Multiple-use management of forests often requires imposition of spatial constraints on the selection of units for harvest. To satisfy such constraints, harvest units must be treated as integral units. A biased sampling search technique is used to find integer solutions to operationally sized problems. Solutions found for the sample problems are within 8% of the upper bound of the corresponding linear programming solution and less than 4% below the upper bound on the true optimum as defined by a confidence interval estimator.
No takes yet. Share an insight, caveat, or question.
O'Hara et al. (1989) studied this question.