Decision Support Systems
Decision Support Systems
Broadly speaking, decision support systems are a set of manual or computer-based tools that assist in some decision-making activity. In today's business environment, however, decision support systems (DSS) are commonly understood to be computerized management information systems designed to help business owners, executives, and managers resolve complicated business problems and/or questions. Good decision support systems can help business people perform a wide variety of functions, including cash flow analysis, concept ranking, multistage forecasting, product performance improvement, and resource allocation analysis. Previously regarded as primarily a tool for big companies, DSS has in recent years come to be recognized as a potentially valuable tool for small business enterprises as well.
THE STRUCTURE OF DECISIONS
In order to discuss the support of decisions and what DSS tools can or should do, it is necessary to have a perspective on the nature of the decision process and the various requirements of supporting it. One way of looking at a decision is in terms of its key components. The first component is the data collected by a decision maker to be used in making the decision. The second is the process selected by the decision maker to combine this data. Finally, there is an evaluation or learning component that compares decisions and examines them to see if there is a need to change either the data being used or the process that combines the data. These components of a decision interact with the characteristics of the decision being made.
Many analysts categorize decisions according to the degree of structure involved in the decision-making activity. Business analysts describe a structured decision as one in which all three components of a decision—the data, process, and evaluation—are determined. Since structured decisions are made on a regular basis in business environments, it makes sense to place a comparatively rigid framework around the decision and the people making it.
Structured decision support systems may simply use a checklist or form to ensure that all necessary data are collected and that the decision making process is not skewed by the absence of data. If the choice is also to support the procedural or process component of the decision, then it is quite possible to develop a program either as part of the checklist or form. In fact, it is also possible and desirable to develop computer programs that collect and combine the data, thus giving the process a high degree of consistency or structure. When there is a desire to make a decision more structured, the support system for that decision is designed to ensure consistency. Many firms that hire individuals without a great deal of experience provide them with detailed guidelines on their decision making activities and support them by giving them little flexibility. One interesting consequence of making a decision more structured is that the liability for inappropriate decisions is shifted from individual decision makers to the larger company or organization.
At the other end of the continuum are unstructured decisions. While these have the same components as structured ones—data, process, and evaluation—there is little agreement on their nature. With unstructured decisions, for example, each decision maker may use different data and processes to reach a conclusion. In addition, because of the nature of the decision there may only a limited number of people within the organization qualified to evaluate the decision.
Generally, unstructured decisions are made in instances in which all elements of the business environment—customer expectations, competitor response, cost of securing raw materials, etc.—are not completely understood (new product and marketing strategy decisions commonly fit into this category). Unstructured decision systems typically focus on the individual who or the team that will make the decision. These decision makers are usually entrusted with decisions that are unstructured because of their experience or expertise; it is their individual ability that is of value. One approach to support systems in this area is to construct a program that simulates the process used by a particular individual. In essence, these systems—commonly referred to as "expert systems"—prompt the user with a series of questions regarding a decision situation. "Once the expert system has sufficient information about the decision scenario, it uses an inference engine which draws upon a data base of expertise in this decision area to provide the manager with the best possible alternative for the problem," explained Jatinder N. D. Gupta and Thomas M. Harris in the Journal of Systems Management. "The purported advantage of this decision aid is that it allows the manager the use of the collective knowledge of experts in this decision realm. Some of the current DSS applications have included long-range and strategic planning policy setting, new product planning, market planning, cash flow management, operational planning and budgeting, and portfolio management."
Another approach is to monitor and document the process used so that the decision maker(s) can readily review what has already been examined and concluded. An even more novel approach used is to provide environments specially designed to give these decision makers an atmosphere conducive to their particular tastes. The key to support of unstructured decisions is to understand the role that individuals experience or expertise plays in the decision and to allow for individual approaches.
In the middle of the continuum are semi-structured decisions—where most of what are considered to be true decision support systems are focused. Decisions of this type are characterized as having some agreement on the data, process, and/or evaluation to be used, but are also typified by efforts to retain some level of human judgment in the decision making process. An initial step in analyzing which support system is required is to understand where the limitations of the decision maker may be manifested (i.e., the data acquisition portion, the process component, or the evaluation of outcomes).
Grappling with the latter two types of decisions—unstructured and semi-structured—can be particularly problematic for small businesses, which often have limited technological or work force resources. As Gupta and Harris indicated, "many decision situations faced by executives in small business are one-of-a-kind, one-shot occurrences requiring specifically tailored solution approaches without the benefit of any previously available rules or procedures. This unstructured or semi-structured nature of these decisions situations aggravates the problem of limited resources and staff expertise available to a small business executive to analyze important decisions appropriately. Faced with this difficulty, an executive in a small business must seek tools and techniques that do not demand too much of his time and resources and are useful to make his life easier." Subsequently, small businesses have increasingly turned to DSS to provide them with assistance in business guidance and management.
KEY DSS FUNCTIONS
Gupta and Harris observed that DSS is predicated on the effective performance of three functions: information management, data quantification, and model manipulation. "Information management refers to the storage, retrieval, and reporting of information in a structured format convenient to the user. Data quantification is the process by which large amounts of information are condensed and analytically manipulated into a few core indicators that extract the essence of data. Model manipulation refers to the construction and resolution of various scenarios to answer 'what if' questions. It includes the processes of model formulation, alternatives generation and solution of the proposed models, often through the use of several operations research/management science approaches."
