本研究應用兩階段隨機規劃法(Two-stage Stochastic Programming)針對具有不確定性之封閉式綠色供應鏈(Closed-Loop Green Supply Chain)問題,建構隨機規劃模型(Stochastic Programming Model),考量本研究之問題特性包含流量配置、潛在逆物流廠址選擇、產能擴充方案選擇、承擔運輸風險單位賠償成本、人口密度與有害物質汙染範圍等,並加入需求量、回收率及運輸風險率等不確定性因素,透過混合整數規劃模型,衡量流量配置、廠址與產能擴充的狀況。 本研究驗證兩階段隨機規劃模型之可行性,將確定性模型與不確定性模型進行比較,其結果顯示不確定性模型較為穩健。根據參數分析有下述三項結果:(1)銷售價格與總利潤呈正向關係,當銷售價格提升時,完美資訊期望價值與隨機解價值也隨之上升;(2)再製率與總利潤呈正比關係,當再製率提升時,以致於環境危害下降並能有效降低總風險成本;(3)在不同成本之下,以承擔運輸風險之賠償成本為最具影響且容易導致決策改變,由上述結果可知,決策者進行決策前必須優先考慮運輸風險之相關因素。藉由本研究分析結果,可提供決策人員在運輸風險的變動下,決定流量配置、逆物流廠址與產能擴充的決策依據。
In this paper, we develop a two-stage stochastic programming approach for closed-loop green supply chain network design problem under uncertainty. The features of the research problem include: traffic configuration, potential reverse logistics site selection, capacity expansion plan selection, transportation risk bearing unit’s compensation costs, population density, contamination scope of harmful substances, etc. The uncertainty occurred in customer demand quantity, recycling rates, transportation risk rates. This study verified the applicability of the two-stage stochastic programming model and then compared with the model of certainty. The result shows the green supply chain that considers the uncertainty of demand, recycling rates and transportation risk rates is more robust. The analysis results in this study shall serve as reference for deices-makers when deciding on traffic configuration, reverse logistics site, and capacity expansion with changes in transportation risks.