| Enterprise operating management is to make decisions on the entire production and operation activities of the enterprise,which has an important strategic significance and role in the development of the enterprise,and the right way of operating management can lead the enterprise to success.With the continuous development of enterprises,the scale is getting bigger and bigger,and a good management style is conducive to further improving the economic efficiency and productivity of enterprises and helping to realize the sustainable development of enterprises.With the advent of the information age,the continuous development of software development technology,and the rise of deep learning and big data technology,more and more enterprises are actively adopting emerging information technology to complete the upgrading and transformation of various internal management systems and develop business management systems that are suitable for the enterprises themselves.M is a technology company focusing on the manufacture of intelligent products in all areas of life,with over 5,000 product categories on sale and annual sales revenue accounting for approximately 30% of the entire Group’s revenue,the production and sale of products has become an important business and one of the key sources of revenue within the Group.However,at present,the company still adopts the original business management model,with business management work done offline by employees,which is not only inefficient but also difficult to meet the demand.Therefore,it is urgent to establish an intelligent business management system.This thesis combines the company’s own business situation and the company’s demand for its business situation analysis,based on software engineering ideas,combined with deep learning technology,to design and implement an intelligent enterprise integrated management system.In order to realize the development of the management system,Java language is chosen and the design pattern of front and back-end separation is adopted.The back-end uses Spring Boot and My Batis framework;the front-end uses VUE framework,Element-ui as UI framework and style standard,Spread JS to display spreadsheet interface;the database uses My SQL.This system includes project development stage management module,new product management module,report management module,price change management module,personal center module and system management module.In order to improve the processing efficiency of financial data analysis,an automated report information recognition algorithm is designed and implemented in the report management module of this system.The algorithm is mainly divided into two parts: report image text detection and recognition,combined with OCR technology.First,the CTPNbased report image text detection algorithm is designed and implemented,and the Res Net50 network is introduced as the feature extraction network based on the original algorithm.Then,the CRNN-based report image text recognition algorithm is designed and implemented,introducing the Res Net residual network for feature extraction and adding an attention module to the recurrent layer for relevance weighting of the feature vectors.Finally,the experimental validation is carried out on the homemade report image dataset separately compared with the original algorithm,and the experimental results verify the effectiveness of the algorithm designed in this thesis.The system has now completed the development and testing of all modules and has been successfully put into use within the company.This system meets the company’s new demand for business management,improves business operation efficiency,reduces manual workload,and helps realize data-based decision-making,thus ultimately improving the company’s overall economic efficiency and achieving healthy,rapid,and sustainable development. |
PDF Full Text Request