Article Preview
TopIntroduction
Decision Support Systems (DSS) are computer-based interactive systems that support human decision makers through the use of data and models (Eom et al., 1998, p. 109). Decision Support Systems consist of three parts: data base, model base and user interface (Eom & Lee, 1994, p. 335). Data base contains the input data used in decision making. Model base contains possibly several analytic or simulation models that use input data to produce suggested optima. The decision maker interacts with data and models through the User Interface.
Decision Support System research has been ongoing since the 1970s and one of the first areas of application was logistics systems optimization (Eom & Lee, 1990a, p. 67). Literature contains several reports from the 1980s where small minicomputers have been programmed to solve mixed integer programs involving distribution networks (Geoffrion, Graves, & Lee, 1982). Supply Chain optimization has since moved to MS/OR domain where the focus has been in designing better algorithms (Thomas & Griffin, 1996). Decision Support System research requires that three parts are explored in detail: a semi-structured or unstructured problem; human-computer interface and the nature of computer-based support given to humans to make decision on the problem; the system landscape required to support the decisions (Eom & Lee, 1990b). As such DSS research has produced a plethora of research papers on design science and uses of DSS in specific applications (Eom & Lee, 1990a).
Recently, DSS research has been facing a crisis of confidence (Arnott & Pervan, 2005, p. 82). Professional relevance has suffered since DSS research is not addressing Data Warehousing and Business Intelligence adequately. The gap between DSS theory and practice is widening, and a need exists for case studies, especially interpretive ones, to bridge the gap (Arnott & Pervan, 2005).
An area of DSS research which has not received enough attention is the validation of results given by a particular DSS against the actual values. Many DSS’s are only haphazardly validated against outputs that are produced by the real-life system (Adelman, 1991; Kleijnen, 1995). Showing that the DSS results may be validated through a company’s own reporting system adds credibility to the decision support system (Kleijnen, 1995, p. 159). Literature on Supply Chain Management research suggests that supply chain cost is among the most important performance measures along with order fulfillment lead time and delivery performance (Gunasekaran & Kobu, 2007; Gunasekaran, Patel, & McGoughey, 2004; Gunasekaran, Patel, & Tirtiroglu, 2001). Measurement of non-financial metrics such as cycle times and order fulfillment lead times is common in the industry today (Gunasekaran & Kobu, 2007; Gunasekaran, Patel, & McGoughey, 2004; Gunasekaran, Patel, & Tirtiroglu, 2001). Still, the effect of Supply Chain Management on financial metrics is discussed only on high-level, with e.g., cash flow and operating profit (March & Hevner, 2007). The problem is that only high level executives possess information on cash flow, liquidity projections and ratios; Operational level managers typically do not have a way of assessing the impact of their day-to-day decisions on the company’s aggregate financial measures (Gunasekaran & Kobu, 2007; Gunasekaran, Patel, & McGoughey, 2004; Gunasekaran, Patel, & Tirtiroglu, 2001). Without clear visibility to the effects of one’s decisions, people are not optimally attached to their jobs.