A GIS - based Pipelining Using Fuzzy Logic and Statistical Models

Oil and gas industry is an important pat of Iran's national economy. In this paper, geospatial information system (GIS) application in optimum route finding in raster model has been considered. Effective criteria have been weighted using knowledge driven and data driven methods. Index overlay, fuzzy logic and statistical methods have been used for generating cost surface. For creating accumulated cost surface, isotropic, partially anisotropic and full anisotropic models have been investigated. Since oil and gas pipeline route finding depends on location and direction of slope which varies in every direction, full anisotropic model has been implemented. In case study, for calculating the weight of factors, a pipeline between Bidboland (in Chahar Mahale Bakhtiari Province) in central part of Iran and Farah (in Kohkoloyeh and Boyer Ahmad Province) with 25 kilometers length has been selected. Cartographic and topographic maps at a scale of 1:25000 and geology map at a scale of 1:100000 were collected. Factor layers have been weighted using knowledge and data driven methods and then integrated using a number of integrated methods. A number of evolution models including distance to existing route, calculated cost and the number of effective criteria have been tested. Using the three evaluation models, it is found that the path determined from fuzzy gamma cost surface is the best route. By error propagation method, the spatial error with 5.065 Meters has been calculated. The value of fuzzy gamma has been determined in Bidboland-Farah route. These values have been implemented in Ahvaz-Maron gas pipeline and the analysis indicated that it would be 31% cheaper than the existing pipeline route.