Abstract
A new hybrid sensor technology integrating existing Weigh-In-Motion (WIM) axle configuration data combined with inductive signature data obtained from advanced Inductive Loop Detectors (ILDs) is gaining interest due to its potential to provide detailed classification of truck body types as well as anonymous tracking of truck movements on freeways. This paper investigates two proposed strategies for optimally deploying this new technology on California freeways based on actual truck GPS trajectories: (1) A flow-interception approach to maximize the total amount of net origin-destination (OD) flows captured using integer programming; and (2) A truck re-identification approach to maximize insights into origins and destinations of sampled truck trips, as well as routes of those trips using Genetic Algorithm. The flow-interception model is capable of selecting locations emphasizing different body types with flow-based weight factors. The truck re-identification model investigates the best locations to identify heavy truck movement by selecting pairwise locations, and is shown to be sensitive to the re-identification performance uncertainty.