Where imagination meets mathematics

As technology develops, the business environment also changes. This is where companies think of more and more innovations to survive and win in a competitive world. Ansys, a simulation software company, comes to the help of engineering firms, with its simulation solutions that help them reduce the complexities involved in developing a product - be it a car, a sports accessory, a turbine, a boiler, an aircraft or an aircraft carrier.

Gautam Dutta, country manager — Sales and Support, Ansys, talks to Sudhakaran of CIOL, what is the role of simulation in a technology-driven world. Excerpts:

What is the basic concept behind simulation-driven product development (SDPD)? And what are its benefits compared to the conventional methods?

Gautam Dutta: We produce analysis solution software in four areas - structural, fluids, thermal and electronic devices.

To know the functionality of a product, there are two ways. One is to make a prototype of the product and test it, or simulate and then converge the problem. The second is called the simulation-driven product development, where the validations are done on the computer design using various simulation techniques. When these validations are done very early in the product development cycle and are performed at every step of product development, then it is called simulation-driven product development or SDPD.

How does SDPD differ from the conventional product development?

Suppose you want a thumb drive to be developed. You give a set of requirements, both function-wise as well as design-wise. In normal case, the prototypes are designed and redesigned till it gets approved. Then, we have to test it to know whether the product will work or not. Here, my point is, every time when we are designing something, we should be able to simulate and see whether it is going to work.

In normal case, the functions and problems of a product can be known only when it is manufactured.

Also, while making a product, there should be a proper balance between the functionality and economics of the product. All these things can be effectively done with simulation.

For example, suppose you are designing a car. You want to know whether the air conditioner will cool the car in one minute once it is standing in a parking lot open to the sun. How can we know this?

How it is done today? A car is put to stand in the parking lot. An engineer switches on the AC and notes the time taken to cool the car. If there is an issue, then you go back and redesign the air conditioner and test it again! Instead, if we use simulation, money and effort can be saved considerably, because the actual product is produced, tested and manufactured only after the requirements are satisfied in simulation.

{#PageBreak#}

But how reliable is it?

My point is, anyway you are going to test it in real-life situation. Instead of doing it ten times, do it on the computer 7-8 times and then confirm it once in real life so that you can save considerably on time and efforts.

Simulation is not eliminating the physical testing, it rather reduces the number of testing. There should be a convergence between computer models and physical models. But when the number of physical model is reduced, it reduces cost.

It costs nearly thirty lakh rupees to make one prototype of a car, and suppose four prototypes are made before the final product, it costs huge amount. But with simulation, many factors can be verified with the computer design and finally a prototype can be designed based on the results of the simulation, which can be done in a matter of hours. This way the months-long efforts on designing prototypes can also be saved apart from saving money in the long run.

Simulation is more about reliability. Without technology, you will not be able to know whether your product would be reliable.

But here too there is a certain level of hypothesis...

Sure. Not just hypothesis. We are also giving some plus minus percentage in the simulation results. The smaller the percentage that means you are gaining more knowledge into brining the physical life into the mathematical life. This is where the mathematical codes that we write become more important, and these codes become more and more precise as we learn more about technology.

Here I would like to give you another interesting example. Suppose a bomb is being dropped by an aircraft. We have to know what is the air resistance and also other pressures that the bomb goes through. But it cannot be tested in real life. Here comes the use if simulation.

You can solve this issue in three steps — First you do meshing, which means breaking the air into tiny elements. Then using computer technology you find out how much is the fluids creating stress in every element around the bomb. After that, you integrate it mathematically, which would come in a color code on your computer. These steps are called pre-processing, solving and post processing. All the three steps put together give you the result.

Here, what you do is to create a 3D model using CAD, then create a mesh, then you solve the problem and then analyze it. That's how innovation works. an Engineer gets more time to ask more “what if” questions and gets to check-out results of his imagination and knowledge and that drives innovation.

Today every CEO is asking whether I am getting the returns on the money I am spending on expectations of innovation? Simulation is the only way to answer as to how efficient is your innovation spend.

