影响图的扩展综述

影响图是单决策人基于不确定信息表示和求解复杂决策问题的图模型,已成为一种流行的标准建模工具。由于决策环境纷繁复杂,决策问题多种多样,没有任何一种建模工具能够普遍适用于各种决策问题。为了表示复杂决策问题,增强影响图的表达能力,研究者对影响图进行了多种扩展。从建模无限制决策问题、非对称决策问题、涉及连续变量的决策问题、涉及非精确变量的决策问题及多Agent决策问题等方面对影响图的扩展进行了较为全面的回顾与分析,指出了有关影响图的进一步研究方向。关键词:决策;影响图;无限制决策;非对称决策;多Agent决策     

基于增广输入变量的T-S 模糊模型建模

针对现有T-S 模糊模型建模精度与计算效率之间的矛盾, 提出一种利用增广输入变量进行T-S 模糊模型建模的方法. 对输入变量进行多项式增广处理后, 以核模糊?? 均值聚类算法配合聚类评价指标自适应获得最佳聚类数及相应的模糊划分, 并通过递推最小二乘计算得出T-S 模糊模型的后件参数. 提出可利用后件参数反推断前件结构的方法来快速有效地确定前件结构. 最后通过仿真验证了上述方法的有效性.

     

考虑提前期压缩的Newsvendor型产品供应链协调模型

在假定制造商的单位生产成本是关于提前期的减函数的前提下,将提前期视为可控决策变量,建立了带有提前期压缩的Newsvendor型产品供应链协调模型,讨论了提前期压缩对供应链及其成员收益的影响,并利用由回购和回扣/惩罚所组成的联合契约实现了供应链的协调.数值仿真分析结果验证了结论的正确性.     

组织学习与组织结构模式选择的仿真分析

为研究组织外部学习对组织结构模式选择的影响,采用基于代理的计算实验方法,建立了区域零售市场中两个零售企业之间存在竞争时,采用集中决策和分散决策模式下的仿真模型。计算结果表明,企业的外部学习导致竞争更为激烈,从而集中决策模式在市场差异较小时明显优于分散决策模式,而随着市场差异程度的增大采用分散决策模式则可以为企业带来更高盈利。另外,随着市场容量的增大,集中决策模式也逐渐优于分散决策模式。     

考虑内在影响关系的不安全行为组织管理控制模型

针对不安全行为组织管理体系难以长期稳定运行的问题,在充分考虑管理变量之间影响关系的基础上,提出一套不安全行为组织管理的控制模型.由于变量之间影响关系的存在,待改进变量及其前置变量的不足均应被识别为组织管理漏洞,且变量的重要性应兼顾自身重要性与变量之间的影响强度.基于此研究思路,首先,结合行为安全“2-4”模型与反馈控制理论,确定控制模型的范围、变量、构建步骤及运行流程;然后,通过三角模糊数与WINGS(weighted influence non-linear gauge system)算法的结合,在考虑主观判断存在不精确性的同时,实现变量之间影响关系的刻画以及变量自身重要性与变量间影响强度的融合,进而给出识别管理漏洞及确定控制措施次序的方法.案例分析表明,该控制模型有助于提高管理漏洞识别的全面性与控制措施次序的合理性.     

关联变量分组的分解多目标进化算法研究

含有大规模决策变量的优化问题是当前多目标进化算法领域中的研究热点和难点之一。在解决大规模变量问题时,目前的进化算法并没有寻找决策变量之间的关联信息,而都只是将所有变量视为一个整体来进行优化。但随着优化问题中决策变量的增多,“变量维度”成为瓶颈,从而影响算法的性能。针对上述问题,提出关联变量分组策略,通过识别决策变量间内在的关联信息把关联变量分配到同组中,将复杂高维变量的优化问题分解为简单低维的子问题来求解。该策略通过增加关联变量分配到同组中的概率来使算法尽可能地保留变量之间的关联性,减少分组后子问题间的依赖性,从而提高子问题最优解的质量并最终获得最佳的Pareto最优解集。将该算法在标准测试函数上进行变量扩展后再进行仿真对比实验,采用性能指标对算法的收敛性和多样性进行对比分析。实验结果表明,该算法在解决大规模变量的多目标优化问题中,随着决策变量维度的增加,比经典的多目标进化算法NSGA-II、MOEA/D以及RVEA具有更佳的收敛和更好的分布性能,所求得的Pareto解集质量更高。     

