Hybrid Predictive Framework for Expiry Date Estimation of Live-Attenuated Vaccines Using Stability Testing and Machine Learning

Transport sustainability is a major issue for most pharmaceutical agents, especially live-attenuated vaccines, which are highly sensitive to the environmental factors like temperature, humidity, light, and oxygen which is why there is need of of serious steps to be taken in the supply chain. In pharmaceutical supply chains, the most important part is to enhance the predictions related to stability for better storage and control over the inventory and minimized waste. Even though the supply chain helps us in tracking the Vaccine, the condition of it is still unknown. The stability of the vaccine is tested via these stability tests. Any deviation from the optimal order at any stage in the supply chain can compromise stability and make vaccines ineffective or unsafe. This study takes care of this by performing stability testing predicting the drug expiry according to the environmental factors. The predictive framework is a combination of machine learning regression models and the Arrhenius Equation to accommodate environmental variables affecting degradation. Ambient temperature and humidity were identified as the most significant environmental factors affecting degradation rates. The hybrid model comprising both ElasticNet and LSTM showed better prediction performance with a mean absolute error (MAE) cut off under 4.5% as compared to other conventional methods. For example, Gradient Boosting Regressor (GBR) attained an R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> score of 0.85, the conventional methods scoring between 0.60 and 0.75. The model also accounted for a 12% decrease in the errors of expiration prediction when exposed to exposing temperature fluctuations during transportation. This is essentially intended to be produced in terms of standard recommendations for stability testing, which are crucial to the life cycle for the safe and effective use of medications.