A Business Benefits Argument for CFD

Stephen Ferguson, CD-adapco

 

In common with other Computer Aided Engineering disciplines, Computational Fluid Dynamics has often suffered from a misconception that the technique is both complicated and time-consuming to apply. Often, this had led potential adopters of CFD simulation tools to come to the incorrect assumption that investing in CFD technology is a high risk proposition with an uncertain return on investment.

The research presented in this article demonstrates that, far from being a risky proposition, introducing CFD technology into an existing product development process has a high probability of delivering one or more tangible business benefits, the financial return on which is likely to outweigh the total investment CFD technology.

The basis of this research is a survey, conducted in 2007, of industrial users of CD-adapco’s CFD simulation technology. The survey was conducted  order to gain a better understanding of our users, and the ways in which our CFD software is deployed in industry. The results of the survey allowed us to quantify the business benefit that users of our software gain through using CFD simulation on a day-to-day basis.

In this article we explore some of those benefits, quantifying the magnitude of benefit achieved and calculating the probability that a new user of our software will achieve that benefit. Of the 250 respondents to the survey, drawn from a wide range of industries, 96% realised at least one of the benefits described below.

Benefit 1: Improved Product Quality

Increasing product quality is a strategic objective of every company involved in product design or manufacture. Despite the fact that improvements in product quality are notoriously hard-won, increased product quality is the most frequently achieved benefit of using CD-adapco’s CFD technology, according to the users that responded to our survey. In their responses, 66% stated that they had managed to achieve increased product quality as a direct result of applying CFD simulation in their product design process. The average increase in quality estimated by these users was 19%.

By most standard definitions, quality is the extent to which a product meets or exceeds a customer’s expectations. American business quality guru Joseph M. Duran defines quality as “fitness for use”, where fitness is defined by the customer. What constitutes a high quality product is hard to define because quality, like beauty, is in the eye of the beholder, or more specifically, in the eye of a potential consumer. Although quality is difficult to define, most people can instantly recognise a quality product:

“What I mean (and everybody else means) by the word ‘quality’ cannot be broken down into subjects and predicates. This is not because quality is so mysterious but because quality is so simple, immediate and direct.” Robert Pirsig, Zen and the Art of Motorcycle Maintenance (1974)

Better quality products are more desirable to the consumer, and therefore more saleable. A product that is perceived (by the customer) to be high in quality will also typically command a premium over a lower quality alternative. Increasing the quality of an existing product will, almost always, result in increased revenue for that product and, provided that the improvements can be delivered at a reasonable cost, will deliver an increased profit.

If making better quality products were easy then everyone would do it. Any improvement in product quality is usually the result of significant investment in design, engineering, manufacturing and marketing. In ‘CAE-mature’ industries, such as the automotive sector, utilising CFD is a competitive necessity since everyone sees benefits in using it. For these industries, small increases in the quality of an already highly engineered product (for example a 1% increase in fuel economy of an automobile) are likely to command a significant premium in the marketplace.

However, applying CFD simulation to the redesign of a product that has not previously been the subject of CAE analysis can deliver immediate and sizeable benefits. Within a few days of installing STAR-Works, CD-adapco client VIASYS healthcare managed to implement design changes to a neonatal ventilator which they estimate will lead to a tripling of sales: from 400,00 units/year to around 1.2 million units, a potential top-line revenue increase of approximately $16 million.

Benefit 2: Reduction in the number of physical prototypes

The traditional product development process is built upon on an iterative “design-build-test” principle in which the influence of successive design changes is quantified by experimentation on a physical mock-up of the product. Although almost universally adopted, a problem with this approach is that the physical prototypes used in testing are usually both expensive and time consuming to construct.

In the earliest stages of the iterative design process, physical prototypes are also likely to involve a degree of simplification, neglecting some of the physics of the problem or perhaps being constructed at a smaller scale than the actual product.
Increasingly, CAE is being used to replace some of these physical tests, reducing the number of physical prototypes required in the product development process and replacing a number of ‘design-build-test’ iterations with much quicker ‘design-simulate’ iterations. Of the respondents to our survey, 63% claimed that by using CFD technology they had reduced the amount of experimental testing, on average by around 19%.

An MIT survey of over a hundred respondents estimated the average cost of a single prototype to be over $85,000 dollars. However, the net benefit of reducing the number of prototypes is likely to significantly higher than this: product development time saved can be invested in either securing a faster time to market (see below) or in further design iterations that increase the quality of the product (see above).  The rapid turn-around time of numerical simulation (in which simulations can often be performed within the CAD package) also promotes a more interactive design process in which a designer can often visualise simulation results within a few hours of submitting a design change, promoting full and thorough investigation of the design space.

