How Industrial Internet of Things and ML Drive Industry 4.0

In the 1870s, with the widespread use of machine tools, Bessemer steelmaking technology, telegraphy, electrification, internal combustion engines, and oil and natural gas, industrial development accelerated again. Then, in the late 1960s, due to the emergence of digital computers, the third era of industrial subversion began, realizing computer-aided design (CAD), unified manufacturing of complex components, higher automation level and higher productivity

The Industrial Revolution began in Britain around the 1760s. It brought about an era in which coal-fired steam engines were used as the driving force to improve the level of mechanization and productivity, and transformed the society mainly agricultural into a society dominated by manufacturing. In the 1870s, with the widespread use of machine tools, Bessemer steelmaking technology, telegraphy, electrification, internal combustion engines, and oil and natural gas, industrial development accelerated again. Then, in the late 1960s, due to the emergence of digital computers, the third era of industrial subversion began, realizing computer-aided design (CAD), unified manufacturing of complex components, higher automation level and higher productivity.

Industry 4.0 is in full swing

Now, we find ourselves in "Industry 4.0", which is characterized by large-scale automation, intelligent technology, extensive machine-to-machine communication (M2M), and machine learning (ML) for traditional manufacturing and industrial practice. Although the key of Industry 3.0 and Industry 4.0 is digital information, the difference between them is that Industry 3.0 uses digital information to make better decisions, while Industry 4.0 can use the same information (and more information) to optimize things, basically without human intervention.

The "Industrial" Internet of Things (IIoT) is the core of this new manufacturing and production development stage. It is a platform supporting automation, M2M and ML operations. IIoT uses a feedback loop, in which the sensors monitor the operation process, and then use their data to control, improve and improve the machine operation.

The importance of high accuracy

Optimizing operations is important because manufacturing depends on high accuracy and repeatability. For example, the manufacturing of auto parts must meet the strict tolerance requirements, so that they can be fixed in thousands of specific models of cars with bolts and can work perfectly for many years. The smaller and more refined the products, the higher the precision required in the manufacturing process. Think of the mechanical devices of advanced watches, the windings of micro motors or the welding of smart phones.

Making the product right the first time can reduce potential site failures and costly maintenance claims. This can also save a lot of resources and money, especially if the parts are made of external materials or by machine tools with higher purchase and operation costs. Make the product right for the first time, which is also more environmentally friendly, saving the energy and carbon emissions required to produce new products to replace defective products.

Use ML in manufacturing

Process control is the key to ensure high accuracy again and again. Sensors and cameras can monitor the machine, measure the finished parts, find any small deviation of the product, and correct the production process before the parts deviate from the tolerance. In addition to the manufacturing process itself, many other factors will also affect the manufacturing process, so other sensors are used to track vibration, temperature, humidity and air quality.

However, continuous and large-scale manufacturing processes will produce a large number of sensor data, and most of these data show little change. Transferring and analyzing a large amount of data is time-consuming and expensive, and takes up a lot of energy consumption. On the contrary, sensors now begin to incorporate on-board edge processing and ML functions to learn how to find trends in basically unchanged data streams. When the trend of change is detected, the information will be sent to the cloud for analysis and action.

ML can also prevent problems that may arise from external factors in advance, such as increased humidity caused by workers coming to work, air flow caused by opening doors and windows, and temperature changes during the day and night. By using this information, you can adjust the process before potential problems occur.

Connect front desk and factory workshop

IIoT has not only changed the way products are manufactured, but also the way products are designed. Industry 3.0 makes full use of computers, and people engaged in design no longer need to communicate with people engaged in manufacturing. The engineer will obtain CAD output from the design room and use this information to program the machine tool. In addition to the advantages of labor-intensive and error-prone, Industry 3.0 missed the opportunity to improve design to help product manufacturing easier, cheaper and faster.

Industry 4.0 links the front desk with the factory workshop. With M2M communication, the design computer can talk with the machine tool and directly program to manufacture parts. The machine tool can talk with the design computer to let them know the bottleneck in the manufacturing process, so that the product can be redesigned and the manufacturing process can be simplified without affecting the function. The central computer can use all the design and manufacturing data to find the best way to manufacture future products, making them durable, repairable and easy to recycle at the end of service life.

IIoT ushers in the industrial 5.0 era

Investment in IIoT is expensive. However, wireless technology does reduce the cost of installation and wiring, and as the factory changes and expands, the network can be easily reconfigured. Moreover, with the improvement of productivity and the reduction of product failures, the long-term savings brought by better design and manufacturing are very prominent.

Industry 4.0 has arrived, and the development vision of Industry 5.0 is clear, which is sustainability. The EU pointed out that the next leap will promote industry "not only to take efficiency and productivity as the sole goal, but also to strengthen the role and contribution of industry to society. At the same time, the key is to respect the production limit of the earth." This is a commendable development goal, but also a huge challenge. IIoT will help people solve this problem.