| In recent years, with the development of national economy and the rapid growth of electricity demand, the competition among low-voltage electrical manufacturers have been intensified. The enterprises within industry actively adopt various marketing strategies to obtain high-quality and stable customer base, so more and more executives focus their attentions on the research of quote decision which is an important part of marketing strategies. This paper is based on the development of the Information System for low-voltage electrical manufacturers and researches the application of artificial intelligence in the quotation system. It first analyzes the characteristics and the factors affecting the quotation within the industry, focusing on the characteristics of customers, and then designs a workflow model in order to facilitate system development. At last, this paper classifies quick quote methods into many kinds, which are evaluated and analyzed against each other, and finds the best method adapting to the industry. The quotation system which is improved by the above method could establish the quotation model to regularize operations and also realize the algorithm to ensure accurate decision.The works done in this paper are as follows:(1) Offering the analysis of process-related factorsThis paper analyzes the features of low-voltage electrical products and quotation. And the further analysis of the factors about customers in quote also proceeds in it. In this paper, the idea of improvements to change the three-dimensional model based on the value to the model based on the value and role is suggested and enforced.(2) Modeling the workflow of quote processIt analyzes the current workflow of quote process within the industry and elaborates the workflow model by introducing the customer interface. In this paper, the customer-oriented workflow is analyzed and the petri net is established. With the diagram of the petri net we could analyze the structure of system processes and the situation of resource allocation.(3) Analyzing the algorithm used for quick quoteIt analyzes the common method of quick quote, focusing on analysis of mathematical statistics algorithms (the probability distribution algorithm, regression analysis algorithm, etc.) and artificial intelligence algorithms (neural network algorithm, fuzzy sets algorithm, etc.). With studying the principle and occasions of application, it shows the advantages and disadvantages of various algorithms.(4) Studying on method of quick quote based on BOMIt analyzes the principle of the configurable BOM of low-voltage electrical product. Research of clustering method for quote based on the DBOM (design BOM) is demonstrated in this paper. By analyzing FKM (Fuzzy K-Means) algorithm and SOM (Self-Organizing Map) algorithm, it shows the advantages and disadvantages of these algorithms, and researches on optimized FKM algorithm based on the optimized value of K which is obtained by AHP (Analytic Hierarchy Process) algorithm. Above algorithm is also demonstrated in the paper.(5) Fulfilling the functions of systemIt analyzes the customer requirements and the objective of quotation system and chooses the appropriate platform of the development and the operating environment according to the system architecture. It also designs the structure of database and the specific function modules, and completes implementation of system and testing of the logic of the code and operational efficiency.Through the actual inspection of enterprises, the application of the offer system has achieved good effect. This paper tests the feasibility of the quotation system and the practicality of the algorithm, and shows that the system has nice prospects. |
PDF Full Text Request