A hybrid decision support system in medical emergencies using artificial neural network and hyperbolic secant grey wolf optimization techniques

The distribution of vaccines poses a critical challenge, particularly during public emergencies. This paper introduces a decision support framework to address the requirement of vaccine supply distribution in medical emergencies by optimizing an artificial neural network (ANN) model with hyperbolic secant grey wolf optimization (HSGWO) techniques. The proposed hyperbolic HSGWO-ANN model offers a method for optimizing vaccine requirement by efficiently allocating resources, thereby improving decision-making processes in vaccine supply and enhancing healthcare outcomes. Utilizing HSGWO-ANN model, our research aids decision makers or governmental bodies in strategically allocating vaccines to high-risk locations, mitigating the impact of pandemics. Initially, HSGWO employs a hyperbolic secant function to explore search space and enhances exploration and exploitation capabilities. Subsequently, HSGWO is integrated with ANN to optimize weights and minimize training error for accurate prediction. Furthermore, we compute normalized regression residuals for each location with the highest risk for vaccine deployment. We evaluate the performance of the proposed HSGWO-ANN model on two numerical datasets using error measures such as Root mean squared and Mean absolute percentage errors, and correlation coefficients such as \(R^2\) $R^2$ and Pearson’s. The results demonstrate that the HSGWO-ANN model outperforms existing models, exhibiting lower error rates and stronger correlations, thus offering a more efficient solution for vaccine distribution.