基于遗传算法的装备采购决策优化研究

在装备采购中，由于需求单位地域分布和担负的任务各不相同，对装备的品种、数量、时限要求也就不一样，如何使装备采购科学化、合理化，是一个涉及多变量、多目标的复杂系统问题。在综合考虑装备采购各项因素的基础上，构建多约束条件下的多目标模糊指派模型，提出了基于遗传算法的解决方案，最后通过案例进行仿真实验，验证该算法的可行性和有效性，解决了采用传统优化方法难以解决的装备采购优化决策问题。     

Risk Modeling, Assessment, and Management of Lahar Flow Threat

The 1991 eruption of Mount Pinatubo in the Philippines is considered one of the most violent and destructive volcanic activities in the 20th century. Lahar is the Indonesian term for volcanic ash, and lahar flows resulting from the massive amount of volcanic materials deposited on the mountain's slope posed continued post-eruption threats to the surrounding areas, destroying lives, homes, agricultural products, and infrastructures. Risks of lahar flows were identified immediately after the eruption, with scientific data provided by the Philippine Institute of Volcanology, the U.S. Geological Survey, and other research institutions. However, competing political, economic, and social agendas subordinated the importance of scientific information to policy making. Using systemic risk analysis and management, this article addresses the issues of multiple objectives and the effective integration of scientific techniques into the decision-making process. It provides a modeling framework for identifying, prioritizing, and evaluating policies for managing risk. The major considerations are: (1) applying a holistic approach to risk analysis through hierarchical holographic modeling, (2) applying statistical methods to gain insight into the problem of uncertainty in risk assessment, (3) using multiobjective trade-off analysis to address the issue of multiple decisionmakers and stakeholders in the decision-making process, (4) using the conditional expected value of extreme events to complement and supplement the expected value in quantifying risk, and (5) assessing the impacts of multistage decisions. Numerical examples based on ex post data are formulated to illustrate applications to various problems. The resulting framework from this study can serve as a general baseline model for assessing and managing risks of natural disasters, which the Philippines' lead agency-the National Disaster Coordinating Council (NDCC)-and other related organizations can use for their decision-making processes.     

一种求解双目标flow shop排序问题的进化算法

提出一种求解双目标flow shop排序的递进多目标进化算法.算法采用改进的精英复制策略,在实现精英保留的前提下降低了计算复杂性;通过递进进化模式增加群体多样性,改善了算法收敛性;通过群体进化过程中对非劣解集进行竞争型可变邻域启发式搜索,增强了算法局部搜索性能.采用新算法和参照算法NSGA-II对31个标准双目标flow shop算例进行优化.研究结果表明,新算法在所有算例的求解中均获得了优于NSGA-II的非劣解集,验证了算法的有效性.     

Evaluation of multi-objective optimization approaches for solving green supply chain design problems

This paper evaluates the applicability of different multi-objective optimization methods for environmentally conscious supply chain design. We analyze a case study with three objectives: costs, CO₂ and fine dust (also known as PM – Particulate Matters) emissions. We approximate the Pareto front using the weighted sum and epsilon constraint scalarization methods with pre-defined or adaptively selected parameters, two popular evolutionary algorithms, SPEA2 and NSGA-II, with different selection strategies, and their interactive counterparts that incorporate Decision Maker׳s (DM׳s) indirect preferences into the search process. Within this case study, the CO₂ emissions could be lowered significantly by accepting a marginal increase of costs over their global minimum. NSGA-II and SPEA2 enabled faster estimation of the Pareto front, but produced significantly worse solutions than the exact optimization methods. The interactive methods outperformed their a posteriori counterparts, and could discover solutions corresponding better to the DM preferences. In addition, by adjusting appropriately the elicitation interval and starting generation of the elicitation, the number of pairwise comparisons needed by the interactive evolutionary methods to construct a satisfactory solution could be decreased.     

