Don't buy CFD...Yet

Before spending any money on CFD (software, hardware or consulting) ask yourself these 4 questions to see if it can solve a real problem for you.

Question Time

This is based on an exercise I go through with prospects on our first call. A process I’ve described as “due diligence for your CFD-related spend.” It doesn’t go deep into engineering problems. We don’t discuss flow physics or turbulence models. And we definitely don’t dissect the feature list of any particular CFD software. It’s far too early for all that on a first call. Plus, discussing engineering problems only leads, to discussing engineering solutions and we don’t even know what the real problems are yet.

There’s a CFD code, strategy or tactic that can cope with your engineering problem (almost) guaranteed. So at this point in the conversation we aren’t trying to establish which flavour of CFD is best for you. We’re trying to figure out if you should be bothering with CFD at all?

When you’re eager to get to a “solution” it’s tempting to skip difficult questions like the ones below. But they should help you to figure out the value of what you’re thinking of doing with CFD. With that value you can get an indication of whether you’re going to manage a return on your proposed CFD investment.

So here’s a DIY lite version of my discovery call. Next time CFD spending comes around, don’t get your wallet out unless you have a good answer for these 4 questions…

Are you solving a business problem?

Engineering problems are relatively easily spotted (especially by engineers). However these problems don’t always get first dibs on budgets. If something doesn’t work as well as it could (or should) AND there’s some budget left in the pot AND we have some spare resource — then it MIGHT get fixed. CFD is pretty good at addressing engineering problems.

Business problems, on the other hand, directly impact your business’ goals and can be measured against your business’ metrics. They’re usually very close to the money in some respect. As such, they’re the ones that Directors & Customers pay attention to. But they might not always be as visible to Engineers. Perhaps it’s a problem that’s reducing sales, or hurting renewals, or increasing your customer’s cost of ownership or reducing your margin? These problems often carry much more urgency than engineering problems. Solutions to these problems tend to float much higher on the “to-be-fixed “ list.

However, not all business problems have an engineering aspect. Similarly, not all engineering problems are business problems. And, to top it off, not all engineering problems can be solved in CFD — tricky.

“The sweet spot - a business problem, rooted in engineering, with a CFD solution”

So, for your sweet spot. You’re looking to identify a business problem, that’s rooted in an engineering problem, that has a CFD solution. These problems present an opportunity to generate a return on your CFD investment. They also have the greatest chance of getting the attention and resources they deserve.

Examples of business outcomes, rooted in engineering, with a CFD solution could include:

Is it an expensive problem?

Even if you can identify the type of problem we’ve just outlined, does fixing it have enough upside to justify investment in CFD? This often comes down to margins and volumes. Do you have either the margin &/or the volume over which you can spread the CFD cost?

For example, what if we used CFD to design a custom air management kit for your semi-trailer? It’s probably not economical to do this work for a single unit. It may never recover its cost, let alone turn a profit. But the same kit applied to a fleet of 500 trailers is a different proposition. If you don’t have the margin or volume to support and generate a return on your investment then it might not be the best subject for a CFD study.

Do you have an alternative?

Having identified your target problem, you need to consider your alternatives for finding a solution. You have no decision to make unless there are reasonable alternatives AND a degree of uncertainty as to which is best. Without those components you have a false dichotomy, not a real decision. Usually business problems aren’t short of uncertainty. As for reasonable alternatives, can this problem stay unresolved now that it has been identified? Is that an option? The more expensive the flow problem the less likely it is to be ignored.

However, what about other ways of solving the engineering flow problem that don’t involve CFD? Maybe you can get a better return by prototyping? Making and testing a few options is often a viable option for manufacturers changing their own products. Maybe not so viable for end-users/operators.

Can you act on the outcomes?

It’s quite likely that the outcome of any CFD project will call for a change of some description. Maybe it’s a design change? Maybe it’s an operating procedure change? Are you in a position to implement these changes? Manufacturing companies are much more likely to be able to change a design than an operator can. Conversely, an operator can change their operating procedures directly, where manufacturers can only really make recommendations.

If you can’t action the change then you won’t ultimately be able to solve the problem. So if there are areas that you don’t control, either due to regulations, operation or some other restrictions then now is a good time to get those out on the table.

Did you manage 4 YESes?

Going through this type of questioning before starting the talk of CFD codes, software features and engineering solutions, is a useful exercise to help focus on the right things. And by right I mean the problems that CFD can have a real impact on. It might not always be the problem you first thought of. Or even the most interesting problem. But if it can cover your CFD investment (a return is always nice too) then you can look at the more interesting or more obvious stuff next time around. That is, if you’re not too busy wondering what to spend your extra profit on ;-)