The technological advancement of the 4th Industrial Revolution is creating a domino effect in the digital transition across all sectors of society, and churning out a great deal of data in its wake. Not even the shop floor environment, which has traditionally been brick-and-mortar in nature, is exempt from its influence, as all eyes set upon the big data generated under a broad range of work conditions. The expectation is that in analyzing data across the whole process – from actual manufacturing to the adoption, transport, storage and input of materials – and discovering where such data can be applied, innovation will be made possible also in the manufacturing industry.

The Industrial Data Engineering & Analytics (IDEA) Laboratory headed by Professor Cho Hyunbo at the Department of Industrial and Management Engineering, POSTECH, focuses on developing artificial intelligence (AI) through the use of data generated in the actual field to optimize both quality and productivity on the shop floor. The machines at the manufacturing sites produce a significant scope of data – ranging from machine temperatures and electric current levels to completed work tasks. The IDEA Lab gathers this data and, with the help of AI, pinpoints the precise adjustments needed to improve a company’s productivity.

The Lab emphasizes the importance of field-driven research: researchers identify problems, collect data to develop AI, and perform verifications all within the shop floor environment. “I never encourage my students to solely engage in academic discussions”, Professor Cho asserted, adding “We find all too often that the content or theories learned in textbooks simply do not work in reality. This is precisely why field experience is so encouraged at the Lab”.

A perfect example of this is the digital transition research being conducted in partnership with LG Electronics. While the global electronics maker normally develops production plans to deliver on customer orders, it happens that the accuracy of such plans declines. The IDEA Lab is developing AI to quickly detect such decreases in accuracy to help the company modify its production plans. AI’s role here is to promptly spot and report any plan errors in order to minimize production losses. The Lab is working jointly with LG’s production engineering research laboratory and has been awarded the title of ‘joint laboratory on intelligent manufacturing’.

The necessities of the shop floor dictate the Lab’s next research initiatives. In collaborating with a machinery processing company, the IDEA Lab has developed technology to implement AI in swiftly detecting tool abrasion to improve productivity and quality. Teaming up with a startup that promotes cosmetics on Youtube, researchers at the Lab have engineered and delivered AI capable of identifying the most queried face makeup parts within specific time windows by analyzing the rate of which specific Youtube videos are viewed.

The IDEA Lab sees to it that its students in their 4th year are given the opportunity to study at the National Institute of Standards and Technology in the US to effectively assimilate its advanced AI technology. Professor Cho stated, “This is a win-win experience as the students absorb advanced AI expediently in partnership with IDEA’s US partner, while the Lab is able to share the invaluable insights gained from actual experience in the field”.