一种新型混合算法在混流装配调度中的应用  

Application of new hybrid algorithm in mixed model assembly scheduling

作　　者：娄高翔 蔡宗琰[1] 刘清涛[1] LOU Gaoxiang;CAI Zongyan;LIU Qingtao(School of Construction Machinery,Chang＇an University,Xi＇an 710064,China)

机构地区：[1]长安大学工程机械学院,西安710064

出　　处：《计算机工程与应用》2018年第16期254-259,共6页Computer Engineering and Applications

基　　金：国家自然科学基金青年科学基金项目(No.51305042);中央高校教育教学改革专项经费建设项目(No.16049;No.170501)

摘　　要：针对含有缓冲区的混流装配中同时存在的生产成本和库存成本问题,提出了一种基于遗传算法和差分进化算法的混合框架,并将其用于混流装配调度的实际问题中。通过融合遗传算法有效处理离散变量及差分进化算法有效处理连续变量的优点,在综合考虑降低生产成本和缓冲区库存的同时,兼顾了每个型号产品生产的顺序及数量。计算机仿真结果表明,与传统算法相比,该算法在混流装配调度上具有收敛速度快、优化能力强、算法可靠等优势。该混合算法可以显著改善多参数、高度非线性问题的优化结果,提高计算效率。This paper proposes a hybrid of Genetic Algorithm（GA）and Differential Evolution algorithm（DE）for actual mixed model assembly scheduling to address the production and inventory costs in mixed model assembly with buffer area.This hybrid algorithm comprehensively considers not only the reduction of production cost and the inventory at the buffer area but also the production order and quantity of each model of products by integrating the advantages of the GA in the effective processing of discrete variables and that of the DE in the effective processing of continuous variables.The computer simulation results demonstrate that the proposed hybrid algorithm has a higher convergence rate,stronger optimization ability,and better reliability in mixed model assembly scheduling than the traditional algorithms.This algorithm can also significantly improve the optimization and computational efficiency of multi-parameter and highly nonlinear problems.

关 键 词：差分进化 遗传算法 混流装配 调度 缓冲区 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]