Pemodelan Sistem Pakar Berbasis Komputasi Lunak untuk Evaluasi Kesesuaian Lahan Komoditas Serealia: Studi Kasus Kabupaten Bogor
Sitanggang, Imas Sukaesih
MetadataShow full item record
Nowadays, the threat of food shortages are happen in Indonesia. Most of crops are cereals commodities. Cereals cultivation often experience some problems in defining if the land is good or not for the plant. Expert system can help researcher and practitioners to choose the suitable kind of cereals for land. In this research, a model of fuzzy rule within fuzzy inference system using artificial neural network and genetic algorithm was built. The model was used to develop an expert system of land suitability for cereals commodity. This model implement soft computing which combine fuzzy system, artificial neural network and genetic algorithm. In this research, an expert system model of land suitability for cereals plant were built. The model implement soft computing methods for inference engine which combine fuzzy system and genetic algorithm. There are 16 parameters to define the land suitability which consists of 12 numeric parameters and 4 categorical parameters. The kind of cereals that will be used are wetland paddy and corn. Trapezoid membership functions was used here. Genetic algorithm was used for tuning the membership function of fuzzy for land suitability which is consist of very suitable (S1), quite suitable (S2), marginal suitable (S3) and not suitable (N). This research has produced a model of soft-computing-based expert system. The experimental result by entering some of test cases in the system and compared with the results of the expert show that the system was built in accordance with the results of the expert with accuracy 85% for rice and 90% for corn. Errors of land suitability class grouping occurs because the input value was on the edge of the membership function. We suggest the usage of actual data for future research, so that the output of the model appropriate to real condition. Each parameter should be weighted based on its importance.