As automobile and airline manufacturers work on the embryos of their future models, the first thing they set out to decide is the number of prototypes they need to develop in order to test the feasibility of their design becoming a final product.
With the prototypes, for instance, they would test if their prospective model can stand the test to assess its robustness in various conditions.
However, with the advent of Virtual Product Development (VPD) using Finite Element Analysis (FEA) based simulation, much of the testing of an engineering design has been trusted to the computers and the ever-improving efficiency of the software. These software applications re-create various conditions on the screen using different variables and help analysts of designs understand to what extent they can meet the criteria set for the product.
So efficient are these applications that several VPD vendors today are gung-ho about the technology’s ability to completely eliminate prototyping from the development cycle of engineering products.
Dr Swami Narayanaswami, Chairman and Founder of CSM Software, an expert in Finite Element Analysis, muses over this claim for a minute and says it can’t be completely refuted. “Indeed, much of the prototyping has been made redundant with the arrival of VPD tools,” he agrees.  
“The accuracy and greater analytical capacity that VPD and other simulation and modeling software offer today have certainly brought down the number of prototypes from 10 or 12 to just 1 or 2.
But, using computational/algorithms based analysis can’t totally replace prototyping, which will still have to be done to physically verify the design’s perfection.”
Degree of freedom
The fact that today’s simulation technology can scale up to handle problems of a millions of degrees of freedom of a model shows they have achieved the possible levels of sophistication and accuracy.
For a comparison, Dr Swami points to the earliest simulation tool that solved 2000 degrees of freedom of a static object and 100 degrees of freedom of an object in motion.
Whereas, the best tool in the business today can handle models of up to 10 million degrees of freedom including movements and vibration.
“Imagine each point of an object, say, a car, being stretched in different direction to see its limits or the extent to which it can vibrate (while in motion)?” Dr Swami asks.
“The ability of the software to determine this allows analysts to accurately understand the strength of a design.”
Dr Swami points out that the emergence of computers with increased process power and advancements in software technology coincided with the quest of the scientific community to find out ways of improving accuracy, cost effectiveness and better time to market for engineering products. 
Recipient of the prestigious National Research Foundation Fellowship (at NASA Langley Research Centre), Dr Swami was one of the first scientists to explore FEA and CAE (Computer Aided Engineering) that sowed the seeds for today’s advancements.
“FEA was then a hot and an emerging field,” Dr Swami noted. “And the enormous wastage of time and resources to introduce innovations in their machines were hurting even bigger institutions like NASA, who had decided to create an open and general purpose software code for design simulation and modeling.
It was a real privilege for me to be a part of the team that developed it.”
In the last three decades, Dr Swami says simulation technology has matured considerably. In fact, it is ready to make its foray into newer areas of engineering like simulating large and open areas to analyze the impact of climate change. “I think software applications in this realm are more or less accomplished in terms of their efficiency,” says Dr Swami.
“However, the challenge in future for researchers is to find out the areas in which it can make a qualitative difference.”