In addition, new technologies for instructional display, such as Augmented Reality (AR), are entering the market. However, these systems are still static, so that each operator is presented with the same amount of information. This does not meet the operator’s need for customisation and therefore slows down the production process.
Deployment of Artificial Intelligence
Artificial Intelligence (AI) is used to directly use the measured data from the process (productivity, human errors) and feedback from the operator. AI can then generate the optimal number and type of work instructions for the respective operator and adjust them in real time.
What challenge does it solve?
Technically skilled employees are scarce in today’s labour market. The adaptive support system provides employees with work instructions tailored to their skills and experience levels. This reduces the training and supervision time of new employees. This makes it possible for companies to quickly and effectively deploy both experienced and inexperienced employees as well as both more and less technically skilled personnel to complex tasks.
What will the use case teach us?
Within the project, a prototype system is developed for adaptive work instructions, which automatically adapts to:
- The skill and experience level of the operator.
- The task performance of the operator for a product or order.
The prototype of the adaptive support system will be made available in demonstration set-ups in the field labs Flexible Manufacturing at the BIC in Eindhoven and in RoboHouse in Delft for companies interested in the possibilities of AI.
First result
In July 2021, a lab experiment was conducted in RoboHouse for the purpose of developing the AI model for adaptive operator support. Using a total of 30 participants, each participant built a number of products using work instructions, with varying degrees of detail. The assembly time, the quality and also the feedback from the participants will be recorded (anonymously, of course). These data will provide input for the AI model of the adaptive support system. In the autumn, a demonstration of the adaptive system is planned in RoboHouse in Delft and also in the Fieldlab Flexible Manufacturing at the Brainport Industries Campus.
Parties involved:
This project is part of the APPL.AI programme of TNO and is partly funded by the start impulse that the NL AIC received from the Dutch government for research and development of AI applications. TNO is collaborating in this project with the companies Omron and Thomas Regout International. These companies will evaluate and test the prototype system of the adaptive operator support, both in a lab and in an industrial setting. Other partners of the Fieldlab Flexible Manufacturing (BIC in Eindhoven) and of the Fieldlab Robohouse (Delft, SMITZH) are also involved in evaluations.
