May 11, 2012
|Processes implemented by the best-in-class companies to support their ability to get it right the first time. Source: Aberdeen Group|
Analyzing and modifying design parameters early and often can help companies engineer better products.
Today's business world is characterized by increased demand for innovation, shorter product lifecycles, and pressure to launch new products quickly. At the same time, R&D teams face cost-cutting challenges. Yet no business can afford to sacrifice robust design. Products must perform as expected in the real world, every day and in every situation.
Engineering simulation, with its ability to design, prototype, and test products in the low-risk virtual world, provides a fast, cost-effective way to create robust designs. These advanced solutions also support parametric analyses—in which specific design parameters are modified and the effect of variations is studied iteratively. By understanding the impact of each change, an organization can improve product development speed by a factor of 10.
Parametric analysis: High speed, high integrity
A 2011 study of companies using computational fluid dynamics (CFD)—Getting Product Design Right the First Time with CFD from the Aberdeen Group (Boston)—demonstrates the high value that organizations place on "getting it right the first time." The top 20% of participants deemed "best in class" by Aberdeen showed a higher propensity to focus on getting design right from the start (47% versus 39% for all others).
These companies understand that if a non-optimized product is introduced, there may not be a second chance to fix the design of the product after a recall. The design must be flawless upon launch. Why? A business down the road—or across the globe—will quickly steal market share and come up with a better design. Engineering simulation has helped businesses in every industry meet that challenge by enabling virtual design, prototyping, and testing through in-depth design investigation to reduce risk.
According to Aberdeen's findings, best-in-class companies are 81% more likely to use simulation to make design tradeoffs than others at 66%.These companies also use simulation to assess multiple design criteria simultaneously at a much higher rate than other businesses.
In addition, the Aberdeen study found that by varying model parameters through simulation, engineers can quickly make the highest-impact design tradeoffs that maximize overall product quality and performance, as well as simultaneously investigate multiple design criteria. Many companies use these tools strategically, creating high-fidelity models by identifying the right combination of geometric meshing, numeric, and calculation schemes to obtain reliable and accurate results. By varying parameters of complex models—including materials properties, operating conditions, and geometries—engineering teams can rapidly quantify the impact of any modified parameter on a product's overall behavior.
In fact, engineers can virtually evaluate the best possible designs against a range of real-world scenarios—for example, how a wind turbine holds up to hurricane force winds or how an airplane engine functions at an altitude of 10,000 ft. In some cases, they identify critical areas that could jeopardize product integrity—and that require more specific validation—or help the engineering team to understand which parameters really matter. This improves the ability to define cost-effective and targeted design of experiments for final physical testing.
A case in point: Dyson
U.K.-based Dyson develops common household products such as the vacuum cleaner, but they also espouse an innovative, best-in-class engineering process. Dyson routinely leverages parametric analysis to bring products to market rapidly, while also ensuring product integrity. In designing its unique Dyson Air Multiplier bladeless fan, engineers had to develop and optimize an original product without the benefit of any previous design experience.
To complement experimental testing and minimize development time, engineers used simulation software and a parameter-based approach to evaluate up to 10 different designs per day. Dyson's engineers steadily improved fan performance to 2.5 times the original concept design. The team investigated 200 different design iterations using simulation, which was 10 times the number that would have been possible with physical prototyping (Image below).
Making parametric analysis business as usual
Parametric design investigation can accelerate development processes while still protecting product integrity. Even so, only the best-in-class companies—as described in the study—leverage the benefits of varying model parameters for what-if analyses, goal-driven optimization, and trade-off decision making.
Systematic use of engineering simulation should be business as usual, and comprehensive parameter-based investigation should be mandatory in any design document. Project managers should ensure that teams identify important design parameters as well as investigate and quantify possible impacts on product functionality.
Challenging decisions and assessing tradeoffs through variations of parameters should be made at every step of the design process—not just after building a complete virtual prototype. Every design decision may benefit from modifications that produce better quality environmental benefits, or cost savings that do not compromise overall product integrity.
Companies can seamlessly integrate virtual testing at the core of the design process, without slowing it down (Image below). Designs can be modified in the early stages when there is more flexibility and fewer negative consequences. Whether applied to complex systems or individual components, parametric investigation has the power to change the entire product development process.
|Engineering amplification requires an organization to systematically adopt simulation at each stage of the smart product system design model (V shape). Source: Aberdeen Group|
Preparing for a new future
A number of important developments are making parametric analysis more feasible. Computer processing speed and memory capabilities have been obstacles to the widespread use of analysis. However, large-scale multicore and multiprocessor computers form a new technology infrastructure to implement this winning strategy.
Simulation software has made a number of rapid advances that support the growth of parametric analyses. There is a probabilistic nature associated with every parameter-based study because of natural variations of geometries, material properties, and operating conditions; but future software improvements will deliver tighter tolerances and advanced probabilistic descriptions of expected product behavior, as well as risk-of-failure predictions.
When simulation-driven product development was first introduced, there was a performance gap between the leaders, who quickly adopted this practice, and followers, who favored a wait-and-see attitude. The leaders quickly and cost-effectively launched innovative, new products with a high degree of confidence in their results, and the followers lost market share and profits. As the global engineering community systematically adopts parametric analysis, the same type of performance gap is expected between the early adopters and companies who are more hesitant to embrace the future.
Source: R&D Magazine