اثر به اشتراک گذاری و پیش بینی اطلاعات در زنجیره های تامین صنعتی توانمند شده: مطالعه موردی

This paper models and analyses the effect of information sharing and forecasting on the performance parameters of an actual industrial supply chain consisting of Small-To-Medium sized enterprises. The paper reports on the industrial supply chain studied, which was undergoing a Business Process Re-engineering (BPR) exercise. The aim of the BPR exercise was the streamlining of existing unstructured processes, ultimately culminating in the introduction of an ERP system into the organisation to improve information sharing between the supply chains echelons. The paper reviews previous work in this area and expands this work to address the issues posed by a more complex real industrial example. The model itself has been developed for a complex supply chain structure. This supply chain has multiple customers, distributors and product families, with customers and distributors face differing demand patterns. This model and its associated experimentation highlights the significant benefits achievable through the use of improved information sharing and forecasting techniques on the supply chain performance parameters. Potential total supply chain cost savings of up to 9.7% have been shown, with increased savings occurring with reduced system capacity. The model also quantifies the impact of collaboration between all partners in the study and shows that gains are achievable by all parties in the supply chain.

Supply Chain Management (SCM) is made up of the control of both material and information flow among suppliers, manufacturers, distributors and customers. SCM involves the management of these flows both within and between companies and organisations. However, to coordinate the supply chain, it is necessary for these supply chain partners to share information. It is widely recognised that advances in technologies in the areas of information, manufacturing, and distribution systems have driven much change through the supply chain and logistics management services. This has particularly been the case with improving information technology enabling instantaneous global information sharing with more powerful information processing. Traditionally, the management of information has been somewhat neglected. The method of information transfer and forecasts carried out by members of the supply chain consisted of placing orders to the member directly above them, termed their preceding ‘echelon’. This causes many problems in the supply chain. These included excessive inventory holding and shortages, increased lead times and reduced service levels. In addition, increased demand variability or the ‘Bullwhip Effect’, compounded forecasting problems and led to difficulties in echelons further up the supply chain. Thus, as SCM progresses, supply chain managers are realising the need for utilising improved information sharing and forecasting throughout the supply chain in order to remain competitive. This is because companies increasingly operate and compete in an ever-expanding global economy. It is widely recognised from studies in the area that improving information sharing, forecasting and general supply chain collaboration will lead to supply chain gains. For some recent papers advocating these points the interested reader is referred to the following: Yao et al. (2005), Holweg et al. (2005), Lee et al. (2000), Kelle and Akbulut (2005), Chandra and Grabis (2005) and Liberopoulous and Koukoumialos (2005) However, Raghunathan (2001), has suggested that the study by Lee et al. (2000) overestimates the benefit of demand information sharing by the retailer in the two-level supply chain studied. They suggest that the reason for the overestimation is the assumption that the manufacturer uses only the current period's retailer order quantity to forecast that of the next period. However, they do accept that information sharing is useful in this model when there is an element of demand which cannot be predicted by the manufacturer using order history. Things such as promotions, price reductions and advertisements by the retailer can cause these demand uncertainties. Many of the studies found in the literature have analysed simplified supply chains using, in the majority of cases, analytical techniques. While analytical models are computationally efficient, they tend to be highly modified versions of reality in order for them to be tractable. Such models can be envisaged as being restrictive in an industrial setting and are therefore only useful to gain simple insights. For such studies to be of real industrial use, they must have the ability to incorporate the details found in today's detailed supply networks and have the flexibility to change within a changing environment. In other words, such models should be reusable and adaptable by operational personnel on an ongoing basis. The purpose of this study is to explore the effect of utilising information sharing and forecasting in a capacitated supply chain, where minimal forecasting and information had been shared in the past. The model developed is that of a real industrial supply chain, with multiple products and multiple echelons with significant product interactions. In this paper, we consider two information sharing strategies, two forecasting techniques and three capacity levels that were being examined by the case study company in question. In order to model the situation in detail, a computer simulation model of the supply chain has been developed. The aim of this study is to expand existing theoretical studies to include a real-life complex case study in an attempt to provide practitioners with realistic expected performance improvements consequent on initiatives. The next section reviews the related literature. Section 3 formulates the problem and outlines the model. Section 4 details a numerical study with experimentation on information sharing, forecasting and capacity constraints and Section 5 discusses the results and concludes.

This paper presents a simulation model of a supply chain, which has been used to highlight the potential savings from utilising improved forecasting and information sharing techniques in a real industrial study. The work extends earlier analytical studies such as those presented by Gaonkar and Viswnadham (2001) and work carried out on small hypothetical supply chains such as those presented by Zhao et al., 2002a and Zhao et al., 2002b. This study extends these idealised, but useful studies to incorporate real world complexities into the analysis. In this case, the supply chain model was developed using a dedicated simulation package. Such specific modelling packages allow for a more flexible environment where objects representing system parts can be easily manipulated to the industrial supply chain condition. Thus, the building blocks from such a model can be manipulated to address numerous extensions to the current system in relation to differing and diverse supply chain questions and structures. With this in mind, this work extracts useful techniques from the study, which can be applied to a complex industrial systems and quantifies the benefits from such strategies. It is felt that this study is not only useful in the single case as presented in this paper, but also in the more general sense as the results here correlate with those results published in relation to these more restrictive models. The model presented here incorporates multiple products flowing through multiple echelons with capacity constraints in a complex network structure. Each of these products has different processing times and has detailed reordering policies and bills of material. This study provides practitioners with further proof that there is benefit in using improved forecasting and information sharing techniques, but does so with real industrial data in a complex supply chain structure. It can be seen from the information given in this particular study that both the distributors and Company X have benefited significantly in all cases examined and in most cases approximately equally, thus providing motivation for all participants in the supply chain to adopt and cooperate with improved information sharing and forecasting techniques and processes. From analysis carried out on the results from this study it was found that the information sharing technique has a greater impact on the supply chain performance particularly in relation to the potential supply chain cost savings. The distributors experienced up to 9.7% savings (equating to €5 million) and Company X experienced up to 6.3% savings (equating to €3.5 million) using an improved periodic information sharing technique. It should be noted that these potential savings were achieved on the system operating with a CT of 1.05. When additional capacity was added to the model (using a CT of 1.33), savings of €1.3 million and €1.6 million for the distributors and company, respectively, were achieved using the same policies as above. This point highlights the fact that more capacitated systems benefit to a larger extent from improved information sharing. However, it should also be noted that in this case it was assumed that there was no cost associated with this extra capacity. In other words, the resources were already in place and could be used at no additional cost, which in general will not be the case. Although such conclusions have been drawn in the past they have only been drawn on more idealised supply chain structures. This study provides results for a real case study thus providing encouragement and potential incentives for companies to pursue such strategies in the real world. This study has taken an industrial example and has quantified the impact of collaboration between partners in relation to information sharing and the use of intelligent forecasting using this improved information. It can be seen from this study that all parties have benefited with the distributors coming out slightly better when information is shared instantaneously, thus encouraging distributors to share such information. This is primarily due to the timely receiving of orders and hence the reduction in backorders due to the production being triggered by actual customer demand.