This paper presents an optimal kinetostatic design method for a general 3PRS (Prismatic-revolute-spherical) spatial parallel manipulator by formulating a multi-objective optimization problem considering the performance indices are as the objective functions. Three performance criteria—Global conditioning index (GCI), Global stiffness index (GSI) and workspace volume-were formulated and the effect of actuator layout angle on the performance indices was studied. A multi-objective evolutionary algorithm based on the Control elitist non-dominated sorting genetic algorithm (CENSGA) was adopted to find the final approximation set. The optimal geometric parameters that yield minimal compliance with larger workspace volume and improved dexterity are suggested for a general 3PRS parallel manipulator. For the optimal design, it is shown that global isotropy and global stiffness of the platform is improved at the cost of workspace reduction.