Author(s): Dr. T Srikanth, Akinapalli Poornima, Dayyala Balamani and Chaliganti Sadha

Abstract: For scheduling and quality cost assessment, it is necessary to anticipate the remaining shelf life of fresh vegetables and fruits (FFVs) during transit. The IoT allows for the real-time processing of measured environmental factors. The translation of environmental observations into dynamic RSL estimations, however, requires a verified, real-time computing approach. There are a lot of qualitative, intrusive, and static generic RSL models out there for FFVs. Under unpredictable and ever-changing logistical circumstances, this research develops a general RSL model for FFVs. The model's foundation is an approximation of the product's anticipated respiration rate, which is then used to estimate the present general decay rate. This estimate is then integrated with regard to time. You won't need to do accelerated shelf-life tests on samples before using it, and it won't damage or invasively alter the sample in any way. Furthermore, a surrogate model was suggested to enable the model's implementation in rapid, real-time applications for 'Edge IoT,' since the initial design is computationally demanding. Using three fresh products—strawberries, apricots, and spinach—in a residential refrigerator, the model was experimentally validated. The original model had a maximum variation in prediction error of 1.3 days, whereas the surrogate model had a deviation of 2.95 days. However, even at the 0.01% significance level, neither the original nor the surrogate models' predictions were statistically different from the samples' actual shelf life.

DOI: 10.33545/26633582.2024.v6.i2a.127

Pages: 62-66 | Views: 1613 | Downloads: 862

Download Full Article: Click Here