An Adaptive Parameter Controlled Ant Colony Optimization Approach for Peer-to-Peer Vehicle and Cargo Matching

With the development of the sharing economy, a lot of vehicle-cargo matching service platforms have been built to reduce supply-request information asymmetry and enhance the use of idle vehicles. It is a critical but challenging task for a logistics service platform to create matches automatically between vehicles and cargos and achieve high vehicle utilization. Conventional vehicle-cargo matching such as one-to-one matching and one-to-many matching cannot recommend multiple vehicle and cargo sources for both drivers and shippers simultaneously. In this article, we investigate an intelligent approach to generate many-to-many vehicle-cargo matches and recommend them to the corresponding cargo owners and vehicle drivers. To this end, we develop a mathematical model for maximizing the platform's matching rate and matching profit and propose an innovative ant colony optimization based on parameter tuning to solve the matching problem. A clustering technology combining bacterial foraging chemotaxis with the k-means algorithm is used to judge the state of the ant colony, and the parameters are adjusted adaptively to make the algorithm converge rapidly to the neighborhood of the global optimal solution. We then apply the randomicity and ergodicity of chaos to adjust the parameters, aiming to jump out of local optimum. Numerical results show that the proposed algorithm can achieve recommendation results that are accurate and stable compared with those of some other search methods and can provide satisfactory solutions for vehicle drivers and cargo owners.