The industrial Internet can connect all the equipment, sensors and robots, so that we can have a better understanding of the equipment itself, and more importantly, how to continuously improve the production process with this information. From the perspective of manufacturing life cycle, the industrial Internet can bring changes to the three main aspects that factories pay attention to: productivity efficiency, uptime and product qualification rate, and provide new ideas for all links of the whole manufacturing life cycle.
For example, when an electronic factory assembles mobile phones, computers and other products, the general factory practice is that workers rely on precision tooling to ensure the assembly accuracy. Each assembly link needs to pass the test, and the test results determine whether the previous process can pass. General factories can't trace back to the previous process, but the industrial Internet can adjust the parameters needed in the assembly process. Taking the mobile phone as an example, it is assumed that the assembly accuracy of various components inside is 30 microns. According to the final test results, the tolerance of the components has always deviated to 50 microns on one side. Using the industrial Internet, this production data can be fed back to the design layer through the form of industrial Internet. By analyzing, determining and adjusting the parameters of a certain assembly link in the previous process, systematic errors can be eliminated. This shows that the actual data in the production process obtained through the industrial Internet can improve the final production quality, which reflects the value of the industrial Internet both from the perspective of the capability ladder and the life cycle.
Application of artificial intelligence in manufacturing
Artificial intelligence is widely used in manufacturing. The factory comprehensively judges the possibility of robot problems based on the historical data of thousands of robots, and makes preventive diagnosis on the operation of the whole equipment. The system uses machine learning algorithms to make judgments based on a large number of historical data, and can carry out preventive maintenance on the equipment operation status. ABB began to connect robots to servers in 2007 to share data such as potential problems and equipment operation. After more than ten years of data accumulation, we have mastered a large number of operation data of various factories around the world. In the future, we will further use machine learning to launch cloud platform based preventive diagnosis and maintenance services through data analysis. In addition to preventive maintenance, AI can also bring some ideas to solve the bottleneck problems in the whole production process, such as the body welding process production line of an automobile factory. The most important point is that in the field of human robot interaction in the future, artificial intelligence will have great achievements. At present, human-computer interaction based on production equipment is still at a relatively traditional stage, requiring people to input instructions to realize the interaction process. Artificial intelligence technology can make the interaction between people and intelligent robots more natural in the future.
Development trends and application scenarios of future robots
With the change of external factors, the development speed of industrial robots in the past 10 years is somewhat surprising, both globally and in China. Globally, industrial robots maintain an annual growth rate of 15% to 20%. In China, according to abb, the growth rate of China's industrial robot market exceeded 50% in 2017.
From the perspective of products and technologies, the structure and application technology of industrial robots have not changed much since the 1970s. Most industrial robots are used to complete repetitive, simple, boring and even dangerous work. At present, industrial robots are mainly used in large-scale production with production capacity and output demand, such as automobile, electronics, food and beverage and other industries. Due to the obvious scale effect of the automobile industry, the automobile industry has always been the most widely used industry for industrial robots. Since last year, the electronics industry has become the largest user of industrial robots due to the increase of demand in the Chinese market. At the same time, robots are also used in traditional industries such as food and beverage, metal products and plastic products.
In terms of application, the logistics and retail industry will become a new application field of robots in the future due to the high demand for human resources and the rapid development of industrial scale. Sorting work required by both the warehouse and the logistics industry; Whether it is loading, replenishment or retail shelf management, it is suitable for robot application scenarios. Therefore, the logistics and retail industry will be the next emerging industry, and also the beginning of the penetration of robots from industry to service industry.
Due to aging and rising labor costs, in Europe, the demand for robots has gradually penetrated from large factories to small and medium-sized factories and even small workshops. For small and medium-sized enterprises, the production is characterized by small batches and multiple varieties, and the production process is constantly switched. Using traditional industrial robots will consume too much switching time. Therefore, small and medium-sized enterprises need small and flexible products, and the ease of use of robots is the key.
If compared with the development of the computer industry, industrial robots are still at the stage of "supercomputers", and the era of "personal computers" for robots has not yet arrived. Looking back at the history of computer from invention to popularization, it is found that the price reduction, the volume reduction, the application is easy to operate and the user-friendly graphical interface are the three important factors that ultimately make the computer enter thousands of households from the laboratory. Similarly, cost, human-machine cooperation safety and ease of use are the limiting factors for robots to penetrate from industry to other fields. In the process of the penetration of robots from industry to consumption, human-computer interaction is one of the factors restricting the development of robots. Whether in industry or other scenarios, how can machines better interact with people? How to better assist people to complete the work in the work and production process? Artificial intelligence brings the possibility to solve these problems. In terms of the reliability of human-computer interaction, there is still a breakthrough to be made in technology. In terms of industrial robots, robots in factories now can execute instructions exactly without making mistakes, because engineering design, installation and commissioning need to operate on the production line through instructions. The ideal situation in the future is that robots can interact with people in a more natural way like apprentices, and can change from apprentices to mature workers under the guidance of people.