A. Alonso Ayuso, L. Escudero, M. Guignard, A. Weintraub
The forest-harvesting and road-construction planning problem basically consists of managing land designated for timber production and divided into harvest cells. For each time period in the given time horizon one must decide which cells to cut and what access roads to build in order to maximize expected net profit under a risk manageable scheme to control the negative impact of the solutions of the unwanted scenarios (i.e., the so-called black swans) on the objective function value. A MIP stochastic optimization model is presented; this model includes different risk-averse measures that avoid the risk neutral optimal (expected profit) solutions with a high variability in the scenario profit. Those risk-averse measures are compared computationally with the risk neutral one under different price, demand and probability of the scenarios under consideration, as well as with a risk-averse measure that is a mixture of both.
Palabras clave: Forestry, stochastic optimization, risk management, CVaR, time consistency
Programado
X05.5 Grupo de Localización
7 de septiembre de 2016 12:30
Aula 21.06
