A grouped storage method for minimizing relocations in block stacking systems

When storage and retrieval times of inventories are uncertain and the storage space for the inventories is limited, it is an important problem to assign the inventories to storage spaces so that the total expected number of relocations is minimized. This paper addresses a dynamic location problem as well as a static location problem. Mathematical models are proposed for obtaining the optimal solution. To overcome the excessive computational time of optimizing methods, a genetic algorithm is suggested. Also, simple heuristic rules are suggested to solve the dynamic location problem. Performance of various solution methods are compared with each other. Received: June 2005 / Accepted: December 2005     

Automated Discovery Systems and Scientific Realism

In the paper I explore the relations between a relatively new and quickly expanding branch of artificial intelligence –- the automated discovery systems –- and some new views advanced in the old debate over scientific realism. I focus my attention on one such system, GELL-MANN, designed in 1990 at Wichita State University. The program's task was to analyze elementary particle data available in 1964 and formulate an hypothesis (or hypotheses) about a `hidden', more simple structure of matter, or to put it in contemporary terms –- the discovery of quarks. The central thesis of my paper is that systems like GELL-MANN not only discover (or rediscover) the hidden structure of matter, but also provide independent strong evidence in favor of scientific realism about entities involved in that structure. I make an attempt to show how an argument for scientific realism about sub-microscopic entities can be constructed that would parallel Ian Hacking's `argument from coincidence' presented with respect to microscopic objects in his famous book Representing and Intervening.     

A Multi-Objective Genetic Algorithm for Solving Assembly Line Balancing Problem

In this paper, a multi-objective genetic agorithm to solve assembly line balancing problems is proposed. The performance criteria considered are the number of workstations, the line efficiency, the smoothness index before trade and transfer, and the smoothness index after trade and transfer. The developed genetic algorithm is compared with six popular heuristic algorithms, namely, ranked positional weight, Kilbridge and Wester, Moodie and Young, Hoffmann precedence matrix, immediate update first fit, and rank and assign heuristic methods. For comparative evaluation, 20 networks are collected from open literature, and are used with five different cycle times. All the six heuristics and the genetic algorithm are coded in C++ language. It is found that the proposed genetic algorithm performs better in all the performance measures than the heuristics. However, the execution time for the GA is longer, because the GA searches for global optimal solutions with more iterations.     

一种松耦合的生物医学命名实体识别算法

生物医学命名实体识别（Bio-NER）是生物医学文献挖掘利用的基础工作。针对目前Bio-NER存在的困难和问题，提出了松耦合的Bio NER算法LCA，该算法利用启发规则过滤器、词性模板匹配及改良的隐马尔科夫模型（HMM）识别生物医学命名实体。在GENIA corpus 3.02语料库上进行的实验表明，LCA可以达到80%的准确率和89%的召回率，优于相关工作中的结果。     

基于字典技术的本体整合系统

针对本体整合中缺少语义理解的问题，使用字典技术对同义词进行识别，并使用启发式规则计算本体中实体间的相似度。以此为基础，开发了一个半自动化的本体整合系统。该系统为用户整合本体提供帮助，减轻了手工整合本体的工作量，同时提高了本体整合的准确性。     

连续约束蚁群优化算法的构建及其在丁烯烷化过程中的应用

经典蚁群系统只适用于离散问题，缺少处理约束的专门机制.基于蚁群觅食的生物学行为，以搜索最优食物源为目标，将约束纳入食物源优劣评价的启发式规则，采用成群募集和海量募集两种方式，并辅以局部搜索，以此引导蚁群寻找可行域中的最优解，构建为适用于连续约束优化问题的蚁群系统（constrained ant colony system, CACS）.测试实例表明，CACS具有良好的适用性及全局优化性能，将它应用于丁烯烷化过程的约束优化，取得了令人满意的结果.     

基于遗传算法的作业车间调度优化

将遗传算法和启发式调度规则相结合，研究了具有柔性加工路径的作业车间的智能优 化调度问题，调度规则的引入使该算法具有较高的搜索效率，遗传算法的引入保证了解的全 局最优性，对照算例，表明该算法在求解性能和效率两方面均具有显著的优势．     

单阶段多产品批处理过程的短期调度2. 模型的简化

本文第一部分建立了具有并行处理设备多产品 单阶段批处理过程短期调度的基本数学模型．根据该模型和多产品单阶段批处理过程的特点 ，本文这一部分引入一些启发性规则，并将这些启发性规则融入到模型中．合理地使用这些 启发性规则不但能减小模型的整数变量、连续变量和约束的数量，使得模型表达更紧、求解 速度加快，而且能得到最优解．大量计算表明该模型求解速度快，尤其对包含多个同种订单 的调度问题更为显著．     

一种启发式的用哼唱检索音乐的层次化方法

“用哼唱检索音乐”是一种友好的基于内容的音乐检索方法，它已经引起了广泛的研究兴趣；在对音乐库做了统计分析的基础上，总结了一些启发式规则，帮助对哼唱输入进行基音检测、音符分割，哼唱输入表达为音高轮廓图和节奏，音乐库中的音乐按音乐的节奏类型分为不同的节奏区域，并从每首音乐中抽取旋律轮廓图和节奏信息，用递归神经网络记忆旋律轮廓，音乐库的索引是神经网络的权值矩阵，将哼唱输入与音乐库中的音乐匹配的过程就是计算神经网络的输出过程。实验结果显示了所提方法的有效性。     

工艺路线可变的双资源双目标车间调度优化

将遗传算法与启发式调度规则相结合，研究了工艺路线可变的双资源双目标的作业车间调度优化问题。在探讨过程中，不仅考虑到了每个工件有几条可行的工艺路线，而且考虑到了工件的调度受到机床、工人等资源的制约，以及在加工过程中发生的储存费用、机床的加工费用和工人的劳动费用对工件调度的影响，设计了以生产周期和生产成本综合优化为目标的适应度函数。启发式调度规则使该算法具有较高的局限搜索效率，遗传算法保证了了解的全局最优性。最后给出了算例，并对计算结果进行了分析和讨论。