A kind of modal partitioning method to improve ergonomic reliability based on human senses for the multi-modal product

In multi-modal interaction systems, the number, type, and coordination of modalities are critical determinants of human reliability. Due to the inherent limitations of human attention, poorly configured sensory modalities across different channe ls may impair users’ cognitive processing and operational performance, ultimately reducing human reliability in product interaction. To address this issue, a multi-modal segmentation approach is proposed, which is based on Fault Tree Analysis (FTA) and an improved CREAM model refined using DEMATEL. This approach aims to mitigate cognitive and operational overload caused by excessive sensory modalities in multi-modal product design. Firstly, FTA is employed to identify various sensory perceptions in the specific interface operation process, considering contextual factors, to determine which sensory channels should receive the information for cognitive operations. Secondly, the CPC weights of the CREAM model are adjusted using DEMATEL to enhance its applicability for human reliability analysis in multi-modal systems, followed by the calculation of human failure probability for the basic events in FTA. Finally, based on the probability importance and critical importanc e analysis derived from FTA, this approach facilitates the precise identification of which modality causes operational obstacles, enabling targeted design improvements rather than relying solely on subjective judgment from designers.