基于辩论的多agent商务谈判辩论产生和决策过程研究

由于在自动谈判中加入辩论成分能够允许Agent在谈判过程中通过辩论的形式交换和交流额外的信息,进而影响对方的信念、偏好、意向或目标,使谈判向着有利于达成协议方向发展,并提高谈判效率,降低交易成本,因此已经成为多Agent商务谈判领域的热点发展方向.本文针对如何产生辩论这一重要问题进行研究,提出了"候选辩论集的产生策略"和"基于冲突分析的辩论目标产生模型",分析了基于辩论的多Agent商务谈判决策过程,并通过原型系统进行了验证.本文的研究成果将促使多Agent商务谈判系统向着更为实用方向发展.     

一种基于 Agent 劝说分类的让步模型及实现

随着电子商务的迅速发展以及Agent理论和方法的引入,出现了关于Agent自动谈判等方面的研究。 A-gent劝说由于能节约商务谈判成本,提高商务谈判效率,使谈判过程更加理性,解决谈判中出现的许多问题,受到了广泛的关注。本文结合人际劝说理论,对Agent劝说进行了分类;结合形式化理论,构建了相应的形式化表述模型和交互机制;然后将两者相结合,在让步影响幅度的基础上,构建了基于Agent劝说分类的让步模型;最后,通过算例和系统实现对模型的可行性、有效性进行了验证。     

基于Agent与分解协调的综合生产计划研究

以作业单元为局部决策Agent,车间管理者为全局协调Agent,引入生产节点间的内部结算价格,基于多Agent系统,建立了综合生产计划的分布式决策模型.通过将局部Agent决策目标的总和与全局Agent决策目标进行对比,证明了所引入的内部结算价格就是全局Agent目标函数关于物流平衡约束的Lagrange乘子.基于Lagrange分解协调原理,设计了局部作业单元Agent和全局协调Agent的迭代协调算法.该迭代算法以上次计算的中间结果作为对其它作业单元生产需求的估计,从而能将各个生产单元Agent的决策模型分离,实现了分布建模与求解.在算例研究中使用启发式规则来确定Lagrange乘子迭代的步长系数,保证了较好的收敛性,证明模型和算法是有效的.     

基于NSGAII的协同产品开发项目自动协商决策

供应链环境下的协同产品开发项目中,核心企业与合作企业之间,需要对项目工期、预算和质量进行协商。科学高效的协商决策方法有助于协同产品开发项目参与各方实现双赢的协商目标。采用多Agent系统建立了协商问题的框架,构建了协商Agent以自身效用最大化和与对方建议相似度最大化为目标的多目标反建议模型;提出了基于NSGAII的自动协商决策算法,算法采用了正整数和小数混合的实数编码方式,并在遗传操作中增加了约束限制,剔除不可行个体。计算机仿真实验结果表明了算法对提高协商成功率和协商效果的有效性,说明了协商Agent可接受的最低效用值对于协商结果的影响。     

基于MAS的多模式分布式资源约束多项目调度

针对实践中分布式多项目的活动往往具有多种执行模式,提出多模式分布式资源约束多项目调度问题。在项目动态到达环境下,考虑活动不同的执行模式,以工期最短和多项目延期成本最小为目标分别构建局部单项目调度模型和全局多项目决策模型,采用改进变邻域搜索算法求解初始局部调度计划,并设计基于模式调整的全局协商调度算法求解全局决策模型,通过双层算法实现分布式多项目调度中局部单项目调度与全局多项目调度系统性协调,减少项目中断和多项目延期成本。基于构建的多模式测试集进行的多项目数值实验表明：本文设计的双层算法可有效求解多模式分布式多项目调度问题,并且对不同规模问题求解具有良好的适应性。     

基于灰色关联分析的犹豫模糊多属性决策模型构建及其应用

本文构建了基于灰色关联分析的犹豫模糊多属性决策模型,将之应用于中外大学评价体系相关性研究。针对隶属度是由离散且数值不等的犹豫模糊数组成的犹豫模糊集,引入灰色系统理论,建立了基于灰色关联的犹豫模糊多属性决策模型,应用该模型对中国大学评价体系与ESI评价体系的相关性进行分析,并应用统计学中的相关分析进行实证检验,发现两种方法所得结果一致,表明了本文所建多属性决策模型的可行性和有效性。     

基于模糊决策模型的高速公路环境风险评价

本文构建了非常规突发事件下的高速公路环境风险评价的目标体系,将模糊逻辑与多目标决策、层次分析法相结合,提出一种基于模糊多目标决策模型。它与传统的多目标决策模型相比,将输入参量转化为语言变量后,获得一个集专家知识、系统变量间关系为基础的优势得分排序规则,可对不确定环境下的多目标及其子目标进行决策评价。通过设计一个高速公路路段环境评价的算例,并对其目标及子目标进行敏感性分析,验证这种决策方法在评价高速公路环境风险上的科学性。     