Entrepreneurs and owners of established enterprises are urged to make certain that their business needs a DSS before buying the various computer systems and software necessary to create one. Some small businesses, of course, have no need of a DSS. The owner of a car washing establishment, for instance, would be highly unlikely to make such an investment. But for those business owners who are guiding a complex operation, a decision support system can be a valuable tool. Another key consideration is whether the business's key personnel will ensure that the necessary time and effort is spent to incorporate DSS into the establishment's operations. After all, even the best decision support system is of little use if the business does not possess the training and knowledge necessary to use it effectively. If, after careful study of questions of DSS utility, the small business owner decides that DSS can help his or her company, the necessary investment can be made, and the key managers of the business can begin the process of developing their own DSS applications using available spreadsheet software.
DSS UNCERTAINTIES AND LIMITATIONS
While decision support systems have been embraced by small business operators in a wide range of industries in recent years, entrepreneurs, programmers, and business consultants all agree that such systems are not perfect.
Level of "User-Friendliness"
Some observers contend that although decision support systems have become much more user-friendly in recent years, it remains an issue, especially for small business operations that do not have significant resources in terms of technological knowledge.
Another limitation that decision makers confront has to do with combining or processing the information that they obtain. In many cases these limitations are due to the number of mathematical calculations required. For instance, a manufacturer pondering the introduction of a new product can not do so without first deciding on a price for the product. In order to make this decision, the effect of different variables (including price) on demand for the product and the subsequent profit must be evaluated. The manufacturer's perceptions of the demand for the product can be captured in a mathematical formula that portrays the relationship between profit, price, and other variables considered important. Once the relationships have been expressed, the decision maker may now want to change the values for different variables and see what the effect on profits would be. The ability to save mathematical relationships and then obtain results for different values is a feature of many decision support systems. This is called "what-if" analysis, and today's spreadsheet software packages are fully equipped to support this decision-making activity. Of course, additional factors must be taken into consideration as well when making business decisions. Hard-to-quantify factors such as future interest rates, new legislation, and hunches about product shelf life may all be considered. So even though the calculations may indicate that a certain demand for the product will be achieved at a certain price, the decision maker must use his or her judgment in making the final decision.
If the decision maker simply follows the output of a process model, then the decision is being moved toward the structured end of the continuum. In certain corporate environments, it may be easier for the decision maker to follow the prescriptions of the DSS; users of support systems are usually aware of the risks associated with certain choices. If decision makers feel that there is more risk associated with exercising judgment and opposing the suggestion of the DSS than there is in simply supporting the process, the DSS is moving the decision more toward the structured end of the spectrum. Therefore, the way in which a DSS will be used must be considered within the decision-making environment.
Processing Model Limitations
Another problem with the use of support systems that perform calculations is that the user/decision maker may not be fully aware of the limitations or assumptions of the particular processing model. There may be instances in which the decision maker has an idea of the knowledge that is desired, but not necessarily the best way to get that knowledge. This problem may be seen in the use of statistical analysis to support a decision. Most statistical packages provide a variety of tests and will perform them on whatever data is presented, regardless of whether or not it is appropriate. This problem has been recognized by designers of support systems and has resulted in the development of DSS that support the choice of the type of analysis.
Carlson, John R., Dawn S. Carlson, and Lori L. Wadsworth. "On the Relationship Between DSS Design Characteristics and Ethical Decision Making." Journal of Managerial Issues. Summer 1999.
Chaudhry, Sohail S., Linda Salchenberger, and Mahdi Beheshtian. "A Small Business Inventory DSS: Design, Development, and Implementation Issues." Computers & Operations Research. January 1996.
Gupta, Jatinder N. D., and Thomas M. Harris. "Decision Support Systems for Small Business." Journal of Systems Management. February 1989.
Kimball, Ralph, and Kevin Strahlo. "Why Decision Support Fails and How to Fix It." Datamation. 1 June 1994.
Kumar, Ram L. "Understanding DSS Value." Omega. June 1999.
Laudon, Kenneth C., and Jane Price Laudon. Management Information Systems: A Contemporary Perspective. Macmillan, 1991.
Muller-Boling, Detlef, and Susanne Kirchhoff. "Expert Systems for Decision Support in Business Start-Ups." Journal of Small Business Management. April 1991.
Parkinson, Chris. "What If? Decision Shaping Systems." CMA—The Management Accounting Magazine. March 1995.
Raggad, Bel G. "Decision Support System: Use It or Skip It." Industrial Management and Data Systems. January 1997.
Raymond, Louis, and Francois Bergeron. "Personal DSS Success in Small Enterprises." Information and Management. May 1992.
Hillstrom, Northern Lights
updated by Magee, ECDI
Decision Support Systems
Decision Support Systems
DECISION SUPPORT SYSTEMS FIRST TESTED IN PORTFOLIO MANAGEMENT
DECISION SUPPORT SYSTEMS IN PORTFOLIO MANAGEMENT
PORTFOLIO MANAGEMENT DSS IN PRACTICE: AN ILLUSTRATION OF THE INVESTOR SYSTEM
Decision support systems (DSS) are computer information systems that perform complex data analysis in order to help users make informed decisions. In general, a DSS retrieves information from a large data warehouse, analyzes it in accordance with user specifications, then publishes the results in a format that users can readily understand and use. DSS applications are interactive, and they are valuable in a wide range of business settings.
According to ComputerWorld Magazine, the rise of the decision support system can be traced to the late 1960s when businesses started to use computer mainframes. These mainframe computers first enabled businesses to interactively query data so they could enhance their previously-static reports. Other experts generally agree that computerized decision support systems became practical with the development of minicomputers, timeshare operating systems, and distributed computing.
