Heat-related illness (HRI) is a common occupational injury, especially among workers in the construction industry. Methods need to be developed to predict and minimize HRI risk. The objective of this study was to investigate whether differences in heart rate (HR) can be used to predict HRI risk. We surveyed construction workers during the period from May 1 to October 30, 2020. The physiological data of 79 workers were recorded during their working hours for a period of 3163 person-time. The resting HR was defined as the lowest reading taken within the first hour of wearing the sensor, and HRI risk was determined using the following formula: 180 – (0.65 × age). The rate of increase in HR from rest was calculated using the following formula: (maximum HR – resting HR)/resting HR. On these 3163 person-time, HRI risk was observed at 368 person-time (11.6%). After analyzing the receiver operating characteristics curve, the cut-off value was 132.9%, with a sensitivity of 75.5% and specificity of 85.0%. Our results showed that the difference in HR from rest is a good indicator for predicting HRI risk. Furthermore, continuous physiological monitoring using a wearable sensor can aid in detecting early signs of HRI risk.