When is a rigid pelvis–trunk approximation sufficient? An output-specific model-reduction protocol applied to lumbar mobility in multibody biomechanics

This preprint presents an output-specific computational model-reduction protocol for evaluating when simplified lumbar mobility is sufficient in multibody biomechanics. The primary study compares mass-equivalent exact-lock, axial-yaw-only and three-coordinate lumbar architectures using the public ModelicaHumanBodyPArts v0.10.0 RC1 library. It preserves the six mirrored nominal cases, 45-case operating sweep, 20 passive-lumbar sensitivity simulations, 18 numerical-refinement runs and analytical two-inertia benchmark. This revision adds the verified rigid head and upper-limb inertia sensitivity campaign: an analytical known-answer check and 45 simulations across 15 conditions and three architectures. At the stated 10% locked-normalized criterion, primary pelvis/thorax yaw adequacy is 15/15 each. With restored upper-body inertia, pelvis yaw meets the criterion in 13/15 conditions and thorax yaw in 15/15. The two pelvis-yaw exceedances are 11.19% and 11.95%, and both satisfy 20%. Free-side and support/contact outputs are more sensitive. Adequacy is specific to the output and operating conditions. The manuscript is a preprint and has not undergone peer review. This is a computational model-reduction study and does not establish human or clinical validation. Complete reproduction data, source code, logs and portable tests are archived at https://doi.org/10.5281/zenodo.23175126. The unchanged public software release is archived at https://doi.org/10.5281/zenodo.22941517. This version preserves the collaborator review markers; the publication-field patch updates only the archive filename, SHA-256 and version-specific dataset/preprint DOIs.