基于多目标规划的中国可持续发展模式优化研究

本文以人口(Population)、资源(Resource)、经济(Economy)、环境(Environment)和科技(Science & Technolo gy)为核心,构造了PREEST系统模型。以PREEST系统为对象,运用主成分分析法对1987-2001年度中国的可持续发展水平进行了综合评价;借助多目标规划理论,构建了中国可持续发展多目标集成模型,并以"十五"计划纲要为基准,开展了中国可持续发展规划研究的实证分析。在此基础上,通过对"十五"期间中国可持续发展传统模式和规划模式的对比分析,就中国人口、资源、经济、环境和科技等领域的协调发展提出了政策建议。     

利用和声分散搜索算法解决动态共乘的乘客选择研究

动态共乘作为一种配合解决城市交通出行难题的新模式近年来引起了人们越来越多的关注，然而在较大范围内选择合适的乘客，以便获得最佳的综合服务效果却具有相当大的挑战性。本文正是针对这一问题，建立了以乘客效用最大化和司机总行程最短为目标函数，以满足司机与乘客的时间要求和司机参与约束为限制条件的多目标0-1规划共乘模型，用于帮助司机选择最合适的乘客。根据该模型的特点，构造了加入了分散搜索机制的新的和声搜索算法。在仿真实验时，针对司机和乘客效用的两种产生方式，在较大规模的路网环境下利用该算法分别对模型进行了求解，得到了Pareto最优解集。仿真结果不仅表明了模型的合理性和算法的可行性，而且还指出基于效用函数可以发现更多合适的潜在乘客。最后，通过与文献中其它算法的对比进一步展示了本文算法的有效性。     

用模糊多层次综合评判优选深基坑支护体系

在分析深基坑优选影响因素的基础上,提出了方案评判的指标体系,应用模糊综合评判理 论建立了基坑支护方案的模糊优选模型.最后,应用实例说明模型的可行性及有效性.     

海南省高新技术产业发展规划研究

高新技术产业的发展已经成为21世纪国际经济竞争和科技竞争的焦点。对我国高新技术产业的发展进行预测与规划研究,具有重要的理论意义和现实的指导意义。本文运用系统工程学的多目标决策分析方法对海南省高新技术产业的发展建立了多目标规划模型,而后利用历史数据进行仿真运算;在此基础上,对海南省未来高新技术产业的发展提出了可行性建议。     

考虑进口配额的跨国供应商选择与订单分配模糊多目标优化决策

针对跨国供应商选择与订单分配问题,在考虑进口配额限制和需求不确定的情境下,构建了一个实现核心制造商利润最大化和供应商交货质量最大化的模糊多目标混合整数规划模型。该模型考虑汇率、关税等多个全球因素,采用一种交互式模糊多目标规划方法对模型进行求解,并应用一个算例进行验证。研究表明:1.存在进口配额限制时,制造商会优先考虑低成本高质量的海外供应商,达到配额限制后再选择本土供应商补充剩余订单;2.人民币汇率正向影响制造商海外采购。当人民币汇率上升时,制造商进口成本降低,其会将更多订单分配给海外供应商;3.关税负向影响制造商海外采购。当关税上升时,制造商会减少关税高的供应商的进口数量,选择替代供应商来满足其需求。由此获得如下管理启示:1,制造商企业在选择跨国供应商时,应重点关注本国的进口配额政策,使供应商多样化以减少针对单一供应商配额政策剧烈波动对制造商采购成本和企业利润的不利影响。2.汇率时刻都在变动,制造商企业应关注本国和供应商所在国的长期汇率政策以及短期突发事件对汇率的影响,及时调整订单分配决策。3.制造商企业在选择供应商时,需要对供应商所在国的短期和中长期关税政策进行综合评估。     

军队技术人才培养规划的鲁棒优化研究

本文针对军队发展带来技术人才需求的动态变化,通过将技术人才各方面素质需求的不确定性离散成三种不同情景,采用鲁棒优化方法构建了军队人才培养的鲁棒优化模型,以军队信息化建设人才需求为背景,以某部人才队伍为研究对象进行了实证求解,并比较了基于单一情景的规划结果与鲁棒优化模型的结果。研究结果表明,鲁棒优化模型对于军队人才培养规划制定具有更好的稳定性和可行性。