Multiple-load carriers are widely introduced for material delivery in manufacturing systems.The real-time scheduling of multiple-load carriers is so complex that it deserves attention to pursue higher productivity and...Multiple-load carriers are widely introduced for material delivery in manufacturing systems.The real-time scheduling of multiple-load carriers is so complex that it deserves attention to pursue higher productivity and better system performance.In this paper,a support vector machine(SVM)-based real-time scheduling mechanism was proposed to tackle the scheduling problem of parts replenishment with multiple-load carriers in automobile assembly plants under dynamic environment.The SVM-based scheduling mechanism was trained first and then used to make the optimal real-time decisions between“wait”and“deliver”on the basis of real-time system states.An objective function considering throughput and delivery distances was established to evaluate the system performance.Moreover,a simulation model in eM-Plant software was developed to validate and compare the proposed SVM-based scheduling mechanism with the classic minimum batch size(MBS)heuristic.It simulated both the steady and dynamic environments which are characterized by the uncertainty of demands or scheduling criteria.The simulation results demonstrated that the SVM-based scheduling mechanism could dynamically make optimal real-time decisions for multiple-load carriers and outperform the MBS heuristic as well.展开更多
A new method for topology optimization of truss-like structures with stress constraints under multiple-load cases(MLCs)is presented.A spatial truss-like material model with three families of orthotropic members is ado...A new method for topology optimization of truss-like structures with stress constraints under multiple-load cases(MLCs)is presented.A spatial truss-like material model with three families of orthotropic members is adopted,in which the three families of members along three orthotropic directions are embedded continuously in a weak matrix.The densities and directions of the three families of members at the nodes are taken as the design variables.An optimality criterion is suggested based on the concept of directional stiffness.First,under each single-load case(SLC),the truss-like structure is optimized as per the fully stressed criterion.Accordingly,the directional stiffness of the optimal structure under an SLC at every node is obtained.Next,the directional stiffness of the truss-like structure under MLCs is determined by ensuring that the directional stiffness is as similar as possible to the maximum directional stiffness of the optimal structure under every SLC along all directions.Finally,the directions and densities of the members in the optimal truss-like structures under MLCs are obtained by solving the eigenvalue problems of the coefficient matrix of the directional stiffness at every node.Two examples are presented to demonstrate the effectiveness and efficiency of the method.展开更多
基金the National Natural Science Foundation of China(Grant No.71471135).
文摘Multiple-load carriers are widely introduced for material delivery in manufacturing systems.The real-time scheduling of multiple-load carriers is so complex that it deserves attention to pursue higher productivity and better system performance.In this paper,a support vector machine(SVM)-based real-time scheduling mechanism was proposed to tackle the scheduling problem of parts replenishment with multiple-load carriers in automobile assembly plants under dynamic environment.The SVM-based scheduling mechanism was trained first and then used to make the optimal real-time decisions between“wait”and“deliver”on the basis of real-time system states.An objective function considering throughput and delivery distances was established to evaluate the system performance.Moreover,a simulation model in eM-Plant software was developed to validate and compare the proposed SVM-based scheduling mechanism with the classic minimum batch size(MBS)heuristic.It simulated both the steady and dynamic environments which are characterized by the uncertainty of demands or scheduling criteria.The simulation results demonstrated that the SVM-based scheduling mechanism could dynamically make optimal real-time decisions for multiple-load carriers and outperform the MBS heuristic as well.
基金The research reported in this paper was financially supported by the Natural Science Foundation of China(No.11572131)the Subsidized Project for Postgraduates’Innovative Fund in Scientific Research of Huaqiao University(No.17011086002).
文摘A new method for topology optimization of truss-like structures with stress constraints under multiple-load cases(MLCs)is presented.A spatial truss-like material model with three families of orthotropic members is adopted,in which the three families of members along three orthotropic directions are embedded continuously in a weak matrix.The densities and directions of the three families of members at the nodes are taken as the design variables.An optimality criterion is suggested based on the concept of directional stiffness.First,under each single-load case(SLC),the truss-like structure is optimized as per the fully stressed criterion.Accordingly,the directional stiffness of the optimal structure under an SLC at every node is obtained.Next,the directional stiffness of the truss-like structure under MLCs is determined by ensuring that the directional stiffness is as similar as possible to the maximum directional stiffness of the optimal structure under every SLC along all directions.Finally,the directions and densities of the members in the optimal truss-like structures under MLCs are obtained by solving the eigenvalue problems of the coefficient matrix of the directional stiffness at every node.Two examples are presented to demonstrate the effectiveness and efficiency of the method.