Design is a process, not a talent
When people hear "design" they usually think of creativity or innovation. Those things are part of it. But in engineering, design means process - a systematic way to apply technical knowledge to real-world, open-ended problems.
A real-world problem is meaningful for people, society, or the planet. An open-ended problem has no single right answer and no single route to get there. Most engineering problems are both. Your job as an engineer is to understand the problem, then find the best solution you can.
This is completely different from a textbook problem, which has one right answer, is already well-defined, and tells you which method to use. Textbook problems are good for learning specific techniques. Design problems are what you'll actually face.
What the novice gets wrong
Most people, when presented with a problem, immediately try to solve it. They pick the first idea that comes to mind and start building. When it doesn't work, they fix it. When it still doesn't work, they fix it again.
This trial-and-error approach might work fine when you're building with Lego. It does not work well when you're designing a bridge, a medical device, or a solution to an infrastructure problem. The costs of fixing mistakes rise sharply the further into the project you are. And the result is almost never optimal.
The 5-stage engineering design process
Here is the process you'll be working with on May 16. Each stage has a specific purpose:
Iteration is not failure
The stages are not a straight line you walk through once. In practice, you are constantly moving backwards. You discover something in Stage 3 that sends you back to Stage 1. A constraint you missed changes which solutions are viable. A new idea surfaces in Stage 4 that's worth revisiting in Stage 3.
This is expected and normal. The design process handles iteration systematically. An unstructured approach treats every backtrack as a failure. The formal process treats it as information.
Why early stages are the most important
Consider two cost curves running alongside each other through the project timeline.
The first is actual costs - what you're spending. Early stages are cheap. Speaking with stakeholders, sketching concepts, and studying the problem costs very little compared to building and testing a solution in Stage 4.
The second is the costs committed curve - future spending that is effectively frozen by decisions you make now. Even in Stage 1, your choices about what the problem is and whose needs matter will determine most of what you'll need to spend later. Choose the wrong problem in Stage 1 and Stage 4 will be very expensive.
The novice approach skips the cheap stages and commits enormous resources based on incomplete understanding. That's why it almost never produces an optimal result.