While health insurance systems have gained traction in both developed and developing nations, underdeveloped countries continue to strive for cost-effective healthcare solutions. Designing effective insurance predictive models is critical, relying on variables such as age and average monthly medical costs. Researchers are exploring supervised and unsupervised models to accommodate diverse data and prevent financial strain on all stakeholders due to high medical expenses. This study contributes to this effort by evaluating the efficiency of various supervised models in predicting insurance costs. We assess their model performance and provide insights for developing robust insurance systems.