Integrating Information Systems Using Normalized Systems Theory: Four Case Studies

Abstract

New technical and business opportunities lead to the adoption of new information systems. Integrating these new systems with the existing application landscape poses many challenges, especially when legacy systems are involved. These systems may rely on completely different architectures or assumptions regarding functionality. Normalized Systems Theory addresses this issue by stating that information systems should behave as stable systems. It identifies the main obstacle to achieving stability, so-called combinatorial effects, and proposes four design theorems to avoid these combinatorial effects. Based on these theorems, a set of exemplary software elements are created which can be used to generate applications. In this paper, we demonstrate the real-world usage of these software elements in legacy integration projects by describing four case studies. From a theoretical perspective, these cases validate the theorybased approach of Normalized Systems Theory. From a practical perspective, these cases demonstrate how adoption of Normalized Systems software could be approached: first, by integrating existing systems, and second, by building future applications adhering to the theory.