Manish Sharma Lab
Menu

Lab note · 2026-05-09

What AI Needs Before Recommending LMD Repair

I would not recommend Laser Metal Deposition from a vague part description alone. The minimum useful context includes material, geometry, damage type and depth, operating conditions, tolerance, and inspection requirements.

Good AI behavior is conservative: it separates known facts from assumptions, asks for missing information, and explains uncertainty. If material grade is unknown, the recommendation should stay tentative.

Local damage on a large, high-value part may be a good LMD candidate because the process can add material only where needed. That signal is still not enough by itself. Access, heat sensitivity, post-machining allowance, and inspection feasibility all matter.

For safety-critical parts, the answer should move toward stronger inspection and expert review rather than confident automation.

Related pages

Related source notes

These source notes are placeholders for future citation links. They are not invented citations.

Source link to be added: review literature on in-situ monitoring in metal additive manufacturing. Profile planned. Further reading to add
Source link to be added: ISO/ASTM additive manufacturing terminology. Profile planned. Further reading to add
Source link to be added: public Exafuse LMD case study. Profile planned. Further reading to add