Optimization of Crops Allocation Planning in Cianjur Involving Water Cost Constraints Using Genetic Algorithm

To become urban agriculture, Cianjur needs to create land-use planning base on its agricultural situation. In this work, an optimization model to compute optimal crop allocations for each sub-district is proposed and solved by a genetic algorithm (GA) to find the best solutions. The land utility is formulated as the margin between the possible cost and revenue base on the data collected from the Indonesian Central Bureau of Statistics data. The result analysis can support decision-making for allocation crop production planning in practice to increase agricultural revenue in the Cianjur regency. A genetic algorithm is used to find the optimal solution that maximizes the margin function with some realistic conditions obtained from the government's allowed land restriction. The computation results show that the highest optimal value is 30.65 on the log scale. The fitness constant to this asymptotic rise is after 17 generations, so that it converges quickly for some experimental GA parameter setting. In the optimal allocation results, the sweet potato gains the smallest portion, followed by cassava that is only around 2% of the total plantation area. Meanwhile, the largest portion of land allocation is suggested to be corn, based on the optimal solution returning the maximum profit margin. For all crops, the Cidaun sub-district appears to be the most expansive land area in Cianjur.