Industry 4.0 is the global manufacturing trend. According to the report of the research institute, the market scale of the Internet of Things in the global manufacturing field in 2015 is estimated to be 52.9 billion U.S. dollars. It is estimated to grow to 1,332 billion U.S. dollars in 2020, with a compound annual growth rate of 20.3%. The industry The sensor output also increased from 2.54 billion U.S. dollars in 2016 to 4.05 billion U.S. dollars in 2020. Internet of Things era, sensing is king, all information must be sensed by sensors to measure and collect data. Internet of things want to go farther, depending on the degree of sensor development, whether it is RFID, voltage, temperature and humidity, gas and other environmental sensors, or security in the field of various images, thermal, infrared, indoor positioning, alarm sensor applications , And even the information collaboration, measurement and analysis to control of the machine. Only through these sensing components can we construct the IoT and the extremities of the IoT to complete the real-time data collection, retrieval, operation, analysis and analysis of the big data cloud database Data intelligence, both of which are key cornerstones to building an intelligent factory. Data collection and implementation of plant monitoring and management To science and technology plant, to improve product yield is its top priority, even if only to improve the process of 1% are willing to actively try, after all, human error is the main reason for the yield decline. For example, the machine tool can be embedded with various sensors that can detect the motor load, speed, voiceprint, vibration, current, etc., to enable the machine and equipment to intelligently collect information: When the Fork assembly precision components Accurately and steadily sensing the tiny force in its axial direction, grasp the change of force when the cassette is taken in and out, provide logical judgment and find the defective product. And when the aging of machinery and equipment, leading to unexpected downtime or failure, with the sensor to retrieve a variety of environmental information, including temperature and humidity, pH, gas, dust, etc., to analyze and determine the tolerance of their operating environment in order to achieve preventive measures The function, make the production line to be able to operate smoothly. In addition, industrial robots as the next important production tool, bear the brunt of human-computer interaction. When people work with machines at the same time, many sensors need to be installed to provide man-machine collaboration sensing mechanism to determine whether their mode of operation will cause harm or impact on people. Each displacement is located from the position of the robots, Of the security design to master the peripheral state of the robot arm; when the staff into the scope of common work, the arm will automatically slow down until the staff left, and then restore the original speed, so you can work without interruption, strengthen the safety of human-computer collaboration Sex. With the positioning tracking sensor, we can confirm that the personnel in the operation or other production-related equipment are in the right place. With the tracking of these personnel and objects, and the interaction between the two, managers can monitor the production process in an all-round way. This time to judge, which process needs to be adjusted. In fact, the reason why the plant monitors the data through a variety of sensors is nothing more than trying to master the quality of production and reach a certain process condition. With data acquisition during manufacturing, back-end platform operations can be further utilized for big data analytics to help businesses make money. Big data analysis to enhance competitiveness Who can quickly respond to market changes and produce diversified products is the winner, while big data is the best application for accelerating decision-making and predicting the future. Data collection will always be the focus of plant intelligence. However, if the information is not processed, it is a waste of storage space, and the data must be stored in the cloud through a combination of virtual information and the Cyber-Physical System. Analysis, the formation of decision-making, and then come back to guide the production. According to the long-term measurement, statistics and analysis of the relevant data of the production process, preventive maintenance, repair and even early warning system of advanced equipment control can be developed as early as possible when process quality gradually deviates and precision of equipment deteriorates. Achieve self-monitoring and forecasting capabilities to maintain product process stability and improve yield. The practical application of big data, you can Gou "turn on the lights factory" as an example: Foxconn through the Internet of things, accumulated many years online production of a variety of machine data, and the use of image recognition and machine learning and other technologies for these devices even connected Neural and brain big data analysis, come to the conclusion of each manufacturing process in the production process results; coupled with a variety of intelligent sensing and sensing network import, we can do to allow the machine to operate independently in the dark can also be completed production . Prof. Jane Zhanfu, a professor at Tsinghua University who built smart production for TSMC, once said: "In industry 4.0, the key lies in the decision-making behind the data." In the process of intelligent manufacturing, robots are not intended to completely replace manpower and human beings Roles are no longer "operators" of the workforce, but rather are promoted to the bottom line through big data analytics at the back end to become "designers", "decision makers" and "managers" of the process. Integration of the actual situation to create business value Identification and analysis are an extremely important part of security and production, and sensors and big data are the core technologies that help make smart manufacturing. Through the factory inside the high precision and high stability of the sensor to capture the required information, pull into the software platform to establish industrial IoT system, and then use the big data analysis to complete the integration of the actual situation, predict the performance of the system equipment and the future, improve risk control Transparency and effectiveness, and ultimately achieve zero failure, optimize production goals. How to make good use of the Internet of Things with innovative thinking, real-time monitoring and big data forecasting technologies, and independently optimizing the resource allocation in the production environment will be the key to success of the manufacturing plant for science and technology and will help China to enhance its international competitiveness and explore new market opportunities. .