Decision Support Systems - Assignment Example

DECISION SUPPORT SYSTEMS INTRODUCTION Decision making is an integral part of our day-to-day life. Everyday, we take lot of decisions which are simple or complex in nature depending on the problem. But we all know how difficult it is to take a decision when the stakes associated with it are higher. Many studies have shown that human judgment and decision making capability can be far from perfect, and it deteriorates even further with increase in complexity and stress. Scholars have been trying to come up with methods to aid human decision making since long. With the advent of computer technology, many techniques based on cognitive psychology and artificial intelligence have been tried in the form of computer programs. They are being used either as stand-alone tools or as integrated computing environments for complex decision making. Such techniques are collectively known as decision support systems (DSSs). Thus decision support systems are the techniques and methods which help the humans to take an optimal decision. In a paper on history of decision support systems, Power (2007) provides the growth of these systems since last 40 years (from late 1960s to 2000s) He provides five broad categories of DSS models based on the technology -- communications-driven, data-driven, document driven, knowledge-driven and model-driven decision support systems. While at the technical level, Power (1997) differentiates between enterprise-wide DSS (This is linked to large data warehouses and serves many managers) and desktop DSS (This is a single-user DSS that resides on an individual manager’s computer system) In the following section, we present different works done by different authors on Decision support systems and try to evaluate them. LITERATURE REVIEW: In an article on Decision Support Systems, Druzdzel and Flynn (2002) provide a comprehensive discussions on the use and components of DSS. Their study focuses on the core of a DSS that directly supports modeling decision problems and identifies the best alternatives. They discuss the different types and characteristics of decision problems and try to explain and find out how these decisions can be supported through the use of computer programs. Various component which define DSS and their roles are also discussed while focusing on normative systems (i.e., DSSs based on sound theoretical principles), and decision-analytic DSSs. They have defined three major components of decision making models -- : a measure of preferences over decision objectives, available decision options, and a measure of uncertainty over variables in influencing the decision and the outcomes. They proposed Decision support systems to overcome this problem of uncertainty in decision making process. They defined the three components of DSS as – database management systems, model based management systems and dialog based management systems. According to this paper, the Normative systems which are based on graphical probabilistic models have been proven to show reasonable results in handling the human judgmental biases and provide a theoretically correct and appealing way of handling uncertainty and preferences in decision problems. They go on to argue that the user interface is the most important part of a DSS and unclear and difficult user interfaces result in unpopular DSSs which are not used by the managers at all. But the above authors did not consider the actual manufacturing environment and provided just a theoretical framework. Most of the early works on decision support systems were done on the manufacturing sector and specially in production planning and control areas. Actually, this is the area where most of the complex decisions need to be taken by managers and that affect the whole performance of a company. Major decisions such as expansion and diversification plans of an organization are taken based on the economic feasibility studies which evaluate the performance, functioning and design of the manufacturing system. In 2003, a major study on Manufacturing system design, with flexibility and related issues was undertaken (Choudhary, Bopanna and Kanda, 2003), and they worked on the use of decision support systems to select the optimal solution from a wide variety of available system configurations and control strategy alternatives in the light of several criteria (flexibility, quality, productivity, costs etc.), many of which are difficult to quantify. Such decisions are difficult to make because of the high stakes involved and the uncertain and changing market conditions. Accommodating the fluctuations in the product and market demand requires a flexible approach to manufacturing. Thus various manufacturing flexibilities need to be identified and measured to be able to improve performance in the critical areas. Such critical decisions require multi criteria decision making (MCDM). This multi-criteria decision making is a process by which a selection foe a course of action is taken from amongst the alternatives which are mutually conflicting in an unstructured and complex environment and the alternatives are said to be conflicting when satisfying one criteria leads to non meeting of the other. Thus they developed a Flexible manufacturing Decision Support System (FMDSS) by combining integrated manufacturing performance measurement, multi-criteria evaluation and ranking, and a knowledge based expert system approaches to help high level decision making process. They identified data, model and dialogue bases as the major components of Flexible manufacturing DSS as was done by Druzdzel and Flynn in 2002 in their paper. This was thus a combination of communication driven and knowledge driven DSS. Evan levy discusses the alternatives for Data Integration within the scope of data driven warehousing systems.