The Engineering Design Process (Years 6–9) — Science

Defining a problem, identifying criteria and constraints, generating multiple designs, and building, testing and iterating a solution.

Ask an engineer to protect a raw egg dropped from two metres, using only ten straws, 50 cm of tape and a sheet of paper, and they won't reach for glue straight away. They'll define exactly what "success" means, sketch several different ideas, build the most promising one, drop-test it, and then make it better. That structured process — not a lucky first guess — is engineering.

Why this matters Engineering design is a distinct process from scientific investigation, and it is tested separately in the NGSS Science and Engineering Practices. Science investigates how a variable affects an outcome, using a fair test to find out what is true. Engineering starts from a problem and optimises a design against criteria and constraints — there is often no single "correct" answer, only better and worse solutions. Bridges, phone cases, water filters and rockets are all designed this way.

Define the problem

Every design starts with a problem stated clearly enough to work with. "Protect an egg" is too vague. A proper design brief is specific: "Design a package, built only from the given materials, that stops a raw egg from cracking when dropped from 2 metres onto a hard floor." A clear brief tells you exactly what you are trying to achieve before you touch any materials.

Criteria and constraints

Every brief splits into two kinds of requirement, and mixing them up is a common mistake:

Criteria — what the solution must achieve to count as successful. For the egg drop: the egg must not crack; the package must be reopenable to check the egg without destroying it.
Constraints — the limits the design must work within. For the egg drop: only the given materials may be used (no extras); the package must fit in a 15 cm cube; it must be built within 30 minutes.

Criteria (must achieve)	Constraints (limits)
Egg survives the drop intact	Only straws, tape and paper allowed
Package can be opened to inspect the egg	Must fit inside a 15 cm cube
Package is reasonably lightweight	Must be built in 30 minutes

Generate more than one design

A common beginner mistake is building the very first idea that comes to mind. Good design practice is to sketch several different concepts and weigh their trade-offs before committing materials: a cushioning design (egg wrapped in a soft padded nest of paper), a structural cage design (a straw framework that absorbs impact by bending or breaking instead of the egg), or a parachute/air-resistance design (a large paper canopy that slows the fall). Each has strengths and weaknesses — cushioning is simple but heavy; a cage is light but risky if it doesn't flex enough; a parachute needs enough drop height to work. Comparing ideas before building saves wasted materials and time.

The design cycle is not a straight line — testing feeds back into improving and redesigning, again and again.

Build, test and iterate

Once a design is chosen, it is built as a working prototype and tested against the criteria — for the egg, that means an actual drop test from 2 metres. The critical step most beginners skip is what happens after the test: the results must be used to iterate — improve the design and test again — rather than accepting the first attempt as final. If the egg cracks, a good engineer asks exactly where the impact force reached the egg and why the padding didn't absorb enough of it, then changes that specific part of the design before re-testing.

Put these steps of the egg-drop engineering challenge in the correct order:

Write a clear design brief: what must the package achieve?
List the criteria (must achieve) and constraints (limits)
Sketch several different possible designs and compare them
Choose the strongest design and build a prototype
Test the prototype by dropping it from 2 metres
Use the test result to improve the design, then test again

Worked example. A team's first egg-drop prototype is a straw cage with no padding — the egg cracks on drop 1. Explain what an engineer should do next, and why. They should examine exactly how the egg failed (e.g. the impact force reached the shell directly through a rigid strut) rather than discarding the whole idea, then make one specific, targeted change — such as adding a soft paper cushion layer between the cage and the egg to absorb impact energy — and re-test. This links the test result directly to a design improvement, rather than randomly trying something completely different or assuming the first failure meant the whole concept was wrong.

Common mistake: confusing this process with a scientific fair test. Fair testing (see Variables & Fair Testing) investigates how one variable affects an outcome, changing only the IV and controlling everything else, to find out what is true. Engineering design instead optimises a solution against a set of criteria and constraints — there may be several workable designs, and the goal is the best one for the brief, not a single correct fact.

Memory hook: Define, Design (several!), Build, Test, Improve — "Dizzy Ducks Build Tall Igloos". And never treat your first prototype as your final answer: the test is there to make the next version better.