Optimal forest harvesting policy using dynamic programming method (Case study: Shafarod forest)

The aim of this research is to determine the optimal harvesting using dynamic programming in Shafarod forests at Guilan province. First of all, a logistic growth function was used to predict the growth model. Then, wood price function and variable harvesting cost function were estimated based on an autoregressive model and time series data. Finally, the optimal cutting maximizing the expected present value was computed using growth, price and variable cost functions. LINGO software was used for cutting optimization. Minimum acceptable volume, growth model, wood price at forest road side, variable harvesting cost, planning horizon, interest rate, and moving of cutting groups and harvesting equipment werecoded in LINGO software. Results indicated that the optimal cutting varies between 6 to 12 m3 per hectare in various stocks during 100 years regarding to the volume average of the study area (490 m3/ha) and the minimum acceptable volume (100 m3/ha).The maximum expected present value was estimated 323326 thousands Rials during 100 years. Sensitivity analysis showed that the optimal harvest could fluctuate over the planning horizon based on the volume average of the study area and the minimum acceptable volume even which in some cases harvesting is not required.