| dc.contributor.advisor | Budiastra, I Wayan | |
| dc.contributor.author | Pramesya, Ketrina | |
| dc.date.accessioned | 2024-07-10T06:10:55Z | |
| dc.date.available | 2024-07-10T06:10:55Z | |
| dc.date.issued | 2024 | |
| dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/153407 | |
| dc.description.abstract | Tomat ceri merupakan salah satu jenis tomat yang memiliki harga jual relatif lebih tinggi dibandingkan tomat sayur, serta memiliki kandungan gula dan asam organik yang lebih banyak. Parameter mutu penting dalam pemilihan tomat ceri yaitu berdasarkan sifat fisiko-kimianya yang meliputi kekerasan, Total Padatan Terlarut (TPT), dan total asam. Penentuan mutu tomat ceri selama ini masih dilakukan secara destruktif. Near Infrared Spectroscopy (NIRS) dikembangkan sebagai salah satu metode alternatif yang lebih efisien untuk penentuan mutu buah karena bersifat tidak merusak (nondestructive). Penelitian ini bertujuan untuk memprediksi sifat fisiko-kimia pada buah tomat ceri secara nondestruktif dengan parameter kekerasan, total padatan terlarut, dan total asam buah menggunakan NIRS dengan metode PCR dan PLS. Bahan yang digunakan sebanyak 90 buah tomat ceri dengan tingkat kematangan yang berbeda, yaitu mature green, pink, dan red. Alat yang digunakan adalah spektrometer NIRFlex N-500 fiber optic solids untuk memperoleh data pengukuran gelombang dengan panjang 1000-2500 nm. Pengukuran secara destruktif dilakukan dengan menggunakan rheometer untuk mengukur kekerasan dan refractometer untuk mengukur TPT dan total asam. Pengolahan data yang digunakan untuk kalibrasi dan validasi spektra NIR dengan data destruktif sifat fisiko-kimia tomat ceri yaitu metode Partial Least Square (PLS) dan Principal Component Regression (PCR). Pre-treatment yang digunakan untuk meningkatkan hasil prediksi NIRS yaitu Normalisasi, Standard Normal Variate (SNV), Multiplicative Scatter Correction (MSC), De-Trending (DT), dan Smoothing Savitzky-Golay (SSG). Prediksi terbaik untuk kekerasan diperoleh dengan metode PLS menggunakan pre-treatment SSG faktor 6 dengan nilai r = 0,96; SEC = 2,34 N; SEP = 2,66 N; CV = 18,25%; RPD = 3,14; dan konsistensi 87,97%. Prediksi terbaik untuk TPT diperoleh dengan metode PLS menggunakan pre-treatment SNV faktor 10 dengan nilai r = 0,94; SEC = 0,32 oBrix; SEP = 0,38 oBrix; CV = 6,35%; RPD = 2,44; dan konsistensi = 84,01%. Prediksi terbaik untuk total asam diperoleh dengan metode PLS menggunakan pre-treatment SSG faktor 6 dengan nilai r = 0,53; nilai SEC = 0,23%; SEP = 0,21%, serta nilai CV = 28,41%; RPD = 1,12; dan konsistensi = 106,16%. Prediksi kekerasan dan TPT buah tomat ceri secara nondestruktif dapat dilakukan dengan NIRS, namun belum layak digunakan untuk memprediksi kandungan total asam karena menghasilkan nilai r yang jauh mendekati 1 dan nilai Ratio of Prediction to Deviation (RPD) yang masih kecil di bawah 1,5. Metode PLS menghasilkan keakuratan yang lebih tinggi dibandingkan metode PCR. | |
| dc.description.abstract | Cherry tomatoes are a type of tomato that has a relatively higher selling price than vegetable tomatoes, and contains more sugar and organic acids. Important quality parameters in selecting cherry tomatoes are based on their physico-chemical properties which include hardness, Total Dissolved Solids (TPT), and total acid. So far, determining the quality of cherry tomatoes is still carried out destructively. Near Infrared Spectroscopy (NIRS) was developed as a more efficient alternative method for determining fruit quality because it is nondestructive. This research aims to predict the physico-chemical properties of cherry tomatoes nondestructively with the parameters of hardness, total dissolved solids, and total fruit acids using NIRS with PCR and PLS methods. The ingredients used were 90 cherry tomatoes with different levels of maturity, namely mature green, pink and red. The tool used is a NIRFlex N-500 fiber optic solids spectrometer to obtain wave measurement data with a length of 1000-2500 nm. Destructive measurements are carried out using a rheometer to measure hardness and a refractometer to measure TPT and total acid. The data processing used for calibration and validation of NIR spectra with destructive data on the physico-chemical properties of cherry tomatoes is the Partial Least Square (PLS) and Principal Component Regression (PCR) methods. The pre-treatments used to improve NIRS prediction results are Normalization, Standard Normal Variate (SNV), Multiplicative Scatter Correction (MSC), De-Trending (DT), and Smoothing Savitzky-Golay (SSG). The best prediction for hardness was obtained by the PLS method using SSG pre-treatment factor 6 with a value of r = 0,96; SEC = 2,34 N; SEP = 2,66 N; CV = 18,25%; RPD = 3,14; and consistency 87,97%. The best prediction for TPT was obtained by the PLS method using pre-treatment SNV factor 10 with a value of r = 0,94; SEC = 0,32 oBrix; SEP = 0,38 oBrix; CV = 6,35%; RPD = 2,44; and consistency = 84,01%. The best prediction for total acid was obtained by the PLS method using SSG pre-treatment factor 6 with a value of r = 0,53; SEC value = 0,23%; SEP = 0,21%, and CV value = 28,41%; RPD = 1,12; and consistency = 106,16%. Nondestructive prediction of hardness and TPT of cherry tomatoes can be done using NIRS, but it is not yet suitable to be used to predict total acid content because it produces an r value that is much closer to 1 and a Ratio of Prediction to Deviation (RPD) value that is still small below 1,5. The PLS method produces higher accuracy than the PCR method. | |
| dc.description.sponsorship | | |
| dc.language.iso | id | |
| dc.publisher | IPB University | id |
| dc.title | Prediksi Sifat Fisiko-Kimia Buah Tomat Ceri secara Nondestruktif Menggunakan Near Infrared Spectroscopy (NIRS) | id |
| dc.title.alternative | | |
| dc.type | Skripsi | |
| dc.subject.keyword | cherry tomatoes | id |
| dc.subject.keyword | NIRS | id |
| dc.subject.keyword | nondestructive | id |
| dc.subject.keyword | PCR | id |
| dc.subject.keyword | PLS | id |