With the exploration of the fourth industrial revolution gradually go deeper, data driven elements of the role of the new digital economy has become increasingly prominent as compared with traditional elements, have a stronger data reproducibility, easier to share, and unlimited growth and supply the limited natural resources to break supply constraints on economic growth possible, is important for fostering economic development new momentum and open up the development of new roads. Manufacturing is the main foundation of the national economy and the depth of both the development and utilization of the physical development of the digital economy sign industry data, but also the difficulties the digital transformation of traditional industries and key. With the proliferation of a new generation of information technology and depth of applications, industrial Internet platform came into being, the acquisition of industrial products and production activities throughout the life cycle of data, storage, analysis, sharing, application, and other value-added services is showing rapid vitality It spawned a series of new products, new models, new formats and push the manufacturing sector accelerated to digital, networked, intelligent change. Industrial Internet platform to make it possible to release the full value of the data, a number of enterprises by the depth of excavation consumer behavior data, has achieved remarkable results in the development of business change in consumer internet, and spawned a new batch of data-driven business mode. But in the industrial field, because the data collection more difficult, more complex types of data, more data application of professional difficulty, development and utilization of data applications appear to be lagging behind. The development of industry Internet platform enables industry has witnessed tremendous change in the data collection, analysis and application ways to create value for industrial data has opened up a vast space. First, the Internet industry platform to bring a fundamental change in the data collection. Traditional industrial data acquisition mainly through manual entry, surveys, telephone follow-up, etc. This embodiment obtained a small number of data collection, poor aging, low precision and high cost. With a bar code, two-dimensional code, RFID, industrial sensors, industrial automation systems, CAD / CAM / CAE / CAI, 5G technologies are widely used in industry, industrial data collection methods emerged nature of change CONTROL ENGINEERING China Copyright , industrial enterprises available data sources richer, fuller, more timely, more types of data (available only from IT information system data, can be obtained from the equipment, staff, production environment the OT data), laid Fresher innovative applications for industrial data, accurate, Integrity of the data base. Second, the Internet platform for accelerated industrial innovative breakthrough industrial data analysis methods. Industrial equipment can produce continuous data, and can achieve data acquisition millisecond frequency of large amounts of data, and the type of data involved, a variety of data formats and data structures, complex professional relationships, traditional data analysis methods can not effectively deal with. In addition, conventional data analysis is often the data is stored at the edge or in a separate system, so that the data network, exchange, interoperability is blocked, and can not effectively extract the relevant data CONTROL ENGINEERING China Copyright collaborative desired application , hindering the relevant data analysis, system utilization and conversion value. Internet-based industrial platform , the innovative application of edge computing, big data, artificial intelligence and other emerging technologies can break the causal paradigm and subjective cognitive limitations of traditional data analysis, data from the mass industrial the correlation between identification data, in-depth tap the potential value of the associated data, the decision to form the new paradigm and new insights. Third, the Internet industry to become the best platform for carrier data value creation. Most of the traditional industrial data applications approach is based on historical data to carry out part of the local business analysis, it will refine some individuals experience into causal rules, and promote the use. Emergence of the Internet industry platform to achieve industrial data online, real-time, dynamic, cross-border analysis applications, providing a more comprehensive, faster and more accurate data support for large industrial enterprises of goods and services, production and management decision-making, the rules of cause and effect into a personalized based on accurate decisions related data. On this basis, the industrial Internet platform enables organizations to accelerate to break the traditional business chimney model of development, greatly improving the level of automation of data flow, to achieve leapfrog area, across the organization’s devices, services, innovation and collaboration market data applications, to build an open development model ecosystem and innovation. Clear results thanks to innovative applications based on industry data acquired Internet platform in industrial data acquisition, analysis, changes in application methods, data-driven applications for the promotion of industrial innovation and the value of the Internet platform floor and show the effectiveness of rapid development. According to the data flow range and complexity of the innovative applications of industrial internet platform has produced clear results at different levels of equipment, enterprise and supply chain. First, the device-level data applications for online, real-time device management. Device-level data should beWith the most basic data applications, whether industrial enterprises for the advanced analysis of networked devices running information from, or device manufacturers to understand their own products in the whole life cycle, management, diagnostics and maintenance, need to rely solely on the manual from the original experience into a more scientific and more efficient decision-making based on the data obtained. Device-level data applications through interconnection industrial equipment, the operating state of the device into a digital, visualization, and further analysis of the data of the industrial device based on collected to data, such as fault diagnosis by plant operation related information whether recognition of its technical condition is normal, to determine the nature and location of failure to find fault causes, forecasting fault trends and propose appropriate countermeasures; predictive maintenance is based on historical data and real-time data in anticipation of the impending