Unit commitment (UC), a critical power system operation problem, involves high complexity due to numerous constraints. This paper presents a novel approach using the seeker optimization algorithm (SOA), an evolutionary method inspired by human search behavior. The proposed SOA employs integer coding to efficiently handle minimum up/down constraints and binary coding for spinning reserve requirements. By leveraging population-based search, SOA ensures optimal convergence while addressing UC’s combinatorial challenges. Simulations across diverse test systems demonstrate algorithm superiority in convergence speed and solution quality compared to conventional methods. The integer-coded framework enhances computational efficiency, particularly in managing unit minimum uptime/downtime constraints. Results implemented on 10 to 100 generating units validate SOA’s effectiveness in optimizing large-scale UC problems, compared to the other existing approaches, offering a reliable tool for power system operators. This approach bridges the gap between metaheuristic adaptability and UC’s rigid constraints, paving the way for cost-effective and operationally feasible solutions.