Sustainable and Robust Control of Cyclic Pallet Systems

Artikel i vetenskaplig tidskrift, 2013

A proper control of a system to get a desired function and increase the system lifetime is a crucial step towards the sustainable paradigm. In this paper, such a control is designed for a cyclic pallet system to achieve a minimal force on its drive unit, meet safety conditions on the system chain tension force, and the momentum of pallets, and fulfill a desired production rate. The optimal values of control parameters, namely, number of pallets, conveyor velocity, and part set schedule, are obtained through solving a mixed integer linear optimization model. The objective function in the model defines the average force on the drive unit in a cycle production. In addition, the related constraints characterize the pallet system properties such as cyclic and dynamic behavior, buffer size, constant work in process, and safety specifications. This optimization model strongly suffers from the time complexity due to the binary decision variables defining the part set schedule. To reasonably handle the computation time, a heuristic search strategy based on a modified form of the weighted profile fitting algorithm is introduced. Furthermore, the robustness of the optimal control and the system design is analyzed, using worst control and worst but safe control strategies. The optimal control and the robustness analysis are applied to some case studies, and the results are evaluated and discussed.