VIASYS Healthcare managed to replace a whole year of physical testing with just a few days of CFD calculation, reducing the time to market and significantly increasing product quality1:

“Within three days we had reduced the supply pressure required to the unit by 48%. This level of improvement previously took approximately 1 year to achieve. STAR-Works saved us $250,000 in physical testing alone, and paid for itself in three days!" Steve Han, VIASYS Healthcare

Unlike testing of physical prototypes, CFD simulations are typically carried out at full scale (the computer model has the same dimensions as the actual product rather than those of a smaller experimental model). This has the considerable advantage that results can be interpreted directly and do not have to undergo scaling, a process that can introduce a significant uncertainty, especially for transient phenomena or those involving a number of fluids.

Replacing physical prototypes with numerical simulation demonstrates increased confidence in CAE technology. When describing how Daimler-Chrysler had managed to reduce the number of full-car mock-ups used in the vehicle production from five to three using CAE simulation (including CFD with STAR-CD), Daimler-Chrysler’s CAE manager outlined the conditions under which CAE becomes a viable replacement for a physical test2:

“For a CAE simulation to replace a physical test, the CAE tool must deliver results of required quality within given time frame. In Virtual Product Development, a late result is not much better than no-result at all” Walter Bauer, Daimler-Chrysler AG

Benefit 3: A Faster Time to Market

If two products, equal in quality and cost, are released to the market six months apart, the revenue generated by the first product during that time is incremental profit. Beyond this, being first to market generally helps to increase both brand awareness and market share while simultaneously maximising the potential lifetime of the product.

A faster-time-to-market is an obvious benefit of reducing the amount of physical prototyping required to bring a product to fruition, but also a direct benefit of the availability of simulation data early in the design process. This allows designers to rapidly eliminate poor design variants, allowing them to focus their efforts on a smaller number of potentially more productive designs. In safety critical or politically sensitive applications, simulation results can also help to demonstrate compliance with legislative requirements, ultimately overcoming one of the final barriers to product release.

In their responses, 39% of our users claimed that, by applying CFD simulation to product design, they had managed to reduce time-to-market, by on average 11%.  Although the length of development cycle varies by product and by industry, the development process for most products is typically measured in years rather than months, for which an 11% speed-up represents a significant competitive advantage over a similar product that was not engineered using CFD.

Benefits 4 and 5: Fewer Field Failures and Avoided Product Recalls

In the perception of most consumers, durability is a key factor in determining product quality. Market research3 has conclusively demonstrated that consumers are not only capable of accurately estimating product durability, but that their perception of durability is often linked to the price that they are willing to pay for a product.

While all products have a finite lifetime, the failure of a product in service can have serious consequences, particularly in the case of safety critical applications, in which unforeseen failure can result in injury or loss-of-life. Even in less serious circumstances, the unexpected failure of a product can act to de-motivate consumers from further purchase, damaging brand reputation, and potentially incurring large warranty expense.

As companies such as Mattel and Firestone will testify, although product recalls are rare, when they do occur, the cost can be enormous, in direct financial terms (the cost of executing the recall, performing repairs, providing replacements and compensating consumers), but more importantly in terms of lost reputation. 26% of our survey respondents indicated that they had reported fewer product failures as a result of applying CFD; 13% reported that they had avoided product recalls.

Benefit 6: Increased Satisfaction of External Customers

Customer satisfaction is the bigger picture. While the benefits listed above might help to increase margins and satisfy internal customers, the biggest benefit of any process improvement occurs when it makes a tangible difference to the end user of the product. More satisfied customers are also more loyal, generally more likely to offer you repeat business or to forgive any temporary lapse in service. According to the survey, 44% of users managed to increase the degree of satisfaction experienced by their external customers.

Multiple Benefits
Realised individually, any of the six benefits described above is likely to yield significant bottom-line benefits for an organization that successfully adopts CFD or CAE technology. However, the real benefit of CFD simulation is that even if you are seeking to realise a single specific benefit, the ancillary benefits of increased engineering insight will inevitably lead to a better overall product. In our survey, 76% of respondents realized two or more of the benefits, 11% four or more, and 4% all six.

The Cost of Inaction
Finally, although the focus of this article is a positive one, it is worth remembering that, in a increasingly competitive marketplace, the benefits described above are available to your competitors.
96% of respondents realised at least one of the benefits described above. Should your competitors adopt CFD technology into their process, when you do not, there is a 96% probability that they will achieve a significant increase in productivity, an increase you might not be able to replicate.