Simultaneously minimizing energy consumption and cycle time under varying payload conditions is challenging in industrial robot operations. Traditional trajectory planning often prioritizes speed which leads to increased energy consumption and wear. This paper introduces a hybrid NSGA-II MOPSO framework for multi-objective trajectory planning of industrial UR10 robots operating under varying payloads. Joint-space motion is parameterized by cubic spline interpolation through a small set of joint-space waypoints. It is optimized to minimize energy consumption and cycle time while maintaining joint position, velocity, and torque limits for varying payload conditions. The methodology initially utilizes MOPSO to populate an external archive and then refines the resulting Pareto front using NSGA-II with elitist selection, crossover, and mutation. Operations in Robotics System Toolbox of MATLAB show that the proposed hybrid consistently attains Pareto fronts with larger hypervolume, lower generational distance, and improved spread compared with singular uses of MOPSO and NSGA-II. Detailed analysis of the selected knee solutions reveals how increased payload steepens the trade-off between energy and time, amplifies actuator torques and energy consumption even for the smallest of cycle times. An automatic knee-point detection technique introduced by the hybrid delivers practically relevant, energy and time efficient trajectories that can be directly implemented in industrial robotics. This will play a vital role in supporting scheduling, energy budgeting, and preventive maintenance decisions in high-throughput manufacturing.