In the 1970s, however, decision support systems experienced a huge boom. Query systems, what-if spreadsheets, and rules-based software were developed. The advent of packaged algorithms made it easier to get better, faster decisions. As technology evolved, new computerized decision support applications were developed and studied. Researchers used multiple frameworks to help build and understand these systems. Today, one can organize the history of DSS into the five broad DSS categories, including: communications-driven, data-driven, document driven, knowledge-driven, and model-driven decision support systems.
Trends in all these categories are emerging. Data-driven DSS continuously use faster, real-time access to larger, better integrated databases. Trends suggest that model-driven DSS will grow more complex. Systems built using simulations and accompanying visual displays are becoming increasingly realistic. Communications-driven DSS provide more real-time video communications support. Document-driven DSS access larger repositories of data; the systems present appropriate documents in more useable formats. Finally, knowledge-driven DSS are usually more sophisticated and comprehensive. The advice from knowledge-driven DSS is often considered better, and the applications cover broader domains.
Technology advances continue to make it easier and more efficient to collect relevant data. However, collecting, analyzing, correlating, and applying these massive amounts of data pose a challenge to businesses. Even so, companies are eager to respond in real-time to customer queries. They strive to anticipate customer needs, create opportunities, and avoid potential problems, for the end goal is to establish a predictive business.
The airline industry provides a good example of using data to instantaneously respond to customer queries. In the past, most customers called the airlines to purchase their airline tickets—a process that typically took about twenty minutes. That all changed with Web transactions, which can provide more information, more quickly. Ultimately, these types of DSS enable customers to book a ticket in just a few minutes.
With decision support systems, companies correlate information about their operations and performance with information about expected behavior and business rules. Decision makers anticipate and respond to threats and capitalize on opportunities before they occur. This ability makes predictive business, which is considered the next step in the evolution of a real-time enterprise, a reality.
Decision support systems were first tested in portfolio management, which poses one of the most essential problems in modern financial theory. It involves the construction of a portfolio of securities (stocks, bonds, treasury bills, etc.) that maximizes the investor's utility. The process leading to the construction of such a portfolio consists of two major steps. In the first step, the decision-maker (investor, portfolio manager) has to evaluate the securities that are available as investment instruments. The vast number of available securities, especially in the case of stocks, makes this step necessary, in order to focus the analysis on a limited number of the best investment choices. Thus, on the basis of this evaluation stage, the decision-maker selects a small number of securities that constitute the best investment opportunities. In the second step of the process, the decision maker must decide on the amount of the available capital that should be invested in each security, thus constructing a portfolio of the selected securities. The portfolio should be constructed in accordance with the decision-maker's investment policy and risk tolerance.
The portfolio theory assumes that the decision-maker's judgment and investment policy can be represented by a utility function that is implicitly used by the decision-maker
in making his investment decisions. Thus, the maximization of this utility function will result in the construction of a portfolio that is as consistent as possible with the decision-maker's expectations and investment policy. However, it is quite difficult to determine the specific form of this utility function.
The founder of portfolio theory, Harry Markowitz (recipient of the 1990 Nobel prize in economics), developed a framework according to which the decision-maker's utility is a function of two variables: the expected return of the portfolio and its risk. Thus, he formulated the maximization of the decision-maker's utility as a two-objective problem: maximizing the expected return of the portfolio and minimizing the corresponding risk. To consider the return and the risk, Markowitz used two well-known statistical measures, the mean of all possible returns to estimate the return of the portfolio, and the variance to measure its risk. On the basis of this mean-variance framework, Markowitz developed a mathematical framework to identify the efficient set of portfolios that maximizes returns at any given level of allowable risk. Given the risk aversion policy of the investor, it is possible to select the most appropriate portfolio from the efficient set.
This pioneering work of Markowitz motivated financial researchers to develop new portfolio management techniques, and significant contributions have been made over the last decades. The most significant of the approaches that have been proposed for portfolio management include the capital asset pricing model (CAPM), the arbitrage pricing theory (APT), single- and multi-index models, as well as several optimization techniques.
The concept of decision support systems (DSS) was introduced, from a theoretical point of view, in the late 1960s. DSS can be defined as computer information systems that provide information in a specific problem domain using analytical decision models and techniques, as well as access to databases, in order to support a decision maker in making decisions effectively in complex and ill-structured problems. Thus, the basic goal of DSS is to provide the necessary information to the decision-maker in order to help him or her get a better understanding of the decision environment and the alternatives available.
A typical structure of a DSS includes three main parts: the database, the model base, and the user interface. The database includes all the information and data that are necessary to perform the analysis on the decision problem at hand. Data entry, storage, and retrieval are performed through a database management system. The model base is an arsenal of methods, techniques, and models that can be used to perform the analysis and support the decision maker. These models or techniques are applied to the raw data in order to produce analysis or more meaningful output for the decision maker. A model base management system is responsible for performing all tasks that are related to model management, such as model development, updates, storage, and retrieval. Finally, the user interface is responsible for the communication between the user and the system, while it further serves as a link between the database and the model base. The appropriate design of the user interface is a key issue towards the successful implementation of the whole system, so as to ensure that the user can take full advantage of the analytical capabilities that the system provides. Advances in computer hardware and software have enabled user-friendly graphical user interfaces (GUIs) to serve this function.
