Aerospace & AeronauticsDoctoral Program

Aerospace & Aeronautics PhD

Doctoral
Mission Studio

Start with a research question, make an independent judgment, and build a result another researcher can check. The same evidence becomes part of your first-year review, candidacy dossier, and dissertation defense.

Research method

From judgment to evidence

Start with a mission brief that names who or what the research could serve. Use a benchmark in the first year and carry the evidence forward as your dissertation question takes shape.

  1. 01

    Work independently

    Set a baseline and a decision rule. Record an initial judgment before using AI.

  2. 02

    Investigate with AI

    Use AI to propose a test or challenge a conclusion. Log material claims and how you checked them.

  3. 03

    Reproduce the result

    Ask a second researcher to rerun the analysis or examine restricted evidence through an authorized review.

  4. 04

    Defend the claim

    Explain what the evidence supports, where the conclusion breaks down, and what would change your mind.

Aerospace Engineering

When should you distrust a simulation?

Use the NASA wall-mounted hump case to test how closely a published model recovers a measured feature of separated flow.

Research brief

  1. Choose an acceptable error in reattachment location before examining the worked result. Put the tolerance in units of x/c, distance normalized by chord.
  2. Estimate reattachment in each series by interpolating across the downstream crossing of skin friction from negative to nonnegative. Record the two sampling points that bracket each estimate.
  3. Challenge the illustrative AI claim, “Grid resolution alone explains the disagreement.” Identify what this comparison actually tests and design one follow-up capable of testing that claim.
Worked result and tolerance check
Experiment
1.116
CFL3D model
1.267
Difference in x/c
0.151

The model reattaches farther downstream in this comparison. The experiment crosses within x/c 1.09–1.15, and the model crosses within 1.263–1.269. These are sampling brackets. The analysis uses published solver output and leaves experimental and numerical uncertainty to a follow-up study.

The downloaded script reproduces the comparison from both input tables.

Aviation Practice

Does an AI tutor improve retention?

Use a synthetic comparison of AI-assisted and standard ground-school practice. Examine whether the answer changes when you measure learning a week later.

Research brief

  1. Specify day-seven retention as the primary outcome before looking at the scores. Define the smallest difference that would matter in a real study and explain why.
  2. Compare the arms at day seven. Check immediate scores and confidently incorrect responses as separate outcomes. Report denominators and flag the limits of the comparison.
  3. Challenge the illustrative AI claim, “Higher immediate scores demonstrate lasting learning.” Explain what the delayed measure shows and specify the additional evidence needed to evaluate a real tutor.
Worked result by outcome

The teaching dataset contains 12 synthetic records per arm. The AI-tutor arm scores 14.2 percentage points higher immediately and 2.5 points lower at day seven. These figures illustrate why a protocol defines the primary outcome in advance.

The synthetic comparison includes no participants or product performance data. Use it to practice interpreting outcomes and designing a study.

Faculty review

What the PhD assesses

The same rubric applies to both pathways. Submit the artifact and independent checks alongside your capability record. Use the qualifying practice questions and worked answers to rehearse how you will defend them.

Independent reasoning
Explain the baseline, decision rule, and initial judgment. Show which parts of the reasoning you can defend without assistance.
Evidence and uncertainty
Connect each major claim to the appropriate data or outcome. Identify limitations that materially affect interpretation.
AI verification
Record material AI assistance, one claim you checked, and how the check changed or confirmed the conclusion.
Reproducibility and defense
Provide inputs, instructions, and review evidence. Explain failures or restrictions and defend the scope of the result orally.

Reproduce the worked examples

Download both datasets, a Python script, and notes on the sources and methods. The script runs locally with the standard library.

Download the mission kit

Degree progression

One record across the doctorate

Use the mission record to show how your methods, evidence, and independent judgment develop through Calibration, Elevation, and Activation.

  1. 01

    First-year review

    Submit a baseline artifact and record the advisor’s feedback, the next test, and its review date.

  2. 02

    Candidacy

    Defend the baseline, the evidence record, and the dissertation proposal under the published qualifying process.

  3. 03

    Dissertation

    Update the record at each annual review and bring reproduction or authorized review evidence to the final defense.

  4. 04

    Research dissemination

    Prepare the dissertation for deposit and share permitted materials. Eligible original work can enter Aviation Futures’ editorial process.