In this study, a simulation of a three-link robotic manipulator has been analyzed for pouring liquid metal during casting. The robotic manipulator is controlled by a proportional integral derivative (PID) controller, a tilt integral derivative (TID) controller, and a fractional-order proportional integral derivative (FOPID) controller. The controller parameters are optimized by a novel meta-heuristic nature-inspired optimization algorithm called grey wolf optimization (GWO), and a comparative analysis is performed with particle swarm optimization (PSO). This robotic manipulator has been designed in SolidWorks and simulated in MATLAB Simulink/sim-mechanics environment. The end-effector trajectory during the casting and comparative joint trajectory has been studied. The FOPID controller optimized by GWO shows better results by giving less integral time absolute error (ITAE) compared with FOPID-PSO and other controllers. Numerically GWO decreases the error by 3.85%, 9.09%, and 41.89% using PID, TID, and FOPID controller respectively as compared to PSO.