作者: T.A. Arentze , H.J.P. Timmermans
DOI:
关键词: Albatross 、 CHAID 、 Machine learning 、 Empirical research 、 Situational ethics 、 Decision tree 、 Operationalization 、 Heuristics 、 Process modeling 、 Engineering 、 Artificial intelligence
摘要: This paper describes the conceptual development, operationalization and empirical testing of Albatross: A Learning Based Transportation Oriented Simulation System. activity-based model activity-travel behavior is derived from theories choice heuristics that consumers apply when making decisions in complex environments. The model, one most comprehensive its kind, predicts which activities are conducted when, where, for how long, with whom, transport mode involved. In addition, various situational, temporal, spatial, spatial-temporal institutional constraints incorporated model. decision tree proposed as a formalism to represent an exhaustive set mutual exclusive rules each step CHAID induction method used derive trees activity diary data. case study develop test indicates performance very satisfactory. We conclude therefore methodology this article useful computational process models behavior.