Improved Solution Techniques for Multi-Period Area-Based Forest Harvest
Scheduling Problems
David M Ryan
Department of Engineering Science, University of Auckland, Private Bag
92019, Auckland, New Zealand (Email: [email protected])
Juan Pablo Vielma
Departamento de Ingenieria Industrial, Universidad de Chile, Casilla 2777, Santiago, Chile (Email: [email protected])
Alan Murray
Department of Geography, The Ohio State University, 1036 Derby Hall, 154 North Oval Mall, Columbus, Ohio 43210, USA (Email: [email protected])
Andres Weintraub
Departamento de Ingenieria Industrial, Universidad de Chile, Casilla 2777, Santiago, Chile (Email: [email protected].)
Area-based forest harvest scheduling models, where management decisions are made for relatively small units subject to a maximum harvest area restriction, are known to be very difficult to solve by exact techniques. Previous research has developed good approaches for solving small and medium sized forestry applications based on projecting the problem onto a cluster graph for which cliques can be applied. However, as multiple time periods become of interest, current approaches encounter difficulties preventing successful identification of optimal solutions. In this paper we present an approach for elasticizing timber demand constraints, which lends itself to an efficient solution technique. It is also possible using this approach to examine trade-offs between objective value performance and maintaining demand constraints.