Single Scenario Planning

My daughter’s birthday is fast approaching, so it’s time to plan a party. A nice gathering at the park sounds great, but what if it rains? We could have a backup location indoors, but that would change the party games that we could play. We’d also have to account for the extra time it might take people to get to the alternate location, and the risk that some people may no longer want to attend.

“When you assume…”

Mine planning is just like this (except with less cake). The problem is that most of us plan by making decisions until our objective is met, and then we stop and move on to the next job. There is often no thought of contingency plans, no consideration of what may or may not go wrong, or a very limited defined strategy for risk management. These are deemed either as unimportant, or as unworthy of the additional time required to consider alternatives to our initial assumptions. How do we know if the assumptions and choices made lead to an overly-optimistic plan that will fail if conditions worsen?

How do we know if they lead to a pessimistic plan that might be missing value if conditions improve? How do we choose a plan that avoids these extremes and mitigates risk?

“What If” Scenario Planning

Multiple Scenarios

The best way to ensure that we understand the various scenarios is to investigate them. While we can’t look at every possible scenario, we can experiment with various critical variables. If we are unsure of market/commodity prices, we might consider the lowest and highest possible prices, or perhaps a transition from the current price to the lowest or highest over a number of years. If there are cycles in the market, we may model some example cycles.

In addition to market conditions, we may wish to experiment with other variables. We should consider anything that would have a significant impact on our costs, revenue, or ability to meet constraints. For example, we can account for grade variability by running scenarios for more than one conditionally simulated model. Budget policy or restrictions and capital limitations can also lead to different scenarios with radically different results.

By running multiple scenarios that cover our most critical variables, we can determine which scenarios have the largest impact and, importantly, which scenarios introduce the highest risk to our plans.

Best Case vs Worst Case

After even a little experimentation, we should have a reasonable idea of what our best and worst case scenarios are. In most cases, we don’t want to base our plan on either of these scenarios. The worst case is usually too pessimistic, and doesn’t allow for opportunity to make a real profit. The best case is usually too risky, and — while profitable — doesn’t leave much room for recovery when conditions are anything but optimal.

So, once we have plans for different scenarios, how do we choose which plan to follow? When we compare their profitability, we should see that some plans fit within the bounds of all our investigated scenarios. These plans are preferable, as they are more robust and risk-averse. Plans that produce extreme results tend to fall outside the bounds of other scenarios, and are therefore the most vulnerable to changes in circumstance.

Once we select a candidate plan, we should evaluate it against alternative scenarios and then calculate its profitability if costs rise or commodity prices fall. In the next step we should determine whether it is still possible to meet constraints if the mill doesn’t perform as well as expected. Our ideal plan should be as resilient to these variables as possible, or it should be adaptable to alternative plans without high risk of failure.

Your Case vs Reality

After selecting and following a plan for a while, we will gain new knowledge and information. Perhaps the market has fallen, costs have gone up, or our grade control shows that our orebody isn’t as originally estimated. If we have selected the right plan, based on consideration of these possible scenarios, we should be in a reasonable position to re-plan for the new circumstances.

The cycle of accepting new information and replanning will not only mitigate worst case scenarios, but also allow us to capitalise on narrowed possibilities. By incorporating new information, we can determine what scenarios are most likely as the project develops. Scenarios — both good and bad — will become more or less likely. At each iteration, we can better predict the future. This leads to better planning, and a process of continuous improvement.

Experiment, Evaluate, Mitigate

At each iteration of planning, the only real way to find the perfect plan is to experiment until we understand how all of the critical variables will affect it. The more scenarios we can evaluate, the more we will know and the better equipped we will be to mitigate the inherent risk of our project.

To include more scenarios, we need to establish a process for evaluating them quickly, as manually planning multiple scenarios is simply not a viable or effective option. Optimization, on the other hand, allows us to configure the project and produce plans and schedules quickly. By modifying the parameters of these optimizations, we’re able to very quickly get results for each of the scenarios that we’re interested in. For pit optimization, this might be ranges of values for costs and revenues, along with multiple models resulting from conditional simulation. For schedule optimization, it might be different values for costs, revenues, or constraints.

By using optimization to increase the rate at which we generate plans, we are able to consider more scenarios, and do so more often. This leads to a more informed planning process, which will result in a more effective plan, and a more successful business.

Did you know that the latest version of Minemax Planner includes multivariate risk analysis?

Download a demo of Minemax Planner and try it out for yourself. Are you interested in running various scenarios for your strategic scheduling? Minemax Scheduler provides a framework for modelling your mine, and allows you to quickly optimize updated scenarios as you respond to changing circumstances.