Document Type : علمی - پژوهشی
Authors
1
Ph.D Candidate in Environmental Science at Dep. of Environmental Science, Natural Resource Faculty, University of Gorgan, Gorgan, Iran
2
Assistant Prof., Dep. of Environmental Science, Natural Resource Faculty, University of Gorgan, Gorgan, Iran.
3
Assistant Prof., Dep. of Electrical Engineering, Faculty of Engineering, Golestan University, Al-ghadir Blv., Gorgan, Iran
4
Assistant Prof., Dep. of Environmental Science, Natural Resource Faculty, University of Gorgan, Gorgan, Ira
5
Assistant Prof., Dep. of Industrial Engineering, Faculty of Engineering, Golestan University, Al-ghadir Blv., Gorgan, Iran
Abstract
In the fast growing world of today, land use planners frequently face situations in which various uses compete for the same piece of land. Hence, the final result heavily depends on the decision maker’s capabilities to select the best use among different conflicting land uses. Taking this approach, the present study aims at providing the best allocation solution for multiple land uses including agriculture, forest, range and development in Gorgan Township, Golestan Province of Iran with respect to minimizing allocating cost and maximizing compactness and contiguity as shape criteria of landscape metrics. To aim these objectives Linear Programming as an exact method in combination with Ant Colony as metaheuristic algorithm have been used. Since land use planning is NP-Hard problem with respect to its size (132 rows in 127 columns) and the mentioned objectives, LP-Relaxation and Branch & bound method have been used to solve it. Results indicate the superiority of the hybrid model (linear programming in combination with ant colony) to employment of each of the models separately in every objectives including allocating cost, compactness and contiguity. Additionally, comparing the results of proposed hybrid model with the results of MOLA algorithm in IDRISI shows the superiority of hybrid model against MOLA. In hybrid model cost, compactness and contiguity levels after standardization are respectively 0.03, 0.1 and 0.07 better than MOLA. Furthermore, using the proposed approach, it is possible to consider both suitability and landscape metrics or even more objectives
Keywords
)