Modeling Car Allocation Decisions in Automobile DeficientHouseholds

摘要

Computational process modeling has been introduced as an alternative approach tornutility-maximizing framework to deal with the complexity of activity-based models ofrntravel demand. ALBATROSS, a rule-based system, used data mining algorithms tornderive choice rules underlying activity-travel patterns. In the context of a project thatrnattempts to better include household as opposed to individual decision making into thernoriginal model, this paper describes the results for the car allocation decisions. ThernCHAID algorithm is applied to derive a decision tree for the car allocation decisions inrnautomobile deficient households using a large activity diary data set recently collected inrnthe Netherlands. The results show a satisfactory improvement in goodness of fit of therndecision tree model compared to the null model. The probability of the male getting therncar is considerably higher than the female getting the car in many condition settings. Inrnonly 16% of the condition settings, the female has the highest probability of getting therncar. Accessibility of the work location by car relative to slow mode appears to be thernmost influential factor when both male and female work.