When Does an Engineering Project Become Research?

A working build becomes research when it answers a focused question with evidence—not when it simply adds more features.

1. Frame a Testable Question

Narrow a broad interest into one question that names the system, condition and outcome you can actually measure.

2. Map the Existing Evidence

Compare prior problems, methods, datasets, assumptions and findings in a literature matrix. Look for patterns, not a pile of summaries.

3. Validate the Gap

A gap is not merely “few papers.” Show a credible limitation in performance, validation, robustness, scalability, sustainability or deployment.

4. Align Problem, Questions and Objectives

Make each objective produce evidence for a research question, and make every question trace back to the defined problem.

5. Design a Fair Method

Specify variables, conditions, instruments, calibration, data and repeats before testing. Control what must stay constant.

6. Compare Against a Baseline

Test the proposed approach and a credible baseline under the same conditions. Use metrics that match the claim you want to defend.

7. Analyze and Document Honestly

Report results, uncertainty, limitations and unexpected outcomes. Record the setup and decisions so another engineer could reproduce the work.

Research Is a Chain of Evidence

Question to gap. Gap to method. Method to comparison. Comparison to a defensible conclusion. Keep every link visible.