Intelligent Parking-Guidance System Based on Connected-Vehicle Sensor Networks
Professor of Industrial and Operations Engineering, College of Engineering and Adjunct Professor of Law, Law School
Research Associate Professor, University of Michigan Transportation Research Institute, Associate Professor of Industrial and Operations Engineering, College of Engineering and Associate Professor of Public Policy, Gerald R Ford School of Public Policy
Real-time parking-occupancy information is critical for a parking management system to enable drivers to park more efficiently. Recent advances in connected and automated vehicle (CAV) technologies allow sensor-equipped cars (probe cars) to detect and broadcast available parking spaces when moving through parking lots. In this paper, we evaluated the impact of market penetration of probe cars on system performance, and investigated different parking-guidance policies to improve the data-acquisition process. We adopted a simulation-based approach to impose four policies on an off-street parking lot, influencing the behavior of probe cars in parking in assigned spaces and, in turn, the scanning route and the parking-space-occupancy estimations. The last policy we proposed is a near-optimal guidance strategy that maximizes the information gained of posteriors. The results suggested that an efficient information-gathering policy can compensate for low penetration of CAVs. We also highlighted the policy trade-offs that occur while attempting to maximize information gained through explorations and improved assignment accuracy through exploitations.
Studied an infrastructure-independent intelligent parking system based on impacts of CAV technologies, with results that can assist urban policymakers in designing and managing smart parking systems.