The iron giant has thrown some serious mathematical thinking at its rail maintenance programs to make sure the tracks are being optimally maintained. It is also searching for 200 additional train drivers to add to its fleet.

BHP WA iron ore asset president Brandon Craig told the Austmine 2021 Conference that the company would look to employ the drivers as trainees to "address this critical skills shortage in WA".

It will likely take some time for BHP to train up those 200 additional train drivers, even if they have some previous train driving experience. As a rule of thumb it takes about one year to train a metropolitan commuter train driver to a standard where they can manage a 250-plus carriage train.

It is not so much the size of the train that is the problem but rather the terrain it needs to traverse.

Going through the Pilbara ranges with fully loaded carriages requires a deft hand on the throttle. Too much or too little can lead to a decoupling leaving half the train heading to port and the other half rolling down the opposite side of the range back towards the mine it came from.

"Our train drivers are a vital part of our pit-to-port value chain and the new traineeships will create rewarding careers for those who choose to join us," Craig said.

"This program will help us strengthen our business while advancing critical skills within WA."

BHP, like most other Pilbara iron ore miners, has previously ruled out automating its trains. They all argue that the benefit gained from the investment needed for such an automation effort would not stack up.

Rio Tinto has been the outlier, spending about $1 billion to create its Autohaul autonomous train system.

However, Rio Tinot has the longest hauls in the Pilbara. Fatigue management guidelines meant train drivers would often have to be changed over mid haul. With Autohaul system that is no longer an issue.

What BHP has automated in its rail system is some of the decision making around its track maintenance.

"Across virtually every railroad maintenance program you will find rail grinding is the cornerstone of maintaining the integrity of these assets," Craig said.

"You need to grind the track to maintain optimum rail performance and rail life. If you grind too much you consume too much track time and track life. If you grind too little you risk defects on the track, which creates a safety and reliability issue.

"Grinding the track accounts for about 30% of all track possession activities - the highest amount of hours that takes our rail systems offline.

"How we make decisions around grinding determines the optimal level of performance through the railway system.

"There are many factors that come into play when you plan this work: the size and complexity of the network; the grinding machine availability; and constraints such as crew size, shutdown work, and even extreme weather conditions.

"Our team took this physical challenge and translated it into a mathematical model that found new solutions.

"Through this they were able to adjust the grinding plan - providing a 45% reduction in our annual grinding hours and about 94% increased compliance."

BHP's data scientists took a similar approach to optimising BHP's car dumpers at Port Hedland.

The logistics at Port Hedland are complex covering five car dumpers, eight stackers, five reclaimers, three lump rescreening plants and eight ship loaders, all connected by more than 99 different conveyor routes.

Working with the operations team the data scientists and mathematicians used algorithms to identify the high and low performing routes at the car dumpers.

This reduced dump times and positively impacted vessel line ups.

"We were able to lift outflow capacity by 1.4 million tonnes simply by optimising the way we select our routs to transport iron ore, with no capital investment required," Craig said.