Developing Optimal Road Alignment Using Geospatial Technologies: Case Of Ginchi To Chullute Of West Shoa Zone, Oromia Region

Abstract

Route selection is a critical first step in the process of design and construction of a road. Especially, route planning analysis using geospatial technology provides strong decision support in searching for optimal routes. Even though optimal route selection has been the focus of several studies, only a few efforts have been made on optimal route selection in a landslide-prone area using geospatial technology, in Ethiopia. Therefore, this study aims to develop optimal road alignment connecting Ginchi to Chullute using geospatial technology. The study employed Sentinel 2A satellite image, ALOS DEM, geological map, population data, SPOT 5, and intensive field visit to drive the major factors for optimal road alignment. Analytical Hierarchy Process (AHP) multi-criteria decision-making was used to drive the weight for the factor. Thematic maps were prepared for four cost surfaces by using weight overlay in spatial analyst tools. Four cost surfaces generated from engineering, environmental, landslide, and hybrid factors were used for modeling alternate least-cost paths. The engineering cost surface was generated from five major factors; namely, slope, stream order, lithology, lineament, and soil. AHP was used to drive weight based on importance, and cost surface was generated by using weight overlay. Likewise, the environmental cost surface was generated from three major factors, LULC, slope, and proximity to the river, and the cost surface was generated by the same procedure with engineering one. The landslide cost surface was generated by using the frequency ratio method and the results were validated by using receiver operating characteristics (ROC) and Area under the curve (AUC). The result shows an AUC of 0.756, which indicated that the model has very good performance. The hybrid cost surface was generated by an equal weight of all the three generated cost surfaces. The least-cost path function in spatial analyst tools was used for modeling the most economical path. The first cost distance and cost backlink were created by using cost surface and origin (Ginchi). Afterward, cost paths were generated by using destination (Chullute), cost distance and cost backlink has been used as input for modeling the least-cost path. The generated least-cost path has been optimized by using integrated criteria of engineering, environmental, and social perspectives. The hybrid least cost path was found to be the optimal route connecting Ginchi to chullute as it had overall advantages over the others. Besides, the selected and the remaining modeled least-cost paths were more feasible than the existing routes. The existing route is 122 km long and the proposed route is only 103.46 km. Moreover, the existing road is prone to landslides; however, the proposed route is relatively free from the landslide. The environmental, engineering and landslide-based setups (104.3, 105, and 98.3 km long, respectively) were also more feasible and less susceptible to landslide compared with the existing road. Therefore the study concludes that using integrated geospatial tools in route planning is useful. It is therefore recommended to include the technology as a planning tool for analysis of transportation and other linear structures, such as drainage, pipeline, and railway.