Cornering Speed Analysis

Taking Haul Road Modelling to the Next Level

In mining, even the smallest details can make or break operational efficiency. One area that has long been overlooked? Corners and intersections on haul roads.

• What is this project?

Haul road models generally lack nuance, as roads move frequently, and are often represented as straight regardless of their actual shape, leading to digital models poorly representing or lagging behind the actual roads.

This project created an algorithm that leverages machine learning techniques and the data available on mining sites to automatically detect and measure the shape of haul roads, specifically to improve the modelling of corners and turns.

This enables:

✅ More precise speed and travel time estimates on curved segments.

✅ A realistic understanding of haulage performance versus what is actually achievable on the ground.

✅ Automated road shape modelling without the endless manual adjustments.

• Impact of project to mining industry

Cornering speeds and intersections can be automatically identified and accounted for in design time calculations, allowing for more efficient and effective planning and maintenance processes.

Haulage offers great potential for improved efficiency within mining sites, and an accurate view of the underlying haulage network is vital for those improvements to be made most effectively.

• Results it brought / any use case

On testing data design times/speeds were up to 30% more accurate on curved roads than previous estimates.

Road representations can be automatically aligned with reality, the need for manual adjustment/manipulation can be greatly reduced.

The Future of Haulage Performance

This is more than just a technical improvement — it’s a step toward smarter, safer, and more sustainable mining operations. By using ML driven modelling, we turn guesswork into actionable insight, unlocking real value across mine sites.