A Graph-Based Scheme Generation Method for Variable Traffic Organization in Parking Lots

Abstract

To deal with the traffic congestion issues caused by the imbalance between supply and demand in parking lots, this study proposes a graph-based scheme generation method for variable traffic organization in parking lots. A graph-based methodological framework is developed to dynamically generate feasible traffic organization schemes and adapt the road networks of parking lots based on fluctuating demands. First, we design a parking lot-tailored enhanced primal approach by adding a directedness attribute while maintaining road continuity to ensure correspondence between generated graphs and traffic organization schemes. A graph generation algorithm is then designed to generate all feasible schemes in the scenario, deploying the depth-first search algorithm to check the connectivity of each graph and narrowing down feasible options based on domain knowledge. Finally, the initial parking space distribution and parking demand are used as inputs to calculate the total vehicle cruising time under each scheme, serving as the key indicator to select the optimal organization scheme. A single-level parking lot model is developed to verify the performance of our method under six initial parking space distributions. This model is built using the multi-agent simulation platform AnyLogic version 8.8.6, which enables the quick transformation of organization schemes by customizing the behavior of different agents. The results show that the optimal organization scheme generated by the proposed method can reduce vehicle cruising time by 15–46% compared to conventional traffic organization, varying according to parking space distributions. The significance of this study lies in its potential to mitigate traffic congestion in parking lots, thereby enhancing overall efficiency and user satisfaction. By dynamically adapting to fluctuating parking demands, this method provides a robust solution for urban planners and parking lot operators aiming to optimize traffic flow and reduce unnecessary delays.