Human Systems Management

This editorial provides an overview of two emerging technologies that have wide implications for decision making and organizational issues. These technologies are decision support systems (DSS) and expert systems (ES). Both technologies are defined and discussed as stand-alone systems and as integrated systems. A perspective is provided to enable better integration of the papers that follow in this special issue.

This paper examines the potential organizational impact of expert systems (ES's), in particular, the impact on decision making, organizational structure, degree of decentralization, level of organizational effectiveness, content of organizational roles, leadership and power, communications and information flow and the personnel requirements of the organization. Generally, the nature and magnitude of the impact can be described as a function of many factors, four of which are considered in this paper: the suitability of tasks to ES's, the purpose of use of the ES, the ES tools used and the ES computing environment.

The development of DSS has traditionally addressed individual and group decision support systems. These so-called specific decision support systems tend to support only one dimension of the decision making process in an organization. The need for an organizational DSS has been addressed by many researchers recently. This artilce develops the basic framework in which the organizational DSS becomes the foundation for developing specific DSS. A process-oriented approach to developing the organizational DSS is proposed and the tasks involved in implementing this approach are presented. The relationship of the organizational DSS to individual and group DSS is discussed. A three-tier model for generating specific DSS from organizational DSS is proposed. An example of an organizational DSS is presented.

First and foremost, decision support systems must be accepted by their users if the systems are to be used. Recently, user satisfaction with DSS has been studied in several different ways. This paper explores the relationship between organizational subcultures and users of DSS, in order to discover how subcultures can be useful in explaining user satisfaction with decision support systems. Active DSS users from three functional areas of a large midwestern financial institution participated in the study. The existence of organizational subcultures was determined through multiple methods and questionnaires were used to determine user satisfaction for the same population of DSS users. The three subcultures were more successful than demographic variables in explaining variation. Implications of this study include utilizing information about subcultures to design decision support systems acceptable to the various subcultures.

During the emerging information age, organizations will have to adopt improved decision-group technologies and structures. One such technology will be expert systems. The purpose of this paper is to present a framework to ascertain the applicability of an expert system to a particular decision area. Attributes such as quality of a decision, the structure of a problem and the necessary expertise and information influence this decision. Before this framework is presented, issues concerning the ability of an organization to assimilate such a technology and transfer that technology throughout are discussed.

An important difference between expert systems and more conventional DSS is that many expert systems contain explicit representations of metaknowledge. Metaknowledge is information about the content and structure of an expert system – for example, a description of the information contained in the system or an explanation of how the system works. This information may be useful in helping a user to interpret the output of the system or otherwise to use the system more effectively. We examine here the organizational implications of a particular type of metaknowledge – knowledge about the varietey of information sources available to a manager that may help him to solve a particular decision problem. This information may come from people, organizational units, and decision support systems, the latter in the form of stored data, data analysis procedures, text files, decision models, and knowledge-bases. Thus, a knowledge-based DSS may help a manager to interact more productively with a network of people and computers. We are concerned here with the ways in which this might be accomplished.

Humans do not apply formalistic scaffolds of fixed rules of ‘knowledge’ to integrate the a priori given objective world of data ‘out there’: they do not compute the world. Regardless of some ‘knowledge’-modeling assumptions, just the opposite is true: humans use their subjectively perceived world of turbulent circumstances to bring forth (create, recreate and adapt), again and again, knowledge as an autopoietic network of relations through which they coordinate their actions. Such knowledge brings (through language) coherence and coordination to the otherwise turbulent and chaotic world of human action. Knowledge is not ‘processing of information’ but a coordination of action. As a consequence, any management support system (DSS, AI, ES, etc.) claiming knowledge as its purpose or its base, cannot be of the symbolic computation type à la Simon.