Sistem Pendukung Keputusan Spasial untuk Evaluasi Kesesuaian Lahan Bawang Putih

Abstract

Direktur Jenderal Hortikultura Indonesia saat ini melakukan program pembangunan berswasembada pada komoditas bawang putih. Program yang dilakukan untuk meningkatkan luas lahan bawang putih menjadi 72.249 Ha pada tahun 2019 dari semula yang hanya 23.900 Ha pada tahun 2018. Peningkatan luas lahan dikarenakan jumlah produksi bawang putih yang menurun 2,5% per tahun. Namun, jumlah tersebut belum maksimal untuk menutupi jumlah konsumsi bawang putih di Indonesia yang meningkat mencapai 10,6% per tahun. Pengambilan keputusan yang tepat dalam evaluasi kesesuaian lahan bawang putih sangat penting dalam perencanaan peningkatan luas lahan bawang putih. Keputusan yang baik akan diperoleh apabila data pendukung yang dibutuhkan terpenuhi dan menggunakan metode yang tepat. Penelitian ini bertujuan untuk mengembangkan suatu sistem pendukung keputusan spasial, yang dapat digunakan untuk melakukan evaluasi kesesuaian lahan bawang putih di Kabupaten Magetan dan Kabupaten Solok. Sistem yang dibangun menggunakan model pohon keputusan spatial decision tree. Model spatial decision tree dihasilkan pada penelitian sebelumnya dengan akurasi 94,36%, dengan jumlah aturan 66 pada Kabupaten Magetan dan menghasilkan akurasi 60,29% dengan jumlah aturan 33 pada Kabupaten Solok. Pada pengembangan sistem berbasis web ini digunakan metode iterative system development cycle. Pengembangan sistem pada penelitian ini terdiri atas empat tahapan pokok yaitu perencanaan, analisis, desain dan implementasi. Penelitian ini berhasil membangun sistem pendukung keputusan spasial (SPKS) untuk evaluasi kesesuaian lahan bawang putih. Hasil pengujian menggunakan blackbox testing menunjukkan bahwa fitur utama yaitu menu klasifikasi lahan dan peta visualisasi lahan. Serta fitur pendukung dalam SPKS kesesuaian lahan bawang putih yaitu menu data master, kriteria lahan, menu basis aturan dan menu bantuan mampu berjalan dengan baik sesuai dengan analisis-analisis kebutuhan sistem

The Directorate Jenderal of Horticulture (Indonesia) carried out a self sufficient development program on garlic commodities by increasing the garlic land area to 72.249 Ha in 2019 from 23.900 Ha in 2018. The expansion of cultivated land caused by the fact that gralic production in 2017 decreases 2,5% per year, meanwhile this number couldn't cope with the garlic consumption rate in Indonesia which increased 1,06% per year. The proper decision making in evaluating the suitability of garlic cultivation land is crucial for the expansion planning. Good decision will be achieved if the supporting data are complete and developed using the proper method. This study aims to develop a decision support system for garlic land suitability in Magetan and Solok districts spacial data. This system is developed using the spatial decision trees. On the Magetan dataset, the best model has 33 rules with accuracy of 94,34% and relief variable as the root node. Whereas on the Solok dataset, the best model has 66 rules with accuracy of 60,29% and soil texture variable as the root node. This web-based system is developed using the iterative system development cycle method. System development in this study consist of four main phases: plan, analysis, design and implementation. This study has successfully produced a spatial decision support System for Evaluation of Garlic Lands Suitability. Knowledge base of the system is created based on the spatial decision tree resulted in the previous study. The results of the blackbox testing show that the five features of the system: land criteria menu, rule base menu, data master, classification menu, visualization map menu, and help menu are working properly.The Directorate Jenderal of Horticulture (Indonesia) carried out a self sufficient development program on garlic commodities by increasing the garlic land area to 72.249 Ha in 2019 from 23.900 Ha in 2018. The expansion of cultivated land caused by the fact that gralic production in 2017 decreases 2,5% per year, meanwhile this number couldn't cope with the garlic consumption rate in Indonesia which increased 1,06% per year. The proper decision making in evaluating the suitability of garlic cultivation land is crucial for the expansion planning. Good decision will be achieved if the supporting data are complete and developed using the proper method. This study aims to develop a decision support system for garlic land suitability in Magetan and Solok districts spacial data. This system is developed using the spatial decision trees. On the Magetan dataset, the best model has 33 rules with accuracy of 94,34% and relief variable as the root node. Whereas on the Solok dataset, the best model has 66 rules with accuracy of 60,29% and soil texture variable as the root node. This web-based system is developed using the iterative system development cycle method. System development in this study consist of four main phases: plan, analysis, design and implementation. This study has successfully produced a spatial decision support System for Evaluation of Garlic Lands Suitability. Knowledge base of the system is created based on the spatial decision tree resulted in the previous study. The results of the blackbox testing show that the five features of the system: land criteria menu, rule base menu, data master, classification menu, visualization map menu, and help menu are working properly.