New Trend in Intelligent Manufacturing: Wireless Device Networking and Data Analysis Across Processes

The TPCA Show 2019 has just come to an end, and this year's exhibition continues to focus on intelligent manufacturing issues, especially in the PCB industry. Since 2018, the PCB industry has established demonstration teams and launched a series of intelligent manufacturing solutions for soft and hard boards as well as equipment, targeting small and medium-sized enterprises as the main target to drive their upgrading and transformation. So far, it has achieved considerable results. However, in fact, not only the PCB industry, but also the communication industry and traditional manufacturing industry, intelligent manufacturing is still the focus of attention from all walks of life at present. As the previous seeds have gradually begun to bear fruit or advanced to the next stage, Xie Peihong, the head of the Digital Service Innovation Research Institute of the Institute for Information Industry, has also pointed out several development trends worth paying attention to in the future of intelligent manufacturing.

If we look at the development of intelligent manufacturing from the perspective of devices, production to operation, undoubtedly, the networking and automation of devices are the basic development trend in terms of devices, especially as the device side gradually adds wireless technology in response to the development of 5G in the future. Xie Peihong observed that although the development was not yet complete in the past few years, the market did not pay much attention to whether the device side networking would move towards wireless. However, recently, the demand for wireless networking in the market is gradually growing.

This is mainly due to the complex factory environment issues, where the wiring itself needs to be maintained and the distance is limited. The wiring problem remains a major challenge for operators. Secondly, in response to the demand for M2M networking, not only does the production data collection equipment need to be networked, but the production equipment also needs to be integrated and communicated with peripheral automation equipment such as robots and AGVs. Especially for AGVs, after breaking away from magnetic stripe track guidance and increasing their deployment flexibility, they also rely on wireless signals for positioning and navigation. Mobile devices themselves are also relatively difficult to deploy online. Therefore, with the increasing demand for intelligent functions in the factory, the development trend of wireless networking on the device side is expected to gradually mature in the next 3 to 4 years.

In addition, collecting device side production data and digitizing and visualizing it is still a major application in intelligent manufacturing. However, it is worth noting that the focus of data collection is not on a large amount of "collection", but on how to use the data and what its significance is. Xie Peihong believes that visualization applications are closely related to the quality, resolution, and traceability of data collection, which are also closely connected to backend AI analysis applications. Therefore, in the future, there will be a demand for more refined data collection on the device side, resulting in higher data resolution and more diverse data collection in response to data analysis.

On the production side, through visual production management, there is now a more comprehensive integration from equipment to personnel, and even further with management processes. The traditional manual scheduling method has gradually been transformed into dynamic scheduling based on the system. It can be seen that visual applications, which are quite basic, have gradually been extended from production line applications to integration with management processes. Due to the fact that there are usually more than one type of production equipment on the production site, the production mode should adapt to single machine, system to whole line. In the future, data analysis on the production end will also move from single station to cross process analysis, and AI applications such as process simulation and quality prediction will gradually deepen.

However, it is worth noting that when companies are determined to invest in the development of intelligent manufacturing, in fact, exploring the demand side and implementation side are two different operational directions. Xie Peihong pointed out that in fact, enterprises should first consider the overall development direction from the operational side in order to clarify the purpose of data collection in the future. Therefore, this part must integrate the concept of business model transformation in order to establish how to adjust operations or business models by integrating internal and external data information. After establishing the enterprise goals, further planning is carried out on how to transform and upgrade the production site to match the new operating model.

However, when it comes to practical implementation, it is necessary to start the first step from the device side intelligence, gradually introduce the entire factory production management process, and finally enter the operational level, using data to assist managers in intelligent decision analysis. Therefore, in the future, focusing on the development of the operational side will clearly lead to a gradual upward push from the development of applications at the bottom level. Although many enterprises currently have relevant information systems to assist them in decision-making, when it comes to some more complex supply chain management decisions in the future, IT can no longer be alone. Through the integration of data across production processes at the bottom level and the convergence of information between IT and OT, it is possible to make overall operational decisions more accurate and real-time.

Author: DIGITIMES Liao Jiayi

Editor He Yang of this website/Organizing