Stochastic forestry harvest planning under soil compaction conditions

Area of Science:

• Forestry Science
• Environmental Engineering
• Operations Research

Background:

• Soil compaction from heavy machinery poses ecological and economic risks in forestry.
• Current planning methods struggle to account for compaction risks, leading to operational disruptions and increased costs.
• Understanding the soil's hydrological balance is crucial for assessing compaction risk.

Purpose of the Study:

• To develop an advanced planning model for annual forestry harvesting that incorporates soil compaction risk.
• To improve the accuracy and efficiency of forestry planning by analyzing hydrological balances and rainfall uncertainty.
• To present a novel stochastic model that outperforms traditional deterministic approaches.

Main Methods:

• Development of a stochastic model analyzing monthly and biweekly hydrological balances (rainfall and evapotranspiration).
• Incorporation of rainfall regime uncertainty through scenario analysis.
• Application of the Progressive Hedging method to manage computational complexity in large-scale scenario problems.

Main Results:

• The proposed stochastic model provides superior results compared to existing deterministic planning methods.
• A biweekly model formulation offers a more dynamic system representation, enhancing planning accuracy.
• The Progressive Hedging method effectively decomposes the problem, enabling high-quality solutions within practical timeframes.

Conclusions:

• Stochastic modeling of hydrological balances is essential for effective forestry harvesting planning under soil compaction risks.
• The developed model offers a robust and efficient approach for optimizing forestry operations.
• This research provides a valuable tool for sustainable and economically viable forest management.