【作者】 何蓓； 吴敏； 三间均； 周意诚；

【Author】 Bei He, Min Wu, Hitoshi Mitsuma, Yicheng Zhou (School of Information Science and Engineering, Central South University, Changsha, 410083) (Tokyo Electric Power Systems Corporation, Tokyo, Janpan, 105-0004)

【机构】 中南大学信息科学与工程学院； 株式会社东电系统；

【摘要】 本文针对大规模的One to One营销优化问题,提出一种基于最小度排序的线性规划(LPMDP)算法,它在不可解内点算法的基础上对约束矩阵进行列最小度排序,从而加强其结构稀疏性,大幅度减少计算量以及内存消耗。仿真实验表明,LPMDP算法不仅能获得最佳营销方案,并且能解决大规模One to One营销优化问题。更多还原

【Abstract】 This paper presents a linear program algorithm based on minimum degree permutation(LPMDP) for the large scale One to One marketing optimiztion problem. In this algorithm, on the base of infeasible interior point algorithm, the constraint matrix is permuted according to column approximate minimum degree. So as to form a sparser matrix, and reduce the computational time and memeroy consumption. The results of the simulation show the LPMDP algorithm not only can obtain the optimization marketing plan, but also can solve the larger scale One to One marketing optimiztion problem.更多还原