence, a method successfully applied in one area could be unsuitable for use in others. The literature suggests misused in travel demand modelling process could result in errors up to 60 per cent. The sources of errors are not only from a lack of information related to the parameters that the model tries to estimate but also due to the absence of sharply defined criteria of class membership that can play important roles in human thinking, for which qualitative variables may be better representations. Fuzzy Set Theory is suggested in this study as one approach to tackle the computation of such variables. The Neural Network Approach has a unique ability which, it is claimed, can capture unseen or hidden relationship among the spatial interaction data. A new method, called here the Fuzzy-Neuro approach, is proposed for modelling travel demand, focusing on trip distribution and mode choice, with the expectation that it can improve the accuracy of the resulting models.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 4 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.