Decision Support Systems

Every day, managers are faced with decisions about how best to run their organizations in order to gain or maintain a competitive edge. Although some decisions are simple, many are quite complex and require the manager to consider many variables. To help managers in decision making processes, many organizations employ the use of computer-aided decision support systems. These systems can help managers make decisions about semi-structured and even unstructured problems where not all the information is known in advance. Several different types of decision support systems are available. These include model-driven systems, data-driven systems, knowledge-driven systems, and group support systems. These decision support systems can be quite effective in aiding managers in making decisions and improving the effectiveness and performance of organizations.

Keywords Artificial Intelligence (AI); Database; Decision Support System (DSS); Executive Information System; Expert System; Groupware; Spreadsheet; Systems Theory

Information Technology > Decision Support Systems

Overview

The world around us is constantly changing. Market needs are shaped by numerous factors including political realities, advances in technology, and changing cultural expectations. To be competitive in this environment, businesses cannot be static, but need to grow and change to meet the needs of the marketplace. To do this, businesses need to be able to make good decisions in response to the changing needs of their environment. In some cases, these are simple decisions: Do we produce more widgets in response to increase market demand? Should we order more raw materials so that we can continue to produce gizmos? In other cases, however, the decision is not so simple. The demands of the marketplace may not be easily deciphered. As exemplified by the 8-track tape systems and the Beta video recorder of the latter half of the 20th century, decoding trends and translating them into business strategies is neither a simple nor an obvious process. In addition, management decisions can have far reaching impact within the organization. For example, the decision to increase the production of widgets may mean that more employees are needed for the widget production line. Management is now faced with another decision: Should new production workers be hired or should employees be transferred from the gizmo production line? If the former decision is made, where does the money come from to hire the new workers? If the latter decision is made, what is the impact on the gizmo production line? Considerations of the production rate for new gizmos or even the viability of the gizmo line also need to be taken into account in order to understand the ramifications of the decision in context.

In most complex business situations, it is important to understand the factors of decision making in context. Organizational behavior theorists tend to view the organization as a system, where changes in one subsystem can have far ranging impact throughout the organization as a whole. In this view, an organization is an interactive system in which the actions of one part influence the functioning of another. Rather than merely focusing on the needs of one part or subsystem within the organization in isolation (e.g., the widget production line) when making a decision, it is important to look at the impact of the decision on all subsystems and levels within the organization.

According to systems theory, organizations are systems comprising numerous subsystems. Rather than viewing an organization as a collection of disassociated parts acting independently, systems theory posits that the functioning of each subsystem impacts the functioning of the other subsystems. For example, in the illustration above, the decision to increase production of widgets can affect other parts of the organization. If more production workers are hired for the widget line, there might also be a concomitant need to hire more human resources and accounting personnel to take care of issues and tasks related to a larger work force. Other considerations in the decision might include whether or not suppliers could deliver sufficient raw materials to make the widgets or sufficient boxes to package the widgets so that they could be sold to consumers. In addition to hiring more workers for the widget production line, the organization would also more than likely also have to hire additional supervisors or managers for the new production lines. The increase in workers for the widget line might also cause conflict between the new workers and the workers already working on the widget production line or between the widget workers and the gizmo workers. Such conflict or job dissatisfaction could affect the functioning of the organization as a whole. Personnel issues are not the only factor that needs to be taken into account in this decision. Management would also have to consider where the new production lines would be located. If there was insufficient room in the current production plants, new facilities would need to be bought or leased. Similarly, new lines would probably require additional equipment. If, on the other hand, the organization decides that it is not cost effective to incur the expenses associated with starting new production lines, there would be other impacts on the organization. If workers are transferred from the gizmo line to make widgets, what happens to the gizmo product line? Even if the organization's management decides to convert some of the gizmo lines into widget lines and transfer some of the existing employees to the new line, these employees would have to be trained to become widget line operators; a necessary activity that would also cost the organization money. Each subsystem within the organization (e.g., the various production groups, accounting, human resources, management) affects the ability of the other groups — as well as of the organization as a whole — to do their jobs. Systems theory also recognizes that the organization is not only made up of interlocking subsystems, but is also part of a larger system itself that depends on inputs of raw materials, human resources, and capital and that needs to export goods or services, employee behavior, and capital in order to continue to the viability of other organizations.

