Fourteen of the eighteen ‘complex systemscharacteristics’ identified by Lucas (2000) are placed in three groups, whichare set up with an aim to understand construction from a complexity point ofview. These groups are then used as the tools for analyzing construction from acomplex perspective.
Autonomous agents comprise ‘constructioncomplexity characteristics’ that are not identical. All of these agents areequally valuable in the operation of the system and no executive or directingnode exists by design in the system. Therefore any control structure orleadership must emerge by self – organization (Lucas, 2000).
The undefined values indicate that thepurpose of the system interface with the environment is not initially specifiedbut must evolve. This requires that a communication is created dynamically bythe system as a result of environmental interaction (Lucas, 2000).
Finally, complex systems are non – linear.Their outputs are not proportional to their inputs. This means that taking theproperties of each part and adding them will not give a valid solution tooverall fitness. In other words the whole is different from the sum of theparts (Lucas, 2000).
It is crucial to be understandable that ‘complexity’ is a multi – dimensionalphenomenon, so it cannot be treated as something simple. If someone would liketo simulate complexity then this simulation would definitely lead in apolyhedron geometrical shape with a dynamic movement or rotation. According toBaccarini (1996), the two most important dimensions of complexity are theorganizational and the technological or technical one, and each one of them canbe further subdivided into more categories according to its differentiation orinterdependency level. However, because of the polyphony and the differentperspective of each researcher, more complexity dimensions are named andexamined, such as managerial, operative (Gidado, 1996), social (Wild, 2002),etc. These are some of the most important dimensions that compose ‘projectcomplexity’. Therefore, there are developed non-linear dynamics, usuallyclassifying the complexity phenomenon in the edge of chaos (Bertelsen, 2002).
In a project context, measurement givespeople the information they need to make decisions and to manage a project to asuccessful conclusion. A measurement is an observation that identifies anddescribes reality and is expressed as a quantity (Kerzner, 2011). Measurementis in fact a relationship or mapping between the thing being measured andnumbers (Kerzner, 2011). A reasonable question that arises is if ‘constructionproject complexity’ is a measurable concept. Some researchers studiedcomplexity from their own perspective proceeding in the measurement of thecomplexity level, in order to classify and name the degree in which the projectis complex and therefore some of them to further investigate project’s potentialfailure and the possible way of avoiding it. Two brief examples of such kindsof attempts are given:
The first is an attempt of measuringcomplexity for building projects, known as ‘Delphistudy’. The aim of the study was to identify complexity measures especially forbuilding projects for the Republic of China people. The used methodology inorder this to be done was based on a three round Delphi questionnaire survey,for the purpose of identifying the key parameters that measure building projectcomplexity. Six key parameters – metrics – were found and weighted. For thefinding of the most appropriate parameters, the opinions of expert professionalpanels were asked. The importance weighting for each identified metric -parameter calculated and validated afterwards (Xia, Bo & Chan, Albert,2011). Because of the fact that building project complexity was hard to be preciselyquantified, the research team focused on identifying and weighting factors /aspects related to project complexity.
Another example of researcher (Ion, 2001)gives examples of mathematic models that calculate project complexity ingeneral. The critical thing is that these formulas do not measure the overallcomplexity but the complexity of a specific product; this is either simple orformed by an aggregation of diversity elements.
The important point to mention, whichbesides is obvious from the above examples, is that each researcher follows hisown perception about project complexity. All of them agree on the fact that inorder complexity to be measured each time, either qualitatively orquantitatively, interdependencies and interactions between different butsimultaneous developed phenomena into the project’s bay must be combined andvaluated (Baccarini, 1996). In the current thesis, an effort will be made inorder all these to be combined so that to conclude in a more generic measuringmodel’.