Automatic Recognition of Construction Labor Activity: A Machine Learning Approach to Capture Activity Physiological Patterns Using Wearable Sensors

The first step toward developing an automated construction workers performance monitoring system is to initially establish a complete and competent activity recognition solution, which is still lacking. This research proposes a novel approach to recognize workers’ activity based on their physiological signals (i.e. Blood Volume Pulse (BVP), Respiration Rate (RR), Heart Rate (HR), Galvanic Skin Response (GSR) and Skin Temperature (TEMP)), broadcasted from wearable sensors to a tablet application developed for this particular purpose. The proposed approach will utilise a Machine Learning classifier that captures physiological patterns to determine whether a worker is idle or not and determine the type of activity the worker is performing (e.g., casting concrete or cleaning moulds) on a real-time basis. A pilot test of this method conducted against three light-stone factory workers throughout three full working shifts showed a promising result of up to 88% accuracy level for activity-type recognition. The proposed method complements previously proposed labor tracking methods that focused on monitoring labor trajectories and postures, by using additional rich source of information from labors physiology, for real-time and remote activity recognition. Ultimately, this paves for an automated and comprehensive solution with which construction managers could monitor, control and collect rich real-time data about workers performance remotely.