Machine-Learning-Based Spectroscopic Technique for Non-Destructive Estimation of Shelf Life and Quality of Fresh Fruits Packaged under Modified Atmospheres

Abstract

The safety and quality of fresh fruits deserve the greatest attention, and are a priority for producers and consumers alike. Modern technologies are crucial to accurately estimating and predicting fresh fruits’ quality and shelf life, to optimize supply chain management. Modified atmosphere packaging (MAP) is an essential method that maintains quality parameters and increases the shelf life of fresh fruits by reducing their ripening rates. This study aimed to develop a cost-effective, non-destructive technique using tiny machine learning (TinyML) and a multispectral sensor to predict/estimate the quality parameters and shelf life of packaged fresh dates under the natural atmosphere (Control), vacuum-sealed bags (VSBs), and MAP with different gas combinations: 20% CO2 + N balance (MAP1), and 20% CO2 + 10% O2 + N balance (MAP2). The shelf life and quality parameters of the packaged fresh dates (pH, total soluble solids (TSSs), sugar content (SC), moisture content (MC), and tannin content (TC)) were evaluated under different storage temperatures and times. A multispectral sensor (AS7265x) was utilized to correlate the fruit quality parameters with spectrum analysis under the same storage conditions, to prepare the dataset to train the prediction models. The prediction models were trained in the Edge Impulse Platform, and deployed to an Arduino Nano 33 BLE sense microcontroller unit (MCU) for inference. The findings indicated that the vacuum and MAP1 efficiently increased the shelf life and maintained the quality parameters of the packaged fresh fruit to 43 ± 2.39 and 39 ± 3.34 days, respectively, at 5 °C. The optimal neural network consisted of the input layer with 20 nodes (the packaging type, storage temperature, and 18 channels of the spectral sensor data at 410 to 940 nm wavelengths), two hidden layers with 20 and 12 nodes, and an output layer with one node for the target quality parameter or shelf life. These optimal prediction models efficiently predicted the shelf life with R2 = 0.951, pH with R2 = 0.854, TSSs with R2 = 0.893, SC with R2 = 0.881, MC with R2 = 0.941, and TC with R2 = 0.909. The evaluation of the developed prediction models under each packaging condition indicated that these models serve as powerful tools for accurately predicting fruit quality parameters, and estimating the shelf life of fresh dates.

TinyML Algorithms for Big Data Management in Large-Scale IoT Systems

“…This approach has been effectively implemented in the development of a smart edge computing robot (SECR) [82]. Such methodologies are increasingly important in sectors like supply chain management, where they play a crucial role in predicting product quality parameters and extending the shelf life of perishable goods, including fresh fruits in modified atmospheres [83]. Moreover, the growing complexity in communication systems, spurred by diverse emerging technologies, underscores the need for AI and ML techniques in the analysis, design, and operation of advanced communication networks [84].…”

Impact of Modified Atmosphere Packaging Conditions on Quality of Dates: Experimental Study and Predictive Analysis Using Artificial Neural Networks

“…These capabilities, along with the ANN ability to achieve higher predictive performance as evidenced by the higher R 2 and lower RMSE values in this study, demonstrate that ANNs are a more powerful and suitable technique for establishing prediction mod-els involving complex systems with unknown mathematical equations relating multiple variables, such as the relationships between the storage conditions and quality parameters of date fruits. Srinivasagan et al [79] and Mohammed et al [80] utilized ANN regression models to accurately predict and estimate the shelf life of stored date fruits. This application of NN models offers several advantages.…”