电子中介中的多属性匹配研究

Research on Multi-attribute Matching Problem of Electronic Broker

研究了电子中介处理个人之间单件物品交易时的多属性匹配问题。建立了多目标匹配模型,推导了用理想点法求解该模型时,求解其一次距离最小等价于求解原各目标相应系数直接相加所得的单目标指派问题。用有指导随机搜索算法求解了距理想点的二次距离最小。仿真实验表明,变量规模小于65时可以求二次距离最小并得到满意解;而大于65时求一次距离最小更为合适。

The purpose of this paper is to research the multi-attribute matching problem of electronic broker when it is used in the C2C single commodity exchanging market,and a multi-objective mathematical model is constructed.It is verified that,when using the ″ideal solution method″,minimizing the Hamming Distance between final solution and the ideal solution equals optimizing the single objective assignment problem whose coefficient is the sum of all the coefficients of the previous objectives.A guided stochastic searching algorithm is used to solve the minimum Euclidean Distance.Experiments in real estate market show that,when the variable size is under 65,people can use the Euclidean Distance model,when over 65,Hamming Distance model is appropriate.