Risk-aware scheduling for post-wildfire salvage logging under inventory estimation uncertainty

Abstract

In post-wildfire salvage logging, accurately estimating the available inventory of harvestable timber is subject to high uncertainty. This paper integrates chance constraints into a time-indexed mixed-integer programming framework to address these inventory estimation errors. By leveraging a grid-based linearized Chernoff-bound approximation for the model’s chance constraints, the proposed approach ensures numerical stability and robust scheduling, minimizing the risks associated with the overestimation or underestimation of salvageable wood.