During the last four decades, DSS have been developed and implemented to tackle a variety of real world decision-making problems, in addition to financial problems and portfolio management. The portfolio management process involves the analysis of a vast volume of information and data, including financial, stock market, and macroeconomic data. Analyzing a continuous flow of such a vast amount of information for every available security in order to make real time portfolio management decisions is clearly impossible without the support of a specifically designed computer system that will facilitate not only the data management process, but also the analysis.
Thus, the contribution of DSS to portfolio management becomes apparent. They provide an integrated tool to perform real-time analyses of portfolio-management-related data, and provide information according to the decision-maker's preferences. Furthermore, they enable the decision maker to take full advantage of sophisticated analytic methods, including multivariate statistical and econometric techniques, powerful optimization methods, advanced preference modeling, and multiple-criteria decision-making techniques. DSS incorporating multiple-criteria decision-making methods in their structure are known as multicriteria DSS, and they have found several applications in the field of finance.
The Investor system is a DSS designed and developed to support the portfolio management process and to help construct portfolios of stocks. The system includes a combination of portfolio theory models, multivariate statistical methods, and multiple criteria decision-making techniques for stock evaluation and portfolio construction. The structure of the system is presented in Figure 1.
Financial Data. The database of the system includes four types of information and data. The first involves the financial statements of the firms whose stocks are considered in the portfolio management problem. The balance sheet and the income statement provide valuable information regarding the financial soundness of the firms (e.g., sales, net profit, net worth, liabilities, assets, etc.). The system contains such financial data spanning a five-year period, so that users can reach informed conclusions about the firms' financial evolution.
Qualitative Information. In addition to these financial data, information on some qualitative factors is also inserted in the database. The management of the firms, their organization, their reputation in the market, their technical facilities, and their market position affect directly the operation and the performance of the firms; thus, they constitute fundamental factors in the analysis of the firms whose stocks are considered in the portfolio management problem.
Market Data. The third type of information included in the database involves the stocks' market histories. This information involves the stock prices, the marketability of the stocks, their beta (b) coefficient (a measure of risk representing the relationship between the changes in the price of individual stocks with the changes in the market), the dividend yield, the price/earnings ratio, and so forth.
Macroeconomic Data. Finally, information regarding the macroeconomic environment is also included. Inflation, interest rates, exchange rates, and other macroeconomic variables have a direct impact on the performance of the stock market, thus potentially affecting any individual stock. The combination of this information with the financial and stock histories of individual firms enables portfolio managers to perform a global evaluation of the investment opportunities available, both in terms of their sensitivity and risk with respect to the economic environment, and to their individual characteristics.
Analysis Tools. The analysis of all this information is performed through the tools incorporated in the system's model base. Two major components can be distinguished
in the model base. The first one consists of financial and stock market analysis tools. These can analyze the structure of the financial statements of the firms, calculate financial and stock market ratios, apply well-known portfolio theory models (e.g., the market model, the CAPM, the APT), and present several graphical summaries of the results obtained through these tools to facilitate drawing some initial conclusions about the stocks' performance.
The second component of the model base involves more sophisticated analysis tools, including statistical and multiple-criteria decision-making techniques. More specifically, univariate statistical techniques are used to measure the stability of the beta coefficient of the stocks, while principal components analysis (a multivariate technique) is used to identify the most significant factors or criteria that describe the performance of the stocks, and to place the stocks into homogeneous groups according to their financial and stock market characteristics. The criteria identified as most crucial can be used to evaluate the stocks and thereby construct a portfolio that meets the investment policy of the investor/portfolio manager. Of course, the portfolio manager interacts with the system, and he or she can also introduce into the analysis the evaluation criteria that he or she considers important, even if these criteria are not found significant by principal components analysis.
The evaluation of the stocks' performance is completed through multiple-criteria decision-making methods. Multiple-criteria decision-making is an advanced field of operations research that provides an arsenal of methodological tools and techniques to study real-world decision problems involving multiple criteria that often lead to conflicting results. The multiple-criteria decision-making methods that are incorporated in the model base of the investor system enable the investor to develop an additive utility function that is fairly consistent with his or her investment policy, preferences, and experience. On the basis of this additive utility function a score (global utility) is estimated for each stock that represents its overall performance with respect to the selected evaluation criteria. The scores of the stocks are used as an index so they may be placed into appropriate classes specified by the user. Thus, the portfolio manager can develop an evaluation model (additive utility function) to distinguish, for instance, among the stocks that constitute the best investment opportunities, the stocks that do not have a medium-long term prospect but they can be considered only for the short run, and the stocks that are too risky and should be avoided. Of course, any other classification can be determined according to the objectives and the policy of the portfolio manager.
On the basis of this classification, the investor/portfolio manager can select a limited number of stocks to include in the actual portfolio, which represent the best investment opportunities. Constructing the portfolio is accomplished through multiple-criteria decision-making
techniques that are appropriate for optimizing a set of objective functions subject to some constraints. The objective functions represent the investor/portfolio manager's objectives on some evaluation criteria (return, beta, marketability, etc.), while constraints can be imposed to ensure that the constructed portfolio meets some basic aspects of the investment policy of the investor/portfolio manager.
