Automatically Generating the Technology Compatibility Matrix Using Graph Transformations

Conference Paper (2019)
Author(s)

M.N. Roelofs (TU Delft - Flight Performance and Propulsion)

R Vos (TU Delft - Flight Performance and Propulsion)

Kristian Amadori (Saab Aeronautics, Linkoping)

Christopher Jouannet (Saab Aeronautics, Linkoping)

Research Group
Flight Performance and Propulsion
Copyright
© 2019 M.N. Roelofs, Roelof Vos, Kristian Amadori, Christopher Jouannet
DOI related publication
https://doi.org/10.2514/6.2019-2886
More Info
expand_more
Publication Year
2019
Language
English
Copyright
© 2019 M.N. Roelofs, Roelof Vos, Kristian Amadori, Christopher Jouannet
Research Group
Flight Performance and Propulsion
ISBN (print)
9781624105890
ISBN (electronic)
978-1-62410-589-0
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

One of the first stages during technology evaluation and selection is constructing the technology compatibility matrix, which indicates the compatibility of each pair in a technology portfolio. Construction of the technology compatibility matrix usually involves subjective expert judgment, which inhibits assessing each pair consistently and storing the decision rationale. Therefore, this study develops a method based on graph transformations, allowing for a formal theoretical description of technologies. In order to automate this process, two algorithms use these graph transformations to deduce technology compatibility and dependencies between technologies. The first checks whether changes made by a pair of transformations are compatible with each other. The second defines dependencies for each technology and attempts to resolve these using other technologies. The method is put into practice for a technology portfolio within Saab Aeronautics and the automatically generated compatibility matrix is compared to a pre-existing expert-judgment-based matrix for the same portfolio. Roughly 90% of the entries in the algorithm-based TCM are identical to the expert-based TCM. Therefore, the automated method is capable of capturing much of human reasoning in this context. However, in order to fully agree with human expertise, physics and design reasoning should be included to expand the scope of inference.

Files

6.2019_2886.pdf
(pdf | 0.853 Mb)
License info not available