Skip to Main content Skip to Navigation
Theses

Optimisation multicritères de gammes d'usinages

Abstract : The objective of this work is to develop a Computer-Aided Process Planning (CAPP) methodology to allow aeronautical subcontractors to face the current productivity and competitiveness problems. To start, a state of the art on the existing CAPP systems is presented and introduce the limits of the current systems. A four-step methodology is then proposed, allowing the user to obtain optimized machining ranges respecting his know-how and experience and introducing innovation. This methodology is based on a representation of the decision-making behavior of the process planners in a given situation as well as the risk of industrialization and broadens the formalization of the performance by taking into account performance criteria other than the machining time or cost. Thus, a state of the art on multicriteria decision-making methods helps to select the method for the aggregation of performance in relation to the industrial problem. After a geometrical and technological description of a mechanical part and the creation by the process planners of elementary machining operations, a genetic algorithm is used to generate optimized ranges. In a final step, the methodology presents the best process plans generated to the process planners and the use of the social choice theory allows to target the most efficient ones to choose, integrating a risk criterion based on tool deflection. This methodology was applied to a Titanium alloy machined landing gear part and allowed a 25% reduction in machining time while increasing reliability. Finally, prospects for improving the decision-making tool at various stages of the methodology are proposed.
Complete list of metadatas

Cited literature [173 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-02683330
Contributor : Abes Star :  Contact
Submitted on : Monday, June 1, 2020 - 3:03:51 AM
Last modification on : Monday, June 1, 2020 - 3:45:34 AM

File

2017CLFAC019_DELOLME.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-02683330, version 1

Citation

Laurent Delolme. Optimisation multicritères de gammes d'usinages. Génie mécanique [physics.class-ph]. Université Clermont Auvergne, 2017. Français. ⟨NNT : 2017CLFAC019⟩. ⟨tel-02683330⟩

Share

Metrics

Record views

63

Files downloads

68