Cruising-for-parking is a common phenomenon in many urban areas worldwide. Properly understanding and mitigating cruising can reduce travel times, alleviate traffic congestion, and improve the local environment. Most of the existing studies estimating cruising traffic are based on empirical data and/or detailed simulation models. Both approaches have large data requirements, and the detailed simulation models tend to have high computational costs. In this paper, we present a case study of an area within the city of Zurich, Switzerland, using a recently proposed macroscopic model to analyze the current conditions of cruising-for-parking. The results are validated with empirical data. The macroscopic model, inspired by a bottleneck model, reproduces the dynamics of both, the parking and the traffic system, as well as their interactions. As such, it calculates the delays encountered by drivers while waiting for parking, and the impact of such delays on the overall traffic stream, which involves not only the searching traffic but also the through traffic. It is shown that the macroscopic parking model could, additionally, incorporate the data generated by agent-based models, cooperatively producing valid and trustworthy results of cruising estimations, while requiring comparatively few data inputs and relatively low computational costs. The study shows that in a small area of Zurich (0.28 km2) with a demand of 2687 trips in a typical working day, cruising-for-parking generates 83 h of additional travel time and 1038 km of additional travel distance. Surprisingly, the worst conditions are observed at noon, corresponding to a maximum number of 30 searchers with an average search time of 13 min. Additionally, four types of parking policies are discussed, and their potential impacts on traffic performance are either quantitatively or qualitatively illustrated. The four policies include: the adjustment of the parking supply, the adjustment of parking time controls, the adoption of dynamic parking charges, and the provision of parking forecasts.
Read full abstract