Shelf-Life Prediction of Oil Palm Fresh Fruit Bunches Using Artificial Neural Networks

Indonesia, one of the world’s main producers of oil palm fresh fruit, boasts a Crude Palm Oil (CPO) export performance index of 0.94. Predicting the shelf life before processing oil palm fresh fruit is critical. However, the challenge lies in accurately estimating when ripe fruit will spoil, leading to potential rot during storage or distribution. This research introduces an innovative approach for accurately predicting the shelf life of oil palm fruit by utilizing a Multilayer Perceptron-type artificial neural network. The shelf-life prediction is calculated based on measurements from electrical resistivity, gas, and color sensors on oil palm fresh fruit processed using an Arduino Nano. The application of artificial neural networks in predicting the shelf life of oil palm fruit achieves an accuracy level of 98.62%, with a Mean Square Error (MSE) value of 0.0069. Consequently, the system can predict when ripe oil palm fruit will rot.