Predicting Maximum Work Duration for Construction Workers

One of the most common health problems that threaten transportation infrastructure construction workers in Hong Kong is heat stress. An effective way to reduce this problem design a proper work–rest schedule, and key issue predicting maximum working duration given different conditions surrounding environment, which research question study. Air temperature, an important input feature, also determined by itself, i.e., feature function prediction target. Therefore, model developed from ordinary models hard solve standard statistical or machine learning models. For process, trial-and-error algorithm proposed derive solution based on two theorems are rigorously proved; there exists unique solution, within certain range model. The its approach were constructed validated using simulated data; temperature data collected Observatory. results showed mean squared error (MSE), absolute percentage (MAPE), R2 test set 0.1378, 0.1123, 0.8182, respectively, showing performance was generally accurate. This study can help practitioners governments rationally schedules thus protect them risks brought about