Prototyping

In one line: A prototype is an approximation of the product built to answer a question. Decide what you need to learn first, then build the cheapest thing that answers it.

Core idea

Ulrich and Eppinger define a prototype as an approximation of the product along one or more dimensions of interest [1]. That covers everything from a sketch or a spreadsheet model to a full working unit. They classify prototypes along two axes, which together give a useful map for choosing what to build:

How it works

Four purposes of prototypes:

• Learning: will it work, and how well does it meet needs?
• Communication: showing the idea to managers, customers and partners.
• Integration: making sure subsystems work together.
• Milestones: demonstrating that the product has reached a defined level of functionality.

Principles:

• Physical prototypes reveal unanticipated phenomena that analytical models miss.
• Prototypes can reduce the risk of costly iterations, so build early when the cost of a late surprise is high.
• Prototypes can expedite other steps, for example by giving manufacturing or test something real to plan around.
• They can restructure task dependencies, letting work proceed in parallel.

Planning a prototype:

1. Define its purpose.
2. Establish its level of approximation.
3. Outline an experimental plan.
4. Create a schedule for procurement, construction and test.

In practice

A practical exercise is to take a prototype from your own field and place it on the two-axis map, then ask which purpose it served. Engineering organizations often build integration and milestone prototypes on schedule but skimp on cheap learning prototypes early, when a quick breadboard or a simple model would have exposed the problem months sooner.

Questions to ask yourself

• What specific question is our next prototype meant to answer?
• Is there a cheaper, more focused prototype that would answer it sooner?
• Which of our risks would one early physical prototype retire?

Connections

• Related methods: Concept Testing, Design of Experiments, PDCA Cycle.
• Partner discipline: Industrial Design. Purpose: Risk Management.
• Capabilities exercised: Resourcefulness, Realizing the Vision, Inquiry.

Sources

• Ulrich and Eppinger - Product Design and Development [1]

[1]

K. T. Ulrich and S. D. Eppinger, Product Design and Development, 6th ed. New York, NY, USA: McGraw-Hill Education, 2016.