(2005) he argued that requiring a centralized platform for integrating systems is not mandatory. Though it was the most prevalent system, it was not necessarily simplified or time effective. For these situations, he advocated the use of distributed database architecture. He defined two categories of questions such as Detailed Inquiry. and Content-Bound Analytics and emphasized that they required only a fraction of record to be fetched from the database and not the whole system which was being done at that time. He was one of the first to advocate the use of EAI and EII and SQL databases and Query system to handle small queries and to identify which data to move and find that data at its operational source. Marc Demarest discussed the importance of data legibility as an important concept in DSS (2006). This article utilizes the concept of data-driven decision support systems. He argued that DSS environments must provide readable, clear, and navigable data to the end user otherwise it will not be useful to him. And to do so, the organization needs to adopt or formulate a data legibility strategy such as reschematization or aliasing. The choice of strategy will depend on the requirements of the organization and to the skill set of end user to some extent. To qualify as legible data, it should be easily understood, assimilated, and manipulated. He summarizes that it is more preferable to use reschematization strategy to produce legible data as it is easier to deploy and more cost effective. This article was the first systematic treatment of the concept of multi-dimensional schema (MDS), also known as data cubes or star schema, and the relationship between MDS design and large-scale decision support. In 1994 and then subsequently in 2004, the relationship between GIS (geographical information systems) and DSS was explored and its use was defined in Spatial Decision support systems (SDSS) (Keenan, 2004). The easy availability and wide spread use of spatial data has led to the increasing interest and popularity of GIS. Since GIS can assist in the collection or organization of spatial data used by the decision maker, some authorts described it as another DSS. But this is not true. GIS does help in providing data but this is true only for relatively structured spatial information. Thus in this article, Keenan discusses the use of Geographic Information System (GIS) as a Decision Support System (DSS) generator to create Spatial Decision Support systems. He identified three categories of users who can use GIS and SDSS for decision making process. One such use is in geology, forestry, and land planning. Second area is routing or location analysis while demographic data can find good use in marketing and advertising fields. Thus we can see that the relevance of spatial information as a component of the information needed for a wide range of decisions is constantly increasing and it is special since it will allow decision makers to incorporate a spatial dimension in their decision making. Research on Web based Decision support systems started in 1990s and Power (1998) defined a Web-based decision support system as a computerized system that delivers information or data which aids decision making using a "thin-client" Web browser like Netscape Navigator or Internet Explorer. In 2002 , he published his paper on DSS using web based systems in which he discusses the use, concept and building of such systems.(Power and Kaparthi) In this paper, he explores DSS within the framework of five categories – models driven, data driven, communications driven, document driven and knowledge driven. Since now most firms have established intranets and have web based data warehouses, it makes sense to build web based DSS as well. They argue that user interface become quite important as many users with different levels of skills are using these DSS. This approach is similar to that given by Druzdzel and Flynn (2002). They also indicated the importance of user interfaces in DSS. They go on to discuss the use and importance of newer web technology such as HTML, XML and Java in developing tools for DSS. The web based DSS is advantageous in the sense that it allows enterprise wide DSS to be implemented at a low cost at geographically dispersed locations and it is quite a useful tool for mobile managers since it can be accessed form anywhere. CONCLUSION Thus we see how the concept of decision support systems has developed over the years and moved form one technology to another thereby increasing its advantages and usage day by day. During the last decade the evolutionary change of those models is related with the better insight and growth in new technologies in decisions to be taken with respect to interventions and the acceptance of large uncertainties in the external environment. The future is pointing towards a big change with most emphasis on the predictive mode such that the decision support systems will integrate the monitoring data in its ambience and the use of pre-established scenarios will also be possible. According to a survey, computer-based DSS are widely applied in both profit making (about 72 per cent) and non-profit organizations (about 28 per cent). Its use is widespread in corporate functional management fields such as production and operations management, management information systems, marketing, finance, strategic management and multifunctional areas. And with organizations going global, many organizations will increasingly design and implement group/organizational/global DSS. Global management support systems (MSS) will emerge as a key element in management decision making and as an essential weapon against global competitors. References Chowdary, Bopanna and Kanda, (2003), Decision Support System for Flexibility in Manufacturing, Global Journal of Flexible Systems Management. Demarest, M., (2006) Improving Data Legibility in Decision Support Systems, DSSResources.com. Druzdzel, M., Flynn, R. (2002) Decision Support Systems, Encyclopedia of Library and Information Science, Second Edition, Allen Kent (ed.), New York: Marcel Dekker, Inc Keenan, P. (1996) "Using a GIS as a DSS Generator", in Perspectives on DSS, edited by J. Darzentas, J. S. Darzentas and T. Spyrou, University of the Aegean, Greece, pp 33-40. Updated in 2004 Levy, Evan, (2005) Beyond the Data Warehouse: Architectural Alternatives for Data Integration, DSSResources.com. Power, D. and Kaparthi, S. (2002) Building Web based decision Support SystemsInformatics and Control, Vol. 11, No.4 Read More