<< Back to 'Fields of Research'

Motivation

Planning, designing, introducing, and maintaining corporate information systems is a complex endeavour. Beyond demanding a deep understanding of a company's current situation, it has to be remembered that introducing advanced information technology has to allow for new ways of organizing the business. It is an aspect emphatically stressed by the numerous authors who recommend "business redesign" or "business process redesign". Like system design, analyzing and redesigning a corporate strategy and a company's organisation respectively are complex tasks on their own. Management Science and organizational theory offer a wide range of dedicated approaches for analyzing and shaping a firm's strategy as well as for organizational (re)design. Often they are based on graphical models which are introduced to illustrate essential concepts and interrelations - and to communicate them to others who should be involved. Organizational models cover a wide range from rather prosaic to more formal representations. This is similar to models for strategic planning. They usually stress a more abstract view with highly aggregated data. Strategic and organizational models are usually based on different concepts. Furthermore, they generally have nothing in common with conceptual models used in software engineering. While there is certainly need for specialization, such a separation of concerns can be seen as a major inhibitor of efficient information systems. Enterprise models are aimed at providing a medium to foster discourses between people with different professional backgrounds. At the same time they serve as a conceptual basis of highly integrated information systems. 

Concepts

To support the development of powerful enterprise models, MEMO offers a framework that allows structuring an enterprise to a high degree of abstraction. It corresponds to abstractions common in management science and software engineering. The framework suggests three main perspectives on the enterprise: strategic, organizational, and via the information system. The perspectives themselves are further detailed by different aspects (resource, structure, process, and goal). For instance, business process models would be assigned to the perspective "organisation" and the aspect "process". To support conceptual models that are semantically rich and intuitive, MEMO includes an extensible set of specialized modelling languages. The modelling languages are defined within a multi-layer language architecture.

Macro Process

Our work in Philosophy of Science is mainly inspired by the question whether research goals and methods of the information systems discipline are essentially different from those of the neighbour discpilines. We found that informations systems research is essentially different in one respect which is clearly expressed by an approach that we call "constructivist" (sometimes referred to as "design science"). It recommends to construct reality by introducing new ways of coordinating cooperative work. Where the inductivist approach assumes that the variance in using information and expressing it through languages is a necessary reflection of the variety of tasks to be taken into account, the constructivist approach relies on the presumption that variance in actual information use and related coordination mechanisms is the result of an - at least partially - arbitrary process. For this reason, reducing variance by introducing new common concepts to handle information would not necessarily cause dysfunctional effects. Moreover, if the processes they are to be used in were thoroughly designed, they would contribute to more efficiency. In any case, those common artefacts would allow for a high level of integration and reusability.

Despite the prospects of the constructivist approach, it bears a severe epistemological problem: In order to allow for a rational growth of knowledge we need criteria to comparatively evaluate competing constructions. Usually, these artefacts, like models and modelling languages, cannot be validated against reality in a straightforward fashion. We have worked on methods to evaluate artefacts (especially modelling languages and conceptual models). Other work includes reflections upon the appropriate language for information systems research, methods of empirical research and the relationship between theory and praxis.