Details

Autor(en) / Beteiligte

• Gu, Shuning
• Zhao, Wenqi
• Qin, Xiao

Titel

Evaluation & Decision Models for Forest Management

Ist Teil von

• Proceedings of the 10th International Conference on Communications and Broadband Networking, 2022, p.66-70

Ort / Verlag

New York, NY, USA: ACM

Erscheinungsjahr

2022

Quelle

ACM Digital Library

Beschreibungen/Notizen

• This paper introduces a system for forecasting carbon sequestration and helping make deforestation management. Based on Principal Component Analysis data optimization, we construct Back Propagation Neural Network to predict the value of carbon sequestration under certain conditions. Thus, we can use Simulated Annealing Algorithm to make plans for CO2 sequestration. For sake of the situation of, where a nature-friendly management plan may not be adopted by government considering it posing brakes on social development, we need to determine whether the plan deviate from the actual situation. Long Short-Term Memory Model is trained to fit the future tree loss according to past dataset. Since there're evidence to determine the applicability of our decision-making model. Dynamic Programming maximizes global benefits in one state within a carbon emission limit. Specifically, one state has its carbon emission quota and will allot some units of emission to counties under its jurisdiction. The state would earn economic development in reward for its allotting, while counties’ revenue function is discrete and different. Based on this problem abstraction, we apply Dynamic Programming method to seek the approximate optimal decision for forest management. We state the applicability of the models and talk about where our models won't fit well.