This thesis explores strategies for integrating Lockout-Tagout (LOTO) procedures into an intelligent manufacturing environment, aiming to enhance health, safety, and environmental conditions during the transition to Industry 4.0. LOTO is a safety procedure used to ensure that machines are properly shut off and cannot be restarted before maintenance work is completed. The primary objective is to provide concrete solutions to the challenges faced in the manufacturing industry regarding the planning, preparation, and execution of LOTO procedures.
A joint policy for production, maintenance, and operational LOTO has been proposed, with the development of a model based on the stochastic optimal control theory and using Hamilton- Jacobi-Bellman equations. The results show a significant reduction in production costs due to the inclusion of operational LOTO. Additionally, a management tool for production and maintenance has been developed.
The assistance in drafting LOTO procedures was addressed using machine learning and deep learning. Multi-task classification algorithms were employed to predict the devices to lock out based on machine names. The time required to develop LOTO procedures decreases, but this drafting assistance is limited to simple machines. Complex machines still require human intervention due to the criticality of LOTO.
The fusion of LOTO procedures was automated using a multi-objective mixed-integer nonlinear programming optimization model, planning the routes of maintenance teams based on the machines to lock out. Using vehicle routing problem-solving algorithms reduced worker travel and locking time by more than 30%.
Intelligent LOTO strategies adapted to the manufacturing environment were also developed. The first approach, based on the geographical position of devices, allows teams to geographically distribute the devices to lock out, maximizing human resource utilization and reducing locking time by 20%. The second approach is a dynamic LOTO approach, where procedures adapt in real-time to unforeseen events, reducing locking time by 30% in the most challenging scenarios.
Intelligent technologies, in the context of Industry 4.0, such as the Internet of Things and Artificial Intelligence, can significantly improve LOTO procedures by increasing the efficiency of operations. A significant reduction in lockout times, better management of human resources, and improved production continuity are possible with the proposed strategies. The results have been developed and validated with real case studies from industrial partners.
| Date | 15 Aug 2024 |
|---|
| Original language | French |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Lucas Hof (Supervisor) & Jean-Pierre Kenné (Co-supervisor) |
|---|
Delpla, V. (Author),
Hof (Supervisor) &
Kenné (Co-supervisor),
15 Aug 2024Student thesis: Doctoral thesis › Doctorate in Engineering: Engineering