What is Decision Support System?
In an organization, decisions are made on a daily basis. Some of these decisions consist of allocating jobs to employees, deciding daily targets, prioritizing daily tasks, etc. In small organizations, a manager is able to easily take such decisions.
However, in a large organization, it is difficult for a manager to take instant and spontaneous decisions such as selecting a suitable vendor for raw materials or purchasing a machine.
In these cases, managers need to assess the different options available, such as different vendors or machines and select the best option. DSS is used by organizations to achieve this purpose because it helps in storing and analyzing the huge databases of large-scale organizations.
Table of Contents
- 1 What is Decision Support System?
- 2 Evolution of DSS
- 3 Components of DSS
- 4 User Interface Management System
- 5 DSS Architecture
- 6 Analytical Models in DSS
- 7 Characteristics of DSS
- 8 Types of DSS
- 9 Tools and Technologies Supporting DSS
- 10 DSS and Outsourcing
- 11 Group Decision Support Systems
DSS is an interactive information system that delivers information to business experts and managers for making effective decisions. In other words, it is a computerized information system designed to help in the decision-making process. DSS provides support to organizations in the decision-making process by providing useful information.
It helps in gathering, analyzing, interpreting, and reporting information for the selection of the best solution to the business problem. DSS supports data integration from different sources within the organization and supplies analytical information, needed to make better business decisions.
The advantages of DSS are as follows:
- It helps the decision-making process of an organization and enables it to make a better selection from the available outcomes.
- It allows the organization to perform a ‘what-if’ analysis, which shows a logical view of decision-making. DSS not only displays alternatives but assesses them in coordination with dominant situations.
- It saves the time and effort involved in a business process of an organization, which leads to enhancement in the productivity of the organization.
The figure shows a DSS:
DSS is applied in different fields for different purposes. Some of the applications of DSS are as follows:
- Clinical DSS (CDSS) is a DSS used for medical diagnosis.
- DSS is also used in business and management. Executive dashboards and other business performance software help in taking decisions faster, identifying negative trends, and allocating business resources in an efficient way.
- DSS can be applied in the production and marketing of agricultural products. For instance, the DSSAT4 package enables a quick assessment of several agricultural production systems for supporting decision-making at the farm and policy levels.
- Modern DSS addresses all the aspects of forest management. These aspects include log transportation, harvest scheduling sustainability, and ecosystem protection.
- Another application of DSS is seen in the form of the Canadian National Railway system. With the help of DSS, Canadian Railway is able to reduce the incidence of derailments at the same time other companies were experiencing an increase.
Evolution of DSS
Do you know when DSS made its first appearance in the market? The concept of DSS was introduced by Meador and Ness (1974) in their article, ‘An Application to Corporate Planning’. In ‘A Study of Computer Aided Decision Making in Organisations’, Keen has stated that there are two main areas of research from where concepts of decision support have developed.
The first area comprises research about organizational decision-making, and the second contains research in the field of technology, which is associated with interactive computer systems, between the 1950s-1960s.
The initial involvement in this field turned out as a classic management information system, which was able to provide pre-defined management reports to support decision-making processes. This development introduced DSS with ad hoc and interactive support in decision-making by 1970. The research was conducted in this field to update the DSS with advanced technology.
The rising of micro-computers and advanced operating systems by the 1980s modified DSS with more interactive system development. Using artificial intelligence, the trend of a knowledgeable system was developed.
In 1987 a breakthrough was achieved with the successful application of Gate Assignment Display Systems (GADS) in the aviation industry. In this system, the airport schedule was integrated, which helped in reducing travel delays.
In the 1990s, changes occurred in technology from mainframe-based DSS to client-server-based DSS. The desktop OLAP (Online Analytical Processing) tools were introduced. Other than this, an object-oriented technology was also introduced by the vendors to create a technology for reusable decision-support capabilities.
In 1994, the network infrastructure was upgraded by many organizations. According to Powell, Database Management System (DBMS) vendors realized that decision support differs from Online Transactional Processing (OLTP).
They started the use of real OLAP capabilities in their databases. In 1995, data warehousing and the World Wide Web started to influence practitioners and academics who were interested in decision-support technologies.
Components of DSS
Nowadays, many organizations are dependent on numerous decision-support tools, techniques, and models to perform their daily business operations. DSS is able to facilitate the decision-making process of an organization due to the various components involved in it. These components perform different types of work and facilitate the decision-making process of an organization.
