Abstract
Most traditional mode choice models are based on the principle of random utility maximization derived from econometric theory. Alternatively, mode choice modeling can be regarded as a pattern recognition problem reflected from the explanatory variables of determining the choices between alternatives. The paper applies the knowledge discovery technique of rough sets theory to model travel mode choices incorporating household and individual sociodemographics and travel information, and to identify the significance of each attribute. The study uses the detailed travel diary survey data of Changxing county which contains information on both household and individual travel behaviors for model estimation and evaluation. The knowledge is presented in the form of easily understood IF-THEN statements or rules which reveal how each attribute influences mode choice behavior. These rules are then used to predict travel mode choices from information held about previously unseen individuals and the classification performance is assessed. The rough sets model shows high robustness and good predictive ability. The most significant condition attributes identified to determine travel mode choices are gender, distance, household annual income, and occupation. Comparative evaluation with the MNL model also proves that the rough sets model gives superior prediction accuracy and coverage on travel mode choice modeling.
Highlights
Within the transportation field there exists many informative and detailed datasets that reveal a great deal about the travel behavior of households and individuals
Many of these models suffer from the property of independence of irrelevant alternatives (IIA), which implies that the effects attributes of an alternative are compensatory and result in biased estimates and incorrect predictions in cases that violate the IIA property [2], significant improvements on eliminating the IIA property have been made
This paper has demonstrated the successful application of a relatively new technique in the area of knowledge discovery to the well-studied problem of understanding and predicting traveler’s mode choices
Summary
Within the transportation field there exists many informative and detailed datasets that reveal a great deal about the travel behavior of households and individuals. Since the multinomial logit (MNL) model [1] was developed in the 1970s, the parametric model family including different logit models with different structures and components has become the most widely used tool for mode choice analysis Many of these models suffer from the property of independence of irrelevant alternatives (IIA), which implies that the effects attributes of an alternative are compensatory and result in biased estimates and incorrect predictions in cases that violate the IIA property [2], significant improvements on eliminating the IIA property have been made. Computational Intelligence and Neuroscience can suggest the relationship between variables it contains using as few probability assumptions and linear structural relationships as possible This information is usually contained in a series of rules that when they are evaluated to be true suggest a definite outcome. The primary objectives of this paper include (a) investigating the capability and performance on mode choice modeling of urban diary travel using rough sets theory, (b) figuring out the significance of condition attributes on mode choices, and (c) to comparatively evaluating the performance of rough sets model and MNL model
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have