Forest road network design using a trade-off analysis between skidding and road construction costs

Abstract

Designing forest road networks in a large forest land is a challenging task because many feasible alternatives exist and need to be analyzed. To provide field managers with an analytical tool that can create and analyze alternative road networks, we have developed a road network optimization model. The model formulates a large network problem in which links represent two timber transportation options from evenly distributed timber locations: on-road transportation via new roads and off-road transportation using skidders. A heuristic network algorithm is employed to solve the network problem and identify cost-efficient road networks for timber harvesting under given cost parameters. To demonstrate our model, we applied it to a 4760 ha forest in the upper part of the Mica Creek watershed in Idaho owned by Potlatch Forest Holdings, Inc. The sensitivity analyses were conducted to verify the model’s performance under various cost and volume settings. The model-generated road network was compared with a road network proposed by experienced forest engineers in Potlatch. The sensitivity analyses confirm that the model appropriately responds to changes in input parameters. Comparisons between the model output and the manually designed road network indicate that the model tends to develop a tree-shape road network to evenly cover the entire management area.