Forest planning models have increased in size and complexity as planners address a growing array of economic, ecological, and societal issues. Hierarchical production models offer a means of better managing these large and complex models. Hierarchical production planning models decompose large models into a set of smaller linked models. For example, in this paper, a Lagrangian relaxation formulation and a modified Dantzig–Wolfe decomposition – column generation routine are used to solve a hierarchical forest planning model that maximizes the net present value of harvest incomes while recognizing specific geographical units that are subject to harvest flow and green-up constraints. This allows the planning model to consider forest-wide constraints such as harvest flow, as well as address separate subproblems for each contiguous management zone for which detailed spatial plans are computed. The approach taken in this paper is different from past approaches in forest hierarchical planning because we start with a single model and derive a hierarchical model that addresses integer subproblems using Dantzig–Wolfe decomposition. The decomposition approach is demonstrated by analyzing a set of randomly generated planning problems constructed from a large forest and land inventory data set.
A fuzzy mid-term capacity and production planning model for a manufacturing system with cloud-based capacity