The customer is an established miner producing direct shipping iron ore for the international market, with shipments in the tens of millions of tonnes per annum. Their supply chain consists of open cut pits, ore stockpiles, a plant for crushing, screening and grinding, multiple lump and fines mine stocks, train loadout, a shared rail network and port with multiple terminal stockpiles.

Every day the customer’s planning team must determine how to best use the current stockpiles and future production outlook to meet their upcoming market demand. With stockpiling, processing, transportation, blending, product assembly, and vessel loading decisions to be made across their supply chain to deliver the correct iron ore lump and fines grade to customers.

The operation has multiple ore source locations, resulting in the extraction of non-homogenous resources that require controlled blending to ensure product quality and consistency. Complex stockpiling operations are required at both the mine and the port. This presents a challenge to the operation as stockpiles must be accurately blended to meet the required quality for shipment. These decisions are crucial to the operation to ensure the timely delivery of on-specification product onto the vessels.

One of the key constraints in the supply chain is the efficient use of a multi-user rail network. Detailed train schedules must be created and forecasted two weeks in advance to be managed by a third-party rail haulage provider. The complicated nature of preparing stocks for timely rail transport makes the supply chain vulnerable to inefficiencies.

One area identified for improvement was the direct loading of vessels. Direct loading involves linking the port in-load and out-load streams so material travels from a train directly to a ship. It relies on multiple aspects of the supply chain working seamlessly together to have the right material delivered to the port at the right time. Without a sophisticated decision support system, planning for direct loading can be challenging, and valuable opportunities for enhancing supply chain efficiencies could be missed.

Previously, the customer coordinated its supply chain via multiple Excel spreadsheets, with separate workbooks for mine and production planning, rail planning and port planning. This siloed planning process caused a lack of visibility across the supply chain, resulting in potential loss of value.

Due to the nature of the operation, plans required revisions every day or two, depending on changes to the supply chain, such as variations in the order of vessels arriving at port. The incumbent process was inflexible and time-consuming to update, making it difficult to incorporate changes into existing plans. This meant that opportunities could not easily be exploited, and setbacks could not easily be planned around. For example, a disruption in the supply chain could lead to large stockpiles remaining at the mine, a potentially costly situation for the operation.

Furthermore, using the incumbent process, planners only had time to produce a single plan per day, with no way of knowing if the plan was optimal.

BOLT, a supply chain optimization software, was deployed to provide decision support and modeling around key supply chain factors. These included processing, stockpiling, blending, material drying time, port in-loading and out-loading, and vessel scheduling. Additionally, BOLT provides insight into where value can be gained in a supply chain whilst honoring the complex financial costs and revenue models of the operation.

Two BOLT modules were deployed - Operational and Tactical, providing decision support on a shiftly and weekly level respectively. Tactical planning ensures value destroying short term focused decisions are avoided and operational planning responds to the current system state.

By utilizing BOLT’s mathematical optimization algorithms, planners can now meet the complicated blending requirements and detailed stockpiling targets for the mine and port. After the BOLT implementation, blends are now significantly closer to product specification than they were when using the incumbent process. This has enabled the customer to produce plans that consider the required contracted tonnes without compromising on giving away too much high-grade material. BOLT also intelligently makes trade-off decisions with the offtake of stockpiles to maximize direct loading opportunities.

BOLT ensured that the right material was available for rail transport to the port, substantially improving direct loading opportunities for the operation. With improved direct loading, dependence on port infrastructure is reduced, as the requirement to stockpile product at the port is minimized. Furthermore, improved direct loading increased ore tonnes to be more efficiently loaded onto vessels and sold, therefore maximizing profits. Using BOLT, the customer was able to increase direct loading opportunities by 44%.

With the implementation of BOLT, all supply chain information is now available in a single interface. Separate, siloed plans for mine, port and rail planning are now integrated.

While the incumbent process was time-consuming and unwieldly, BOLT’s dynamic planning capabilities allow users to adapt rapidly when changes occur within the supply chain. Data can be modified, and new plans created within ten minutes, enabling the operation to quickly react and adapt to changing situations. Additionally, ‘What if?’ scenarios can be developed to better understand the impact of potential setbacks on the supply chain and to identify opportunities to leverage contract bonuses or avoid penalties.

Blends are significantly closer to product specification than they would be using a manual plan, and complex stockpiling questions are answered.

Optimized supply chain plans allow materials to be transported with increased efficiency.

Plans can be modified in a fraction of the time required to modify a manual plan, whilst producing superior results.

All information is included in one place, providing visibility across the entire supply chain.

BOLT’s Operational and Tactical modules provide decision support on a shiftly and weekly timeframe.