Inferring spatial–temporal attributes of vehicle itinerary with Automatic Vehicle Identification data: Methodology and application

Abstract

Daily itinerary, consisting of an individual’s trips and activities on a day, is usually fundamental input for many travel demand models. However, current research lacks effective methods to extract daily itineraries of large-scale samples for a long period. To this end, this study presents a methodology to Infer Daily Itineraries (IDI) of vehicles with Automatic Vehicle Identification (AVI) data. A problem-specific Probabilistic Graphical Model is constructed to define how possible one itinerary is true given its observed AVI data. To seek the most possible itinerary among vast feasible states, a candidate movement state generation algorithm and optimal itinerary searching algorithm are developed. Empirical studies have been conducted based on field-test data. Compared with two benchmarks, the proposed IDI improved the inference accuracy significantly even for activities with missing observations. Sensitivity analyses on the size of traffic area zone and data collection have also been performed, which can provide guidance for administrations and researchers on the partition of the study region and placement of the sensors. As the AVI system captures almost entire samples, vehicle movements inferred by IDI can provide a better representation of traffic patterns. This enables a series of applications related to transportation policy and practice. Traffic congestion tracking and parking demand estimation are introduced as two application examples.