Feature based costing of extruded parts

Torre Nieto, Jose

Permalink

https://hdl.handle.net/2142/16501 Copy

Description

Title

Feature based costing of extruded parts

Author(s)

Torre Nieto, Jose

Issue Date

2010-06-22T19:46:46Z

Director of Research (if dissertation) or Advisor (if thesis)

Philpott, Michael L.

Department of Study

Mechanical Sci & Engineering

Discipline

Mechanical Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• Feature Based Costing
• Extrusion
• Design for Manufacture (DFM)
• Manufacturing Cost Estimation

Abstract

This thesis presents the creation of a model to predict the manufacturing cost of parts made by the extrusion process. Feature Based Costing (FBC) techniques have been used to develop the cost model, which is intended to be used early in the design phase of a part, when the geometry and the material are being defined in a Computer Aided Design (CAD) software by the designer. The cost model was developed specifically for hot extruded parts made of aluminum alloys, extruded in the forward direction using single-hole dies, as this was found to be the case for the majority of extrusion work in the industry. As the cost model is developed, the main Geometric Cost Drivers (GCDs) and Non-Geometric Cost Drivers (NGCDs) of a part made by the hot extrusion process are identified. The main outputs of the model include the Piece Part Cost (broken down into its various recurring costs components), the Amortized Investment (i.e. Amortized Tooling Cost) and ultimately the Manufacturing Cost. The Piece Part Cost estimation includes an algorithm for selecting the optimal combination of press and billet size to extrude a given shape. Extrusion speed, cycle time and yield are all calculated by the model as well. The Piece Part Cost Model was validated against actual quotes data from extruders in the US for a number of different designs. The model predictions were shown to correlate within 3% to the actual quotes. Unlike the Piece Part Cost, the Tooling Cost part of the model is of empirical nature, obtained by Multivariate Regression techniques from actual quotes data provided by extruders in the US. The Tooling Cost Model was validated statistically, and it was shown to explain 84% of the variance observed in the actual quotes. Finally, the complete Manufacturing Cost Model developed herein was successfully implemented in a Computer Spreadsheet, which is included as a supplemental file for this thesis, representing a fantastic tool for design engineers, manufacturing engineers and any other Design for Manufacture (DFM) practitioners in general.

Graduation Semester

2010-5

Permalink

http://hdl.handle.net/2142/16501

Copyright and License Information

Copyright 2010 Jose Torre Nieto

Owning Collections