For instance, the investor/portfolio manager can introduce constraints on the amount of capital invested in stocks of specific business sectors, the amount of capital invested in high-risk or low-risk stocks (high and low β coefficient, respectively), to determine a minimum level of return or a maximum level of risk, and so on. Once such details are determined, an interactive and iterative optimization procedure is performed that leads to the construction of a portfolio of stocks that meets the investor's investment policy and preferences. The results presented through the screen of Figure 2 show the proportion of each stock in the constructed portfolio, the performance of the portfolio on the specified evaluation criteria (attained values), as well as the rate of closeness (achievement rate) of the performance of the portfolio as opposed to the optimal values on each evaluation criterion (the higher this rate is, the closer the performance of the portfolio to the optimal one for each criterion).
Since the development of the portfolio theory in the 1950s, portfolio management has gained increasing interest within the financial community. Periodic turmoil in stock markets worldwide demonstrates the necessity for developing risk management tools that can be used to analyze the vast volume of information that is available. The DSS framework provides such tools that enable investors and portfolio managers to employ sophisticated techniques from the fields of statistical analysis, econometric analysis, and operations research to make and implement real-time portfolio management decisions.
DSS research in the twenty-first century has been oriented toward combining the powerful analytical tools used in the DSS framework with the new modeling techniques provided by soft computing technology (neural networks, expert systems, fuzzy sets, etc.) to address the uncertainty, vagueness, and fuzziness that is often encountered in the financial and business environment.
Business intelligence (BI) practices are often cited as key to the evolution of decision support systems. BI refers to the technologies, applications, and practices used for collecting, integrating, analyzing, and presenting business information. It is the variety of software applications used to analyze an organization's raw data and extract useful insights from it. Therefore, like DSS, business intelligence systems are data-driven. They use fact-based support systems to improve business decision-making, making BI a reporting and decision support tool.
Used at the operational level, BI projects have great potential to transform business processes. For example, well-known companies use BI technologies to improve corporate sales and customer service processes. Used correctly, BI systems can transform companies from regionally-operated businesses to unified global businesses.
Like many technological advances, there are obstacles. A key impediment to BI progress is lack of corporate understanding. Often, companies don't know their own business processes well enough to determine how to improve them. Before commencing a BI project, companies must consider and understand all of the activities that make up a particular business process, how information and data flow across various processes, how data is passed between business users, and how people use it to execute their part of the process. In order to motivate upper management to standardize such processes company-wide, BI systems must have a direct impact on revenue.
Implementation of BI systems requires a change in thinking about the value of information inside organizations. Everyone involved in the BI process must have full access to information to be able to change the ways that they work. This necessitates a trusting working environment.
Well-known firm McKinsey Consulting noted that decision support systems were one of eight technology trends to watch in 2008. DSS technologies will advance as more innovative data collection and processing methods are introduced. This, according to McKinsey, will result in more granular segmentation and low-cost experimentation. The resulting information will help managers acquire more data, make smarter decisions, and develop competitive advantages and new business models.
SEE ALSO Competitive Intelligence; Computer-Aided Design and Manufacturing; Computer Networks; Management Information Systems; Strategic Planning Tools
Elton, E.J., and M.J. Gruber. Modern Portfolio Theory and Investment Analysis. 5th ed. New York: John Wiley and Sons, 1995.
Levinson, Meridith. “Business Intelligence: Not Just for Bosses Anymore.” CIO Magazine 15 Jan 2006. Available from: http://www.cio.com/article/16544/Business_Intelligence_Not_Just_for_Bosses_Anymore.
Lundberg, Alan. “Leverage Complex Event Processing to Improve Operational Performance.” Business Intelligence Journal 28 Mar 2006. Available from: http://www.tdwi.org/Publications/BIJournal/display.aspx?ID=7898.
Markowitz, H. Portfolio Selection: Efficient Diversification of Investments. New York: John Wiley and Sons, 1959.
Phillips-Wren, Gloria E., et al., “A Multiple-Criteria Framework for Evaluation of Decision Support Systems.” Omega August 2004.
Power, D.J. “A Brief History of Decision Support Systems.” DSSResources.com 10 Mar 2007. Available from: http://DSSResources.com/history/dsshistory.html,.
“Supply Chain News: McKinsey's Eight Technology Trends to Watch for 2008” Supply Chain News 16 Jan 2008. Available from: http://www.scdigest.com/text_search.php?keyword=decision+support.
Turban, Efraim, et al. Decision Support Systems and Intelligent Systems. 7th ed. Englewood Cliffs, NJ: Prentice Hall, 2004.
Zopounidis, C., M. Godefroid, and C. Hurson. “Designing a Multicriteria DSS for Portfolio Selection and Management.” In Advances in Stochastic Modeling and Data Analysis. edited byJ. Janssen, C.H. Skiadas, and C. Zopounidis. Dordrecht, Netherlands: Kluwer Academic Publishers, 1995, 261–292.
Decision Support Systems
DECISION SUPPORT SYSTEMS
Decision support systems (DSS) are tools meant to assist in human decision-making (Turban and Aronson 2001). In an increasingly complex and rapidly changing world where information from human, software, and sensor sources can be overwhelming, DSS tools can serve as a bridge between the social and technical spheres. DSS tools offer support based on formal, technical approaches, but do so within a context that is often largely socially mediated.
Most DSS tools are assembled out of hardware devices and software constructs. The hardware devices, in the early twenty-first century, are dominated by digital computers and peripherals such as sensors, network infrastructure, and display and alerting devices meant to interact with these. Historically, many DSS were hard-wired to solve a specific task; control systems in nuclear power plants are an example. DSS hardware is increasingly dominated by physically distributed systems that make use of wired and wireless networks to gather and share information from and with remote sources (Shim et al. 2002). The convergence of remote sensing, sensor-networks, and distributed computational grids using the Internet as a foundation in the late 1990s–early 2000's reflects this trend.
