Robotic arc welding has been popular in the welding industry. Major emphasis has been placed on integrating robots with the gas-metal arc welding process due to its predominant use in industry. But, the programming of robots for arc welding is still a time-consuming process. It needs substantial expertise in both robotics and the arc welding process. Thus much effort in robot arc welding programming is concentrated on off-line programming. The Off-Line Programming (OLP) of robotic arc welding requires sophisticated process planning composed of various stages such as robot and positioner motion planning, collision avoidance, weld sequencing, CAD data interfacing, weldseam tracking, and calibration of the robot arm and robotic workcell. However, existing OLP systems do not provide welding operations planning such as weld sequencing and positioner motion planning. Furthermore, they do not consider both workcell productivity and weld quality in the robotic welding cell, which are closely associated with the robot configuration. This is due to unstructured and compound characteristics of welding operations involving the determination of robot welding configurations, weldline clustering and sequencing, and multi-weldline/multipass welding planning. In this paper, we will present a method for robotic arc welding operations planning that can support the robot OLP; namely, a downhand welding algorithm using a rotating and tilting positioner, a robotic welding stability model associated with weld quality and cell productivity, and operations planning procedures for robotic multi-weldline/multi-pass welding.