In recent years, the demand for mental health services has significantly increased, accompanied by a rise in the severity of mental health cases. This has prompted mental healthcare organizations to identify key variables affecting service delivery, leading to challenges in effectively allocating human resources. While previous studies have explored the influence of therapist performance on patient treatment outcomes, they have not specifically analyzed the impact of assigning each patient to a high-performing therapist based on their specific condition. This research proposes a data-driven optimization model and performance assessment approach to enhance therapist allocation, integrating empirical data, machine learning algorithms, and mathematical optimization techniques. The findings of this study highlight the significance of caregivers in psychiatric settings, suggesting that patients benefit more from being matched with the appropriate therapists for their disorders. The proposed model enhances the effectiveness and efficiency of mental health therapy, leading to a 4.9% improvement in patient outcomes and more equitable access to quality care.