Optimisation of Forest Road Investments and the Roundwood Supply Chain

Abstract: In a pre-study, concerning roundwood freshness, performed by the Forestry Research Institute of Sweden during the spring of 2000, the quality on the forest road network was identified as a major problem for the Swedish forest sector. These problems mostly occur during spring thaw, but also when rains give muddy gravel roads. The Swedish forest company Holmen Skog decided to support the PhD project presented here with harvest and road data from a forest district, and expert knowledge of road investments.   Upgrading of forest roads has been studied in a major part of this thesis, with two case studies at a forest district of average size located around Sveg in Härjedalen. Deterministic scenario analysis has been used and compared to a two-stage stochastic model. Cost functions, assortment mixtures and scenarios have been analysed, and results have been evaluated together with the staff at Holmen Skog. These case studies indicate that the problem can swiftly be solved, at least at average size forest districts.   In the last paper a model of the roundwood supply chain is developed and tested via a sample problem. The problem is restricted to the supply chain with fixed harvesting and fixed demand at the mills. The considered uncertainty is load-bearing capacity on the road network, which depends on the water content in the road body and the bearing layer quality.   In contrast to the road investment which are long-term decisions, this planning problem is short-term, and in our paper between half a year and one year. One difficult part in this model is the estimation of storage costs. This study, therefore, puts forward the need of measurements of storage costs. Otherwise, it is not possible to optimise the roundwood supply chain in a consistent way. Storage costs should, for that reason, be analysed in the future.