Coordinated capacity planning in two-stage thin-film-transistor liquid-crystal-display (TFT-LCD) production networks

Abstract

This paper studies capacity planning problems in two-stage Thin Film Transistor-Liquid Crystal Display (TFT-LCD) production networks. In the TFT-LCD industry, capacity planning must coordinate capacity expansion and capacity allocation decisions within the planning horizon. Capacity planning problems are extremely challenging for the TFT-LCD industry owing to the following characteristics: (1) multi-stage, multi-site, and multi-generation production network with limited configuration flexibility; (2) an economic cutting ratio that varies at different sites; (3) multiple capacity types at each production site; and (4) a complex product hierarchy and rapidly changing demand. Due to the limited configuration flexibility, capacity expansion at different production sites must be coordinated to prevent bottlenecks at other sites. Moreover, capacity allocation decisions must be coordinated with the capacity expansion decisions to generate most profitable product mixes.To address these problems, we develop a mixed integer linear programming (MILP) model to maximize profit. Since large MILP models are computationally intractable, an iterative two-phase shadow-price heuristic with reactive and conservative expansion strategies is proposed to find near-optimal solutions. Through an extensive computational study, the effectiveness and robustness of the proposed heuristic is demonstrated in terms of solution quality and computational time. In addition, by conducting the sensitivity analysis, we suggest the suitable usage timing and choice of the two expansion strategies under different price and cost situations for capacity planners of the TFT-LCD manufacturing.