The software, or algorithmic, component of DSS derives from historical research in statistics, operations research, cybernetics, artificial intelligence, knowledge management, and cognitive science. In early monitoring decision support systems the algorithms were typically hard-wired into the system, and these systems tended to be unchanging once built. Software-based decision support allows for multiple approaches to be applied in parallel, and for systems to evolve either through new software development or via software that "learns" through artificial intelligence techniques such as rule induction (Turban and Aronson 2001).
When used appropriately, DSS tools are not meant to replace human decision-making—they are meant to make it more effective (Sprague and Watson 1996). DSS tools do this by presenting justified answers with explanations, displaying key data relevant to the current problem, performing calculations in support of user decision tasks, showing related cases to suggest alternatives, and alerting the user to current states and patterns. In order to be a support rather than a hindrance, these tools must be constructed with careful attention to human cognitive constraints. As a result, DSS design is a prime area of human-computer interaction and usability research. In many cases, DSS tools make use of adaptive software interfaces; depending on the situation, different contents will be displayed on the interface, so as not to overwhelm the user with secondary or irrelevant information.
Decision Support Tools
Decision support tools fall into two broad classes: those that operate at the pace of the user (for example, to support planning decisions) and those that operate at or near the pace of real-time world events (such as air traffic control systems). The decision-making domain can be further divided into situations in which the system can be completely and accurately defined (in other words, closed and formal systems) and those where this is not feasible, desirable, or possible. The former is not normally considered a prime situation for decision support because a formal situation can be addressed without human intervention, while the latter requires the hybrid human-machine pairing found in DSS. In the case of open systems, heuristic approximations (rules of thumb) are needed in lieu of formal models; these may also be needed in cases in which a formal model exists but cannot be computed in a reasonable amount of time.
Systems that operate at the pace of the user provide support for such tasks as planning and allocation, medical and technical diagnosis, and design. Typical examples include systems used in urban planning to support the complex process of utility construction, zoning, tax valuation, and environmental monitoring, and those used in business to determine when new facilities are needed for manufacturing. Such tools include significant historical case-knowledge and can be transitional with training systems that support and educate the user. Formal knowledge, often stored as rules in a modifiable knowledge base, represent both the state of the world that the system operates on and the processes by which decisions transform that world. In the cases where formal knowledge of state and process are not available, heuristic rules in a DSS expert system or associations in a neural network model might provide an approximate model. DSS tools typically provide both a ranked list of possible courses of action and a measure of certainty for each, in some cases coupled with the details of the resolution process (Giarratano and Riley 2005).
Systems that operate at or near real time provide support for monitoring natural or human systems. Nuclear power plant, air traffic control, and flood monitoring systems are typical examples, and recent disasters with each of these illustrate that these systems are fallible and have dire consequences when they fail. These systems typically provide support in a very short time frame and must not distract the user from the proper performance of critical tasks. By integrating data from physical devices (such as radar, water level monitors, and traffic density sensors) over a network with local heuristics, a real-time DSS can activate alarms, control safety equipment semi-automatically or automatically, allow operators to interact with a large system efficiently, provide rapid feedback, and show alternative cause and effect cases. A central issue in the design of such systems is that they should degrade gracefully; a flood monitoring system that fails utterly if one cable is shorted-out, for example, is of little use in a real emergency.
History of Decision Support Research and Tools
As indicated above, DSS evolved out of a wide range of disciplines in response to the need for planning-support and monitoring-support tools. Management and executive information systems, where model and data-based systems dominated, reflect the planning need; control and alerting systems, where sensor and model-based alerting systems were central, reflect the monitoring need. The original research on the fusion of the source disciplines, and in particular the blending of cognitive with artificial intelligence approaches, took place at Carnegie-Mellon University in the 1950s (Simon 1960). This research both defined the start of DSS and also was seminal in the history of artificial intelligence; these fields have to a large degree co-evolved ever since. By the 1970's research groups in DSS were widespread in business schools and electrical engineering departments at universities, in government research labs, and in private companies. Interestingly, ubiquitous computer peripherals such as the mouse originated as part of decision support research efforts.
By the 1980s the research scope for DSS had expanded dramatically, to include research on group-based decision making, on the management of knowledge and documents, to include highly specialized tools such as expert-system shells (tools for building new expert systems by adding only knowledge-based rules), to incorporate hypertext documentation, and towards the construction of distributed multi-user environments for decision making. In the mid-1980s the journal Decision Support Systems began publishing, and was soon followed by other academic journals. The appearance of the World Wide Web in the early 1990's sparked a renewed interest in distributed DSS and in document- and case-libraries that continues in the early twenty-first century.
Outstanding Technical Issues with Decision Support Tools
DSS tools, as described above, integrate data with formal or heuristic models to generate information in support of human decision making. A significant issue facing the builders of these tools is exactly how to define formal models or heuristics; experts make extensive use of tacit knowledge and are notoriously unreliable at reporting how they actually do make decisions (Stefik 1995). If the rules provided by domain experts do not reflect how they actually address decisions, there is little hope that the resulting automated system will perform well in practice.
A second, related, issue is that some systems are by their very nature difficult to assess. Chaotic systems, such as weather patterns, show such extreme sensitivity to initial (or sensed) conditions that long-term prediction and hence decision support is difficult at best. Even worse, many systems cannot be considered in isolation from the decision support tool itself; DSS tools for stock market trading, for example, have fundamentally changed the nature of markets.
