My Research – Thesis (is your research similiar?)

PERAMALAN PRODUKSI PADI DENGAN ARIMA,FUNGSI TRANSFER DAN ADAPTIVE NEURO FUZZY INFERENCE SYSTEM

ABSTRAK

Angka ramalan produksi tanaman pangan diperlukan untuk mendukung kebijakan pemerintah dalam penanganan isu pangan terutama padi di Indonesia. Angka ramalan produksi padi telah dilakukan oleh Badan Pusat Statistik (BPS), dengan menggunakan teknik peramalan tidak langsung, yaitu peramalan produksi padi melalui peramalan luas panen dan produktivitas padi. Tujuan dari penelitian ini adalah mengembangkan model terbaik dalam meramalkan produksi padi berdasarkan pendekatan Adaptive Neuro Fuzzy Inference System (ANFIS). Hasilnya akan dibandingkan dengan nilai ramalan ARAM I dan dua metode klasik lainnya, yaitu model ARIMA dan fungsi transfer. Data yang digunakan dalam penelitian ini adalah data padi sawah Provinsi Jawa Tengah, Kalimantan Selatan dan Sumatera Utara subround I – III tahun 1983-2010. Tingkat akurasi peramalan yang dihasilkan oleh setiap metode peramalan diukur dengan kriteria Mean Absolute Percentage Error (MAPE). Hasil penelitian menunjukkan bahwa dari metode peramalan yang digunakan dalam penelitian ini, metode ANFIS merupakan metode peramalan luas panen padi sawah terbaik pada Provinsi Jawa Tengah dengan rata-rata nilai MAPE sebesar 6,89%. Sedangkan pada peramalan produktivitas padi sawah, ARIMA merupakan metode peramalan terbaik dengan rata-rata nilai MAPE sebesar 1,83%. Pada Provinsi Kalimantan Selatan, metode peramalan luas panen maupun produktivitas padi sawah terbaik adalah ARIMA dengan rata-rata nilai MAPE masing-masing sebesar 9,96% dan 5,18%. Pada Provinsi Sumatera Utara, model fungsi transfer merupakan metode peramalan luas panen padi sawah terbaik dengan rata-rata nilai MAPE sebesar 2,43%. Sedangkan pada peramalan produktivitas padi sawah, ANFIS merupakan metode terbaik dengan rata-rata nilai MAPE sebesar 1,82%.

Kata kunci: Produksi Padi, ANFIS, ARIMA, Fungsi Transfer

ARIMA, TRANSFER FUNCTION AND ADAPTIVE NEURO FUZZY INFERENCE SYSTEM FOR FORECASTING RICE PRODUCTION

ABSTRACT

Food crop production forecast figure is needed for supporting government on the handling of the issue of food problem especially rice in Indonesia. Rice production forecast figure has been regularly conducted by Badan Pusat Statistik (BPS), Statistics Indonesia, using indirect forecasting technique, i.e. forecasting the rice production through forecasting the harvested area and the rice productivity. The objective of this research is to develop the best model for forecasting the rice production based on the Adaptive Neuro Fuzzy Inference System (ANFIS) approach. The result will be compared to the forecasting results published by BPS and two other classical methods, namely ARIMA and transfer function model. Data about wetland rice in Central Java, South Kalimantan and North Sumatera Province from 1st subround 1983 to 3rd subround 2010 are used as case study. The accuracy performance for each forecasting method is measured by Mean Absolute Percentage Error (MAPE) criteria. The results show that from all of the listed method used in this research, the best forecasting method for harvested area of wetland rice in Central Java Province is ANFIS method with MAPE value 6,89% and the best forecasting method for rice productivity is ARIMA with MAPE value 1,83%. In South Kalimantan, ARIMA is the best forecasting method for both of harvested area and productivity of wetland rice with each MAPE value 9,96% and 5,18%. In North Sumatera Province, the best forecasting method for harvested area of wetland rice is transfer function with MAPE value 2,43% and the best forecasting method for rice productivity is ANFIS with MAPE value 1,82%.

Keywords: Rice Production, ANFIS, ARIMA, Transfer Function

1 comment
  1. Great post. I’m verifying consistently this site and I’m fascinated! Very helpful information and facts particularly the final step🙂 My spouse and i deal with this sort of information a lot. I’m looking for this kind of info for a long time. Many thanks and finest of chance.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: