A Hybrid Intelligent Approach for Co-Scheduling of Cascaded Locks With Multiple Chambers.

Abstract

A complex and typical scheduling problem in waterway transportation: co-scheduling of cascaded locks with multiple chambers (CCLM) is studied. Based on in-depth analysis of the problem properties, the CCLM is handled by separating it into three interconnected subproblems, each with a simpler structure and higher flexibility to be handled. The outer layer and inner layer concerns the sum of lockage number and ship placement, respectively. The interlayer as a connection bridge between the other two refers to lockage direction combination and timetable optimization which is a high-dimensional mixed integer optimization problem. To solve the CCLM problem, a hybrid approach based on iteration which mainly combines quantum inspired binary gravitational search algorithm and modified moth-flame optimization algorithm is proposed. In addition, two different scheduling rules which are usually concerned in practice, the area utilization maximization and first-come-first-served (FCFS) rule, are also tested in the CCLM problem. Experiments are conducted on instances that are extracted from real world data. The scheduling and comparison results verify that the CCLM problem can be well handled by the proposed method.