Our modelling software has optimised Rio Tinto’s West Australian iron ore operations.
An efficient supply and transport planning system
In the Pilbara region of northern Western Australia, Rio Tinto operates Australia's largest private railway – over 1500 kilometres of rail lines that carry some 237 million tonnes of iron ore. Scheduling train movements on the railway is a complex task, with up to 30 trains of high and low grade iron ore moving between 14 mines and three port terminals every day.
To further complicate matters, the two grades of ore are dumped into either live stockpiles, which are part of the production line, or bulk stockpiles, which are used for buffering and storage. At both mines and ports, bulk-in operations move iron ore from live to bulk stockpiles and bulk-out operations move the material in the opposite direction.
The ore in the mine's live stockpiles is transported by rail to the port, where it is dumped onto the port's live stockpiles by rail car dumpers. It is then mixed to achieve the desired grade before shiploaders put the ore onto the ships for export.
The challenge is to create an efficient planning system that optimises train movements and minimises the ore in bulk stockpiles and movements between stockpiles, while ensuring the correct ore types are on hand for mixing to customers' specifications at the terminal.
Working with Rio Tinto and their planners, we developed software to optimise the ore transport and availability between the mines, railway and port terminals.
Prior to this, Rio Tinto carried out the planning manually. A specialist planner armed with a raft of data, a spreadsheet and considerable experience would take five to six hours to arrive at a workable solution for a month's movements. This was no mean feat, as when we analysed the data provided by Rio Tinto, we found there were some 80,000 variables and 50,000 to 60,000 constraints.
The project took six months and delivered not only a solution that reduced planning time, but one that also enabled Rio Tinto's planners to conduct 'what if' scenario evaluations to ensure they had the optimal answer.
Our tool has reduced planning time from five to six hours to as little as 15 minutes and, compared to the manual approach, has resulted in a transport increase of more than 500 kilotonnes of iron ore over an 18-month planning horizon.
Based on the success of this project, we're working with Rio Tinto to create models to optimise other areas of their operations.
This project was a finalist for Rio Tinto's Terry Palmer Innovation Award in 2012. The award acknowledges projects for exceptional innovation and business value delivered within the Rio Tinto Iron Ore Group.
This research was also a finalist for the Association of European Operational Research Societies, Excellence in Practice Award for 2013. The award recognises outstanding accomplishments in operational research.
Do business with us to help your organisation thrive
We partner with small and large companies, government and industry in Australia and around the world.