Coping with Complexity: Exploring Modularity and Flexibility in IT Infrastructure Adaptation

Introduction

Enterprise Resource Planning (ERP) systems have been widely adopted in large organizations recently, due to their claimed ability to improve business processes through the promotion of “best practices” and standardization of key processes (Davenport, 1998). However, a company implementing ERP is faced with the options of either adjusting its processes to conform to the “ERP way” or adjusting the ERP system to support the company’s specific needs (Hong and Kim, 2002). Further, it is difficult to assess in advance how an ERP transition will evolve. On the one hand, it is often assumed that ERP implementations represent radical changes (Lozinsky, 1998), so technological, business, organizational and individual changes must all be dealt with simultaneously (Bancroft et al., 1998; Schneider, 1999). On the other hand, it has also been suggested that ERP packages should be implemented in phases, so that organizational members have more time to understand and assimilate the new software and business environment (Bancroft et al., 1998). The element of IT infrastructures discussed in this chapter is an ERP-based solution and its associated modules. However, it should be recognized that this system is not an isolated artefact, but an integrated part of a larger infrastructure, and that large infrastructures cannot be changed instantly, but only piece by piece and over time (Ciborra et al., 2000; Star and Ruhleder, 1996). This is a complicating factor because when an infrastructure is subject to changes it has to be operational at all times to support the daily activities of the host organization, which severely constrains the introduction of new elements. Hence, the existing infrastructure – the installed base – has a strong influence on the scope for its future development (Grindley, 1995; Hanseth and Monteiro, 1998). Corporate IT infrastructures are thus often emergent (Truex et al 1999) since they are typically established through side-effects of, and spill-over from, the implementation of increasing numbers of installed base elements as well as their closer integration (Rönnbäck et al 2007). It has also been suggested that the complexity of information systems (IS) may increase due to ambitions to build perfect, universal solutions, which lead to systems that are more difficult to manage (Benbya and McKelvey, 2006; Hanseth et al 2006). In this context, ERP systems can be viewed as tools for control and integration, but their implementation may also raise many new issues regarding integration and unforeseen side effects (Hanseth et al., 2001). Indeed, they may provide paradoxical examples of reflexive outcomes, that is, the opposite result of what was intended, and of “technological drift” as a result of striving for increased control. While ERP systems can be viewed as tools for organizational control, they are also complex systems affecting not only technical aspects but also business processes (Ciborra et al., 2000, Hanseth et al., 2001). The introduction of complex systems has often been shown to have negative, unexpected side effects such as decreased instead of increased order and control (Hanseth et al., 2006). Limitation of these side effects is thus an important issue both in theory and practice. This defines the adaptability problem of IT infrastructure design: designers must produce architectures that can cope technically and socially with the growing needs for flexibility so that the IT infrastructure will continue to adapt and grow.