Publications / 2020 Proceedings of the 37th ISARC, Kitakyushu, Japan
Excavators and trucks are important equipment for earthmoving work, which have major contributions to construction productivity. In order to control the work efficiency and productivity of earthmoving equipment, computer-vision (CV) methods have been proposed to monitor equipment operations from site surveillance videos. Existing methods can recognize equipment activities to estimate the working time and idling time; however, they are limited in analyzing the reasons behind the equipment idling and low productivity. Therefore, this research proposes a method to identify the main reasons that cause excavators and trucks idling by analyzing their interactive operations. In this method, the relationships between the excavator and the surrounding truck(s) in each group are analyzed to identify the potential reasons that cause the excavator's idling. The proposed method was validated with a video from construction site and the test results showed its effectiveness and efficiency.