Analysis of flexible robotic cells with improved pure cycle

In this paper, we study the robotic cell scheduling problem with two machines and identical parts. It is assumed that there are flexibility in the robot type, the pickup category, and the cell layout. Also, the objective is to find the robot move sequence that minimizes the cycle time. One of the assumptions in the pertaining literature is that the occupied robot cannot load a part into an occupied machine. In contrast, the authors consider the robot has the swap ability, so the busy robot will be able to load concurrently a part into an occupied machine. It is also presumed that each part can entirely be processed by one computer numerical control (CNC) machine. A wide range of m-unit cycles named pure cycles has been well documented in the literature. However, a new class of the pure cycles which can produce less than m parts is defined in this study. We prove that the improved pure cycles always dominate pure cycles. The performance of these improved pure cycles for linearly and circularly configured robotic cells, based on free-pickup and no-wait categories are examined. Also, the number of machines which should be idle or busy to obtain an optimal cycle is determined. The novelty of this research is the introduction of the improved pure cycles that increase the long run average throughput rate and also reduce the number of machines.