PLANTING PATTERN MANAGEMENT BASED ON RBF NEURAL NETWORK AND OPTIMIZE PROFIT TO DETERMINANE THE TIME PLANTING SEASON ON LOMBOK ISLAND

Purchase PDF

Published: 2015-04-11

Page: 40-58


ALVEN SAFIK RITONGA *

Department of Mathematics, Institut Teknologi Sepuluh Nopember, Faculty of Mathematics and Natural Sciences, Surabaya, Indonesia.

MOHAMMAD ISA IRAWAN

1Department of Mathematics, Institut Teknologi Sepuluh Nopember, Faculty of Mathematics and Natural Sciences, Surabaya, Indonesia.

*Author to whom correspondence should be addressed.


Abstract

Cropping pattern are determination regarding planting schedule, type of planting, and planting area that applied on irrigation area. To obtain ideal planting pattern, we should balancing between water availability and water needed for each plant. Using prediction of hydrological data such as rainfall and climatological data such as temperature, wind speed, solar radiation, and humidity can determine water availability on irrigation area. This prediction method called Radial Basis Function Artificial Neural Network (RBF). The data which used for this research such as rainfall, temperature, wind speed, solar radiation and humidity were taken from the Water Resources Information Center (WRIC) West Nusa Tenggara (NTB) during the last 31 years, which is from 1983 until 2013. This data used to predict hydrological and climatological data on 2014. The result data can determine the water consumption needs for plant (evapotranspiration), effectiveness of rainfall, and preparation of water consumption need for land, and then connected with irrigation water availability volume and how long duration of cropping plant to get cropping pattern design. Cropping pattern optimization design is determined with scheduling of planting for each crop in 2014. By using RBF Artificial Neural Network method to predict hydrology and climatologically data, we can predict that MSE statistical index is 7.0x10-5-9.0x10-5 in average. For network architecture validation, the accuracy can reach up to 99.89% and 0.92% error in average. The result of planting pattern optimization at Lombok Island for 2014 planting season could be seen from increasing the profit of each district, such as: East Lombok increased 16.95%, Central Lombok increased 21.76%, West Lombok increased 4.00%, City of Mataram increased 11.60% and North Lombok decrease 0.55%. In overall there is significant profit increase for Lombok Island about 10.75% from previous year.

Keywords: Artificial neural network radial basis function, cropping pattern management, optimization


How to Cite

RITONGA, A. S., & IRAWAN, M. I. (2015). PLANTING PATTERN MANAGEMENT BASED ON RBF NEURAL NETWORK AND OPTIMIZE PROFIT TO DETERMINANE THE TIME PLANTING SEASON ON LOMBOK ISLAND. Asian Journal of Mathematics and Computer Research, 3(1), 40–58. Retrieved from https://ikprress.org/index.php/AJOMCOR/article/view/104

Downloads

Download data is not yet available.