Artificial intelligence models for yield efficiency optimization, prediction, and production scalability of essential oil extraction processes from citrus fruit exocarps

Abstract

Introduction: Excessive demand, environmental problems, and shortages in market-leader countries have led the citrus (essential) oil market price to drift to unprecedented high levels with negative implications for citrus oil-dependent secondary industries. However, the high price conditions have promoted market incentives for the incorporation of new small-scale suppliers as a short-term supply solution for the market. Essential oil chemical extraction via steam distillation is a valuable option for these new suppliers at a lab and small-scale production level. Nevertheless, mass-scaling production requires prediction tools for better large-scale control of outputs. Methods: This study provides an intelligent model based on a multi-layer perceptron (MLP) artificial neural network (ANN) for developing a highly reliable numerical dependency between the chemical extraction output from essential oil steam distillation processes (output vector) and orange peel mass loading (input vector). In a data pool of 25 extraction experiments, 14 output–input pairs were the in training set, 6 in the testing set, and 5 cross-compared the model’s accuracy with traditional numerical approaches. Results and Discussion: After varying the number of nodes in the hidden layer, a 1–9–1 MLP topology best optimizes the statistical parameters (coefficient of determination (R2) and mean square error) of the testing set, achieving a precision of nearly 97.6%. Our model can capture non-linearity behavior when scaling-up production output for mass production processes, thus providing a viable answer for the scalability issue with a state-of-the-art computational tool for planning, management, and mass production of citrus essential oils.