The Research Of Across-Chain Parts Database And Search System Based On Collaborative Platform Of Automobile Industrial Chains

With the development of cluster supply chain, the methods of cost saving and efficiency improvement in enterprises is no longer limited to their own capabilities and resources. The core enterprise in a single supply chain not only needs to integrate the resources on the upper and lower levels of the chain, but also needs to pay attention to the coopetition between different supply chains, so as to improve the the speed of information exchange and feedback market between supply chain.Accessories supply is an important part of the auto industry chain, different user roles all require to know information about accessories, these problems were solved by constructing the cross-chain database of parts and developing search system.According to the structure features of each enterprise database, started with design the cross chain commodity database.we studied the materialized view technology to improve the response speed of parts database,and used the method of pedigree aggregation to induce the query set which are commonly used, Using the materialized weight formula considering the query cost and maintenance cost, selected appropriate query statements for materialization, and ensured the accuracy and validity of the data by refreshing the materialized view periodically.Then refer to the keyword search technology on relational database to solve the user demand, and collected the information to build the common vocabulary of the parts, and used the word similarity calculation method based on synonyms to expand the key words which the user input,,and choosed a search method based on full text indexed and schema. First, find the tuple containing the user input keywords through the full-text index. Then use the key tuple set to expand the schema of parts database, and generate the candidate network by improved algorithm, we propose a sorting mechanism for tuple connected tree by fusing word similarity, translated into SQL statement in order. Finally, gained results by handing to the database, displayed these results to users in a two-dimensional table. At the same time, the improved candidate network generation algorithm and the original algorithm are tested, and the algorithm effects are compared from three aspects of recall, precision and F1 value.This paper analyzes the process of parts library data parts sales business as an example, to summarize the common complex query aggregation process diagram of lineage, create materialized view to improve the response speed of parts library. The system is developed by using SQLserver database and C# language, and the three layer architecture based on B/S mode is used to design and implement an accessory search system based on industry chain collaboration platform.