To help management make such complex decisions, many organizations use decision support systems. Although there is no universally accepted definition of the term, in general a decision support system is an interactive computer-based system that helps managers and others make decisions. Decision support systems are used both by individuals and groups, and can be stand alone systems, integrated systems, or web-based.

Decision support systems are used when organizations are faced with unique, complex situations where a decision needs to be made. Decision support systems help decision makers better understand the issues underlying the situation and to make decisions in situations where the extent to which certain variables influence the activity or outcome are not initially clear or only part of the information is available in advance. This condition is referred to as an unstructured situation. To answer unstructured questions, decision support systems must be flexible so that the impact of various variables and conditions can be tested and the analysis returned to the user in a form that is useful for the specific situation. The unstructured nature of the problem also means that the use of a decision support system is an iterative process: The answers to the questions are not ends in themselves, but raise other questions for consideration that need to be run through the decision support system.

Decision support systems can help decision makers in such situations by providing the information and structure needed to make a rational decision. The decision support system creates a quantitative model of the situation and then processes data to show the impact of the variables under consideration on the outcomes. Decision support systems can help decision makers answer questions concerning conditions under which an outcome might occur, what might happen if the value of a variable changes, or how many potential customers have certain characteristics. Decision support systems can also help to trim inefficiencies from organizational systems, such as the supply chain, which further can take into account other company priorities such as carbon emmissions reduction. Frequently, decision support systems require the processing of data from multiple files and databases.

Decision support systems are not necessarily esoteric software applications. Spreadsheet software is one of the most frequently used decision support tools. These applications help the decision maker design a table of values arranged in rows and columns in which the values have predefined relationships. Spreadsheet application software allows users to create and manipulate spreadsheets electronically. As a decision support system, spreadsheets enable the decision maker to manipulate variables to see the effect of different values on the variables or of changing the assumptions on the outcome. For example, before making an investment in a new manufacturing plant, a manager could make a model on a spread sheet and analyze the effect of changing interest rates on the risk associated with the venture.

Applications

Several different general types of decision support systems have been identified in the literature. These include model-driven systems, data-driven systems, knowledge-driven systems, and group support systems.

Model-Driven Decision Support Systems

A wide variety of models can be used to form the basis for decision support systems. Model-driven decision support systems use various financial, optimization, or simulation models as aids to decision making. Model-driven decision support systems use limited data and parameters provided by the decision makers, but do not generally require large databases. For example, in 2000, Pfizer merged with Warner-Lambert to form an integrated company with sales over $30 billion. As part of the process of integrating the two companies, a decision support system was used to combine the distribution systems of the two organizations. Three linked simulation models were developed: a distribution center storage and capacity model, a distribution picking and shipping model, and a distribution center facility sizing model. Using these three models, the distribution center capacity requirements for two to five years in the future were studied. Analysts were able to examine the effects on inventory and distribution of a variety of sales forecasts. Warner-Lambert also developed an optimization model that enabled decision makers to analyze the flow of products through the network during stated planning periods. A fifth model was used to analyze inventory investments for the combined organization. The decision support system and its models were used for a series of long-range studies that helped Warner-Lambert make decisions about expansion of its distribution network, implementation of a new pharmaceutical delivery network, and consolidation of the distribution networks after the merger.

Another example of a model-driven decision support system was employed by General Motors in their decision to incorporate the OnStar two-way vehicle communication system in its new vehicles. In 1996, General Motors tested a prototype of the OnStar system, but found that it was both difficult to install and expensive. When the time came to make a strategic decision about the installation of OnStar in General Motors' vehicles, two models were developed. The conservative model posited that OnStar should only be offered as an option and promoted to improve vehicle safety and security. The more optimistic model assumed that OnStar would appeal to customers enough so that they would subscribe to the service, thereby generating significant monthly income from service fees for General Motors. Although the potential income from this latter model was attractive, it also carried with it concomitant increased risk and the need for a greater initial investment. As opposed to the previous example where data were available about the existing systems, the model developed to analyze the potential profitability of OnStar was for a system that had never been implemented. General Motors developed a six factor simulation model of the system incorporating customer acquisition, customer choice, alliances, customer service, finances, and dealer behavior. The model also incorporated feedback loops to make the simulation more realistic. In the end, the decision support system allowed management to determine that customer service was an important factor in the implementation of OnStar technology: If General Motors did not adequately invest in the project, the system would fail. Using the simulation, the recommendation was made for an aggressive strategy to pursue OnStar technology.

