In this column, Aldo Barreto—Founding Partner & Co-CEO of Unilink— reflects on the natural limits of planning in mining supply, even in highly mature operations.

With more than 13 years leading Unilink and a previous career spanning 11 years at SAP Ariba and Quadrem, Aldo puts forward a key idea: not all demand can be anticipated or structured, and recognizing this, accept it and manage it, is key and essential for evolving towards more resilient supply models. 

Traditional model vs. Modern supply model

In recent years, mining has made significant progress in professionalizing and structuring its procurement processes. Robust ERP systems, materials engineering, supply agreements, catalogs, demand planning, and optimization models are now standard in many operations.

That effort has yielded concrete results.

But even in the most advanced environments, there remains a reality that is often uncomfortable to acknowledge: not all demand can be planned or anticipated, two concepts that are closely linked but not the same. 

When planning is no longer sufficient

The traditional supply model is designed for the predictable:

• Recurring expenses

• High turnover items

• Stable patterns

• Sufficient volumes to justify agreements

That model works well. And it will continue to be fundamental. However, actual mining operations are more complex than any theoretical model. In practice, even with mature materials engineering, the following issues always arise:

• Very low frequency items

• Sporadic lawsuits

• New or non-standard requirements

• Modified or discontinued equipment

• Minor projects that do not justify contracts

• Consumption that varies depending on operating conditions

This variability is not an anomaly. It is a structural part of trading complex assets.

The mistake of trying (or forcing) to structure everything

Faced with this reality, many organizations fall into an understandable temptation: trying (forcing) to bring 100% of consumption into a structured model.

But that approach has clear limitations. For part of the demand, the cost of:

• Model the pattern

• Create a code

• Bid

• Negotiate an agreement

• Keep it current

ends up being greater than the profit obtained.  The effort, far outweighs the benefits. 

Not because management is deficient, but because not all demand justifies that level of structuring and effort.

Here's a key point: Supply efficiency isn't measured by how many items are under contract, but by how well the part that isn't is managed.

It is important to state this explicitly: the existence of unstructured demand is not a failureof the supply,operations, or maintenance areas. It is a natural consequence of operating in changing environments, with equipment that ages, adapts, and changes over time.

In fact, the most mature operations are not those that eliminate thisvariability—because they understandthat this is impossible in practice—but those that recognize it and manage it without friction. 

There is a significant conceptual change here. 

There is a significant conceptual change here.

A more realistic approach to MRO consumption

If we look at MRO consumption honestly, it is usually divided into three main areas:

1. Predictable and recurring demand, where agreements work very well.

2. Intermediate demand, which is assessed on a case-by-case basis.

3. Variable or low-frequency demand, where structuring is inefficient.

The problem arises when that third area—which always exists—does not have a clear and efficient channel for management. That's when emails, poorly resolved emergencies, manual searches, and excessive reliance on informal solutions arise.

Not because of a lack of planning, but because planning has a natural limit, and anticipation, or rather the lack of anticipation, is not a deficiency, but rather a consequence of working in complex and multivariable environments. 

The real challenge of modern procurement

The relevant question today is not: How do we eliminate unstructured demand? But rather: How do we manage it efficiently, transparently, and reliably?

Accepting and understanding that limit does not weaken the supply. It strengthens it.

Conclusion: look at reality as it is

In mining, variability is no exception. It is part of the system. Recognizing and understand it does not mean giving up on planning. or a lack of foresight, but rather complementing it with mechanisms that reflect operational reality. 

At Unilink, we work in this area: facilitating market access for unstructured demand, understanding that this is a natural—and permanent—part of mining supply.

Because not everything must (or can) be under contract.

And that's not a problem.