Truck Payload Optimization by IOT and Machine Learning
Matias Pinto, Norman B. Keevil Institute of Mining Engineering, University of British Columbia; Ilija Miskovic, Norman B. Keevil Institute of Mining Engineering, University of British Columbia
Managing the payload is a key to running safe, efficient, and profitable mining operations. Despite the relative simplicity of shovel-truck operations, they are currently not achieving optimum productivity. By overloading trucks, shovel operators can significantly affect the operational efficiency of a mobile fleet and, consequently, reduce the overall profitability of a mining project. If trucks are carrying too much payload, their structural components are under severe stresses, which increases the risk of premature failures and unplanned downtimes. Also, by increasing the wear and tear of haul road surfaces, overloaded trucks have a significant negative impact on the mine infrastructure and related maintenance costs. Modern digital and data technologies (i.e., IoT and AI), offer the potential to factor out this variability and to provide consistent cycle times and optimized shovel-truck performance. Here, we present two approaches for estimating the volume and distribution of the truck payload. The first approach is utilizing a high-performance machine vision system for imaging-based automatic inspection and analysis of the payload. The system is developed and tested on a 1/14 model of a mining truck, and results are visualized in an immersive augmented reality environment. The second approach is based on the utilization of Internet of Things and Machine Learning where data from sensors embedded to different parts of the truck are collected and streamlined to an in-house private cloud for virtualization and processing. A TensorFlow-based ML platform was then used to find correlations between the truck payload and its key performance indicators.
Truck Payload Optimization, Internet of Things, Machine Vision, Machine Learning