Data-Driven Decision Support Systems

Data-driven decision support systems utilize time series data gathered on the factor or characteristic of interest at regular intervals over a period of time. Basic data-driven decision support systems access simple file systems using query and retrieval tools whereas more advanced data-driven systems allow the manipulation of data or analytical processing. Executive information systems are a type of data-driven decision support system. These data-driven decision support systems are designed to support executive decision making by presenting information about the activities of the company and the industry. Executive support systems offer quick, concise updates of business performance for top executives. These systems have powerful processing capabilities in order to summarize and present data in a format appropriate to executive decision making.

Top executives differ from middle managers because they need to take a more global view of the organization rather than focusing on a particular process or product. Top level managers are also particularly concerned with the systems approach and need to know how decisions made regarding one part of the organization effect other parts of the organization as well. Executive support systems take these needs into account. To support top-level managers in their tasks, executive support systems allow executives to look not only at the organization as a system itself, but at the organization as part of a larger system as well. To do this, executive support systems permit scanning of data and information on both internal activities as well as the external business environment. Since executives tend to look at things from a higher level than do many analysts, executive support systems highlight significant data and present information in summary form rather than giving the user all the details and supporting data, although users are allowed access to these data if desired. This "drill down" feature allows the user to go down several levels to acquire the data necessary to make a well-informed decision.

Executive support systems are generally well accepted by top-level managers. Although these systems have relatively simple user interfaces to support the way that most top-level managers work, they can be used for a wide range of purposes. For example, executive support systems can be used to determine why a company's expenses are higher than expected, gain an overview of a competitor's financial picture, be continually updated on various corporate indicators relating to effectiveness and performance, or check for processes that are not meeting expectations.

Knowledge-Driven Decision Support Systems

Knowledge-driven decision support systems are person/computer systems with specialized problem solving abilities that can make suggestions or recommendations to the user. Knowledge-driven systems are expert systems that may include the application of artificial intelligence to the decision making process. Artificial intelligence is a branch of computer science concerned with development of software that allows computers to perform activities normally considered to require human intelligence. In the realm of decision support systems, artificial intelligence applications can be used in the development of expert systems that allow computers to make complex, real world decisions. Applications of artificial intelligence to decision making include medical diagnosis, manufacturing quality control, and financial planning.

An example of a knowledge-driven decision support system is the expert system for dispatchers used by Con-Way Transportation Services. Con-Way offers premium next-day service for less-than-truckload shipments to meet their customers' just-in-time shipping strategies. In the past, Con-Way dispatchers used the common industry model. This approach, however, was labor intensive and required the dispatchers to handle complex logistics, including assigning deliveries to over 400 locations, ensuring overnight, on-time delivery, assigning loads to optimize utilization of trailer space, developing efficient routes including consideration of bad weather, highway hazards, and other complications. To assist dispatchers in these duties, Con-Way implemented an expert system to aid in the scheduling process. The system was designed to optimize routing, loading, and delivery while minimizing costs. The system is constantly updated throughout the day and flags discrepancies so that the dispatcher can resolve issues before the truck is loaded. Con-Way's expert system for dispatchers takes into account multiple factors including elapsed time, delivery times, and customer priorities that factor into route specifications. The expert system uses artificial intelligence to develop an optimal solution for loading and scheduling. By using the expert system, dispatchers can perform duties in minutes that used to require hours. As a result, they are able to accept customer orders later than was previously possible. The system, which cost over $3 million, paid for itself within two years of implementation.

Group Decision Support Systems

In addition to the types of decision support systems discussed above, there are also group decision support systems. These systems enable workgroups to process and interpret information together even when they are not physically collocated. These systems use network and communications technologies to foster collaboration and communication in support of decision making. Communications-driven systems rely heavily on communications technology, including groupware, video teleconferencing, and electronic bulletin boards.

Like other decision support systems, group support systems support decision making in situations that are not fully structured and assist in analyzing problems. However, group support systems are used by workgroups or teams rather than by individuals. To maximize effectiveness, group support systems emphasize communication and idea generation between and among group members throughout the system using communication networks. Group support systems typically require the involvement of a facilitator to keep the group focused on the problem under consideration and to help expedite the development and sharing of creative solutions. This shared information is input into a database rather than a traditional report, and includes questions, comments, and ideas from the group.

