OEE-Improvement by Artificial or Human Intelligence?

In the future, human and machine in the form of human and artificial intelligence or skills will share the work of increasing equipment productivity. Where are the tasks of the humans, where are the tasks of the technology to achieve the best results?

To increase equipment productivity, among others data such as number of pieces produced per time unit and causes for availability, performance and quality losses are recorded as time series data. Time series means the data is collected at regular intervals, e.g. once a minute, recorded and documented. This circumstance quickly creates large amounts of data that need to be analyzed for patterns, anomalies, and trends.

The chain of activities for this optimization process consists of 6 steps that are traversed, and basically can be performed by both – humans and technology. However, if one has the complementary requirement that the processes should be executed precisely, quickly, permanently and at low cost, then human and technology should focus on their respective strengths. Humans can do a lot, but technology can do specific tasks a lot better. The following figure distinguishes who / what performs the activity when no technology is available or when it is available.

Figure: Comparison of the possibilities of implementation between technology and man

At the lower three levels of the process, there is no doubt that the execution of activity by technology is favorable. On the upper two levels humans are unrivaled. No currently available technology has ideas or can implement physical measures. That is the domain of humans.

At level 4, data analysis, rules, statistics and artificial intelligence in the form of machine learning can be used. For example, it can be applied to determine how the OEE or standstill trend in terms of duration and frequency runs, the equipment run is clustered into stable and unstable, the quality of change-over processes are measured, standstills are grouped or classified, or faulty measured values ​​are identified.

This fast and accurate analysis of large amounts of data is a domain of technology. With the availability of these capabilities, analyzing the equipment losses of the past (!) day in morning shop floor management meeting is a thing of the past. Conspicuous data histories are instead identified on-line and made available to the employee live on an Andon board or an industrial smartwatch. Thus, the attention of the equipment team can be directed to the problem immediately during the loss.

This reaction time can only be achieved with a balanced mix of automation, analytics and humans. This combination belongs the future.