The effect of expertise on software selection

Selection of hardware and software is a complicated task involving the consideration of multiple criteria decision making (MCDM) as well as the expertise of the decision maker. Two MCDM studies examined whether or not expertise in database management systems (DBMS) would facilitate selection among DBMS alternatives. The results of the first study suggest that (I) experts seemed to exhibit more agreement on criterion weights than did novices, (2) experts were about twice as consistent in applying the weights they assigned to the choice task compared to novices, (3) novices tended to rate criteria with scores that were closer to the neutral midpoint than did experts, and (4) experts made different choices than did novices. Consistency was assessed using two different quantitative measures; one was based on ordinality and the other on position. In the second study, concurrent verbal protocols showed that experts tended to bring their own experience to the choice task, were distracted less frequently, and used superior strategies in elimination of weakest alternatives. This implies that staff members with significant domain expertise should be used in DBMS acquisition tasks. Future research should determine whether or not these findings generalize to other task domains.

The implementation of expert systems

A descriptive mail survey of managers of work groups using expert system (ES) technology was performed to gain insight into the management of the ES implementation process. This article presents survey results representing 45 different ES projects that have been successfully implemented in organizations. Each of the managers was asked to evaluate the importance of 13 factors to ES implementation success and to report on the organizational climate and practices related to ES implementation. The survey results are generally consistent with earlier implementation research and indicate that successful ES implementation requires management attention to a combination of organizational factors, implementation activities, and technical factors. Implementation practices that were associated with the successful ES implementations were commitment to the ES project, user participation, and management support.

The advantages of electronic data interchange

The purpose of this research was to identify success factors of EDI implementation and the benefits organizations could obtain by using this technology. The success factors found empirically in a field study of 140 Canadian enterprises are the organizational support, the implementation process, the control procedures and the level of EDI integration in the firm. The level of success also depends upon the level of imposition of EDI by partners. The study found evidence that, in order of importance, the benefits of EDI implementation are improvements in terms of information quality, transaction speed, administrative costs, strategic advantage and operations management.

Image transfer

The transfer of digital images between data archives and scientific workstations is likely to consume a significant amount of network bandwidth in the very near future. This paper examines the image transfer problem from an end-to-end perspective, that is, it describes a complete image transfer protocol that takes into account both the nature of digital imagery and the properties of the underlying network. Specifically, it describes a simple algorithm for encoding images into network packets in such a way that the receiver can recover from dropped packets without requiring the sender to retransmit them. Avoiding retransmissions has the advantages of improving response time and eliminating the need to buffer data at the sender.

MIS Success: Why Does it Vary among Users?

Organizations have been developing and implementing computer-based management information systems (MIS) at an increasing rate for the last 35 years. However, evidence indicates that many computer-based MIS are not as successful as they should be and many may be considered failures.

There has been much research to investigate organizational factors, individual differences, user involvement and their relationship to MIS success, and in particular, one indicator of MIS success, user information satisfaction (UIS). Little attention, however, has been paid to whether UIS for any particular MIS varies for users from different organization functions and different management levels within an organization.

This paper reports the results of a field study of MIS in different organizations in Australia which investigates the relationship between type of user, user involvement and user information satisfaction. The results show that MIS success (UIS) varies between users of the same MIS based on their organizational function and that user involvement in the systems development life cycle (SDLC) has a significant positive effect on UIS. The significance of the effect of user involvement varies for users based on their organizational function. Technical users are less satisfied with MIS than administrative users. The relationship between user involvement and user control of the systems development process and UIS is significantly different for administrative and technical users.

Information resource management

Information-based organizations depend upon computer databases and information systems for their ongoing operation and management. Information Resource Management (IRM) is a program of activities directed at making effective use of information technology within an organization. These activities range from global corporate information planning to application system development, operation, and maintenance and support of end-user computing. Numerous approaches to specific IRM activities have been proposed. They remain disjoint, however, and, hence globally ineffective.A significant reason for inability to integrate IRM activities is the failure to adequately define the information resource. What is it that must be effectively managed? This paper addresses this issue. It applies data modeling concepts to the problem of managing organizational information resources. A data model is developed to support and integrate the various IRM activities. This model formally defines the information resource and the data needed to manage it. It provides a basic ingredient for effective Information Resource Management.

Issues in the Design and Administration of Distributed Model Management Systems

Distributed model management systems (DMMS) represent the next frontier in the organizational usage of DSS technology. Due to the growing popularity of distributed computing systems and increased level of modelling activity in most organizations, significant benefits can be realized through the implementation of a DMMS. In several ways, the functions of a DMMS can be viewed as isomorphic to those of a distributed database system.

This paper describes the objectives of a distributed model management system and discusses several important issues that must be considered before constructing the distributed model base. In addition, a layered schema architecture is suggested for designing the DMMS. The proposed architecture facilitates sharing, control, consistency and integration of models, without altering the current model representation schemes. This paper also examines the roles of different individuals responsible for managing the DMMS.

Putting it in its Place: A Polemic for the Nineties

The 1980s were characterized by rapid developments in information technologies, continuously rising IT expenditures in firms and high hopes about the pay-off of IT investments. In many ways., the great expectations of IT seemed to encourage organizations to manage IT as a special case, somehow above the norms of conventional policy and practice. The 1990s look different- IT expenditures are no longer rising, value for money questions are being asked about IT, changes are being made to IT organization structures and IS Directors are losing their jobs. Michael Earl argues for a more realistic approach to managing IT. He calls it “putting the business back into IT”.