توسعه خطر سفارش معوقه چارچوب برنامه ریزی تکمیل دوباره پویا بر اساس استفاده از شبکه اعتقاد بیزی
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|23011||2012||8 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Computers & Industrial Engineering, Volume 62, Issue 3, April 2012, Pages 716–725
Due to the rapid changes in circumstances of cooperates such as globalization, technical innovation and competition, inter-dependence among cooperates which compose supply chain has been intensified. This make cooperates be exposed to various risk and even a small uncertainty can disrupt the balance of whole supply chain. Therefore, in this paper, the framework to develop alternative backorder replenishment plan to minimize the total replenishment cost and expected risk cost has been devised. In order to model the relationship between risks and risk propagation, Bayesian Belief Network has been applied. Moreover, with the fast heuristic algorithm, breath first search and elementary stepwise system based reverse Dijkstra, the alternative backorder replenishment plan can be established. The numerical example shows how to apply the proposed framework and make dynamic backorder replenishment plan considering impact of risk.
Recently, supply chain management becomes more complex in the phase of planning and operation due to the integration and globalization. Therefore, the academia and the industry have shown a growing interest for efficient supply chain management, which come from the rising of manufacturing and transportation cost and the globalization of market economics. These business trends are leading to complex, dynamic supply network. One consequence is that uncertainties are increasing, and shifting around supply network (Harland, Brenchley, & Walker, 2003). Also, the increasing uncertainty requires companies to spend more resources to anticipate for demand, supply, as well as internal uncertainties for better sustainability of their supply chain. Interestingly, such an increasing uncertainty is not solely induced by the external business environments, but also due to the increasing complexity of the supply chain structure and variety of mechanisms initiated by the supply chains in their business (Vanany & Pujawan, 2009). These uncertainties and factors that have negative impact on the business outcomes can be defined as the supply chain risk. Accordingly, the supply chain risk management (SCRM) is rapidly developing into a favoured research area for academicians as well as practitioners in global environment (Manuj & Mentzer, 2008), and it was introduced in the Supply Chain Operation Reference (SCOR) model since the version 8.0 (Supply-Chain Council, 2008). But, these business trends are not the only reason that SCRM has been highlighted. During military operations, it is the vital necessary condition for the victory to design the optimal strategies to flexibly cope with the risk factors like designing optimal route of arms or war supplies. The Gulf war is the most typical example (Matthews & Holt, 1995) . A great deal of researches proposed risk management strategies to develop the optimal counter plans for the supply chain operations in the planning phase to deal with the independently identified risk. And the ultimate object, to enhance the flexibility of the supply chain, can be accomplished by reducing the probability of occurrence, the severity of impact, or both. Here, the more interesting characteristic of supply chain risk not to be ignored is that the risk shifts around the supply network. Also, the supply chain linkage has great impact on the supply chain performance such as benefit and risk reduction (Zelbst, Jr, Sower, & Reyes, 2009). Thus, with the increased complexity of supply chain network, the location of risk has shifted and changed through the whole networks (Harland et al., 2003). The case of Daimler Chrysler may be the most famous practice. It was Hurricane “Floyd” which flooded a plant producing suspension parts in Greenville, North Carolina. As a result, seven of the company’s other plants across North America had to be shut down for seven days (McGillivray, 2000). However, a few researches to proactively cope with risk in the execution phase have been conducted in the limited area such as inventory management considering dynamics of demand (Clark and Scarf, 1960, Iglehart and Karlin, 1961, Kim et al., 2005, Lingxiu and Hau, 2003 and Song and Zipkin, 1993) or coping with disruption risk (Altay and Ramirez, 2010 and Skipper and Hanna, 2009). Therefore, more attention and effort need to be paid to the execution phase of the supply chain management. Especially, impact or expected loss over the entire supply chain when the risk breaks out should be understood. In addition, both internal and external risk should be controlled in the balance. The main objective of this paper is to propose a novel risk management approach that considers the inter-relationship between supply chain risk and the structure of network at the same time. The risk propagation is modelled by using Bayesian Belief Network (BBN), which is widely applied to various areas such as biology, genetics, medical science, enterprise risk management and financial engineering (van der Gaag, 1996). With the backorder replenishment problem, we demonstrate how to practically apply the proposed approach and reduce the expected cost from supply chain risk. The rest of this paper is organized as follows. In the Section 2, basic definition, concepts and previous researches related to BBN and SCRM are introduced and analysed. And then we explain how to model the risk propagation with BBN, to develop dynamic backorder replenishment problem, and to design a heuristic algorithm to find the optimal solution in Section 3. The numerical example is presented to explain the way to apply the proposed framework in Section 4. Finally, we conclude with the contributions and further work in Section 5.
نتیجه گیری انگلیسی
In this paper, we have proposed a new approach of modelling the risk propagation using BBN. The BBN describes a set of causal relationship among sets of conditional independence assumption, and their related joint probabilities. Furthermore, BBN applies the rules of Bayesian inference to propagate the impact of evidence on the probabilities of selected outcomes. In practical supply chain, it is general characteristics that risk is not independent, shifts around the supply chain network and may have impact on the other nodes or arcs that are not associated. We defined this as the risk propagation and modelled it with KRI-BBN by taking advantage of Bayesian inference function of BBN. Also, we described how to develop KRI-BBN and to integrate it into supply chain network in details. To show the way to apply the proposed approach to manage supply chain operation, a backorder replenishment problem, one of the most well-known and important supply chain operations, considering risk cost is developed. With numerical example, we have provided the way to proactively manage risks in supply chain operation. The proposed risk management framework may overcome the limitations of previous researches, the most primitive assumptions that risk is independent and have impact bounded in limited area, only the associated nodes or arcs. In addition, it is able to directly reflect risk measured in real-time to practical supply chain operations, even in the execution phase. Ultimately, it can help managers to make rational decision by suggesting alternatives. In addition, it makes not only the stable supply chain operations possible but also accuracy and efficiency of supply chain plan higher. Nevertheless, some assumptions which have been made for the problem modelling such as single commodity, unlimited transportation capacities are far from the real world. Therefore, these assumptions should be relaxed and another methodology to find the near-optimal solution very fast using Meta-heuristics should be devised.