Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

---

Land use planning is one of the most important parts of urban and regional master planning process. In cases where the study area is large and the number of land uses and /or constraints are high, finding an optimal or near optimal solution in a reasonable time and with acceptable computational costs without applying an Artificial Intelligence based heuristic technique is nearly impossible. In this paper our first aim is to introduce a novel hybrid approach to mathematically model the land use planning problem. Secondly we introduce a new hybrid algorithm to solve the aforementioned mathematical model. Then we apply this approach to a real land use planning case study in which 600 hectares of undeveloped lands belonging to Persian Gulf Science and Technology Park must be allocated to 14 different land uses. In this approach the planning criteria are divided into two main categories. The first group is compensatory criteria. Also we have two non-compensatory objectives that together with the first compound objective function constitute our Multiple Objective Mathematical Programming Model. These objectives include indexes of optimizing economic objectives, traffic volume objective and many other objective functions. Finally we provide best alternative which have been chosen amongst more than 101000 alternatives generated during the process of proposed algorithm. Providing a faster process and providing a decision supporting system in spatial land-use planning make this algorithm a new approach to replace the spatial classical land-use planning approaches.

Keywords: Multiple Attribute Decision Making, genetic algorithm, Land use Suitability Analysis, Variable Neighborhood Search

Full-Text [PDF 481 kb]   (7729 Downloads)    

Add your comments about this article : Your username or Email:

---