Flow control in kanban based multicell manufacturing: I. A structural analysis of general blocking design space

We develop an efficient approach to the design of flow control systems in kanban based multicell manufacturing. The two primary objectives in control system design are: maximization of the expected line throughput and minimization of the expected work-in-progress (WIP). These objectives are usually conflicting, and we formulate them as individual optimization problems in each objective, constrained by an acceptable level of the other. We employ the kanban paradigm of Schonberger (1982) and the general blocking model of Cheng and Yao (1993) as the basis for the design of kanban control systems. We present, in a two paper sequence, a structural analysis of the general blocking model, a control system design framework addressing the above optimization problems and extensive experimental results. In this paper, we present the structural analysis. The analysis consists of two parts: an equivalence analysis and a lattice analysis . The equivalence analysis shows the equivalence between classes of control configurations and their mutual reductions. In lattice analysis, the structure of the design space is studied and a model of generalized neighbourhood spaces is developed. Specializing this model, different classes of lattice-structured neighbourhoods are derived. The structural properties of these lattices and the throughput-WIP characteristics along their lattice paths are developed.