The figure shows the components of DSS:
The different components of DSS are discussed as follows:
Database Management System
A database is a collection of all the data obtained from various internal and external sources by an organization to take sound business decisions. However, managing huge databases manually is not possible for an organization.
Thus, organizations employ DBMS for managing huge databases, so that it can be used while taking various business decisions. It also helps in reducing cost and data redundancy and increasing data control and sharing.
There are five different data models for the database component of DSS. These are record model, relational model, hierarchical model, network model, and rule model. The record of flat file model is common in DSS, which uses time series data.
Knowledgebase Management System
It provides intelligence and support for collecting useful information. A large number of decisions are made on a day-to-day basis, which range from simple to complex. These decisions involve the use of knowledge, which forms the basis of the decision-making process. A billing and document management system is an example of Knowledgebase Management System.
Model Management System
It provides various techniques and skills to produce reliable, insightful, and useful results. A model is the abstract representation for any subject or thing. The modelling component gives decision makers the ability to analyse the problem. It supports by giving access to various models for decision support. Various techniques provided by DSS may include statistical method, sensitivity analysis, and computer simulation.
User Interface Management System
It is a framework where an interaction between human beings and computers takes place. User interface refers to a system that provides a means of:
- Input: Allows users to manipulate the system
- Output: Allows the system to give the results of the user’s manipulation
DSS architecture is formed by including the structure of dialogue management and database management. There are four components of DSS architecture, which are as follows:
It involves multiple dialogues, modelling, and database components, which are able to communicate with each other through a network interface.
It is a standard interface with local dialogue and modelling components, which are able to link modelling and database components remotely.
It consists of a single dialogue and database component, where multiple-model components are linked with the architecture.
It involves more vertical components, called tiers, with tools for data extraction. A DSS tower integrates diverse database components. The rest of the tower architecture is similar to a network structure.
Analytical Models in DSS
Analytical models are mathematical models used for determining the relationship between the variables in a data. In DSS, four types of analytical models are used, which are discussed as follows:
It allows reversing the process done with what-if and sensitivity analysis. This analysis starts with determining the goal that is the desired result and then identifying the values of the variables. It is also called ‘how-can’ analysis, because it facilitates the analysis of how the desired outputs can be achieved with the given variables of input.
It supports the capability of conditional analysis of the available alternatives. Users of DSS can analyse the relationship among variables to arrive at the best productive decision alternative. For example, the change in the level of inventory purchased is related with its excess and shortage. With the use of DSS, a production manager can make decisions about inventory to be ordered.
It is a special type of what-if analysis. This system allows estimating the effect of change in one variable with respect to change in another variable. This process can be repeatedly analysed. These systems are of major help when any of the factors are supposed to be constant, or when decision makers are not sure about the assumptions made in the process of decision making.
For example, while deciding about the inventory level, a decision maker can use one spreadsheet to do a what-if analysis and can use another spreadsheet to check the different options analysed.
It is a complex goal-seeking analysis. In this analysis, there is no specific goal determined. The goal or desired output is to be optimised with the value of variable under certain conditions. This process is repeated till the desired value of variables is derived.
For example, the lowest cost of material purchased can be determined keeping the optimised quality and price of material.
Characteristics of DSS
DSS is a computer-based system that supports the decision making process of an organisation. Some of its characteristics are discussed as follows:
- DSS software is able to interact with the end user while providing information. This feature increases its efficiency from just being a provider of reports to a provider of interactive support for business professionals.
It could select relevant data from among the entire data store to compile it while providing information support.
- DSS is a knowledge-based system, thus, it acts as a storage of shared knowledge from individual users, experts, and various business models for identifying and solving problems, and making decisions.
- DSS helps in maintaining the integrity of data along with flexibility, so that it can match the decision maker’s choice, storing and retrieving the data and ensuring the consistency and accuracy of the collected data.
- DSS stores information in a powerful database, which can be easily distributed and is accessible to individuals throughout the organisation. DSS also helps in an easy access to historical information related to various decisions made by an organisation.
- DSS reduces wastage of time and energy. In DSS, data is directly entered into the software, which helps in reducing the possibility of miscalculation and errors.
- DSS helps in analysing the future in a given set and trend of circumstances with its reasoning capability. In a given trend of market situations, a DSS can forecast a market’s bulls and bears in the coming month.
