Abstract: Enterprise strategic alliance is a new kind of win-win business competition mode. In recent years, its coalition formation problem has aroused more and more concern in the field of academic and application areas. In this paper, based on the analysis of the affecting factors of coalition formation, firstly, we give the weight indexes of these factors by the order relation G1 method, screen and revise these weight indexes by clustering analysis method based on similarity. Then we choose the highest alliance stability as the goal, using λ-maximal compatible classes to simulate the relationship between each enterprise in the coalition, and we construct a λ-maximal compatible classes-based rough programming model using the properties and characteristics of rough planning model. Finally, by a specific example, the performance and characteristics of our modelare verified and the results show that our model enriches the existing theory and provide reference for the correct decision making of coalition formation.

Key words: enterprise strategic alliance, coalition formation, maximal compatible classes, rough programming model

摘要： 企业战略联盟作为一种新型、双赢的商业竞争模式,近年来,其联盟的形成问题一直是学术界与应用领域广泛关注的研究内容。本文在分析影响企业战略联盟形成的各因素基础上,首先利用序关系G1法对各影响因素指标进行了权重确定,并用基于相似度的聚类分析法来对属性权重进行筛选和修正;其次以形成的联盟稳定性最高为目标,以λ-最大相容类模拟联盟中各企业之间的相似关系,利用粗糙规划模型的性质和特点,建立了基于λ-最大相容类的粗糙规划模型;最后结合企业战略联盟的具体实例,分析了该模型的特点和有效性。结果表明,该模型丰富了现有的理论成果,为战略联盟的有效形成提供了正确的决策参考。

关键词: 企业战略联盟, 联盟形成, 最大相容类, 粗糙规划模型