基于NSGA-II的多目标设备动态布局方法

对多个计划期内需求可预测的车间动态设备布局问题进行了研究.针对这一多目标、多约束的问题,以物流搬运和重布局费用之和、非物流关系以及面积利用率作为优化目标,将动态布局问题转化为重布局过程和多个子计划期的静态布局问题,构建了针对不等面积设备的动态多期布局问题的连续型多目标优化模型.采用带精英策略的非支配遗传算法(Non-dominated Sorting Genetic Algorithm II,NSGA-II)进行求解,克服了传统加权法求解多目标问题时加权系数难以确定和无法保证多目标同时优化的缺点,求解得到Pareto解集,供决策者根据企业实际情况优中选优.通过实例验证了本方法的有效性.

NSGA-Ⅱ-enabled multi-objective dynamic facility layout method

Firstly,a dynamic workshop facility layout is conducted on multiple-planning-period demand forecasting.For the multi-objective and multi-constraint problem,the logistics transport cost,reallocation cost,non-logistics relation and area utilization are treated as optimization objectives.By transferring the dynamic layout problem （DLP） into the static layout problem （SLP） with reallocation process and multiple sub-periods,a continuous multi-objective optimization model is then established based on unequal-area multi-period dynamic facility layout.Next,the non-dominated sorting genetic Algorithm Ⅱ （NSGA-Ⅱ）using elitist strategy is used to eliminate the weakness of traditional weighted average method for weight coefficient determination and multi-objective optimization.Afterwards,the optimal solution for decisionmakers is obtained from Pareto solution-set.Finally,the effectiveness of this method is verified.