School of engineering in Physics, Applied Physics, Electronics & Materials
Jean-Marc DEDULLE directs his own company and also teaches at Phelma. His expertise as an R&D engineer are required for teaching students certain advanced disciplines. We explain why below.
Could you introduce yourself in a few words?
After finishing my doctoral thesis at Grenoble's electrical engineering laboratory, I worked for two years in a large industrial metallurgy corporation and then for six years in an SME that developed and distributed scientific computing software. In 1997 I decided, with eleven years of experience in the field of modelling coupled physical phenomena, to set up my own company. My business focuses on designing, training and consulting in the area of optimising and modelling in industrial process engineering.
How do you use your expertise as an R&D engineer in your teaching at Phelma?
As I wanted to share my experience in modelling, I was hired as a part-time associate professor. As part of the post I also joined the Laboratory of Material and Physical Engineering (LMGP), where I have been able to use the my skills and develop modelling in the field of single crystal growth. My experience as an R&D engineer has been put to good use in teaching numerical modelling and methods.
What subject do you teach, in which branch of science and for students on which course?
I teach computer science, electromagnetics and numerical methods on various courses. Given my experience in the world of industry, I am also involved in developing business relations at Phelma.
Why is it valuable for students to be taught by someone working in industry?
In terms of their projects, we give students concrete examples and we can make links with theoretical aspects and highlight the difficult issues that they will inevitably meet as part of their future profession as engineers: checking the accuracy of results, understanding the reasons for erroneous results and providing solutions. In this way, the basics of numerical methods are better understood, without taking the easy route that suggests that, with scientific modelling software now having progressed so much, all you have to do is click something to get a result.
Date of update April 22, 2013