动态背景下考虑顾客退货影响的供应链协调问题

在考虑店铺辅助服务降低顾客退货率的基础上,构建微分博弈模型并研究供应链的店铺辅助策略和供应链的协调问题.首先,利用微分对策理论,给出集中和分散两种决策模式下定价和店铺辅助的最优策略,并对两种决策模式下最优策略进行比较分析;然后,为了提升分散决策下供应链的利润,设计二部收费制契约,实现动态背景下供应链的协调;最后,通过数值算例分析店铺辅助努力和店铺辅助水平效率对供应链最优解及协调的影响.研究发现:集中决策下店铺辅助努力高于分散决策下的相应值,具有较高的店铺服务水平,但两种决策结构下的价格高低依赖于系统参数;契约的协调能力会随着店铺辅助努力或服务水平效率的增加而变强.     

单元机组协调控制系统控制策略研究

针对单元机组协调控制系统多变量、强耦合、时变、滞后大的特点,提出一种采用串联后补偿单向解耦后加入PID控制器自整定的控制策略.该控制策略首先对系统进行串联后补偿实现完全解耦,并通过单向解耦对解耦系统进行简化,然后把多变量的控制问题转化为多个单变量,再对各个单变量加入PID控制器进行自整定.通过对125 MW单元机组协调控制系统的仿真分析可知,这种方法有效消除了输入与输出之间的强耦合带来的不利影响,使系统输出稳定,控制效果好.     

具有不确定参数多目标决策的一类鲁棒有效解

针对具有区间型不确定参数多目标决策问题,探讨了决策人在决策的最优性与决策后果 的不确定性之间的权衡,从而制定决策的方法,提出了鲁棒有效解及ε-鲁棒有效解的概念,并研究 了其最优性条件.     

BP神经网络在机器人足球比赛系统中的应用

足球机器人的决策系统是一个多智能体协调控制系统,控制机器人运动需对机器人未来的方位进行实时预测。为了解决RoboCup小型组比赛系统的延迟和准确定位问题,将BP神经网络应用于决策系统的局势预测中。建立了基于BP神经网络的线性预测模型,确定了神经网络的拓扑结构,并将训练好的网络应用于现有比赛系统,预测机器人的就位效率、协调及配合能力,进行仿真实验。实验证明,方法对机器人的位置、方向等预测比较准确,证明了预测算法的可行性和优越性。     

基于 AI技术的组织决策协调模型研究

分析了传统组织决策协调模型中存在的不足，设计出一种新的决策协调模型——GWB模型。该模型能协调各决策单元的目标，以完成组织的统一目标。GWB模型中采用了遗传算法、世界模型和黑板结构。将GWB模型应用于并行工程环境下多产品开发的组织决策问题，实验结果表明该模型是有效的。     

制造商主导型双渠道供应链协调决策模型

以制造商主导型供应链为研究对象, 分析双渠道供应链协调策略问题. 引入双渠道价格敏感系数和竞争系数两个变量, 分别建立双渠道供应链集中决策、分散决策和协调决策的3 种模型. 研究结果表明, 不同合同策略协调后, 零售商和供应链整体收益增加; 经两部定价和批发价格合同协调, 零售商需要支付固定费用给制造商才能够达到协调, 而经Shapley 值法分配合同协调, 则双渠道供应链可直接达到协调效果.

     

Coordination and Task Interdependence during Schedule Adaptation

This article discusses the influence of organizational and behavioral variables on coordination between planners during plan adaptation. Fast communication and mutual alignment are necessary to maintain schedule feasibility in a situation with several schedulers. Therefore, coordination modes are required that facilitate communication and joint problem solving. Moreover, building on interdependence theory, we hypothesize that the schedulers' perceptions of task interdependence influence rescheduling performance. Experimental results indicate that a group decision‐making coordination mode enforcing cooperation outperforms a distributed decision‐making coordination mode involving emergent alignment. The level of perceived task interdependence explains this better performance. Therefore, perceptions of task interdependence are put forward as an important behavioral factor influencing rescheduling performance, with several important implications for theory and practice. © 2011 Wiley Periodicals, Inc.     

