A GIS-based linear programming model for optimizing agricultural land levelling

Land levelling is an important process of the agricultural land consolidation projects. The land levelling process starts by determining the appropriate levels of the terrain and solves the earthwork allocation for the predetermined terrain levels. The recent attempt, reported in the literature, to find an optimal process is a GIS-based approach which solves the transportation problem by determing the designed terrain level and optimal hauling distances that result in the lowest hauling cost. In this study, the Vogel approximation method (VAM) was adopted in initializing the solution to the transportation problems, and the optimization of the solutions obtained by the VAM was examined using the modified distribution method (MODI). To verify the reliability of this combined method, a case study was carried out based on the proposed land levelling project at the Lixian District of Beijing, China. The results show that the optimal average hauling distance calculated from the proposed linear programming model is 79.53m shorter than that estimated using the empirical method, and the unit cost calculated from the linear programming model is 33.4% lower than the experienced cost. The improvements in the optimization of the hauling distance show that the cost of the agricultural land levelling can be efficiently lowered if an appropriate linear programming model is used.