具有外包选择的无等待流水线干扰修复模型  

A Disruption Recovery Model for No-Wait Flow Shop with Outsourcing Option

作　　者：薄洪光[1] 张鑫[1] 潘裕韬[1] 

机构地区：[1]大连理工大学运营与物流管理研究所,辽宁大连116023

出　　处：《系统管理学报》2015年第4期485-495,共11页Journal of Systems & Management

基　　金：国家自然科学基金重点项目(61034003);国家科技支撑计划资助项目(2015BAF08B02;2013BAF02B03);国家863计划资助项目(SS2012AA040909);教育部人文社会科学研究青年基金资助项目(11YJC630005);中国博士后科学基金特别资助和面上项目(201104592;20100481222)

摘　　要：针对双机成比例无等待流水线环境下最小化完工时间和的调度问题,研究如何基于干扰管理理论和采用作业外包途径来应对机器干扰事件。在证明最短加工时间优先(SPT)最优解定理的基础上,同时考虑最小化工件完工时间和指标(初始调度目标)与最小化工件滞后时间和指标(偏离最小目标),构建了基于SPT规则的干扰修复0-1整数规划模型,提出了基于差分进化全局搜索策略与"插入-交换"邻域搜索机制相结合的多目标混合智能算法。数值实验结果表明,本文提出的机器干扰条件下外包修复模型及算法是有效的。We present a novel disruption recovery method based on disruption management and production outsourcing to deal with machine disruptions for the scheduling problem with the scheduling objective of minimizing the sum of completion-time in the case of proportional two-machine no-wait flow shop.By proving the shortest processing time（SPT）rule producing an optimal solution,we build a disruption recovery 0-1integer programming model based on SPT rule,with the objective of minimizing the sum of completion-time（the original objective）and the objective of minimizing the sum of delay-time（departure revising objective）,respectively.Combining differential evolution global search strategy and ＂insert-swap＂neighborhood structure based on local search mechanism,we propose a heuristic hybrid multi-objective evolutionary solution algorithm.The numerical experiments show that the production outsourcing based on disruption recovery model and its hybrid DE algorithm are effective in condition of machine disruptions.

关 键 词：干扰管理 差分进化算法 无等待流水线 作业外包 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] O223[自动化与计算机技术—控制科学与工程]