Rolling-horizon approach in solving dynamic multisize multi-trailer container drayage problem. (1st September 2022)

• Record Type: Journal Article
• Authors: Bjelić, Nenad; Vidović, Milorad; Popović, Dražen; Ratković, Branislava
• Is Part Of: Expert systems with applications. Volume 201(2022)
• Publication Date: 2022-09-01
• Subjects: Dynamic container drayage problem -- Multisize multi-trailer container vehicle -- Variable neighborhood search -- Intermodal transportation; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2022.117170
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: We consider the dynamic multisize multi-trailer container drayage problem. The problem is mathematically formulated as MILP. We propose heuristic algorithm based on the VNS metaheuristic framework. Numerical results show effectiveness of proposed algorithms. We tested effects of different rolling horizon parameters and strategies. Abstract: We consider the multisize multi-trailer drayage problem in which dynamism is treated by the re-optimization concept. Information on vehicles' current positions and their distances to all task locations are available in real-time. All vehicles in the fleet can be of longer combination sizes. The results of this study provide an efficient solution approach for solving real-size problem instances in a dynamic surrounding of container drayage processes. The solving approach is based on the implementation of a variable neighborhood search based heuristic algorithm. The testing results confirmed the benefits of using longer combination size vehicles by indicating potential reductions in overall traveled distances for up to approximately 30%, and in a total number of required trips for up to approximately 40%. Eventually, we tested the effects of implementing an assignment strategy in which an active task can be taken away from the assigned vehicle before it starts container (un)loading at the task's premise. That strategy proved to reduce tasks' lateness aspect of the problem but on the cost of increased traveling distances.