Using linear programming to optimize rice yields in Battambang, Cambodia

Abstract

Cambodian farmers have been applying traditional techniques for selecting cropping patterns of rice production for generations. Unfortunately, current cropping patterns provide low yields. For the producing maximum rice yields, land resources need to be managed efficiently. Optimization techniques are often used for resource management when there is limited availability of land, capital, labor, and inputs; such as fertilizers and seeds. This study considers how to optimize the use of land resources to produce optimal rice yields. This problem was formulated for three varieties of rice: sen kro ob (fragrant rice), phka romdoul (premium fragrant rice), and IR504 (white rice). A linear programming method was then used to model and determine an optimal solution. The results from this model demonstrate an ideal cropping pattern for the three varieties of rice aimed at achieving a maximum rice yield in the study area of Battambang. A sensitivity analysis was performed on these results by varying the average rice yield of each type of rice in the model. All things being equal, if the average yield of sen kro ob was changed by