Robotic programming is still a challenge in manufacturing. Reasons for this challenge are the time-consuming, complex, primarily manually optimized, and expert-dependent path generation. Therefore, easy and optimized programmable robotic manufacturing systems are required. To make up for this deficiency, this dissertation raises the following research question: How to automatically generate offline optimized paths for robotic manufacturing processes (RMPs) under constraints and process optimization criteria? This research question is answered with the following evaluated thesis: Robotic paths can be automatically optimized by methodologically configuring and deploying sample-based generation algorithms, based on the procedural interpretation of the Product, Process, and Resource model-based components of the RMP. The generalized approach and implemented architecture is named Automatic Optimized Offline Programming. Three different use-cases are presented for evaluating the path planning approach: (1) The optimized collision avoidance in robotic welding, (2) the optimized stiffness for robotic milling, and (3) the optimized sensing strategy for robotic deburring.