Research On Optimization Technology Of Product Service Life Cycle Value Chain

At present,the research on the value chain of the service-oriented manufacturing industry mainly focuses on the value chain of the product life cycle,but the value chain of the product life cycle still has the market demand to determine the customer demand and the optimization method of the value chain less involving big data thinking and technology.Therefore,this paper proposes a value chain of product service life cycle based on customer needs,and uses optimization technologies such as information processing to mine value in each link of the value chain of the whole life cycle of products and services.Feedback the value-added information excavated to the corresponding links in the value chain of the whole life cycle of products and services,so as to realize the optimization and value-added of the value chain of the whole life cycle of products and services.The specific research work mainly includes the following four aspects.(1)The value chain of the whole life cycle of products and services has been constructed.Based on the process of product life cycle,the process of product service life cycle is established.By studying the data flow and value chain structure of the product service life cycle,the value chain of the whole life cycle of products and services is established.(2)The overall optimization scheme of the value chain of the whole life cycle of products and services is designed.Taking the whole life cycle value chain of products and services as the core,and taking customer value optimization technology and inventory optimization technology as the optimization method,the optimization scheme of the value chain of the whole life cycle of products and services based on customer value and spare parts inventory demand is designed.(3)A customer value optimization model based on the combination of the improved RFMT model and K-means algorithm is constructed.For the traditional RFM model,a new dimension of average usage time is added,and a new customer segmentation model RFMT model is constructed.The weights of each index of the RFMT model are determined by the analytic hierarchy process,and the stratification of customer value is realized by using the K-means algorithm.Finally,customers are divided into four customer groups with different values.Then,the improved RFMT model is compared with the traditional RFM model,and the experimental results show that the improved RFMT model has better customer value classification effect.(4)An inventory demand forecasting model based on quantum particle swarm optimization algorithm and support vector regression is constructed.Firstly,the influencing factors of the demand for spare parts inventory are analyzed,a variety of forecasting methods are compared,and the support vector regression forecasting model is selected.The penalty factor and kernel parameters of the support vector regression model are optimized by using quantum particle swarm optimization algorithm,and an inventory demand forecasting model based on quantum particle swarm optimization algorithm and support vector regression is constructed.The experimental results show that,compared with the BP neural network model and the prediction model based on particle swarm optimization algorithm and support vector regression,the average absolute percentage errors of the prediction model based on quantum particle swarm optimization algorithm and support vector regression are reduced by 1.63% and 0.68%,respectively.Therefore,the prediction model based on quantum particle swarm optimization algorithm and support vector regression has better prediction effect.Finally,the layered results of the customer value optimization model and the prediction results of the inventory demand prediction model are fed back to the sales and logistics links of the value chain of the whole life cycle of products and services respectively,which jointly realizes the optimization and value-added of the value chain of the whole life cycle of products and services.