problems, which can effectively shorten equipment unplanned downtime, ensure production plans, reduce operation and maintenance costs; product-service equipment manufacturers shift from selling products to the business model to provide services (such as ROMS, predictive maintenance, etc.) can bring innovative value to customers to provide a new source of revenue for manufacturers. Second, enterprise-class data applications, enterprise lean management. Enterprise data applications means to get through OT enterprise data and IT data, based on data analysis to provide scientific decision-making for the overall business and operational excellence, lean enterprise management. Enterprise data applications were abundant, including quality management, energy management, security management, manufacturing process management. Total quality management is a typical example of enterprise data applications. The main traditional quality management for quality testing after the end of production. This high single link detection mode dependent personnel experienced Copyright Control Engineering , resulting in missed easily, problems such as conflicting standards; this detection system can not recognize the reason for failure of the product and links, making it impossible to enhance the rate of qualified products fundamentally. Data-driven applications can be integrated into a comprehensive quality management equipment, staff, technology, environment, quality control and other multi-party data, to point to the surface to form a comprehensive quality management solution. In the manufacturing sector, for the device (based device management to ensure production quality consistency), employees (based on machine vision to identify employees erroneous operation), the environment (to achieve environmental intelligent monitoring), process (the optimal solution identification parameters based on machine learning) aspect, formed based on a single point of application data quality management; quality control methods employed in the link quality based on machine vision, ensure consistent quality resultsAnd significantly improve the accuracy than other quality control methods, quality control analysis results based on machine learning, the use of intelligent defect tracking to determine the cause of quality problems and links, and to take appropriate improvement measures, forming a closed-loop quality management, a comprehensive quality management . Third-level data supply chain applications, supply chain collaborative dynamic precision. Supply chain applications are industry-level data in the extended data between enterprises, interaction involves multiple companies, suppliers, distributors, customers and other parties Control Engineering Copyright , including planning, procurement, production, logistics and a series of links. Advanced scheduling is a typical example of supply chain application level data. At the same time enterprises to form production plan based on demand forecasts, plan orders can be automatically broken down into the appropriate procurement planning, logistics plans, and exchanging data with suppliers, logistics service providers in the system, according to the actual situation may arise forecast changes (such as the case of a particular product suppliers appear insufficient capacity, freight transportation companies can not normal, etc.), or (such as emergency orders, missing parts, etc.) to adjust the production schedule for unexpected situations arise, accurate dynamic supply chain collaboration. Supply chain application-level data through data sharing between enterprises, transaction data, but also to achieve social manufacturing capabilities open sharing, supply chain, financial, UBI, financial leasing industry and finance innovative model to provide new momentum for China’s economic development. Further accelerate innovative data-driven applications to further accelerate the fall to promote innovative data-driven applications fall promotion, you need to build a sound industrial data management system to pull data value for data applications continue to expand the scale and enhance the value of the data, release of data to maximize performance. First, improving the data management system, we continue to optimize data utilization environment. Robust industrial data transaction, shared systems. Industrial property rights be defined attribute data resources from a legal perspective, improving valuation data, establish and improve the trading mechanism and industrial data resources pricing mechanism to explore the establishment of a national industrial trade data centers; improving industrial data security and privacy protection system. Promote the improvement of industrial grade data security protection system, the establishment of both security and development of industrial data management and security system, strengthen the protection of trade secrets, sensitive data and privacy. The second is to strengthen the value of mining, to expand the scale of development and utilization of data. Strengthen project development and utilization of data values fall. The data value for traction, the big data technologies, mechanism model and analyze the results more convenient, accurate, low costUse of the various business scenarios, to help enterprises to achieve cost quality and efficiency, optimize decision-making and business extension; data collaboration to expand the scope, scale and efficiency for data applications doubled. By open up the industry chain, value chain data, enhance cross-industry applicability of data, data acquisition time, multi-repeated use, achieve economies of scale and value doubled, promote the formation of a new model of industrial data applications, the new format. Third, to conduct pilot demonstration data innovation, speed up the development and utilization of data mode iterative innovation. Typical data applications create a number of cases, lead the way to strengthen the pilot demonstration. Encourage the foundation, there is a demand, companies have an incentive to carry out the depth development and utilization of data, strengthen industrial development model based on the mechanism of generation of information technology, a group based on digital twin, accelerate the formation of large data analysis, artificial intelligence, data-driven and typical cases solutions; foster innovation eco-system data, speed up innovation-driven iterative development model of manufacturing data. On the basis of the data range, scope of applications continue to expand the depth of the development, strengthen the innovation eco-system data elements covering the technical data, applications, industry, capital, and human resources, to explore the formation of innovative data-driven and iterative development model provides manufacturing strong support.