Research On Data Space Construction Of Flexible Packaging Smart Enterprise Based On Big Data

Standing at a new stage of historical development,China’s industrial enterprises are steadily marching towards the direction of Made in China 2025.With the widespread application of 5G,big data,cloud computing and other emerging information technology,flexible packaging production enterprises production and business links will generate a huge amount of data,but most of the printing and packaging production enterprises in China are small and medium-sized enterprises,they have no small difficulties in the storage and analysis of these data.Enterprises in the process of production information technology due to poor information interaction,data is not fully utilized,it is difficult to grasp the direction of production improvement and process optimization direction,to help enterprises to solve the problem of data storage and analysis,from a large amount of data to analyze the information beneficial to the enterprise itself,the development of enterprises is of vital importance.In order to solve the above problems,this study takes a flexible packaging production enterprise as the entity object for the construction of intelligent data space research.The mainstream big data framework and its ecology are applied to analyze the order business data and workshop production data of the flexible packaging enterprise.The main docking enterprise data storage and calculation analysis problems,design and implementation of a big data framework-based enterprise wisdom data space,the flexible packaging production enterprise production site data and production order data related to storage and analysis,the flexible packaging production enterprise workshop production data and business order data combined to achieve the whole process of production product information collection,to improve the enterprise to enhance production decision-making Provide data support to improve the efficiency of production.The main technical difficulties solved.The data is not uniform and there are problems such as missing data.The distributed ETL system is applied to standardize and unify the collected data,so as to improve the accuracy and integrity of the data.The data format is diverse,and it is difficult to store the data uniformly.Design and implement a data warehouse system based on Spark+Hive to determine the dimensional relationship of stored data,including the dimension of time relationship,geography,workshop information,administrator,and order.Rely on Hive’s powerful data encapsulation ability to transform and store data.The problem that enterprise employees find it difficult to accept complex data.Apply the mainstream big data framework and its ecology to refine the data.Through an intuitive form,the data is presented in a simple and clear way to prompt production and maintenance operations.This paper realizes the analysis of order business data and workshop production data of flexible packaging manufacturing enterprises according to the overall architecture design of intelligent data space,and realizes the storage of massive data through HDFS distributed storage system.The Spark computing engine is used to calculate and analyze large-scale data,and the decision tree algorithm is used to realize the prediction of the decision of the flexible packaging production enterprise and the analysis of customers.By differentiating data refinement indicators,the productivity and decision-making ability of flexible packaging manufacturers are effectively improved.The results show the possibility of building an intelligent data space for small and medium-sized flexible packaging manufacturers with a very low budget,which is important for the intelligent development of information technology in flexible packaging manufacturers.