• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International Conference on Mechanical Engineering and Mechanics vol.2; 20051026-28; Nanjing(CN) >Genetic Algorithm Approach for a New Nonlinear Goal Programming Model of FLP
【24h】

Genetic Algorithm Approach for a New Nonlinear Goal Programming Model of FLP

机译：一种新的FLP非线性目标规划模型的遗传算法方法

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Layout generation of facility layout problem (FLP) is a NP-hard optimization problem with multi-objective. Genetic algorithm (GA) is a powerful AI technique for solving multi-objective programming (MOP) model. Because of the discrepancy in the magnitude of different goals, the real intention of critical evaluation in GA for solving many kinds of MOP model of layout generation was distorted. When using GA to solve nonlinear goal programming (NLGP) model of FLP, the value of achievement function z is very important for fitness evaluation. However, in existent NLGP model, z is only the sum of weighted deviation variables and has no important meaning. For weighting method model of FLP, normalization of fitness values can deal with evaluation distortion in a way but act after the event. For multiplicative & divisional method model of FLP, the discrepancy was not considered. All these models consequentially distorted evaluation. To improve it, a NLGP model was put forward, in which the traditional deviation variables d~+ and d~- representing amount of deviation were replaced by the deviated degree variable d. This NLGP model can solve the problem beforehand. Because NLGP model is the general expression of goal programming (GP), the improvement of NLGP can also be applied in GP. The new NLGP/GP model is simple, efficient and meaningful. Then, a NLGP model of FLP and corresponding GA were put forward. This FLP model considers the transportation cost and many kinds of adjacency requirement. It can handle unequal area and fixed facilities too. The GA used real-number encoding, rejection strategy, heuristic crossover operations, shift mutation, and ( μ + λ ) selection. The result testifies the efficiency of the NLGP model of FLP and GA.
机译：设施布局问题（FLP）的布局生成是一个具有多目标的NP硬优化问题。遗传算法（GA）是解决多目标规划（MOP）模型的强大AI技术。由于不同目标的大小存在差异，因此在遗传算法中进行关键评估以解决多种布局生成的MOP模型的真实意图遭到了扭曲。当使用GA求解FLP的非线性目标规划（NLGP）模型时，成就函数z的值对于适应性评估非常重要。但是，在现有的NLGP模型中，z只是加权偏差变量之和，没有重要意义。对于FLP的加权方法模型，适应度值的规范化可以以某种方式处理评估失真，但在事件发生后起作用。对于FLP的乘法和除法方法模型，未考虑差异。所有这些模型都会导致评估失真。为了对其进行改进，提出了一种NLGP模型，其中用偏差度变量d代替了代表偏差量的传统偏差变量d〜+和d〜-。该NLGP模型可以预先解决该问题。因为NLGP模型是目标编程（GP）的通用表示，所以NLGP的改进也可以应用于GP。新的NLGP / GP模型简单，高效且有意义。然后，提出了FLP的NLGP模型和相应的遗传算法。该FLP模型考虑了运输成本和多种邻接要求。它也可以处理不等面积和固定设施。 GA使用实数编码，拒绝策略，启发式交叉操作，移位突变和（μ+λ）选择。结果证明了FLP和GA的NLGP模型的效率。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号