Getting the right prototype

COMPUTER aided engineering (CAE) software is used as an analysis tool where engineers can put new components and sub-system designs through virtual testing to discover any weak points of engineering defects before a prototype or actual production unit can be made. 

In theory, CAE should eliminate the lengthy traditional process of building a physical prototype for testing, followed by redesigning and further prototyping.

But Noor Hisham Ismail, a section manager at Proton’s CAE unit of the engineering division, said that wasn’t the case with the software.

“It turns out that many physical prototypes we created were inaccurate and performed differently from the CAE models and simulation,” he said.

Noor Hisham discovered that this problem was due to the fact that the CAE models on the computer were perfect and accurate but didn’t take into account variations in production techniques and materials composition.

“It is a natural for variations to occur, nothing is 100 per cent perfect,” he said.

“With these variations, we are now able to calculate a more accurate failure rate and would be able to pinpoint a source of failure before putting the component into production.”

Noor Hisham developed his own innovative algorithms to incorporate into Proton’s CAE programs with the use of variations and probability studies to improve the development process of such components.

These algorithms take into account the probability and variances based upon collected data from Proton’s current production process.

With this new understanding in probabilistic analysis, Noor Hisham’s engineering division is now able to create more accurate models for testing and relegate prototyping to the final stage of evaluation rather than part of the design process, thus shortening the time-to-market of its products and allowing engineers to fine-tune their designs.

Noor Hisham presented his studies at the annual MSC.Software ASEAN Virtual Development Seminar last August, where he was awarded the Asia Pacific Innovation Award by the MSC. Software Corporation.

“Right now, the program is in its tenth version and we are still in the process of implementing it fully into the design and engineering of Proton cars,” said Noor Hisham.

“For now, the program is used alongside the normal process of prototyping, but we can expect the development of future Proton cars to be based on this CAE program.”

Noor Hisham hopes that one day the algorithms he developed will be used globally for a more efficient car design process and reliable engineering. 

By Daniel Wong