Types of DSS
DSS is implemented by an organisation for different purposes. For example, an organisation that exists only online requires a system that helps it in taking decisions related to customers and Web technology. Based on the purpose of implementation,
DSS can be categorised into different types, as shown in Figure:
The different types of DSS are discussed as follows:
- Model-driven DSS
- Communication-driven DSS
- Data-driven DSS
- Document-driven DSS
- Knowledge-driven DSS
- Web-based DSS
It manipulates data to generate statistical and financial reports as well as simulation models. This helps decision makers in analysing the decisions and making choices among different alternatives. The model-driven DSS follows what-if analysis as an analytical tool.
This type of DSS is helpful in analysing the effect of change in certain variables towards the efficiency of business. It can be used on a standalone PC, client/server, or the Web. Some examples of model-driven DSS are statistical, financial, optimisation and/or simulation models.
It enhances decision making by facilitating a free flow of information among groups and people. This type of DSS basically supports group decision making. This type of DSS can be implemented by using the Web or client-server technology.
The communication-driven DSS can range from a simple e-mail to a complex Web conferencing application. Examples of communication-driven DSS are online chats, collaboration, and meetings.
It focuses mainly on internal as well as external data for decision making, which is obtained from data warehouses. Managers and other staff members largely depend on data-driven DSS, because they consider the database useful for making different types of decisions.
This type of DSS can be implemented by using a mainframe or client-server technology. This system utilises online analytical processing tools for data analysis. Examples of data-driven DSS are Geographic Information System (GIS), which represents geographical data through maps.
It is common for large user groups that serve the purpose of searching Web pages and documents on a defined set of keywords. This type of DSS converts documents into useful data for business. The document-driven DSS uses data that cannot be easily standardised and stored.
It utilises the different forms of data such as oral, written, and visual. Oral data is derived basically from conversations. Written data is based on all types of written documents, such as reports, e-mails, and other written correspondence, while visual data can be obtained from TV commercials and news reports.
The data obtained from these sources is not standardised. Thus, mangers need DSS tools to convert this data into meaningful information.
It provides advice related to various business decisions. It is implemented by using client/server systems. Knowledge-driven DSS is usually designed to recommend actions to users. It helps in analysing huge amounts of data for determining the hidden patterns and recommendations.
An example of knowledge-driven DSS is a diagnostic system used in laboratories for determining the disease of a patient, which enables the doctor to suggest the best treatment for the disease.
The DSS that uses a Web browser is known as a Web-based DSS. All types of DSS can be Web based. The technologies used to implement a Web-based DSS are client/server system and the Web. These are operated through the interface of a Web browser, even though the data is confined to a data warehouse.
Tools and Technologies Supporting DSS
Various tools and technologies are used for making the working of DSS easy and effective.
Some of these tools and technologies are shown in Figure:
The tools and technologies supporting DSS are discussed as follows:
- Extraction, Transformation, and Loading (ETL)
- Online Analytical Processing (OLAP)
- Relational Online Analytical Processing (ROLAP)
- Multidimensional Online Analytical Processing (MOLAP)
- Hybrid Online Analytical Processing (HOLAP)
Extraction, Transformation, and Loading (ETL)
The ETL process helps in identifying the required data taken from different sources such as database systems and the applications of database. This data is then extracted from a source to transform it for use.
Depending on the source system’s capabilities (for example, operating system resources), some transformations may take place during this extraction process. After extracting data, it has to be physically transported to the target system or an intermediate system for further processing.
Depending on the chosen way of transportation, some transformations can be done during this process too. For example, an SQL statement, which directly accesses a remote target through a gateway, can concatenate two columns as part of the SELECT statement, which leads the manager to take a particular decision.
Online Analytical Processing (OLAP)
It is a significant improvement in model-based management, which manipulates data from a variety of sources that has been stored in a static data warehouse. The software can cause various views and representations of the data, and this helps in making the context explicit.
OLAP can pull the data, create pictures, and make the user see the model run. It creates various models and helps in finding out the best. It facilitates analysis of information presented in multidimensional views and hierarchies.
Relational Online Analytical Processing (ROLAP)
It is useful when we need to handle large data and try to leverage functionalities inherent in the relational database. This methodology relies on manipulating the data stored in the relational database to give the appearance of traditional OLAP’s slicing and dicing functionality.
In essence, each action of slicing and dicing is equivalent to adding a “WHERE” clause in the SQL statement.
