Data Visualisation and Machine Learning for Truck Performance Prediction and Analysis

The Challenge

Electric haul trucks are new to Australian mining. Operators need confidence in how they perform: how far they can travel, when they need charging, and whether they are at risk of overheating. Without reliable data and prediction tools, this confidence is hard to build.

Key Findings

Supported by MRIWA M16044 project funding, Electric Power Conversions Australia (EPCA) built a system that collects and stores data from its E-777D electric haul truck and turns it into performance predictive models. Testing on newly acquired data showed the models work but perform better in some conditions than others; in particular being influenced by hot weather and partial charging. These results identify what additional data is needed to improve the model’s reliability for everyday use.

Benefits to WA

This work provides WA’s mining sector with a practical foundation for adopting electric haul trucks with confidence, through the ability to visualise and predict key operational information. It supports diesel displacement at mine sites, reduces emissions, and builds local expertise in data-driven technology for the state’s METS sector.

Financial Assistance DOI

DOI 10.71342/430042142281

Report DOI

DOI 10.71342/982904340372