Research On Module Parameters Planning Driven By Customer Requirements

In recent years,the customer requirement of mass customization products has been towards to be more diverse and individual,which brought new challenges to the product design.In order to cope with this challenge,enterprises need to develop a variety of products and improve the efficiency of product redesign,while reducing manufacturing costs.Therefore,the modular strategy has become an important method of product design.In the process of modular design,the parameter value planning is an important content in the product portfolio planning.The rationality of the model parameters directly affects customer satisfaction and product manufacturing cost.However,there were some limitations of the existing module parameter planning method,that is to say,the customer requirement analysis was not comprehensive or objective and the problem of multi parameter combination planning could not be solved.To deal with the weakness of existing methods,a three-stage module parameters planning approach driven by customer requirements was proposed.At the first stage,customer requirement information was collected from the enterprise information system,product running state and the Internet these three channels,which was then integrated and preprocessed by the data transformation methods.The importance of customer requirement for module parameters was determined by the random forest machine learning method,and then to access to customer requirements and module parameters correlation matrix.At the second stage,based on the correlation matrix between the customer requirement and the module parameters,the fuzzy C means method was applied to cluster the customer's demand,and the number of cluster was the number of the value of the module parameter.Then the QFD method was applied to determine the range of module parameters,based on which,the customer requirement value of the cluster center was mapped to get the specific value of the module parameter by the customer requirements satisfaction model.At the third stage,a mathematical model to maximize the customer requirements coverage was constructed to get the optimal module parameter combinations.The manufacturing cost and the correlation among parameters were listed as constraints.A main luffing module of crawler crane was planned to demonstrate the rationality and effectiveness of proposed approach.Besides,the sensitivity analysis showed the robustness of this approach.