一般單元製造系統設計偏向在確定的環境下進行，忽略變動因素(例如產品組合、需求不確定、機器(可靠度)故障等)對其影響。本研究專注於直線式單元系統的設計，考慮工件途程彈性、需求不確定、和機器故障、機器佈置等因素，來建構單元群組的確定與不確定性數學模式，本研究重新定義變數和限制式，因此可以使數學模式線性化。本研究以極小化總操作成本和故障成本為目標，為了進一步了解不確定環境下的需求變動、故障率變動對於總成本的影響，因此本研究同時做了一連串的敏感性分析，務求單元設計成果可以符合動態環境下的需求。

This paper presents a stochastic model to deal with the designing of flowline cellular manufacturing systems considering uncertainty in machine reliability and demand lots size variation. The objective is to minimize the total cost consisting of operational cost and breakdown cost in the duration of cycle time. A new deterministic flowline cell formation model is established firstly, and then that is extended to a scenario-based stochastic model. Two index, the expected value of perfect information (EVPI) and the value of stochastic solution (VSS), are employed to evaluate the added-value of incorporating uncertainties into the models. An array of computational experiments is conducted for verifying the model’s effectiveness. The experimental results show the proposed models are available and effective in finding optimal solutions in diverse scenarios, and the figures of EVPI and VSS are in a tendency of getting larger when the number of cells gains increasingly, which reveals the impact of uncertainties is getting more significant in larger scale of flowline manufacturing system than in smaller one. The proposed stochastic model incorporating uncertain issues is capable of designing flowline manufacturing cells which perform well under all possible situations rather than one that is optimal for one possible scenario but does poorly in other possible scenarios.