Aide au chiffrage de pièces usinées en aéronautique : proposition d'un modèle adaptatif basé sur la similarité et le modèle 3D enrichi

Abstract

For every manufacturer, exact cost estimation is both a major priority and a challenge. This routine task is far from being optimized and depends on a very large number of parameters both strategic and technical. It requires a lot of time, researching and analyzing all kinds of documents, including standards, recommended methods or guidelines. To resolve this problem, applications have been developed to help estimate a cost. Most of them are achieving this either using form feature recognition linked to machining strategies and tool paths, databases of tools and CNC parameters, or building models with regression analysis. However estimation is a task that would greatly benefit from a computer application favoring the re-use of knowledge and data found in alphanumeric and geometric documents. Every manufacturing company has switched to numerical data with CAD, CAM or ERP systems but one of the main drawbacks is the lack of interoperability between these applications and the low usage of all this information that contain valuable knowledge and expertise. In the present work we describe the current state of cost estimation and should cost models and propose a new hybrid approach whose purpose is to maximize the re-use of all the available information for machined parts. Our approach is based on a parameterized and customized cost model that utilizes the information found in Model-Based Definition files and related textual documents and uses a descriptor comparator to retrieve similar cases. Our study is based on the analysis of 28 differents parts choosen from one of our industrial partner’s database. Our results show that the model is very efficient to predict cost of machined parts from the aerospace industry. Furthermore it demonstrates that the use of geometrical similarity induces less errors and more robust estimations leading to more stable prices over time.

Date	16 Nov 2017
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	Roland Maranzana (Supervisor)