Predicting preferred motorcycle riding postures to support human factors/ergonomic trade-off analyses within a multi-objective optimisation-based digital human model

Abstract

Digital human models (DHM) can predict how users might interact with new vehicle geometry during early-stage design, an important precursor to conducting trade-off analyses. However, predicting human postures requires assumptions about which performance criteria best predict realistic postures. Focusing on the design of motorcycles, we do not know what performance criteria drive preferred riding postures. Addressing this gap, we aimed to identify which performance criteria and corresponding weightings best predicted preferred motorcycle riding postures when using a DHM. To address our aim, we surveyed the literature to find experimental data specifying joint angles that correspond to preferred riding postures. We then deployed a response surface methodology to determine which performance criteria and weightings optimally predicted the preferred riding postures when using a DHM. Weighting the minimisation of the discomfort performance criteria (an aggregate of joint range of motion, displacement from neutral and joint torque) best predicted preferred motorcycle riding postures.