A decision theory approach to support action plans in cooker hoods manufacturing

Abstract

 Nowadays, Knowledge-Based systems are widespread decisionmaking tools applied in product design and manufacturing 
planning. The series production requires agile and rapid 
decision-making methods to support actions in manufacturing 
lines. Therefore, agent-based tools are necessary to support the 
detection, diagnosis, and correction of accidental production 
faults. The context of Industry 4.0 has been enhancing the 
integration of sensors in manufacturing lines to monitor 
production and analyze failures. The motivation of the proposed 
research is to study and validate decision theory methods 
to be applied in smart manufacturing. This paper shows 
a Knowledge-Based approach to support action decisionmaking processes by a Bayesian network model. The proposed 
method aims at solving production problems detected in 
the manufacturing process. In particular, the focus is on the 
automatic production of cooker hoods. A case study describes 
how the approach can be applied in the real-time control 
actions, after a problem in quality is detected