To know what will be the problem faced by an automobile after it runs 35,000 kilometres, whether it is structural problem, electrical problem, thermal problem, or fluid problem, the only way is to simulate; it cannot be analyzed with a prototype. In a simulation environment you can fast forward an issue and analyze it, which cannot be done in physical modeling.

Also, in a normal conventional product development, the specs get designed.

Then somebody does a top level design, then a detailed designing is done. Then you make the prototype one, test it; then prototype two is made, tested again.

It was in this scenario that CAD came in which reduced the time in the design part, but the time taken for prototypes and testing is the same. Here also there is a problem. How do you know what you have designed is right? How do you know that two prototypes were sufficient to test in all environments? Here comes the importance of SDPD.

Though the time taken in designing is the same, and also time is taken to simulate the design, it saves the time and effort on many prototypes. With one prototype, you can reach the final product once you go for simulation. It gives you a comparatively more reliable product.

{#PageBreak#}

What about the cost factor?

Of course, the cost factor would be comparatively higher. For the CEOs of today cost is of no consequence; it is the returns that matters to them. With simulation, you are ahead of the competition in the market, and that is highly important. The thumb rule is that one month-reduction in time to market given you 33 per cent extra market share. And that itself underscores the financial benefits of SDPD.

When you are ahead in the market, you can charge more. That is what, according to me, Apple does with their products — the first mover advantage. It was this innovation that helped Apple beat Microsoft. This means you can charge more if you are innovative. That is where SDPD assumes significance.

Is SDPD kind of replacing CAD?

Never. It is actually complementing CAD. SDPD is actually using the CAD capability to mesh; it is leveraging on CAD. So there is no question of SDPD replacing CAD. The advantage of SDPD is that, unlike CAD, you can analyze the product in every step of development; not just post-development analysis. In SDPD, we can integrate many real-life issues using virtual prototyping. Reality in comprehensive multi-physics and simulation helps tackle varied perspectives.

Which are the verticals that Ansys is focusing on?

We are targeting 11 verticals including automotive, aerospace, industrial machinery and rotating equipment, energy and power, oil and gas including refineries, consumer goods — that include durables as well as food and beverage, healthcare — both healthcare equipment as well as products, electronics and semiconductor, materials and metal treatment.

Then, we focus on academics, because we believe that engineers have to be ready when they come out of colleges. There is a gap between what the industry requires and what the students coming out of engineering colleges are taught. There should be a bridge somewhere, otherwise they will require a lot of training to get ready for the real-life jobs.

Also, engineering service providers are a major focus area for us. We have a development center in Pune where around 200 people are working.

Who are your major clients in India?

Our clientèle is a mix of government organizations, PSUs, private organizations and engineering service providers - including many big names like, BHEL, Whirlpool, Tata Motors, Aditya Birla, Maruti, Genpact, Siemens, Infosys, Quest, Infotech, L&T, Eaton, Brakes India, Rico Auto, NHPC, NTPC, to name a few. Half of our business comes from South India.

How do you see the India market for SDPD? What is Ansys' revenue share from India?

As of now the revenue share with respect to global revenue is 3-4 per cent.

However, our growth here is above 20 per cent. Our policy is to work across the verticals - from automotive to aerospace to energy to academics. And in all these verticals, SDPD is very relevant and adds a lot of value.

We give high importance to education institutes, and we have association with all IITs thus facilitating the readiness of engineering students for SDPD. As the number of engineering graduates that India produce a year is even higher than that of China, this is a significant component of success of SDPD in India.

I believe that global product development will happen out of Asia, because more engineers are coming from Asia, and as we have more resources, more and more companies would take advantage of this availability. The value chain from verticals to academics would drive growth of SDPD.

What are your immediate expansion/recruitment plans in India?

Many global players are setting up their R&D centers in India and we will continue providing them support from India. Also we will focus on Indian engineering organizations. Then, SMB is also an important area for us, because many SMBs are nowadays adopting simulation technology. And, as I said earlier, academics is another area where we are planning to expand.

Interestingly, in many aspects of real life — from green technology to sports and leisure, simulation applications have a major role and its significance will grow over the years. And that is the potential of this segment.