The Real World Is Not Steady-State

When designing and testing new concepts, the only way to get the complete picture and most reliable predictor of future performance is through transient analysis. And to achieve this within the vehicle development time periods, coupling the appropriate tools is critical.

It is common for engineers to want to optimize the testing environment to save costs. In doing so, many attempt to reach a state of “equilibrium” in the testing environment, or a “steady state.” But the real world is anything but “steady,” and neither are tests performed in it. We live in a transient world, where a car’s performance is influenced by its ability to respond to changing dynamics. Sometimes we drive short distances, with many starts and stops. Other times, we drive long distances at a fairly consistent speed. Sometimes we’re towing a heavy load uphill. Other times, we are running quick errands during the hottest part of the day.

To replicate the real-world conditions an automobile will be under, it is important to perform “transient analysis” that can best replicate reality via simulation — before physical, real-world testing begins. Otherwise, you end up over-designing and over-manufacturing —adding design, labor, manufacturing and material costs.

This is where simulation significantly pays, as opposed to costs, as it allows an OEM to virtually prototype design iterations via technology, as opposed to materially.

To replicate the “real world” via physical testing, conversely, you’d need to develop a prototype — which is expensive in and of itself — place it into environmental conditions such as a wind tunnel (again, costly to set up), and perhaps transport the vehicle and a team of analysts to understand how well it will perform in the real world.

Through simulation, design teams can do most of this virtually through simulation, eliminating most of that time, material, travel and labor cost in a virtual environment. Once you have accounted for most (if not all) of the real-world driving scenarios in the simulation environment, the OEM must then only validate these findings in the physical world at the end of the design/production lifecycle.

Case in Point

Take the typical lifecycle of a model make, which historically may have gotten re-launched every eight years or so. In an attempt to infuse that illusive greater efficiency at lower cost, the norm is now closer to four years for the re-launch of the model. Some are even able to iterate or upgrade every two years.

In order to decrease the time between model launches (be they “facelifts,” full redesigns or design iterations), you need to examine security, safety, performance and efficiency enhancements that will occur as a result of the redesign. But, as demonstrated above, testing all of these scenarios via a physical prototype that needs to be transported to manufactured testing environments eats away at that efficiency and builds unneeded time and costs into the production budget, from materials to manpower. Proper transient analysis methods, representing actual vehicle usage, would reduce the problems earlier, faster and at less cost.

This is where technology comes in, but too many vehicle OEMs are constraining themselves through the limitations of their own simulation technology. To cite just once comparison of efficiency, many organizations are using CFD tools that employ 500 to 1,000 processors in a given simulation, while sophisticated simulation methodology may run only 100 or so processors when coupling the right software.

Based on publications released from one automotive manufacturer, using a simple data set, the following conclusions were made:

• In Scenario 1, alternative transient simulation methodology was able to simulate one driving profile (i.e., set of testing conditions) using 512 processors, and it was able to perform the test and provide results in 30 days.
• In Scenario 2, using advanced transient analysis methodology, they were able to simulate five driving profiles in one week using 100 processors.
• The advanced transient analysis methodology was able to reduce the calculation time by 83%, with an 83% reduction in resources used. 

Scenario 1 – Transient CFD Methodology	Scenario 2 – Advanced Transient Analysis
512 processors	100 processors
30-day simulation cycle	7-day simulation cycle
1 static driving profile	5 dynamic driving profiles
	83% reduction in calculation time
	83% reduction in resources used

In other words, achieving more at less cost, as the data clearly shows.

Simulation Pays, Not Costs

Performing sophisticated simulation via advanced ThermoAnalytics transient methodologies does not come at greater cost to vehicle OEMs. Quite the contrary, it eliminates many costs baked into the methodology that many OEMs currently use. Furthermore, simulation, when done properly and when coupling the right tools and technology, does not add time to the design and production lifecycle — it reduces it!

With intelligent, efficient technology, OEMs and designers will accelerate the vehicle’s production on the development side while dramatically reducing cost on the testing side.

This savings in time, costs, materials and labor can then be reinvested back into the business and into its processes. The savings in time (moving from four years of testing time to, say, two and a half years), the OEM will shave off six to 18 months from the testing, designing, engineering and production time, saving labor and material costs at virtually every turn.

That’s time, money and manpower better spent on innovation, marketing, designing and finding even more creative answers to that age-old question: How can we achieve more at less cost?

For more information about ThermoAnalytics’ transient analysis solutions for the proprietary TAITherm software, please visit our website.