Breakthrough in Digital Manufacturing

In recent years, the need for DX (digital transformation) initiatives has increased in response to the wave of work style reforms and the declining birthrate and aging population in the manufacturing industry.
​Source: White Paper on Information and Communications 2018, Chapter 3 Productivity Improvement and Organizational Reform through ICT (Ministry of Internal Affairs and Communications)()

For example, as a countermeasure against long working hours, I think that the realization of a manufacturing process that does not require human labor has been progressing for some time by promoting the development of manufacturing robots.
Above all, as a concern about the declining birthrate and aging population, the awareness of the problem of technology transfer is higher in the manufacturing industry than the average for all industries. (Fig. 1)

As an urgent measure, many business establishments answered, "We select the necessary people from retired people, extend their employment, re-employ them on a temporary basis, and use them as leaders." In light of this, it is clear that there is a need for measures that can support companies with only the next generation, such as measures in DX. (Figure 2)
​Source: White Paper on Manufacturing 2022, Chapter 4, Section 2 Current Situation of Ability Development of Manufacturing Human Resources (Ministry of Economy, Trade and Industry))(​

Figure 1: Awareness of issues related to skill development and human resource development in the manufacturing industry
Source: “Monodzukuri White Paper 2022” Chapter 4 Section 2 Current Status of Ability Development of Manufacturing Human Resources (Ministry of Economy, Trade and Industry)
https://www.meti.go.jp/report/whitepaper/mono/2022/pdf/honbun_1_4_2.pdf

Figure 2: Efforts to pass on technology in the manufacturing industry
Source: “Monodzukuri White Paper 2022” Chapter 4 Section 2 Current Status of Ability Development of Manufacturing Human Resources (Ministry of Economy, Trade and Industry)
https://www.meti.go.jp/report/whitepaper/mono/2022/pdf/honbun_1_4_2.pdf

Challenges of Digital Manufacturing

ICT (Information and Communication Technology), IoT (Internet of Things), and more recently, AI (Artificial Intelligence) and RPA (Robotic Process Automation). Efforts toward Digital Manufacturing are progressing in various forms using devices such as software and robots while utilizing advanced technology.

However, the purpose of using them is often to improve productivity, and measures have not yet been taken to address the issue of passing on skills to skilled workers, which is directly linked to the declining birthrate and aging population.
​(Source: “White Paper on Manufacturing 2022” Chapter 4 Section 3 Utilization of Digital Technology Related to Manufacturing Human Resources (Ministry of Economy, Trade and Industry))​

In addition, traditional issues remain regarding technology succession.

It means that there are individual differences in the knowledge and skills of the experts who are the target of technology inheritance.
Even if you say "veteran", if the number of years of experience of an individual differs, there will be a difference in the degree of skill, and the accuracy of DX conversion will greatly depend on which individual's skill is used as the basis for AI learning. It will be a current issue.

Conversely, as DX progresses in the future, instead of continuing to rely on the individual on-site, the know-how of experts will be inherited and various people will be able to use it, and a sustainable manufacturing site is expected. increase.

What is Braintech x AI?

There is "Braintech x AI" in the field that we have been working on for a long time.

First of all, we have already introduced the use of EEG in our column, but in recent years, various EEG measuring devices have been developed.
The type and number of brain waves that can be measured differ depending on the measuring device, and the orientation of the measuring device depends on the purpose of EEG measurement.

In particular, evoked potentials, known as event-related potentials, which have been established in existing research, respond to external visual stimuli, so it is possible to decipher human judgments and choices by measuring them.
“Braintech x AI” refers to treating human judgments and cognitions that can be read from such measured brain waves as AI learning data.

Advantages of Digital Manufacturing using Braintech and AI

I explained "Braintech x AI" earlier, but what are the benefits of using human brain waves as AI learning data?

For example, there are subtle judgments that are difficult for non-experts in the field, such as removing abnormalities at manufacturing sites, but in many cases, these judgments are made based on the experience and knowledge of experts. I have.
Their experience and knowledge can be very difficult to verbalize, and some of them make judgments even if they are not 100 % certain.

Such delicate judgments and judgments based on personal experience are difficult to pass down to the next generation, and are a hindrance to technology succession.
Therefore, by measuring and analyzing the brain waves of experts, it is possible to read and understand whether they are making decisions on such events, and by having AI learn it, it is also possible for AI to be responsible for making decisions. .
In addition, the learning data given to AI is usually labeled as 0 or 1, called hard labels, but by using smooth waveform data called brain waves, that is, soft labels, as learning data, it has the effect of preventing over-learning.

From this, you can see that handling brain wave data is advantageous both from the perspective of technology transfer and from the perspective of AI performance accuracy.

InnerEye Product Page Top