Finally, both the DSS tools and the infrastructure on which they operate (typically, computer hardware and software) require periodic maintenance and are subject to failure from outside causes. Over the life of a DSS tool intended to, for example, monitor the electrical power distribution grid, changes to both the tools themselves (the hardware, the operating system, and the code of the tool) and to their greater environment (for example, the dramatic increase in computer viruses in recent history) mean that maintaining a reliable and effective DSS can be a challenge. It cannot be certain that a DSS that performs well now will do so even in the immediate future.
Decision support rules and cases by their very nature include values about what is important in a decision-making task. As a result, there are significant ethical issues around their construction and use (see, for example, Meredith and Arnott 2003 for a review of medical ethics issues). By deciding what constitutes efficient use in a planning support system for business, or what constitutes the warning signs of cardiac arrest in an intensive care monitoring system, these tools reflect the values and beliefs of the experts whose knowledge was used to construct the system. Additionally, the social obligation of those who build DSS tools is an issue. On the one hand, these are tools for specific purposes; on the other hand, many social and natural systems are so interrelated that, in choosing to build an isolated and affordable system, many issues will be left unresolved.
The ruling assumption of efforts to build DSS tools is that decision-making is primarily a technical process rather than a political and dialogical one. The bias here is not so much intellectual as informational: It may overestimate the usefulness of information in the decision-making process. Rather than more information, or ever more elaborate displays, people might need more time to reflect upon a problem. Coming to understand another perspective on an issue is a matter of sympathy and open-mindedness, not necessarily information delivery. Delivering detailed information, cases, and suggested courses of action to a single user is opposed to the idea of community-base processes. While placing these issues outside of the scope of a system design might be a useful design decision from a technical position, it is a value-laden judgment.
In fairness, the decision support literature does occasionally recognize that the public needs a better understanding not only of technology but also of science. There is often little appreciation, however, that decision support is an ethical and political process as much as a technical one—or that the flow of information needs to involve the scientist, the engineer, and the public. Exactly how the political process can be engaged for systems that must by their very nature operate in real time is an open question. Certainly the process of knowledge and value capture for such systems could be much more open than is currently the norm.
A second pressing issue regarding DSS tools is the degree to which the data, knowledge, sensors, and results of their integration represent a limitation on individual freedoms and/or an invasion of privacy. DSS tools based on expert-systems approaches actively monitor every credit card transaction made. Semi-automatic face recognition systems are widespread. Radio-frequency identification tags built into price tags on consumer goods allow consumer behavior to be monitored in real-time. Cell-phone records provide not only who a person was speaking to, but where they were at the time. Decision support tools for national security, market research, and strategic planning integrate information, apply rules, and inform decisions that affect human freedom and privacy every day.
DSS tools will only become more common in the future. The widespread reach of Internet connections and the dramatic decrease in the cost of sensors is driving the creation of decision support tools within governments and industries worldwide. It remains to be seen how these systems may impact on human lifestyles, freedoms, and privacy, and whether these tools can continue to evolve to handle the difficult questions facing decision makers in a complex and changing world.
ROBIN M. HARRAP
Giarratano, Joseph C., and Gary D. Riley. (2005). Expert Systems: Principles and Programming, 4th ed. Boston: Thomson Course Technology. A focussed introduction to expert systems technology from the authors of the open source expert system tool CLIPS, widely used in decision support systems.
Meredith, Rob, and David Arnott. (2003). "On Ethics and Decision Support Systems Development." In Proceedings of the Seventh Pacific Asia Conference on Information Systems, 10–13 July 2003, ed. Jo Hanisch, Don Falconer, Sam Horrocks, and Matthew Hillier. Adelaide: University of South Australia. An examination of the methods of decision support systems in medical practice.
Shim, J. P.; Merrill Warkentin; James F. Courtney; et al. (2002). "Past, Present, and Future of Decision Support Technology." Decision Support Systems 33(2): 111–126. A systematic overview of the development of decision support systems.
Simon, Herbert A. (1960). The New Science of Management Decision. New York: Harper and Row. An early and highly influential look at the automation of decision making in business from the leading researcher of the era.
Sprague, Ralph H., Jr., and Hugh J. Watson. (1996). Decision Support for Management. Upper Saddle River, NJ: Prentice Hall. A broad overview of decision support systems, emphasizing systems used in business practice for financial and organizational planning.
Stefik, Mark. (1995). Introduction to Knowledge Systems. San Francisco: Morgan Kaufmann. A detailed examination of the human and software problems in constructing software-based knowledge systems.
Turban, Efraim, and Jay E. Aronson. (2001). Decision Support Systems and Intelligent Systems, 6th edition. Englewood Cliffs, Upper Saddle River, NJ: Prentice Hall. A broad overview of decision support research and artificial intelligence approaches that support decision making. Includes extensive case studies from the history of DSS tools.
Decision Support Systems
Decision Support Systems
Computer systems that provide users with support to analyze complex information and help to make decisions are called decision support systems (DSSs). In some cases, such systems may predict the impact of decisions before they are made.
Decision support systems are an advancement of management information systems, whose main purpose is to supply information to managers. Although the functions of management information systems are limited mostly to control of storage, organization, retrieval, and maintaining the security and integrity of data, decision support systems include model building and model-based reasoning capabilities. Decision support systems do not have the problem-solving competence of expert systems, however, which are programmed with the knowledge of human experts and can make some final decisions without human intervention. Decision support systems generally help human beings solve complex problems, and provide data that can lead to non-predetermined solutions that are beyond the limitations of expert systems.