Multidimensional Online Analytical Processing (MOLAP)
It stores data in a multidimensional cube. It gives excellent performance and can perform complex calculations. In a management decision support system, the managers need to look into the summary rather than the details report.
Hybrid Online Analytical Processing (HOLAP)
It is another technological attempt to pick out the advantages of MOLAP and ROLAP and combine those in this tool. HOLAP utilises cube technology for performing with speed. Whenever information detail is required, HOLAP is able to “drill through” from the cube into the underlying relational data.
It is the hardware and software component that creates the user interface for the DSS. Software support for dialogue is in terms of packages.
It can be accessed through high-end languages and can be used to construct the user interface and data definition language that describes the dialogue component of DSS architecture. It also focuses on command language and menu dialogues that create a restrictive context for building DSS.
DSS and Outsourcing
Full time resourcing remains viable for only large business organisations, where the range of research and development can justify the ongoing investment. To streamline business processes, organisations can outsource activities that are not considered core business functions.
In many organisations, software development is considered a non-core activity. The term ‘software acquisition’ has been used to describe situations where a customer contracts with a software development organisation for the complete development of a software product.
Outsourcing involves contracting with outside consultants, software houses, or service bureaus to perform system analysis, programming, or other DSS-development activities. The outsourcer should be evaluated as a long-term asset and a source of ongoing value to the organisation.
Some of the benefits of outsourcing of DSS projects are:
- It facilitates low-cost development of products/services.
- It provides access to expertise on new technologies.
- It allows organisations to release resources for other projects.
- It enables organisations to focus on specialised business processes instead of relative business processes.
1Outsourcing of DSS has become risky in sensitive industries, such as defence and healthcare, due to regulations set by the government such as Health Insurance Portability and Accountability. Most of these risks are as follows:
- An organisation relinquishes control of an important capability to an outside organisation.
- Contracts for DSS services may be long term, and this may block the organisation into a particular service provider.
- Relying on external sources for new and complex techniques can lead to low knowledge among in-house staff.
Small and medium-sized organisations can outsource the operation of some decision-support services, provided they maintain control of decision support data. Outsourcing of decision-support services has been there since the early days of DSS.
Time sharing provided access to capabilities when organisations could not afford to provide DSS inhouse. From a managerial perspective, outsourcing avoids the need to directly manage IT, and it may be more cost effective than in-house IT.
However, outsourcing may create strategic vulnerabilities. Increasing the strategic vulnerability of an organisation because of short-sighted outsourcing decisions is definitely undesirable.
Group Decision Support Systems
Group Decision Support System refers to an interactive computer-based system designed for supporting decisions taken by a group or team instead of an individual. It is important for decision makers to select the most suitable method of decision making and maintain proper interaction with other decision makers of the group.
GDSS enables individuals involved in the decision-making process to communicate and exchange knowledge among each other. For example, Hewlett Packard (HP) has used GDSS for solving the communication issues among its engineers.
This arises because they work in different countries and locations and meet once a year. Thus, they are unable to get suggestions or discuss any issues or new technology with each other. GDSS enables these engineers to conduct frequent meetings using electronic conferences.
In such conferences, they can discuss professional and organisational issues, which further help them develop better products for the organisation.
Some of the characteristics of GDSS are as follows:
- An organisation can design GDSS according to the requirements of the group involved in the decision-making process.
- GDSS enables group communication, that is, it provides interactive support to the whole group or team of decision makers. This is more practical, because most of the time, decision making is a group task in business. All the participants in decision making can give comments at the same time.
- GDSS helps the group or team members to comment on an issue without displaying his/her identity.
- GDSS helps the group or team members to comment on an issue without displaying his/her identity.
- GDSS involves an automated system for record keeping and maintaining. GDSS supports decision making by providing records automatically whenever required.
- GDSS helps a group to make efficient decisions without incurring heavy costs.
Applications of GDSS are as follows:
- A common example of GDSS is the file drawer system, which acts as a data access model. Individuals using file drawer system depend on their computers for getting information that can be used for making educated decisions.
- Suggestion model is another example of GDSS. Computer use available data for making suggestions to potential methods for completing a task.
- Accounting models, a GDSS model, is commonly used by groups that perform risk management functions. This system determines the outcomes of various decisions. These systems are used by accountants and financial planners while developing a financial strategy.
- Representational GDSS is used by groups that are able to create simulations of hypothetical situations.