The use of robots in pick-and-place operations has been a well-known and widely used approach in production processes practically since the inception of robotic systems. Currently, however, increasingly stringent requirements are being placed on the control of robotic pick-and-place processes. This requires real-time robot control and the collection of objects of various classes in a random order and location within the robot's workspace. Collaboration between robots and vision systems monitoring objects appearing in front of the robot on a conveyor belt is becoming commonplace in industrial environments. This also necessitates the use of appropriate pick-and-place algorithms in robot control. This article presents an overview of component picking algorithms for robotic production lines. Key features of these algorithms are described. The effect of local object density for a selected object class on sorting performance at varying conveyor belt speeds was analysed. The study focused on the interaction between input flow characteristics and conveyor dynamics, assuming constant robot kinematic parameters and a fixed object placement point location. Furthermore, an approximate relationship describing the maximum allowable conveyor belt speed as a function of object density was derived, defining the stability limit of the SPT algorithm.