A simple observation during a Cairo Metro ride became a small engineering investigation: I reverse-engineered the overhead hand strap, modeled it in SOLIDWORKS, and ran three complementary FEA studies to answer different “what if?” questions.

Assembly (reverse-engineered)
Strap: Nylon 101
Handle: Polycarbonate (PC)
Screws & Loops: AISI 304 Stainless Steel
Connections: Bonded for initial runs; contacts tested during verification
Fixtures: Top mounting points fixed

🎯 1) Static Study — Goal: Can the assembly withstand a worst-case static load?
Setup: 1000 N applied conservatively
Result: Max Von Mises Stress = 102.4 MPa (located on AISI 304 component)
Max Displacement: 2.218 mm

Min Factor of Safety: ≈ 2 → Conclusion: The steel component is within yield limits, showing the metal attachment is structurally safe under this conservative static load.
Insight: Even though the highest stress was on steel (acceptable), verify load paths to ensure plastics are not exposed to damaging local stresses.

🎯 2) Fatigue Study — Goal: Will the product survive repeated daily usage?
Setup: Cyclic load representing repeated passenger grips
Result: Predicted life > 1,000,000 cycles for the defined load case
Insight: The assembly demonstrates good fatigue resistance under the chosen loading assumptions — still, fatigue behavior of nylon/PC interfaces needs close attention in detailed design.

🎯 3) Nonlinear Dynamic Study — Goal: How does the strap respond to transient jerks/impacts?
Setup: Geometry scaled 1/20 to reduce computation; applied force = 5 N (conservative relative to strict geometric scaling)
Result: Max Stress = 36.143 MPa; Max Displacement = 0.044 mm
Insight: Under transient conditions the assembly behaves stably with negligible deformation. Note: I used 5 N (vs theoretical 2.5 N by strict area scaling) intentionally to be conservative and capture dynamic effects.

🔎 Verification & Assumptions:
Max stress occurred on AISI 304, whose yield (~200–215 MPa) gives acceptable safety margin.
I performed mesh refinement and contact checks; still recommend a final mesh convergence study and component-level probes before publishing as definitive.
Scaling used in dynamic run was chosen to balance computational cost and conservative loading; I clearly state this assumption in the post.

It’s a testament to the hidden engineering in everyday life.