Origins of APM and the theory of Complexity in Projects

The ever-changing business needs, drivers and requirements demand project management approaches that are flexible and adaptable to deliver to the market desired products/services faster and satisfy customer requirements (Macheridis, 2009; Weinstein, 2009; Shenhar, 2004). The failure of traditional project management approaches to meet such demands in all situations led to the evolution of agile project management (Augustine and Woodcock, 2008).

Agile project management (APM) methodologies are increasing becoming popular across different businesses (Macheridis, 2009; Griffiths, 2007; Chin, 2004) and hence there is a need to understand their origin, applicability and implications for the other business industries such as consulting firms. This is particularly important if one considers the successes registered in other industries and the recent efforts as well as a rise in interest by scholars to explore the relevance of agile concepts to construction and manufacturing where its implementation improved project managers’ response and effectiveness in an unpredictable environment (Fernandez and Fernandez, 2009; Owen et al, 2006).

Although many scholars agree that APM methodologies emerged from software engineering agile frameworks such as eXtreme Programming (XP) and Scrum in the 1990s (Larman, 2004; Boehm, 2006; Cicmil et al, 2006; Fitsilis, 2008; Hoda et al, 2008; Macheridis, 2009), Aguanno (2004) traces their development to the 1980s when the Japanese automobile manufacturers embraced them in their product development. He mentions that they were initially known as light weight methods before the adoption of the term agile to show their impact on projects experiencing high levels of change. This stance, however, is somewhat controversial because Aguanno combines both lean and agile. According to Augustine and Woodcock (2008) APM principles and practices are hinged on the ‘new science’ theory of complex adaptive systems (CAS). This complexity theory is derived from the ‘chaos theory’ which is defined as the study of how order and patterns arise from apparently disordered systems (Elliot, 2008). It is more concerned with understanding how complex behaviour and structures emerge from simple underlying rules as observed in the flocking of birds and ant colonies (Augustine et al, 2005; Fernandez and Fernandez, 2009).

Complex adaptive systems are such that each ant colony follows simple rules and behaviour at the localised level whilst a collective emergent behaviour is exhibited at the macro-level from their individual actions (Augustine et al, 2005). According to Larman (2004) APM teams are viewed as CAS because they deal with a chaotic system that manifests itself as the project progresses due to the uncertainties surrounding the future. Since consultancies do not exist in a vacuum, it may be argued that they also sometimes face unpredictable and uncertain situations and their accompanying challenges of (1) planning for uncertain outcomes, (2) balancing flexibility with reliability and accountability, (3) balancing decision quality against decision speed and (4) timing scope freeze during rapid change as espoused by Collyer and Warren (2009). Therefore under certain conditions teams from consulting firms may need APM skills from CAS to survive.

Augustine et al (2005) state that CAS have a self-organizing ability and are capable of adapting to environmental changes even though their behaviour is not governed by any central regulation. Since CAS encourages interaction among the sub-parts and/or between the environment and the system to establish an adaptive behaviour, its application to project governance through APM may eliminate some of the weaknesses associated with traditional project management due to its flexibility and context specific approaches to management. It is interesting to note that a complex global behaviour in CAS is a result of simple, local rules that guide the interaction between the semi-autonomous building blocks (Augustine and Woodcock, 2008). Elliot (2008) postulate that if this concept is applied to project governance, it will create more time for project managers to concentrate on more pertinent issues rather than controlling because agents will be semi-autonomous (i.e. requiring minimum control). However, it can be argued that the potential generalisability of such findings has its own pitfalls and is a subject of debate because according to Bryman and Bell (2007) non-human behaviour in these species does not apply equally to humans and the findings should be treated only as unique to that particular species. This is supported by Leybourne (2009) who suggests that only some elements of CAS are applicable to APM. Therefore it may be necessary to adopt such generalisations with the necessary caution that they deserve