| To improve transport efficiency and reduce the workload of formulating a stowage plan,the vehicle arrangement optimization of PCTC for single-deck on multi-port voyages was researched,and heuristic algorithms were adopted to formulate the optimized loading plan step by step.The following works were carried out:(1)Under the constraints of model accuracy and algorithm complexity,the coordinate grid model of the deck and vehicles was established according to the size of deck,the shape of deck,obstacles,lashing holes and the preloaded vehicle information;(2)With space constraints and the vehicle size as the criterion of greedy selection,GRASP(Greedy Randomized Adaptive Search Procedures)algorithm was used to insert the vehicle into the deck to generate the initial stowage plan by dynamically adjusting the search direction and the length of the candidate list;(3)The path length was measured by the vehicle-occupied grids,and the shifting cost was calculated by Dijkstra algorithm;(4)ALNS(Adaptive Large Neighborhood Search)algorithm was used to improve the stowage plan by dynamically selecting strategies for continuous destruction and repair.Taking "VIKING CONQUEST",which is designed by SDARI as an example,the main cargo deck is selected to formulate the stowage plan of a multi-port voyage.Comparing the stowage plan with 7 manual stowage plans,the shifting cost of the former is 109m2 whereas the latter is at least 230m2.It’s turned out that the stowage plan generated by proposed algorithm has a lower shifting cost,which is helpful to improve the transportation efficiency.Meanwhile,the algorithm used in this paper can also reduce the workload of formulating a stowage plan manually. |
