Understanding Institutional Mechanisms of Controlling Information Systems Architecture Complexity: A Simulation-based Research


Over the past decades, we have witnessed an enormous growth of investments in information systems (IS) in organizations. On one hand, constant investments in IS have had a significant impact on organizations’ performance. On the other hand, these investments resulted in a significant complexity of the corporate IS architecture (i.e., the organization’s fundamental IS components, their inter-relationships, and principles governing their design and evolution), mainly due to allocating project ownerships and IS design decision authorities to local units. This manifests in a large and ever-growing number of hete¬rogeneous IS, which are costly to maintain, tightly interrelated, and which lack flexibility with regard to business changes.

With the constantly increasing number of IS, organizations lose control of their corporate IS architecture’s evolution. This encouraged scholars and practitioners to propagate the concept of enterprise architecture management (EAM) for systematically aligning local IS investments with enterprise-wide objectives. Complementary to EAM’s traditional top-down driven approach to control the complexity of IS architectures, recent discussions propose to focus on the local actors and the way they take actual design decisions in IS change projects. However, IS and the complexity of their architecture may not be misconceived as isolated and purely technical phenomena. Instead, IS architecture complexity emerges in a complex socio-technical (ST) system, caused by the diverse interactions among its social and technical components. Therefore, in order to understand how local design decisions are taken in a mutually interactive ST system, how these local decisions add up to the overall IS architecture complexity, and how IS architecture complexity can be controlled, we opt for ST, complex adaptive systems (CAS), and institutional theory as theoretical lenses. While ST and CAS conceptualize the dyadic interplay among social and technical components as well as the non-linear and emergent behaviour of IS architecture complexity, institutional theory postulates control mechanisms to constrain IS architecture’s emergent behaviour.

Such studies, however, are hardly feasible through research methods such as case studies or surveys, since these methods do not allow to isolate and vary the potentially large number of parameters of the respective ST system. Simulation experiments enable researchers to conduct such studies, as they allow for a variation of parameters in a controlled environment, while producing massive amounts of data that, for instance, enable researchers to capture non-linear relations with statistical techniques. Agent-based simulation techniques particularly fit our perspective, as the conceptual model of such simulations focusses on local decision makers. Therefore, through combining a sound theoretical basis with a simulation-based research method, the contribution of this study is twofold. First, this study goes beyond existing static worldviews and deterministic prescriptions to corporate IS architecture through demonstrating its dynamic, emergent behaviour and its inherent complexity. Second, this study provides insights on the effects of institutional mechanisms on the behaviour and control of corporate IS architecture’s complexity.

This project is a pre-study for an ambitious project application, to be submitted for basic research funding.

Additional Informationsunspecified
Commencement Date1 February 2017
Contributors Aier, Prof. Dr. Stephan (Project Manager); Haki, Dr. Kazem (Project Worker) & Beese, Jannis (Project Worker)
Datestamp 31 Aug 2016 07:05
Institute/School IWI - Institute of Information Management
?? Inst IWI RW ??
?? PRS-PPP-PW ??
?? SoM LA / UA ??
Completion Date 31 March 2018
Publications Aleatrati Khosroshahi, Pouya; Beese, Jannis; Yilmaz, Fatih; Matthes, Florian & Winter, Robert: Key Performance Indicators for a Capability-Based Application Portfolio Management. 2017. - Enterprise Distributed Object Computing Workshop (EDOCW). - Quebec City, QC, Canada.
Beese, Jannis; Aleatrati Khosroshahi, Pouya; Aier, Stephan & Volkert, Stefan (2017) Komplexität von IT-Landschaften. Wirtschaftsinformatik & Management : WuM, 9 (2). 40-46. ISSN 1867-5905
Beese, Jannis; Haki, Kazem; Aier, Stephan & Winter, Robert (2018) Simulation-Based Research in Information Systems: Epistemic Implications and a Review of the Status Quo. Business & information systems engineering : BISE, 61 (4). 503-521. ISSN 2363-7005
HSG Profile Area SoM - Business Innovation
Keywords complexity, enterprise architecture, emergent behaviour, agent-based modelling, simulation experiments
Funders HSG – Grundlagenforschungsfonds (GFF)
Id 247311
Project Range Institute/School
Reference Number 2200183
Project Status ongoing
Project Type fundamental research project
Edit Item Edit Item