Analisis Rendemen Individu Menggunakan Near Infrared Spectroscopy (Nirs) Untuk Mendukung Peningkatan Produksi Gula

Abstract

The Indonesian government planned to achieve the sugar self-sufficiency programs in 2014. The program can’t be achieved and Indonesia is now still a sugar importer. The effort to increase sugar productivity is by increasing the average of sugar yield. The low sugar yield is partly due to the poor of cane quality. Fair and accurate measurement of sugar content will encourage the farmers to deliver higher cane quality, and then will impact in increasing sugar yield. The use of Near Infrared Spectroscopy (NIRS) in cane quality analysis will shorten analysis time, and also NIRS can be operated automatically without sample treatment. The research was conducted at the sugar factory in East Java. The used material and instrument are the first expressed juice (FEJ) and FOSS NIR RLA equipped with VISION software, respectively. The brix and pol content of the FEJ samples were analysed by using conventional and NIRS method. The brix and pol from conventional method input into the software, and then processed by statistical treatment using partial least squares. The result from 974 FEJ samples shows that the coefficient of determination (R2 ) of brix and pol were 0.9454 and 0.9569 respectively, with the SEC, SEP and SECV ≤ 2SEL. The model was then applied for 80 FEJ samples, the average value of brix and pol content using NIRSn method (17.77 % and 14.07 %) were not significantly different from the conventional method (17.70 % and 14.08 %). When the data was applied to calculate individual sugar yield, the average value of NIRS method (8.56%) was not significantly different from the conventional method (8.59%). Keyword : NIRS, brix, pol, cane yield, individual cane yield analysis