Adaptive Strategies for The Open-Pit Mine Optimal Scheduling Problem

TL;DR

This paper introduces adaptive index strategies for open-pit mine scheduling, providing bounds for the net present value and a theoretical framework for handling uncertainty and learning in the optimization process.

Contribution

It presents a novel dynamical optimization approach and adaptive strategies for open-pit mine scheduling, including a theoretical framework for uncertainty and learning.

Findings

Index strategies provide bounds for NPV.

Numerical results demonstrate strategy effectiveness.

Framework addresses uncertainty in mine planning.

Abstract

Within the mining discipline, mine planning is the component that studies how to transform the information about the ore resources into value for the owner. For open-pit mines, an optimal block scheduling maximizes the discounted value of the extracted blocks (period by period), called the net present value (NPV). However, to be feasible, a mine schedule must respect the slope constraints. The optimal open-pit block scheduling problem (OPBSP) consists, therefore, in finding such an optimal schedule. On the one hand, we introduce the dynamical optimization approach to mine scheduling in the deterministic case, and we propose a class of (suboptimal) adaptive strategies, the so-called index strategies. We show that they provide upper and lower bounds for the NPV, and we provide numerical results. On the other hand, we introduce a theoretical framework for OPBSP under uncertainty and learning.