Published Date
Abstract
The purpose of forest planning is to support forestry decision-making by suggesting management alternatives, providing information about their consequences, and helping the decision maker to rank the alternatives. In multi-objective forest planning, forest plans are evaluated using various multiple criteria decision support methods and multi-objective optimisation algorithms. Multiple criteria comparison methods help to systematise subjective evaluations whereas multi; objective optimisation seeks the best plan among a huge number of alternatives using automated computer-based search methods. The ranking of alternatives depends on the preferences of the decision maker, both in multiple criteria comparison and in multi-objective optimisation. A careful analysis of preferences is an important step of any multi-objective planning case. The quantitative approach to decision-making suggests that a specific planning model be developed for every planning situation. This model is then solved, the result being a candidate plan that must pass various post-optimisation tests and analyses. There are several ways to prepare a multi-objective planning model, based on linear programming, goal programming, penalty functions or multi-attribute utility theory. The planning model may be solved using mathematical programming techniques or various heuristics. The use of heuristic optimisation has gained popularity in forest planning along the increasing importance of ecological forest management goals, which are often described with spatial variables. Examples of heuristics available to multi-objective forest planning are random ascent heuristics, simulated annealing, tabu search and genetic algorithm. Practical forest plans are produced in a computerised system, which includes subsystems for data management, simulation of stand development, planning model generation and optimisation, and subjective evaluation of alternative plans.
References
For further details log on website :
http://link.springer.com/chapter/10.1007/978-94-015-9906-1_1
Volume 6 of the series Managing Forest Ecosystems pp 1-19
Title
Introduction to Multi-Objective Forest Planning
- Author
- Timo Pukkala
Abstract
The purpose of forest planning is to support forestry decision-making by suggesting management alternatives, providing information about their consequences, and helping the decision maker to rank the alternatives. In multi-objective forest planning, forest plans are evaluated using various multiple criteria decision support methods and multi-objective optimisation algorithms. Multiple criteria comparison methods help to systematise subjective evaluations whereas multi; objective optimisation seeks the best plan among a huge number of alternatives using automated computer-based search methods. The ranking of alternatives depends on the preferences of the decision maker, both in multiple criteria comparison and in multi-objective optimisation. A careful analysis of preferences is an important step of any multi-objective planning case. The quantitative approach to decision-making suggests that a specific planning model be developed for every planning situation. This model is then solved, the result being a candidate plan that must pass various post-optimisation tests and analyses. There are several ways to prepare a multi-objective planning model, based on linear programming, goal programming, penalty functions or multi-attribute utility theory. The planning model may be solved using mathematical programming techniques or various heuristics. The use of heuristic optimisation has gained popularity in forest planning along the increasing importance of ecological forest management goals, which are often described with spatial variables. Examples of heuristics available to multi-objective forest planning are random ascent heuristics, simulated annealing, tabu search and genetic algorithm. Practical forest plans are produced in a computerised system, which includes subsystems for data management, simulation of stand development, planning model generation and optimisation, and subjective evaluation of alternative plans.
References
- Boston, K. and Bettinger, P. 1999. An analysis of Monte Carlo integer programming, simulated annealing and tabu search for solving spatial harvest scheduling problems. Forest Science 45: 292–301.
- Boston, K. and Bettinger, P. 2001. Development of spatially feasible forest plans: a comparison of two modelling approaches. Silva Fennica 35 (4): 425–435.
- Dykstra, D. 1984. Mathematical programming for natural resource management. McGraw-Hill, Inc. 318 P.
- Falcäo, A. and Borges, J. 2002. Combining random and systematic search heuristic procedures for solving spatially constrained forest management scheduling problems. Forest Science 48: 608–621.
- Kangas, J. and Pukkala, T. 1992. A decision theoretic approach to goal programming problem formulation: an example on integrated forest management. Silva Fennica 26 (3): 169–176
- Kangas, J., Loikkanen, T., Pukkala, T. and Pykäläinen, J. 1996. A participatory approach to tactical forest planning. Acta Forestalia Fennica 251. 24 p.
- Öhman, K. and Eriksson, L.O. 2002. Allowing for spatial consideration in long-term forest planning by linking linear programming with simulated annealing. Forest Ecology and Management 161: 22 1230.
- Pukkala, T. and Kangas, J. 1993. A heuristic optimisation method for forest planning and decision-making. Scandinavin Journal of Forest Research 8: 560–570.CrossRef
- Pukkala, T., Nuutinen, T. and Kangas, J. 1995. Integrating scenic and recreational amenities into numerical forest planning. Landscape and Urban Planning 32: 185–195.CrossRef
- Reeves, C. R. (ed.) 1993. Modern heuristic techniques for combinatorial problems. John Wiley and Sons, Inc. 320 p.
- Render, B. and Stair, R.M. 1992. Introduction to Management Science. Allyn and Bacon. 856 p.
- Saaty, R. 1980. The Analytic Hierarchy Process. Planning, priority setting, resource allocation. McGraw-Hill Publishing Company. 283 p.
For further details log on website :
http://link.springer.com/chapter/10.1007/978-94-015-9906-1_1
No comments:
Post a Comment