Publicação
Models and heuristics for forest management with environmental restrictions
| Resumo: | The main focus of this thesis was to develop mathematical models and methods in integer programming for solving harvest scheduling problems with environmental restrictions. Constraints on maximum clearcut area, minimum total habitat area, minimum total core area and inter-habitat connectivity were addressed for this purpose. The research was structured in a collection of three papers, each one describing the study of a different forest harvest scheduling problem with respect to the environmental constraints. Problems of papers 1 and 2 aim at maximizing the net present value. A bi objective problem is considered in paper 3. The objectives are the maximization of the net present value and the maximization of the inter-habitat connectivity. The tree search methods branch-and-bound and multiobjective Monte Carlo tree search were designed specifically to solve the problems. The methods could be used as heuristics, as a time limit of 2 hours was imposed. All harvest scheduling problems were based on the socalled cluster formulation. The proposed models and methods were tested with sixteen real and hypothetical instances ranging from small to large. The results obtained for branch-and-bound and Monte Carlo tree search show that these methods were able to find solutions for all instances. The results suggest that it is possible to address the environmental restrictions with small reductions of the net present value. With respect to the forestry fragmentation caused by harvestings, the results suggest that, although clearcut size constraints tend to disperse clearcuts across the forest, compromising the development of large habitats, close to each other, the proposed models, with the other environmental constraints, attempt to mitigate this effect. |
|---|---|
| Autores principais: | Neto, Teresa |
| Assunto: | Teses de doutoramento - 2018 |
| Ano: | 2018 |
| País: | Portugal |
| Tipo de documento: | tese de doutoramento |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório da Universidade de Lisboa |
| Resumo: | The main focus of this thesis was to develop mathematical models and methods in integer programming for solving harvest scheduling problems with environmental restrictions. Constraints on maximum clearcut area, minimum total habitat area, minimum total core area and inter-habitat connectivity were addressed for this purpose. The research was structured in a collection of three papers, each one describing the study of a different forest harvest scheduling problem with respect to the environmental constraints. Problems of papers 1 and 2 aim at maximizing the net present value. A bi objective problem is considered in paper 3. The objectives are the maximization of the net present value and the maximization of the inter-habitat connectivity. The tree search methods branch-and-bound and multiobjective Monte Carlo tree search were designed specifically to solve the problems. The methods could be used as heuristics, as a time limit of 2 hours was imposed. All harvest scheduling problems were based on the socalled cluster formulation. The proposed models and methods were tested with sixteen real and hypothetical instances ranging from small to large. The results obtained for branch-and-bound and Monte Carlo tree search show that these methods were able to find solutions for all instances. The results suggest that it is possible to address the environmental restrictions with small reductions of the net present value. With respect to the forestry fragmentation caused by harvestings, the results suggest that, although clearcut size constraints tend to disperse clearcuts across the forest, compromising the development of large habitats, close to each other, the proposed models, with the other environmental constraints, attempt to mitigate this effect. |
|---|