Two controlled experiments on model-based architectural decision making

ContextIn recent years, architectural design decisions are becoming more and more common for documenting software architectures. Rather than describing the structure of software systems, architectural decisions capture the design rationale and – often reusable – architectural knowledge. Many approaches and tools have been proposed in the literature to support architectural decision making and documentation (for instance, based on models, ontologies, or templates). In this context, the capturing, organization, and effective reuse of architectural knowledge has gained a lot of attention.ObjectiveHowever, there is little empirical evidence about the supportive effect of reusable architectural knowledge on the effectiveness and efficiency of architectural decision making.MethodTo investigate these aspects, we conducted two separate controlled experiments with software architecture students in which we tested the supportive effect of reusable decision models in decision making and documentation.ResultsOur results show that the use of reusable decision models can significantly increase both the efficiency and the effectiveness of novice architects.ConclusionWe can report, that our findings are in line with similar studies and support the claims regarding reusable architectural design decisions in principle.     

Decision strategies and design of agent interactions in hierarchical manufacturing systems

Decisions in manufacturing planning environments involve various interdependent hierarchical levels, ranging from shop-floor operations over production planning to enterprise coordination. Distributed decision making across these hierarchies reduces the complexity compared to an otherwise monolithic planning approach. We develop a stochastic model with two decision makers (agents) and analyze their strategic interaction using game theory. The mutual influence between agents depends on organizational parameters, which, in a first case, agents themselves can regulate, and, in a second case, an organizational designer can control. The model provides managerial insights on how to achieve cooperation among self-interested agents and how decisions and preferences of agents can be aligned with the goals of the organization. Moreover, we investigate the degree and fidelity of information necessary for optimal decision making. The model is applied to a manufacturing enterprise context.     

非常规突发事件应急决策冲突消解协调方法

针对应急决策冲突性特点和难以形成高一致性决策方案的问题,分析了应急群体决策冲突产生的原因,构建了应急决策冲突消解协调框架,提出了应急决策群体冲突测度方法,以及应急决策冲突消解模型和方法,形成了非常规突发事件应急决策冲突消解协调机制,使应急群决策冲突逐渐收敛以获得冲突程度足够低的应急决策方案。最后通过算例分析验证了冲突消解协调方法和机制的有效性。     

雷达情报组网作战指挥决策模型研究

本文利用改进Petri网建立了雷达情报组网系统作战指挥决策模型,并对决策模型进行了仿真分析,通过仿真,比较了传统模式下与组网模式下,不同的作战指挥决策组织结构对作战指挥决策时延和保障目标效能的差别。     

Negotiation based on constraints in cooperation

This paper presents some aspects of cooperation in organizations. In the first part, we present the importance to coordination processes within an organization. Indeed, the information perceived by the company no longer pertains to the realm of the repetitive, predictable and programmable. In this context of limited rationality, how can one define an efficient and acceptable decision coordination and distribution structure? We argue that the intervention of man in the decisional process remains inescapable on account of the limitations of the coordination process, and define several forms of cooperation between decision centers on an industrial site. In the second part, the assumption retained is that for the management of a production system the decision is made through a network of decision centers. The approach presented puts forward the development of decision and cooperation aid tools only exploiting the information contained in the constraints linking together the decision variables so as to highlight the degrees of freedom effectively available to the decision maker. Finally, we discuss about cooperation and power, where the power issue cannot be disregarded.     

递阶智能控制系统的一般结构

研究递阶智能控制系统的结构问题,介绍了递阶智能机器的一般结构,讨论了与组织级模型有关的决策段结构,并研究了协调级和执行级的控制结构模型。     

Research on the multi-attribute decision-making under risk with interval probability based on prospect theory and the uncertain linguistic variables

With respect to risk decision making problems with interval probability in which the attribute values take the form of the uncertain linguistic variables, a multi-attribute decision making method based on prospect theory is proposed. To begin with, the uncertain linguistic variables can be transformed into the trapezoidal fuzzy number, and the prospect value function of the trapezoidal fuzzy number based on the decision-making reference point of each attribute and the weight function of interval probability can be constructed; then the prospect value of attribute for every alternative is calculated through prospect value function of the trapezoidal fuzzy number and the weight function of interval probability, and the weighted prospect value of alternative is acquired by using weighted average method according to attribute weights, and all the alternatives are sorted according to the expected values of the weighted prospect values; Finally, an illustrate example is given to show the decision-making steps, the influence on decision making for different parameters of value function and different decision-making reference point, and the feasibility of the method.