矩形件排样问题的遗传算法求解

本文研究了求解矩形件正交排样优化问题的遗传算法。同时，将矩形件正交排样问题转化为一个排列问题，提出了求一个排列所对应的排样图的下台阶算法（改进的ＢＬ算法）将下台阶算法与遗传算法相结合，用于矩形件排样问题的求解，给出了该算法的实现。用该算法对文献中的两个算例进行了求解，结果表明该算法获得了比ＢＬ算法更好的解，是一种较为行之有效的方法。     

矩形件优化排样问题的混合遗传算法求解

利用遗传算法结合剩余矩形排样法求解矩形件正交排样问题。通过遗传算法将矩形件正交排样问题转化为一个排列问题，并引人剩余矩形排样算法来惟一确定每一个排列所对应的排样图（即排样方案），两者结合用于求解矩形件排样问题。最后用此混合遗传算法对文献[1]中的两个算例进行了验证，表明了其有效性。     

矩形件优化排样问题的混合遗传算法求解

利用遗传算法结合剩余矩形排样法求解矩形件正交排样问题。通过遗传算法将矩形件正交排样问题转化为一个排列问题,并引入剩余矩形排样算法来惟一确定每一个排列所对应的排样图(即排样方案),两者结合用于求解矩形件排样问题。最后用此混合遗传算法对文献[1]中的两个算例进行了验证,表明了其有效性。     

基于离散粒子群优化算法求解矩形件排样问题

改进了一种近似排样算法,并将改进的近似排样算法与离散粒子群优化算法结合求解矩形件排样问题.设计了应用离散粒子群优化算法求解矩形件排样问题的相关操作和定义,给出了离散粒子群优化算法求解矩形件排样问题的详细步骤,最后通过实验测试,验证了算法的有效性.     

基于蚁群优化算法求解矩形件排样问题

布局问题来源于生产实际,优秀的布局可以提高原料利用率,降低成本,提高经济效益,对许多行业有重要意义。矩形件优化排样是一类具有NP完全难度的组合优化问题。人工蚁群算法是对蚂蚁群体行为的模拟抽象,该算法具有分布计算、信息正反馈和启发式搜索等特点。本文将蚁群算法和剩余矩形法结合用于解决矩形排样问题,首先用蚁群算法将矩形件排样问题转化为一个排列问题;然后通过剩余矩形排样算法排出每一个排列所对应的排样图;最后用算法对文献[9]中的两个算例进行了验证,表明了其有效性。     

一种快速的有约束矩形件优化排样模型

为了有效地解决有约束的矩形件优化排样问题,提出一种快速的求解算法;通过比较待排样矩形件的不同排样模式,选择最优排样方案。算法完全基于解析计算,虽不能寻找理论最优解,但相比于各种启发式算法大大提高了排样速度。实验结果表明,算法能够在较短的计算时间内获得满意的排样效果,是一种效率较高的有约束矩形件排样算法。     

基于两段排样方式的矩形件优化下料算法

针对矩形件下料问题,提出一种基于两段排样方式的优化下料算法。首先构造一 种约束排样算法,生成矩形件在板材上的两段排样方式。然后采用列生成算法依据矩形件剩余 需求量迭代调用上述约束排样算法生成一个虚拟下料方案,按照不产生多余矩形件原则选取虚 拟下料方案中的部分排样方式加入到实际下料方案中,更新矩形件剩余需求量;重复上述步骤 直到矩形件剩余需求量为零。采用文献中基准例题将该算法与2 种文献算法进行比较,数值实 验结果表明该算法下料利用率比2 种文献算法分别高1.61%和0.78%。     

基于模拟退火剩余矩形算法的矩形件排样

针对矩形件优化排样问题,讨论了用模拟退火算法结合剩余矩形法求解问题。首先阐述了矩形件排样问题的数学模型,然后给出了模拟退火剩余矩形算法求解问题的步骤和方法,最后用实例进行了算法验证。实例分析表明,采用模拟退火剩余矩形算法求解矩形件排样问题是适合的。     

二维不规则零件排样问题的遗传算法求解

提出一种基于遗传算法求解二维不规则零件排样问题的方法，通过提取零件的最小包络矩形，将其转变为矩形件的正交排样问题，应用一种有效的解码算法－“最低水平线法”将编码转变为排样图。实例表明，该算法是有效的。     

基于遗传算法的大规模矩形件优化排样

大规模矩形件优化排样是一个典型的组合优化问题，属于NP-hard问题．实际工程中对一个排样方案一般有满足“一刀切”的工艺要求，“一刀切”要求增加了对排样的约束．提出的优化算法，将矩形匹配分割算法作为遗传算法染色体的解码器实现一个排样方案，用遗传算法进行排样方案的全局搜索．算例比较表明，该算法可以求得满足“一刀切”约束的最优解．     

基于重要度的矩形工件优化填充排样算法

大规模矩形件优化排样是一个典型的组合优化问题,属于NP2hard问题。矩形件优化排样已广泛应用于板材切割、瓷砖铺设、服装裁剪等行业。在实际排样工作中发现,决策者对工件的选择不仅要考虑大小、工件费用、铺设利用率等诸多因素,往往还需要考虑颜色、花式、铺设方式等因素。基于这种状况,引入排样属性重要度的概念,提出了基于重要度的矩形工件优化填充排样算法,使用计算机辅助排样。通过实例排样表明了该算法的有效性和实用性。     

