中文網
Taipei
Mon, Aug 8, 2022
15:04
mostly clear
33°C
CONNECT WITH US

AR, AI, IoT all indispensable in implementing Industry 4.0

Sandy Du, DIGITIMES Asia, Taipei 0

In an era when the consumer experience is paramount, AIoT has become the first choice for companies to create services tailored for their consumers. Faced with rapid changes in demand of the consumer market, the global manufacturing industry is making large strides toward Industry 4.0; and AIoT is an important solution for smart manufacturing. Although research institutions define smart manufacturing differently, their definitions all note that it must be able to meet customer requirements, generate optimal decisions, and improve financial investment, reduce the inventory, and enhance contingency management capabilities. Only then can it help to enhance overall competitiveness and lay a solid foundation for the long-term development of companies.

Chao-Lung Yang, professor of the Department of Industrial Management and EMBA Program director, NTUST, said that the goal of production management in manufacturing is to maximize practical capabilities, order fulfillment rate, maintain the high quality level, and minimize the production cycle, decision time, defect rate, and costs. Facing changes in the industrial ecosystem and the market, today's enterprises ought to abandon conventional patterns of thinking that focus on hardware and equipment when developing manufacturing systems; instead, they should draw on the many existing ICT and software application technologies to shape the operation scenarios of Industry 4.0. Considering the unique operational characteristics of enterprises and the shortage of professional worker around the world, enterprises must pay more attention to possess the value of "talent"; they must transfer processes originally performed by humans to automated machines and enhance the value of their workers with the support of technology. By moving from an "operator" to a "controller" and further to a decision maker, a highly integrated cyber-physical system environment needs to be established. In this way, in addition to retaining and passing on valuable experience, the enterprise also needs to maximize the value of precious human resources.

Within the Industry 4.0 trend, cross-campus and cross-field industry-university cooperation has also recently become a very popular cooperation model for academia and industry. Professor Chien-Ching Ma of the Department of Mechanical Engineering of National Taiwan University, Professor Jing-Yuan Chang of the Department of Mechanical Engineering (NTUT), Professor Chao-Lung Yang of the Department of Industrial Management (NTUST), and Professor Po-Ting Lin of the Department of Mechanical Engineering (NTUST), have cooperated to apply high-precision computer vision technology with heterogeneous sensor technology to integrate human posture and action prediction and collaborative robotic arm path planning by using AIoT to build a 5G-AI-mediated human-machine technology. They hope to greatly assist Taiwan's manufacturing industry in the move towards to smart manufacturing.

In addition, AI is also finding wide use in the field of smart manufacturing, such as in abnormality detection, factor analysis, predictive maintenance, and production line adjustment. For example, intelligent maintenance of equipment: when an abnormal event occurs on the production line, the company can dispatch the most suitable personnel to check and fix the event accordingly. Through AI analysis of historical data of the equipment, abnormal signals of machines can be predicted to develop, gradually, a set of predictive maintenance mechanisms. In addition, AI can also identify movements of operators and analyze the relationship between the operation of production line personnel and that of machines, thereby achieving higher industrial efficiency. On the other hand, combining AI with IoT is helpful for the development of Smart Sensing technology; it combines heterogeneous sensor fusion technology between different types of sensors, such as RGB, thermal imaging, LiDAR, millimeter wave radar, and ultrasonic waves, to meet the application needs of different scenarios. When this heterogeneous sensor fusion technology is applied to automated guided vehicles (AGVs) in smart factories, it can help develop a smart logistics system that meets practical needs.

Kai-Lung Hua, director of the NTUST AI Research Center and professor of the Department of Computer Science and Information Engineering, pointed out that AI is an important technology for practicing smart manufacturing. The goal of NTUST's establishment of the center was to assist the industry in the application of AI technology. In addition to raise the overall production efficiency, it can also lead to the birth of more new start-ups and accumulate energy for the development of AI applications in Taiwan's industries. Through the accumulation of practical experience, it can further pursue technological breakthroughs and innovative applications, starting from the application of AI technology to solve existing industry problems. It can also strive towards the goal of becoming an international research center.

In view of the strong demand of enterprises for AI professionals, NTUST is establishing a master program in AI cross-domain ach year. Through the cooperation with international benchmarking companies, multidisciplinary topics are introduced to students, so that the students can gain practical abilities which, after graduation, they can apply to the real-world applications.

(From right)  Director of DIGITIMES Research Jian-Zhi Huang,  director of the NTUST AI Research Center and professor of the Department of Computer Sci

DIGITIMES Research director Roger Huang (left to right), director of the NTUST AI Research Center Kai-Lung Hua, and professor of NTUST's Department of Industrial Management Chao-Lung Yang.
Photo: DIGITIMES Asia, August 2022

Related stories