Determination of the optimum harvest window and quality attributes of oil palm fresh fruit bunch using non-destructive shortwave infrared spectroscopy

Abstract

Determination of the optimum harvest window (OHW) plays a key role in the agro-food chain because the quality of fruit depends on the right harvesting time. In palm oil industry, indices based on destructive measurements are used for this purpose, through visual observation of the number of detached fruitlets. In this study, we proposed a reflectance shortwave infrared (SWIR) method for the evaluation of the OHW of the oil palm fresh fruit bunch (FFB), as well as its standard quality attributes. The experiment involved 150 FFBs in five different maturity stages, followed by a biochemical analysis to determine samples’ quality indices. Reflectance data of the FFB was studied and correlated to its standard quality attributes (ripeness, oil content, free fatty acids (FFA) level, the deterioration of bleachability index (DOBI), and carotene content) based on oil extracted from its mesocarp. The harvest date, as well as the fruits’ quality attributes were modeled using Principal Component Analysis (PCA) and Partial Least Square (PLS). The models showed significant relationships with biochemical changes during FFB maturation and ripening. The investigations showed that the OHW for FFB was indicated by the characteristic drop of reflectance value of certain wavelengths within the SWIR region. Furthermore, the SWIR spectroscopy data could be modeled using PLS regression to determine oil extraction rate (OER), FFA, DOBI and carotene of FFB non-destructively. In this study, five non-destructive evaluation (NDE) models were successfully developed to determine FFB’s OHW, OER, FFA, DOBI and carotene by utilizing the SWIR reflectance. The models delivered rapid results with high accuracy and proven to be a time- and cost-saving solution for NDE of FFB.