改进的蚁群算法求解多目标车间作业调度问题  

An Improved Ant Colony Algorithm for MultiObjective Job-Shop Scheduling Problem

作　　者：王丽红[1] 倪志伟[2] 高雅卓[1] 

机构地区：[1]合肥工业大学管理学院,安徽合肥230009 [2]合肥工业大学过程优化与智能决策教育部重点实验室,安徽合肥230009

出　　处：《计算机技术与发展》2008年第10期49-52,共4页Computer Technology and Development

基　　金：国家自然科学基金重点项目(70631003);安徽省教育厅科研项目(2006sk010);国家863计划(2007AA04Z116)

摘　　要：目前已经有许多解决作业车间调度问题的启发式求解方法,但这些方法多数局限于单目标,因此不能满足现实生活中多目标作业车间调度问题的应用需求。提出一种改进的蚁群算法启发式地搜索多目标车间作业调度问题的近似最优解以满足实际的应用需求。通过对转移概率以及信息素更新方式进行改进,并融合交叉策略,确保算法在加快搜索收敛速度的同时又避免陷入局部最优。仿真实验证明,改进的算法具有较好的性能,能够解决实际生活中的多目标作业车间调度问题。Several heuristic approaches have been proposed to solve the job-shop scheduling problem （JSSP）. But most of them are limited to single objective and fail in real-world applications, which naturally involve multiple objectives. Presents an improved ant colony algo-rithm for solving multi-objective JSSP, searching for near-optimal solutions heuristically and optimizing multiple objectives simultaneously. By improving the transfer probability,pheromone update mode,and combining of cross-optimum to make sure that this algorithm can accelerate convergence rate and avoid falling into local optima simultaneity. Simulation results show that the proposed method is effective, and succeed in solving JSSP of multiple objectives.

关 键 词：多目标优化 作业车间调度 蚁群算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]