Design And Implementation Of Information Service Platform For Manufacturing Enterprises Based On Big Data Technology

In the context of globalization,manufacturing enterprises are facing various competitive pressures at home and abroad,and the production mode of enterprises tends to be more inclined to multi-species.small batch,personalized and customized.The existing information system of enterprises exists some problems,such as delayed data sharing,information asymmetry and lagging information feedback,etc.,it's unable to carry out dynamic identification and real-time response to data,which it has been unable to meet the production requirements of the industry.Therefore,it is urgent to carry out information integration,knowledge sharing,real-time feedback for manufacturing enterprises with the help of big data technology.This paper takes manufacturing enterprises as the research background and W enterprise as the research object,based on big data technology,the design and implementation of enterprise informatization service platform are carried out.The main work of this paper is as follows.1.Overall platform design.Firstly,the total demand of the platform design is analyzed by taking the manufacturing enterprise as the entry point,and the W enterprise is taken as the research object to comb out the problems of low technological process design efficiency and low production rate/output rate in this enterprise.Then,on the basis of in-depth research on the information construction of manufacturing enterprises,the general framework of the platform is designed.Finally,according to the requirements of platform design the data import,data algorithm,data analysis,data storage,data export and data display function modules of the platform and focus on technological process design and production rate/output rate problems that plague W enterprise,the design of data algorithm module and data analysis module are described in detail,at the same time,the development tools used in the platform for the technology selection are carried out.2.Platform-related technologies.This paper analyzes several key technologies in the whole process of building information service platforms such as Spark distributed computing framework,HDFS distributed file storage,HBase non-relational database,Kafka message middleware and deep learning algorithm.Spark is used to realize real-time distributed calculation of data.HBase and HDFS are used to realize the dynamic expansion of the platform and massively and efficiently query of data.Kafka is used to ensure the efficiency and security of data;ACGAN network is used to ensure the image processing of small sample data sets.3.Platform building and module implementation.The Hadoop/Spark distributed framework is used to build an information service platform for manufacturing enterprises,and verify the various functional modules of the platform in combination with the actual production data of W enterprise.The data import module provides the data source for the platform to realize the data collection function;in the data algorithm module,a data-driven model represented by the ACGAN network model was proposed to design the technological process,and the trained model is combined with the big data platform to realize the efficient design of the technological process through the human-computer interaction interface;in the data analysis module,the data in the data import module is captured by Spark Streaming to complete the data pre-processing and index statistics of production rate/output rate;the data storage module completes the storage of the result data processed by the data analysis module;based on the B/S architecture,it realizes the functions of real-time interaction and display of data,it also provides reference for enterprises with the same attributes in data collection,data processing and data storage.Through this study,it is found that the manufacturing enterprise information service platform built by Hadoop and Spark can not only quickly stably realize the design of the technological process,but also accurately monitor the production rate/output rate data in real time,which it verifies the feasibility of offline data and real-time data in the platform,at the same time,it ensures the high efficiency,stability and real-time performance of the platform.