概率犹豫模糊Maclaurin对称平均算子及其多属性群决策模型

本文构建了一种基于概率犹豫模糊Maclaurin对称平均(PHFMSM)算子的多属性群决策模型,使得决策者可以根据决策过程中的风险偏好而选择合适参数值进行决策。首先,基于Maclaurin对称平均和Archimedean范数,提出了概率犹豫模糊Maclaurin对称平均(PHFMSM)算子;其次,详细探讨了PHFMSM算子具有的四种优良性质,并分析了PHFMSM算子的几种特殊情况及其加权形式;最后,建立了一种基于PHFWMSM算子的概率犹豫模糊多属性群决策模型,并且通过实际决策案例验证该模型的有效性和可行性。     

基于拍卖谈判机制的分布式多技能多项目调度

求解分布式多项目调度问题的关键是设计有效的资源协调机制以解决项目间的全局资源冲突。当全局资源为多技能人力资源时,调度项目活动开始时间的同时还需要满足“活动-技能-资源”的匹配关系,从而为多项目制定高质量的调度计划与全局资源指派方案。基于多Agent系统,建立局部调度优化与全局协调决策的双层模型;采用基于正向逆向调度改进的遗传算法求解初始局部调度计划;以人力资源各时段的使用权作为拍卖品,设计考虑人力资源多技能异质特点的拍卖谈判机制协调全局资源冲突。改编MPSPLIB算例集并开展实验研究,研究结果表明：本文设计的拍卖谈判机制与改进的遗传算法相结合可以有效协调不同规模问题的全局资源分配;拍卖谈判机制在测试问题集上的求解结果优于序贯博弈谈判机制;问题规模越大或资源冲突程度越强,项目协调调度难度越大,全局资源分配过程越复杂。     

基于约束放松的电子商务协同谈判模型

在众多的自动谈判模式中，协同谈判是一种新型的多线程谈判形式。文章分析了面向电子商务的协同谈判中的冲突问题，提出了解决谈判冲突的方法——满意度函数法。该方法在谈判模型中用约束网来表示谈判中的变量和约束关系，通过约束放松来解决谈判中的冲突。作者在对约束及约束网进行详细分析的基础上设计了约束传播算法，用它来求解约束网问题。文章通过实例验证了提出的论点和方法的正确性。采用这种基于约束放松的协同谈判模型可以有效地解决谈判中的冲突，提高谈判的效率。该方法的应用可为解决多个谈判的组合问题提供一种新的思路。     

A simultaneous multi-issue negotiation through autonomous agents

Automated negotiation through autonomous agents has become increasingly important since the advent of e-marketplace. A deadlock may arise in which both negotiators refuse to disclose more information; then, mediation is required. We propose an agent-based sealed-bid design in which both agents simultaneously submit their respective offers to the mediate agent, and construct an efficient negotiation strategy which can reach an agreement aiming to maximize their owner’s utility. An important contribution of this paper is that we consider negotiators not necessarily to conflict over all issues, such as quantity. Our experimental results demonstrate that the proposed automated negotiation strategy is efficient in terms of the Pareto-efficiency of the negotiated contracts, the joint utility obtained, and the equality of both agents. In addition, since disclosure of information is reduced greatly, the design can discourage counter-speculation and effectively control fraud and misrepresentation to a certain extent. The method we proposed is simple and stable in the context of a game and the model could be well suited for practical agent applications.     

An intelligent agent negotiation strategy in the electronic marketplace environment

E-commerce will strongly penetrate the market if coupled with appropriate technologies and mechanisms. Mobile agents may enhance the intelligence and improve the efficiency of systems in the e-marketplace. We propose a dynamic multi-lateral negotiation model and construct an efficient negotiation strategy based on a ranking mechanism that does not require a complicated rationale on behalf of the buyer agents. This strategy can be used to extend the functionality of autonomous intelligent agents, so that they quickly reach to an agreement aiming to maximise their owner’s utility. The framework proposed considers both contract and decision issues, is based on real market conditions, and has been empirically evaluated. Moreover, it is shown that in a linear framework like the one we employ, more elaborate ranking mechanisms do not necessarily improve efficiency.     

Negotiation support systems: A multi-criteria and multi-agent approach

This research concerns the development of an Negotiation Support Systems (NSS) based on a multi-criteria conceptual framework of the negotiation and developed according to a multi-agent architecture from Distributed Artificial Intelligence (DAI). A first prototype of such a system, NegocIAD, has already been developed [8], but the weakness of its assistance to the negotiation process have led us to revise the conceptual framework in order to define a more relevant assistance to the negotiation process. This paper presents this new conceptual framework defined in order to develop a new prototype. First, we point out the originality of our multi-criteria and multi-agent approach, the general architecture and the limitations of NegocIAD. Then we present the new multi-criteria conceptual framework mainly based on the definition and the use of projection plans (group Gaia plans) emerging from principal component analysis (PCA) already proposed in a single decision maker context in extension of the Promethee method. In the next part, we develop the possible levels of use of these plans during the negotiation process and the type of assistance provided to the mediator. This assistance is mainly based on the elaboration and the interpretation of group Gaia plans for which we propose a set of interpretation rules and the outline of a method to make use of these rules for a relevant support to the mediator in the management of the negotiation process. Finally, we conclude on the perspectives of our future researches and developments for the new generation of our prototype in a multi-agent architecture context.     