一种求解二维矩形Packing问题的拟人型全局优化算法

针对二维矩形Packing问题,提出了基于占角动作的基本算法。以基本算法为基础,提出了三阶段优化的拟人型全局优化算法。在第一阶段生成初始布局。在第二阶段交替调用邻域搜索子程序和跳坑策略子程序对矩形块的优先级排序进行优化。邻域搜索采用交换式和插入式两种邻域结构,避免单一邻域结构的局限性。当搜索遇到局部最优解时,采用跳坑策略子程序跳出局部最优解,将搜索引向有希望的区域。在第三阶段调用优美度枚举子程序对占角动作的选择作进一步优化。提出了两条优度定理。对于六组benchmark测试用例的实验结果表明,算法的整体表现优于当前文献中的先进算法。针对矩形块方向固定的情形,算法对zdf6和zdf7两个问题实例得到了比已有文献记录更优的布局。     

A Best-first Branch-and-bound Algorithm for Orthogonal Rectangular Packing Problems

In this paper we discuss the problem of packing a set of small rectangles (pieces) in an enclosing final rectangle. We present first a best-first branch-and-bound exact algorithm and second a heuristic approach in order to solve exactly and approximately this problem. The performances of the proposed approaches are evaluated on several randomly generated problem instances. Computational results show that the proposed exact algorithm is able to solve small and medium problem instances within reasonable execution time. The derived heuristic performs very well in the sense that it produces high-quality solutions within small computational time.     

A least wasted first heuristic algorithm for the rectangular packing problem

The rectangular packing problem is to pack a number of rectangles into a single large rectangular sheet so as to maximize the total area covered by the rectangles packed. The paper first presents a least wasted first strategy which evaluates the positions used by the rectangles. Then a random local search is introduced to improve the results and a least wasted first heuristic algorithm (LWF) is further developed to find a desirable solution. Twenty-one rectangular-packing instances are tested by the algorithm developed, the experimental results show that the presented algorithm can achieve an optimal solution within reasonable time and is fairly efficient for dealing the rectangular packing problem. LWF still performs well when it is extended to solve zero-waste and non-zero-waste strip packing instances.     

矩形件优化排样的一种启发式算法

对大规模矩形件正交排样问题,提出了一种快速高效的启发式排放算法。对当前的可排放位置（水平线）,用贪婪算法从未排矩形件中选择可排放于该水平线的最优矩形件组合块;根据各个排放位置与其对应的矩形件组合块的匹配程度,选择最优的可排放位置（最优水平线）优先排放。在排放时,为了便于后续排放,先将待排放位置对应的矩形件组合块从低到高进行排序,再排放。对E.Hopper提供的规模最大的一类实例进行计算,排样率都在99%以上,平均排样率达到了99.38%,平均计算时间只用了1.12秒。与相关文献最好结果进行了比较,结果表明该文算法解决大规模的矩形件排样具有高效性。     

基于遗传模拟退火算法的矩形件优化排样

为了探索更高效的矩形件优化排样方法,提出了一种改进的自适应遗传模拟退火算法。设计了基于矩形件的排样次序及旋转变量的两层染色体编码方法,并采用基于临界多边形的BL定位策略实现矩形件的布局;通过构造启发式算法生成排样初始种群,然后各个种群之间通过相互竞争实现优秀个体的迁移与共享,最终搜索到最优解。标准测试问题的实验结果验证了所提算法的可行性与有效性。     

A heuristic algorithm for optimal placement of rectangular objects

The problem of partitioning a two-dimensional area into pieces having certain sizes with a minimum of wasted space is very important, especially in packing components tightly in the manufacture of very large-scale integrated circuits. The purpose of this paper is to examine the problem of placing rectangular objects in a rectangular area so as to minimize the wasted space, from the viewpoint of establishing maximum empty rectangles rather than the standard linear-programming approach. A comparison of our results with those of the Steudel [5] is reported. Empirical comparisons of our results indicate that our algorithm is very simple and efficient.     

A hybrid simulated annealing metaheuristic algorithm for the two-dimensional knapsack packing problem

The rectangle knapsack packing problem is to pack a number of rectangles into a larger stock sheet such that the total value of packed rectangles is maximized. The paper first presents a fitness strategy, which is used to determine which rectangle is to be first packed into a given position. Based on this fitness strategy, a constructive heuristic algorithm is developed to generate a solution, i.e. a given sequence of rectangles for packing. Then, a greedy strategy is used to search a better solution. At last, a simulated annealing algorithm is introduced to jump out of the local optimal trap of the greedy strategy, to find a further improved solution. Computational results on 221 rectangular packing instances show that the presented algorithm outperforms some previous algorithms on average.     

A genetic algorithm for the two-dimensional knapsack problem with rectangular pieces

Given a set of rectangular pieces and a rectangular container, the two-dimensional knapsack problem (2D-KP) consists of orthogonally packing a subset of the pieces within the container such that the sum of the values of the packed pieces is maximized. If the value of a piece is given by its area, the objective is to maximize the covered area of the container. A genetic algorithm (GA) is proposed addressing the guillotine case of the 2D-KP as well as the non-guillotine case. Moreover, an orientation constraint may optionally be taken into account and the given piece set may be constrained or unconstrained. The GA is subjected to an extensive test using well-known benchmark instances. Compared with recently published methods, the GA yields competitive results.