Marriott Hotels use a group support system to help meet the ever-changing needs of business travelers; one of their biggest customer segments. Periodically, Marriott assembles teams of personnel from various groups involved in providing service to these customers, including managers, housekeeping, front-desk personnel, catering, room-service coordinators, and bellhops. These representatives exchange experiences and ideas via the group support system in an effort to improve service to this market segment. Inputs are made anonymously so that comments can be made frankly.

Terms & Concepts

Artificial Intelligence (AI): The branch of computer science concerned with development of software that allows computers to perform activities normally considered to require human intelligence. Artificial intelligence applications include the development of expert systems that allow computers to make complex, real world decisions; programming computers to understand natural human languages; development of neural networks that reproduce the physical connections occurring in animal brains; and development of computers that react to visual, auditory, and other sensory stimuli (i.e., robotics).

Database: A collection of data items used for multiple purposes that is stored on a computer.

Decision Support System (DSS): A computer based information system that helps managers make decisions about semi-structured and unstructured problems. Decision support systems can be used by individuals or groups and can be stand-alone or integrated systems or web-based.

Executive Information System: A data-driven decision support system designed to support executive decision making by presenting information about the activities of the company and the industry.

Expert System: A decision support system that utilizes artificial intelligence technology to evaluate a situation and suggest an appropriate course of action.

Groupware: Application software that provides tools to help workgroups communicate, coordinate, and organize their activities. Groupware capabilities include scheduling, resource allocation, e-mail, password protection, and file distribution.

Spreadsheet: A table of values arranged in rows and columns in which the values have predefined relationships. Spreadsheet application software allows users to create and manipulate spreadsheets electronically.

Strategy: In business, a strategy is a plan of action to help the organization reach its goals and objectives. A good business strategy is based on the rigorous analysis of empirical data, including market needs and trends, competitor capabilities and offerings, and the organization's resources and abilities.

Systems Theory: A cornerstone of organizational behavior theory that assumes that the organization comprises multiple subsystems and that the functioning of each affects both the functioning of the others and the organization as a whole.

Bibliography

Koh, S., Genovese, A., Acquaye, A. A., Barratt, P., Rana, N., Kuylenstierna, J., & Gibbs, D. (2013). Decarbonising product supply chains: Design and development of an integrated evidence-based decision support system – the supply chain environmental analysis tool (SCEnAT). International Journal of Production Research, 51, 2092-2109. Retrieved October 31, 2013, from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=85286025&site=ehost-live

Liu, S., Leat, M., Moizer, J., Megicks, P., & Kasturiratne, D. (2013). A decision-focused knowledge management framework to support collaborative decision making for lean supply chain management. International Journal of Production Research, 51, 2123-2137. Retrieved October 31, 2013, from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=85286027&site=ehost-live

Lucas, H. C. Jr. (2005). Information technology: Strategic decision making for managers. New York: John Wiley and Sons.

Renna, P. (2013). Decision model to support the SMEs’ decision to participate or leave a collaborative network. International Journal of Production Research, 51, 1973-1983. Retrieved October 31, 2013, from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=85286016&site=ehost-live

Senn, J. A. (2004). Information technology: Principles, practices, opportunities (3rd ed.). Upper Saddle River, NJ: Pearson/Prentice Hall.

Suggested Reading

French, S. & Turoff, M. (2007). Decision support systems. Communications of the ACM, 50, 39-40. Retrieved May 14, 2007, from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=24209666&site=ehost-live

Gadomski, A. M., Bologna, S., DiCostanzo, G., Perini, A., & Schaerf, M. (1999). An approach to the intelligent decision advisor (IDA) for emergency managers. Proceedings of the Sixth Annual Conference of the International Emergency Management Society, Delft, Netherlands: IEMS. Retrieved May 14, 2007, from http://erg4146.casaccia.enea.it/wwwerg26701/TIEMS'99.pdf.

Inmon, B. (2006). Real-time decision support systems? DM Review, 16, 14.

Jones, K. (2006). Knowledge management as a foundation for decision support systems. Journal of Computer Information Systems, 46, 116-124. Retrieved May 14, 2007, from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=22536043&site=ehost-live

Lamont, J. (2007). Decision support systems prove vital to healthcare. KM World, 16, 10-12. Retrieved May 14, 2007, from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=23898467&site=ehost-live

Weng, S. Chiu, R., Wang, B., Chi, R. & Su, S. (2006/2007). The study and verification of mathematical modeling for customer purchasing behavior. Journal of Computer Information Systems, 47, 46-57. Retrieved May 14, 2007, from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=24375428&site=ehost-live