基于PSO-SVM的产品服务系统配置

Product Service System Configuration Based on PSO-SVM

产品服务系统(PSS)配置目的是根据客户需求,从产品模块和服务模块的实例中找出特定的实例组合。为了满足个性化、多样化的客户需求,快速准确地给出相应PSS配置,可将其视为多分类问题,提出基于支持向量机的PSS配置方法。支持向量机对内部参数依赖性极高,为了找出与分类问题相契合的参数,利用粒子群算法全局寻优能力,搜索支持向量机最优参数,将最优参数代入支持向量机模型中,得出用于满足客户需求的最优PSS配置方案。以中央空调产品服务系统配置为例进行分析,得出PSO-SVM正确率为89.84%,比传统支持向量机提高6.75%,验证了该方法有效性。

The product service system(PSS)configuration is to find a specific combination of instances from the instances of product modules and service modules according to customer needs.In order to meet the needs of personalized and diversified customers and provide the corresponding PSS configuration quickly and accurately,it can be regarded as a multi classification problem,and a PSS configuration method based on support vector machine is proposed.Support vector machines have extremely high dependence on internal parameters.In order to find parameters that are compatible with the classification problem,the global optimization capability of particle swarm algorithm is used to search for the optimal parameters of the support vector machine and substitute the optimal parameters into the support vector machine model,the optimal PSS configuration scheme to meet customer needs is obtained.The product service system configuration of central air-conditioning is analyzed as an example,and it is concluded that the correct rate of PSO-SVM is 89.84%,which is 6.75%higher than the traditional support vector machine,which verifies the effectiveness of the method.