An agent-based negotiation model on price and delivery date in a fashion supply chain

Price and due-date negotiation between supply chain members is a critical issue. Motivated by industrial practice, we consider in this paper a make-to-order fashion supply chain in which the downstream manufacturer and the upstream supplier are cooperative on due-date and competitive on price. We propose a two-phase negotiation agenda based on such characteristics, and aim to find an optimal solution to deal with the negotiation problem considering production cost and mutual benefit. We build an analytical negotiation model for a manufacturer-supplier pair, discuss their utilities, and examine the Pareto efficiency frontier from the theoretical perspective. After that, from an application perspective, we build an agent-based two-phase negotiation system where agents are used to represent the two parties to enhance communication. In the cooperative phase, a simulated annealing based intelligent algorithm is employed to help the manufacturer agent and the supplier agent search tentative agreement on due dates which can minimize the total supply chain cost. In the competitive phase, the two parties bargain on the pricing issue using concession based methods. They adjust the reservation value and aspiration value for pricing accordingly based on the integrated utility and the result of the previous phase. Simulation results show that, the proposed negotiation approach can achieve optimal utility of agents and reach a win-win situation for the bilateral parties. Sensitivity analysis is conducted to further generate insights on how different parameters affect the performance of the proposed system.     

Monte Carlo simulation techniques for group decision making with incomplete information

In this paper we deal with group decision-making problems where several decision makers elicit their own preferences separately. The decision makers’ preferences are quantified using a decision support system, which admits incomplete information concerning the decision makers’ responses to the questions they are asked. Consequently, each decision maker proposes classes of utility functions and attribute weight intervals for the different attributes. We introduce an approach based on Monte Carlo simulation techniques for aggregating decision maker preferences that could be the starting point for a negotiation process, if necessary. The negotiation process would basically involve the decision maker tightening the imprecise component utilities and weights to output more meaningful results and achieve a consensus alternative. We focus on how attribute weights and the component utilities associated with a consequence are randomly generated in the aggregation process taking into account the decision-makers’ preferences, i.e., their respective attribute weight intervals and classes of utility functions. Finally, an application to the evaluation of intervention strategies for restoring a radionuclide contaminated lake illustrates the usefulness and flexibility of this iterative process.     

Modeling international investment decisions for financial holding companies

This research analyzes the internationalization process model developed by Johanson and Vahlne and derives two integer programming investment decision models that consider the risk attitudes of investment firms. Johanson and Vahlne’s model provides a starting point for building a model that suits the investment approach and decision making process of financial holding companies. In practice, when firms make an international investment decision, there is a need for a model that can generate outputs based on financial measures such as profit, investment returns, and tolerable levels of risk. Thus, in this paper, Johanson and Vahlne’s concepts are studied and financial managers are interviewed to derive models that match the investment decision procedures of the firms. The model helps firms manage the risks of their investments and derive accurate investment strategies based on investment objectives and constraints.     

Learning about preferences in electronic negotiations – A volume-based measurement method

To create an integrative solution in a bargaining problem, negotiators need to have information about each other’s preferences. Empirical negotiation research therefore requires methods to measure the extent to which information about preferences is available during a negotiation. We propose such a method based on Starr’s domain criterion, which was originally developed for sensitivity analysis in decision making. Our method provides indices for the amount of preference information that can be inferred both in negotiations reaching an agreement and negotiations where an agreement was not (yet) reached. To test the external validity of our proposed measures, we conduct an empirical study which shows that the proposed measures exhibit positive relationships to the success of negotiations as well as to the efficiency of outcomes that would be expected according to negotiation theory.     

基于图模型理论有序偏好下的冲突反问题研究

在冲突谈判中,能获知对手偏好是掌握谈判主动性的重要条件。本文基于冲突分析图模型理论构建了一种获取对手偏好的方法。该方法通过深入分析冲突分析图模型中Nash、GMR和SEQ三种稳定性定义,利用反向思维,建立求解对手偏好最少约束条件的数学模型。该方法能让决策者在预知冲突结局的前提下,得到对手的全部偏好信息。以“云南曲靖陆良县铬污染”冲突事件为例,通过对该事件引发的冲突进行建模和偏好分析,在已知冲突最终结局的前提下,运用数学模型,省环保厅可以得到陆良化工企业的所有偏好序,使其在冲突谈判中做到知己知彼,同时也验证了该方法的可行性和有效性。案例分析过程可以从战略层面为谈判中的一方提供参考。