Generalizing normalized systems theory: towards a foundational theory for enterprise engineering

Date: 24 November 2014

Venue: University of Antwerp, De Kapel Grauwzusters - Lange Sint-Annastraat 7 - 2000 Antwerp

PhD candidate: Peter De Bruyn

Principal investigator: Prof. dr. Herwig Mannaert

Short description: PhD defense Peter De Bruyn - Facultyof Applied Economics

Abstract: Contemporary companies are in need of evolvability (i.e., to adapt their structure and that of their organizational artifacts to a changing environment) and observability (i.e., to derive the internal state of the enterprise from the observable information). Explicit design knowledge on how to achieve this is often missing. In this study, we focus on the utility of Normalized Systems Theory for this purpose as it initially focused on creating evolvable and observable software architectures by means of prescriptive design theorems. Later on, a similar reasoning (or parts of it) has been applied within other application domains (e.g., business processes, enterprise architectures and industrial automation systems).
The first part of this study explores how and to which extent this recurring reasoning can be generalized, given its current use within different application domains. We hypothesize that Normalized Systems Theory is essentially a theory about modular structures and how to make these structures more evolvable and observable. We propose a set of general theorems which prescribe how to decouple modules along different dimensions in order to obtain the envisioned goals, and discuss how elements enable the realistic creation of the fine-grained modular structures implied by the theorems.
The second part of this study investigates the implications of applying this more general Normalized Systems Theory at the organizational level. First, we consider how observability (as considered within Normalized Systems Theory) can be studied at the level of business processes. Second, we argue that the application of Normalized Systems Theory (and its theorems) at the organizational level necessitates the existence of exhaustively defined interfaces at this level as well. The relevance and importance of these issues is illustrated by a set of examples, based on case material and existing artifacts.
In the third and final part of this study, we explore how organizational modules and patterns compliant with Normalized Systems Theory (i.e., the elements) should or might look like. In this respect, we provide an initial set of elements illustrating their viability at the organizational level.