Simulation based design flow
Quite often decision-makers have the impression that simulation is a slow, inaccurate, unreliable or expensive way to develop products. All the benefits of simulation should therefore be viewed as a whole, not as a single component of product development.
Simulation can increase the reliability, quality and performance of the product without affecting the overall product development cost of the project. Using simulation can reduce the number of physical prototypes and thus save on prototype costs.
Why does the vehicle industry use collision or flow simulation? Why do car manufacturers do not just make prototypes, do them collision and wind tunnel tests and develop car designs based on tests? Answer: It is very expensive and untrustworthy to correct the changes that are required by the prototype. This same attitude should also be brought to electronics and RF technology companies.
Since less time is required to produce physical prototypes (less production cycles), the lead times of the product development project can be significantly reduced. My experience is that traditional 13 to 15 months can be shortened up to 8 months using simulation as part of a product development process.
Simulation can also provide information from proto that would be impossible to find out with normal product development. These include materials selections, electromechanical interfaces, and EMC problems due to these choices. Misguided material choices together with weak ground contacts, can produce other passive intermodulation and radiating harmonics.
Let's keep in touch and hopefully we will meet in a project where simulation-based analysis will be done!
I'm old enough that I've watched simulation evolve from one of the early versions of SPICE (run on an 8MHz 8086) to the amazing tools we have today. I wrote one program for sizing the cap bank on a 10KW UPS on a TI59 calculator (and it worked as predicted). Even then it was easier to do a tolerance analysis, than by modifying production parts. Simulation allows you to change parameters you can just modify. It helps identify which are more sensitive to change (thus need more control) and others are less. Also, when you have a problem, you can cut the simulation into smaller pieces. This was very effective to identify a cross-talk issue in a connector. But, simulation is only as good as your understanding of the issues. Keep comparing predicted results with measured results to uncover limitations, and correct assumptions.