Faster parking and less cruise for public parking spot discovery: Modeling and analysis based on Timed Petri Nets

Efficient parking spot search has been a hot research top over the past decades. Most published work is based on the assumption that inter-vehicular communication through VANETs is available for vehicles. Major concerns over VANETs are potential overwhelming messaging and compromise of drivers' privacy. In this paper, we present a new approach that requires only minimal communication between vehicles and parking service providers. Our goal is to help vehicles to get parked with the least amount of time. Therefore, we propose a new parking model based on Timed Petri Nets (TPNs) which considers the parking time, the amount of time spent in a parking lot waiting for an unoccupied spot, and the driving time, the amount of time spent on the road to the chosen parking lot. The proposed approach is low-cost, while modeling with TPNs brings several noticeable benefits: it supports simulation directly; it is a mathematical model so that one can perform timeliness analysis based on the model; it is scalable and refineable. Base on the proposed model, we demonstrate the real-time traffic and the real-time average parking time of a parking lot affect drivers' choice. Finally, we give out some simulations which demonstrate the effectiveness of our proposed model.