With diabetes prevalence rising worldwide and especially rapidly in India due to urbanization and lifestyle transitions, the study underscores the importance of model-based approaches to comprehensive understand of disease progression. Thus, this study explored glucose–insulin regulatory mechanisms and challenges associated with diabetes mellitus among patients. The study further highlights key risk factors contributing to diabetes development, including physical inactivity, obesity, genetic predisposition, viral triggers, and aging.

In this regards, John Holland’s evolutionary computing principles were applied, incorporating a Genetic Algorithm (GA) to address diabetes-related optimization problems. MATLAB’s Genetic Algorithm module was used to implement the computational framework. The proposed mathematical model utilized differential equations to simulate blood glucose and insulin levels.

The results revealed comparative simulations for normal, prediabetic, and diabetic individuals, with optimization achieved through GA, demonstrated using fitness function plots. The findings suggest that Genetic Algorithms offer an effective method for identifying optimal diagnostic parameters and enhancing prediction accuracy.

Overall, the study supports the integration of heuristic computational approaches, such as GA, in diabetes research to tackle complex diagnostic and biological questions. The study may expand the use of biomathematical and evolutionary algorithms to further improve understanding, prediction, and management strategies for diabetes mellitus. Continuous monitoring and adherence to treatment remain essential for individuals living with diabetes.