There are various issues the manufacturing industry has been engaged in such as productivity improvement and quality management as well as a staffing shortage. As a solution for these issues, there has been a growing interest in smart factory transitions utilizing AI and IoT, and the entire manufacturing industry has entered a period of transition.
On the other hand, smart factory transitions cannot be accomplished overnight, and it is also true that there are several obstacles that must be cleared. What kind of issues do the factories actually face and what kind of solutions are they trying to derive? We interviewed Masashi Okamizu and Tsuyoshi Koyama of the Manufacturing Department at the Kanazawa Murata factory, who have been manufacturing using advanced technology, and asked them about the current conditions and future prospects of the manufacturing sites.
Improving the efficiency of processes and accuracy of production plans is essential
According to the statistics “Issues and Outlook Japan’s Manufacturing Industry Faces” issued by the Ministry of Economy, Trade and Industry, “abilities of skilled technicians,” “quality management,” “cost correspondence capability,” and “preservation and improvement capabilities of circuit board technology” are issues the entire manufacturing industry faces. Of the total group of companies, 34.6% feel that “introduction and utilization capabilities of robots, IT, and IoT” are issues related to transitions to smart factories. What kind of issues do these companies aim to solve by introducing robots, IT, and IoT? Let’s have a look at the “purpose of data collection” prior to introduction.
Purposes such as “Improving efficiency of manufacturing processes in general,” “Improvement of quality,” “Improving accuracy of production plans,” and “Reduction of lead time” are listed from the top. On the other hand, what kind of issues and purposes does the Kanazawa Murata factory, which is considering introduction of AI and RPA (robotic process automation) as well as BI (business intelligence), face? Okamizu has stated they are currently handling matters such as “Improving efficiency of manufacturing processes in general” and “Improving accuracy of production plans.”
Okamizu: “We frequently discuss issues of how to improve efficiency for processes in manufacturing and how to improve productivity while we supervise circuit board creation processes at the manufacturing department. For example, if supply from the previous process is delayed, a period of inactivity due to having to wait for the products occurs. We must reduce the lead time as much as possible by eliminating these losses. However, this is not an issue that can be solved at our department alone, so we have discussions concerning the improvement of efficiency in processes as well as optimization of operational performances and equipment with staff from production management and quality management. In the future, we aim to increase our production output to 1.5 times the current output with the same number of staff and equipment.”
Koyama: “I am a supervisor at a different manufacturing site, but I have the same awareness concerning the improvement of process efficiency as well as optimization of operational performances and equipment as Okamizu. About 40% of the processes I supervise are related to prototypes for mass production products, but I believe we must be aware of the profits and losses by inspecting whether or not our equipment and staff are optimal even during the trial manufacturing stage.”
Okamizu: “We handle our work prioritizing delivery time under the assumption that our products will be completed in a high quality if manufactured as designed. I am aware that it is our greatest mission to execute production based on the delivery time. In order to meet the delivery time, I believe it is crucial to improve the efficiency of processes and production plans as well as accuracy.”
AI and IoT use will become solutions for optimization and standardization.
Okamizu has stated that “Skill inheritance” is also a large issue in “Improving efficiency of manufacturing processes in general” and “Improving accuracy of production plans.”
Okamizu: “Trouble such as changes made to the production plan and equipment failure frequently occur at the manufacturing site. Of course we have monthly production plans, but changes such as production increase occur daily. We must think about replacing machines, bearing in mind the number of which machines are to be operated, as well as where to assign which staff member. At this point, if we are understaffed, in terms of ability or simply due to a lack of staff, an issue linked to production output arises.”
“Skill inheritance” is one of our “Purposes of data collection” and we are required to inherit work that has become dependent on independent skill by “visualizing” techniques using elements such as AI.
Okamizu: “Most of our processes, which reach about 50 processes, require manual labor, and for processes such as external inspection, our veteran staff are capable of discovering defects that beginners are unable to detect. On the other hand, if that veteran staff has taken a holiday or retired, we must determine how to proceed with skill inheritance. We seek to implement countermeasures by sharing filmed work processes to eliminate work dependent on independent skill and the craftsman-like idea to gain experience and visually learn from veteran staff.”
Koyama: “We share the staff’s varying skill level and accuracy as the same issue. For example, staff are required to enter daily inspections and confirmation items every morning when starting to operate equipment. This is not a complicated task, but due to the number of items to confirm, some workers may spend a long time accomplishing this. This means that the time it takes for the machine to start running and the product flow will vary and end up affecting the production output. I believe that achieving standardization, by reducing areas that depend on human skill as much as possible, is the solution.”
Okamizu: “I believe it is crucial to assign staff to the required position without simply reducing the number of staff. Products will vary as time goes on, and since delivery times will be shortened, we must review as well as optimize and standardize our current state, in order not to fall behind in competition. In order to achieve this, I believe IoT and AI use is necessary.”
The Kanazawa Murata factory is considering introduction of AI, RPA, and BI as solutions for these issues. This is because back-office issues such as administrative processes and management work, that are not present in the aforementioned data, are included.
Okamizu: “We have many meetings concerning the improvement of productivity, and I am rarely at the factory, instead frequently handling desk jobs or participating in meetings. This is where I feel the complications of data management and analysis. Our work processes have existed for more than 20 years and our supervisors are replaced every few years. Furthermore, data is managed in varying formats depending on the supervisor, making maintenance and update work complicated. Even at the factory, there are burdensome processes such as input of numbers and amount by lot. I expect automatic data conversion and efficient formats through IoT to become a solution.”
So what kind of solutions is the Kanazawa Murata factory working on for these issues? Next, we would like to hear about what kind of efforts the Kanazawa Murata factory is tackling to achieve their transition into a smart factory.
- Murata and ACCESS advancing digital transformation in wide range of industries Available February – support tool for Smart manufacturing sites in Japan
- How to break free from the dependence on intuition and unique skills in maintenance work (Part 2 of 2)
- How to break free from the dependence on intuition and unique skills in maintenance work(Part 1 of 2)
- Issues and solutions in the manufacturing industry from the Kanazawa Murata factory perspective (Part 2)