Implementation of genetic algorithm in GIS for site selection Fatima Mehdipour MSC student of GIS Faculty of Geodesy and Geomatics Eng. K.N.Toosi University of Technology, Iran Email: mehdipour_fa@yahoo.com Mohammad S. Mesgari Assistant professor Faculty of Geodesy and Geomatics Eng. K.N.Toosi University of Technology Email: smesgari@yahoo.com In projects that establish a new activity in an area one of the first and most important question is about its optimum location. Geo-spatial information systems provide various functions and tools for the analysis and integration of different spatial data. Consequently they are often used for site selection for different purposes. Although GIS provides many useful tools, yet it is the role of GIS expert to develop proper models, algorithms and ideas using GIS to find the proper site. In this article the usability of genetic algorithm in GIS environment for site selection is elaborated. Moreover, its implementation difficulties along with its comparison with other methods are described. In general the capabilities of this algorithm in finding solutions for complex and non-structured problems is not properly studied. More specifically, the algorithm is rarely used for spatial data integration in site selection. This is probably because of the complexity of the logic of the algorithm. In this research, genetic algorithm is used for the selection of proper sites for building service areas beside roads. First, the data and analysis requirements of the application are studied. On the basis of this, the components of genetic algorithm are designed. In the next step, the required data are collected, processed and applied into the algorithm. The result is the selected areas for the establishment of service areas. Finally, the output areas are compared with the results of other methods and approaches.