Iterative prototyping to buy down risk

This article was first published in Engineering Magazine (starts on page 26)

The image of the lone mad scientist toiling away to build a fully functional prototype from multiple apparently disassociated components and having their eureka moment couldn’t be farther from the truth. True prototyping is performed by teams of engineers and designers de-risking a system design to develop a product with the greatest likelihood of success.

This is done through building not one successful prototype but many sequential prototypes where success is measured incrementally. Failures during the process inform the design and evaluation approach taken for subsequent prototypes until the major technical risks are effectively mitigated.

But how does the iterative prototyping process buy down technical risk? Learning from early failures on isolated components and subsystems, enabling lateral thinking, and providing necessary course corrections are all critical to success, and if done correctly can help designers land on the perfect solution.

Prototyping creates opportunities to fail early and save money

Failure is not an option; it’s an imperative. By building prototypes early in the design process, the team introduces the opportunity to fail early and to avoid pursuing an unsuitable concept or technology.

Doing this saves considerable time and resources by eliminating options early and allowing for more time and effort to be spent on concepts and technologies with a greater chance of success. These early low-fidelity component and subsystem prototypes also tend to be cheaper to design, build, and execute than later system level tests which further mitigates financial risk.

That is not to say early prototypes always result in failure, but having the mindset that failure - especially a small failure – is good for development, positions the team for greater success and leads to more robust final system designs.

Prototyping enables lateral thinking

A lot of engineering problems tend to be solved with vertical thinking. Performing stress analysis, designing circuits, selecting components all involve sequential reasoning, step by step, until you arrive at a solution. Typically, designing a new product is not like this. Often, the best solution isn’t the most obvious, and the most prudent approach is to expand the solution space rather than prematurely refining a potential solution that may not be the optimum.

This is where quick low-fidelity prototypes can be a powerful tool. By limiting the amount of investment in each idea, the team can explore many more options, with each prototype giving them the opportunity to learn something unexpected that inspires a new direction. Prototyping means you can avoid irrelevant constraints; it reveals creative solutions and optimises the product design.

Subsystem prototypes are powerful

Prototypes don’t need to be full systems. In fact, with high-risk technologies it’s often best to assess features and performance effectiveness individually at the component or subsystem level instead of in a fully integrated system. This gives the engineering team mission-critical insights such as its true performance characteristics and how the component or subsystem affects and interacts the complete system. It also allows for iteration before the team solidifies interfaces between the various subsystems that encompass the entire system.

By spending development time and effort on a component or subsystem, design teams can save money and time by eliminating the risk of a component or subsystem failure rendering a full system prototype unusable; a single early component or subsystem prototype can save multiple full-system prototypes if it identifies interface issues or packaging constraints with a much greater system-level impact.

Similarly, if a subsystem relies on a technology that proves unsuitable for the larger system, evaluating that technology apart from the larger system allows the engineering team to recognize that it’s ineffective and that an alternative technology is needed.

Characterising and evaluating multiple technology candidates helps to determine which works best for the full product. Lastly, evaluating components and subsystems early allows the team to reconcile the lower-level requirements with the system-level requirements, and demonstrate how they’re met ahead of the final release of the larger system design.

Iterating provides more opportunities for course correction

We have all heard the phrase: “If at first you don’t succeed…”, an expression which summarises the foundation of iterative prototyping. Since it’s unlikely that the first prototype will meet all of the requirements, planning for iterations allows the team to develop the design organically. Planning for this also helps to establish stakeholder expectations that the process is an iterative one, and to not expect the final product from the first round.

Each prototype can focus on a different aspect of the final design and build on the failures and successes of the prior prototypes. This means teams can filter the solution space by eliminating features and concepts that prove ineffective and by optimising those features that suit the design.

Early prototypes can also inform the suitability of manufacturing processes for the final product. Focusing on manufacturing methods too early can distract from developing a design that meets the functional requirements, resulting in lost opportunities to choose the appropriate manufacturing method. This proves especially costly with parts that need expensive tooling. It is important to consider manufacturing processes early on, but using stand-ins – like machining a plastic part rather than moulding – allows the team to verify a part’s functionality before maturing its design prior to making costly decisions on how to fabricate it for production.

Iterative prototyping brings clear value to the development process through buying down risk early in the design process where it is most cost effective and enabling the success of a product. By moving through the process of making multiple prototypes, both laterally and sequentially, failures and successes can inform the design and efficiently and cost-effectively develop a product that meets its requirements with minimal technical risk.