Given the current economic conditions, companies must continue to deliver existing products as they develop the products of tomorrow, without increasing the engineering staff. To accomplish this paradoxical challenge, innovative companies have adopted Model-Based Design over traditional development methods.

Historically, new product development has started with requirements from many different sources, including government regulations, engineering standards, and corporate direction. Engineers must then compile this set of disjointed requirements into a unified specification to drive the development process. Assessing the feasible design space from the large set of independent requirements is challenging and time consuming, given the ambiguity of some requirements and outright conflicts between others.

Once the feasibility analysis has been completed, the design requirements are captured in a textual specification that drives the design process. This specification, however, often contains flaws that are then captured in the design and instantiated in the implementation. These flaws in the requirements are then discovered during testing of the hardware implementation, which is performed at the end of the traditional development process.

The late detection of flaws triggers additional testing to determine the root cause of the errors as well as to redesign and retest. As a result, engineers who could be developing the next generation of products are instead reworking designs that were thought to be correct.

Implementing Model-Based Design

Instead of carrying an inventory of design errors throughout the development process, many companies have adopted Model-Based Design. This approach starts with the same set of requirements, but enables engineers to develop executable specifications in the form of models rather than textual specifications.

Engineers use these models to clarify requirements and specifications, quickly evaluate design and configuration choices, and facilitate simulations to verify that a design satisfies the requirements. In one study, engineers uncovered more than three times as many issues with requirements by using Model-Based Design than by using traditional design methods.

These same models let engineers create acceptance criteria and verification tests that can be reused throughout the development process. They can apply these same verification tests to the code, which typically is automatically generated from the models.

By using Model-Based Design to develop flight controls, aerospace companies have reported reductions in development time of 40 to 60 percent from their ability to find and address requirements and design errors early in the life of a program.

What’s next for these companies? They are investigating how to use their additional engineering capacity to win new contracts and develop the next generation of innovations. Their return on investment goes beyond the achieved improvement in quality and time to market; it extends to the additional capacity made available to meet their next design challenge, without increasing the engineering staff.

More Information

For more information on Model-Based Design, visit .