Research And Implementation About The Forecasting For Enterprise Equipment's Life Based On Enterprise Asset Management System

Enterprise assets are the important means of production in the process of enterprise production and operation,the core competitiveness of group enterprises,and the basis for the existence and development of enterprise productivity.The amount and quality of assets in a certain period of an enterprise directly affect its production capacity and profitability.The asset management system is an information platform to realize the full life cycle management of enterprise assets through work order and workflow approval on the basis of informatization and label management of tangible assets.Enterprise equipment is an important part of enterprise assets.Based on the analysis and research of the business process and data flow of the asset management system,this paper abstracted a number of factors affecting the life of equipment,and built a life prediction model of enterprise equipment on the basis of selected characteristic factors.The prediction of equipment life of an enterprise can provide important reference for the determination of spare parts quantity,procurement plan,maintenance and repair plan,asset depreciation calculation and other aspects of the enterprise,so as to indirectly improve the use efficiency of enterprise assets and optimize the management mode of the enterprise.Because of the wide application of BP neural network in prediction and its excellent nonlinear generalization ability,BP neural network is used to construct the equipment life prediction model.The prediction model is constructed by combining theoretical analysis and simulation experiment.The direction of the experiment is determined by theoretical research and the final model is determined by experimental results.Thesis first by many simulation experiments to determine the structure and parameters of BP neural network model,and on the basis of theoretical analysis and experimental model of the improved direction,after using floating-point coding genetic algorithm for global optimization neural network to improve the stability of the model,and finally on the basis of the simulation experiment of the fitness function of GA-BP model and crossover operator is improved,and improve the efficiency of the training of the model,the improved GA-BP model is more suitable for enterprise equipment service life prediction.Based on the determination of factors affecting equipment life and the construction of effective prediction model,this paper designs and implements ETL model based on hadoop,hive,sqoop and other big data processing technologies,and designs and realizes enterprise equipment life prediction system combining python prediction service and javaWeb and other technologies on this model.The design of the system implementation on a certain enterprise asset management system and give full consideration to the asset management system technology implementation and operation status of the original system to meet the low dependence,ensure the forecasting of time-varying and can support large amount of data in the data processing scale,to build assets prediction system on enterprise asset management system provides a reference model of the sample.