<p>Balancing project duration, cost, and safety risk is a critical challenge in construction project management due to the inherently conflicting nature of these objectives. This study proposes a novel multi-objective Jaya algorithm enhanced with super opposition-based learning (SOBL-Jaya) to address the time–cost–risk trade-off problem more effectively. The super opposition-based learning mechanism strengthens both exploration and exploitation by generating strategically opposite candidate solutions, thereby accelerating convergence and improving Pareto-front diversity. The proposed SOBL-Jaya algorithm was evaluated using a benchmark 18-activity construction project and compared with established multi-objective optimization methods, including Non-Dominated Sorting Improved Rao-2 (NDSII-Rao-2) and the Multi-Objective Genetic Algorithm (MOGA). The results demonstrate that SOBL-Jaya achieves strong convergence and diversity, reflected by a high hypervolume value (HV ≈ 0.95) and competitive spread. In particular, SOBL-Jaya consistently identifies safer scheduling solutions that enhance occupational safety and health (OSH) without excessive deterioration in project time or cost. To support practical decision-making, a crowding distance ranking (CDR) mechanism was employed to prioritize Pareto-front solutions and identify representative extreme and compromise alternatives for managerial selection. Correlation analysis further confirms that schedule compression is generally associated with increased safety risk, aligning with established construction management principles. These findings highlight the effectiveness and robustness of SOBL-Jaya as a parameter-free and computationally efficient decision-support tool for construction planners and managers. By enabling safer and more balanced scheduling decisions, the proposed approach offers a meaningful process innovation and methodological advancement for multi-objective optimization in construction project management.</p>