[chuckles]

Typically how it starts is there's this really long problem elicitation phase where that OR expert has to sit down with the mining engineer, understand what iron ore is, how does it go through the process, all of those details.

This is typically like a super, super slow process because, you know, there's a lot of expertise that goes into iron ore smelting that you need to kind of extract.

Once you have that, you also have to get the associated data, right? Like optimization problems run on data, whether that be like constraints relating to how much iron ore you can get through the system, you know, what are the demand distributions, all of that.

So that takes a lot of work too.

And once you have those kind of an understanding of the decision-making problem and the data needed into it, you can then go constructing your OR model.

So we call this like model development.

... not your actual, your forecast.