The concept of decision support systems originated in the late 1950s and early 1960s, when researchers at Carnegie Institute of Technology located in Pittsburgh, Pennsylvania, conducted pioneering studies of organizational decision-making. Some time later Tom Gerrity and a team of scientists performed research on interactive computing at the Massachusetts Institute of Technology. One of their projects was dedicated to supporting investment managers in the administration of their clients' stock portfolios. These original studies were mostly done by graduate students and professors at leading engineering and business schools who had access to the most advanced computer systems of that time. It became obvious at the time that the emerging computer technology could serve as an excellent tool for the development of decision support systems.
Later on, scientists and computer programmers applied analytical and scientific methods for the development of more sophisticated DSSs. They used mathematical models and algorithms from such fields of study as artificial intelligence (AI) , mathematical simulation and optimization, and concepts of mathematical logic.
In recent years, decision support tools like data warehousing and online analytical applications have enhanced the capabilities of decision support systems. Joint efforts of scientists and programmers resulted in the development of the DSS generator—software for the building of decision support systems customized to the user's requirements.
Over the years DSSs have gained popularity in such areas as business management, medicine and health care, the military, environmental policy, and other areas that involve risk management, proper resource allocation, and similar tactical and strategic decisions. Decision-making is particularly important in the areas of business and finance. Very often, distribution of valuable resources and large sums of money are based on human intuitive judgment. DSSs provide aid in such tasks as cash flow analysis, break-even analysis, scenario analysis, and inventory techniques. The number of working DSSs demonstrates the necessity of such systems. The Taxpayers' Assistant System of the U.S. Internal Revenue Service helps its personnel give correct tax information to taxpayers. Also, DSS programs assist credit card company employees in allowing or disallowing certain charges, and advise bank officers regarding loans and mortgage approvals.
In manufacturing, decision support systems are solving design problems by using analytical, statistical, and model-building software, especially in the car- and aircraft-making industries. Such giants of automotive industry as General Motors and Ford Motor Company use comprehensive decision support systems.
Systems for scheduling and customer support are universally available for use in different fields. Their use helps bring down costs and increase service efficiency. American Airlines developed a system to schedule airplanes for required maintenance. As people deal with increasingly complex machinery and sophisticated instrumentation, DSSs become an important component in service and customer support, particularly providing expertise in troubleshooting and systems operation.
Decision support systems are also widely used in environmental science, particularly in air quality impact analysis. For example, a model developed by the Stanford Research Institute computes hourly averages of carbon monoxide for any urban location, while a program developed by the Argonne National Laboratory of Illinois provides meteorological data and classifies sources of emission for the Federal Aviation Administration (FAA).
Health-care professionals are assisted by decision support tools in determining the prognosis of individual patients based on an analysis of clinical data, and in determining whether a patient is eligible for clinical research based on the patient's clinical history. Applications such as alerting systems, critiquing systems, and diagnostic suggestion systems are utilized in hospital operating rooms, intensive care units, laboratories, and other medical facilities. Implementation of these tools can significantly improve the quality and reduce the cost of health care.
These and other DSSs may work actively or in a passive mode. Passive systems are mostly used by physicians for reference purposes, while active systems give advice in certain situations, such as alerting medical personnel when a parameter being monitored in a patient—such as temperature or heart rate—exceeds its designated threshold value.
Whether they are active or passive systems, health care decision support systems are limited to the role of advisor to clinicians and other health care professionals. They do not act as expert systems. Only physicians themselves are responsible for decisions made concerning human health; DSSs act as information tools in this and other circumstances.
An effective decision support system requires reliable data and usually includes an interactive user interface , a knowledge base, and an inference (reasoning) engine.
The knowledge base is the component of a DSS that contains both well-established facts and results obtained by trial-and-error methods. It may include such different elements as formulas (algebraic, logical, or statistical expressions), guidelines, and knowledge of risk and cost of operations. Often knowledge is presented in the form of rules.
The inference engine is a complex computer algorithm, which uses the data in the knowledge base to obtain a solution to a problem. There are two approaches for decision-making systems: under certainty and under uncertainty. Decision-making under certainty most often utilizes a mathematical model that gives an unambiguous response. On the other hand, decision-making under uncertainty uses statistical approaches, such as Bayesian networks , neural networks , and other methods of AI.
The advancement of decision support systems would be impossible without its interaction with a number of related fields. Business analysts and mathematicians develop mathematical models for use in DSS, software engineers provide tools for knowledge maintenance and data mining, and psychologists assist in DSS design by conducting behavioral decision-making research.
As the need for more sophisticated decision-making will increase in the future, it is probable that decision support systems will become more intelligent and user friendly, and applicable to a broader spectrum of professions.
see also Information Systems; Integrated Software; Medical Systems.
Marina Krol and Igor Tarnopolsky
Berner, Eta S., ed. Clinical Decision Support Systems: Theory and Practice. New York: Springler-Verlag, 1999.
Krol, M., and D. Reich. "Development of a Decision Support System for Detecting Critical Conditions During Anesthesia." Journal of Medical Systems 24, no. 3 (2000): 141–146.
Marakas, George M. Decision Support Systems in the 21st Century. Upper Saddle River, NJ: Prentice Hall, 1999.
Turban, Efraim, and Jay E. Aronson. Decision Support Systems and Intelligent Systems, 6th ed. Upper Saddle River, NJ: Prentice Hall, 2001