Geospatial intelligence system for evaluating the work environment and physical load of factory workers.

This study aimed to assess the effectiveness of methods for evaluating the environmental and physical loads on workers in manufacturing plants, considering their locations. Participants were employees of DENSO CORPORATION's manufacturing facilities, and environmental sensors (for temperature and humidity) and BLE beacons were installed to cover the work area. Questionnaires were completed by the participants twice to assess their thermal comfort and fatigue in the work environment. The results showed that a regression prediction model with an adjusted R-squared of 0.418 for fixed-point temperature and 0.495 for perceived temperature was developed for thermal comfort. No linear relationship was found between environmental factors and fatigue, and a decision tree analysis was conducted. Relative humidity and activity level, along with temperature, were selected as predictor variables. The findings suggest that it is possible to estimate the work environment and workload without adding additional measurement-related burdens or challenges. This highlights the usefulness of the proposed method, which takes into account the environmental distribution throughout the work area rather than relying solely on conventional fixed-point observation data, for assessing workers' exposure to the environment and preventing occupational accidents.Clinical Relevance- The proposed approach, combining indoor localization with environmental status, can estimate the condition of workers and is expected to be a good solution for preventing occupational accidents and enhancing workers' health.