Research On Optimization Technology Of Automobile Service Value Chain Based On Third Party Platform

The automobile market has changed from the incremental era to the inventory era,and the competition among automobile enterprises has begun to transform gradually,and the importance of automobile after-sales service has become increasingly prominent.The scale of automobile after-sales market is expanding and the market competition is becoming increasingly fierce.Automobile enterprises need to gain competitive advantages by conducting some business activities related to strategy more cheaply or excellently than competitors in the perspective and way of value chain.With the arrival of the industrial Internet era,digital,networked and intelligent transformation of automobile after-sales service is just around the corner.Only with the help of the new generation of information technology,automobile enterprises can quickly build and reconstruct the value chain of automobile after-sales service that adapts to the changing market environment,so as to meet the challenge of the new era and continuously win the competitive advantage in the after-sales market.Based on the National Key Research and Development Program,the National Science and Technology Support Program,and the National High Technology Research and Development Program(863 Program),this dissertation focuses on the optimization of automobile service value chain under the environment of third-party platform of industrial chain collaboration,proposes the optimization scheme of automobile service value chain under the environment of third-party platform,and studies the relevant theories,algorithm model and key technologies of the supporting scheme.Specific contents include the following aspects:(1)A variety of automobile service models are compared and their problems are pointed out.Based on the automobile service mode and business content of the third party platform of industry chain collaboration,the automobile service value chain model under the environment of the third party platform is constructed.According to the optimization method of manufacturing value chain,the optimization method system of automobile service value chain is constructed,the overall optimization scheme of automobile service value chain under the environment of third-party platform of industrial chain collaboration is designed,and the effectiveness of the optimization method supporting the overall scheme is demonstrated.(2)According to the needs of enterprise organization optimization of automobile service value chain under third-party platform environment,this dissertation analyzes the characteristics of evaluation and optimization of cooperative enterprises in service value chain,and proposes a dynamic optimization scheme of cooperative enterprises in service value chain under third-party platform environment.According to the automobile service value chain cooperative enterprise optimization,not only should different core enterprises have different subjective requirements and standards for cooperative enterprises,but also should take into account the objective characteristics of cooperative enterprises to be evaluated,comprehensive integration model is put forward,based on the analytic hierarchy process(ahp)to determine the subjective weights,with fuzzy rough set method to determine the objective weights,integration of subjective and objective weights to get comprehensive weight coefficient.Based on the extension evaluation method,the advantages of the cooperative enterprise group are evaluated,and the grade division and the order of advantages and disadvantages of each cooperative enterprise group are generated.Taking the optimization of after-sales service providers of a service value chain on a third-party platform of automobile industry chain collaboration as an example,it is verified that the optimization model in this dissertation is superior to other models.(3)Aiming at the customer value optimization demand of the automobile service value chain under the third-party platform environment,the characteristics of customer segmentation in the service value chain are analyzed,and the customer dynamic segmentation scheme of the service value chain under the third-party platform environment is proposed.Taking into account the advantages of clustering integration,semi-supervised learning and spectral clustering algorithm,the semi-supervised spectral clustering integration(SSSCE)model and the semi-supervised spectral clustering integration(CPSSSCE)model based on constraint projection are proposed.According to the purpose and characteristics of specific customer segmentation in service value chain,the RFMD segmentation index model is designed.Based on UCI standard data set,the segmentation accuracy of SSSCE model and CPSSSCE model is better than that of the contrast clustering integration model.Taking the example data of a service value chain of a third-party platform of automobile industry chain collaboration as a sample,and based on the unknown label evaluation index,the subdivision result of the semi-supervised spectral clustering ensemble algorithm is proved to be better.The different customer groups after subdivision are listed with their characteristics of the maintenance advice and guidance strategy.(4)Aiming at the quality optimization requirements of the automobile service value chain under the third-party platform environment,this dissertation analyzes the characteristics of the fault data sample and proposes a correlation analysis scheme based on the total fault data sample.In view of the large amount of fault data in the service value chain and the fast growth of intermediate results,as well as the defects of the traditional FP-growth algorithm in processing massive data,an improved parallel FP-growth algorithm based on balanced grouping and pruning strategy was proposed.The pruning strategy was used to optimize the frequent pattern mining process,and the balanced grouping idea was used to optimize the parallel execution process.Based on the standard data set,the parallel FP-growth algorithm is proved to be superior to other algorithms in time performance.Taking the failure maintenance record of a service value chain of a third-party platform of automobile industry chain as a sample,the single fault parts with high support count are calculated,and a small number of associated fault parts with high confidence are generated according to the frequent binomial set of the fault parts.(5)Aiming at the inventory optimization requirements of the automobile service value chain under the third-party platform environment,this dissertation studies the characteristics of the spare parts inventory of the service value chain,and puts forward a forecast scheme for the spare parts demand of the service value chain under the third-party platform environment.In view of the inventory characteristics and the factors affecting the demand for parts of a particular type of storage point(parts transfer warehouse)in the service value chain,a forecasting model of accessory demand based on support vector regression and quantum particle swarm algorithm is constructed,and the penalty factor and nuclear parameters in support vector regression are optimized by quantum particle swarm algorithm.The "front bumper ontology" data of a specific transfer bank of a service value chain on a collaborative third-party platform of automobile industry chain was used as a sample for simulation analysis.It is verified that the prediction model has certain advantages in prediction accuracy and time performance.