Big Construction, Big Data and Big Chaos
It can be argued that excessive change on mega projects can have unpredictable effects, so how do contractors substantiate this in associated claims for time and money?
You may well think that the term “chaos” isn’t something that should be associated with construction, yet you’d be wrong.
At its simplest chaos theory can be illustrated by the “butterfly effect”, a term coined by the American mathematician and meteorologist, Edward Lorenz.
In 1963 Lorenz created a model of a weather system that led to a powerful insight about the way nature works: small changes can have large consequences. The idea came to be known as the butterfly effect after Lorenz suggested that the flap of a butterfly’s wings in Brazil might ultimately cause a hurricane in Texas. And the butterfly effect, also known as “sensitive dependence on initial conditions” has a profound implication: forecasting the future can be nearly impossible.
The construction industry regularly provides examples of small changes on projects having large consequences. One illustration may be minor changes to a structures design having huge implications on predesigned / preconstructed MEP systems.
Large projects which utilise bespoke elements of design are often executed in dynamic and nonlinear fashions. Such ‘megaprojects’ can involve changes overtime that are hard to predict and are ultimately chaotic in nature. Changes may for example be disorderly, alter existing processes, have small inputs leading to large consequences and require decisions to be made even in the absence of all intended information.
The extent to which a change event on a project can be determined as ‘chaotic’ is however subjective. When considering that contractors are typically constrained by tight timeframes, there are scenarios in which it is not unreasonable to badge the effects of events as being realistically impossible to predict, and therefore chaotic in nature.
If for example, a client was to enhance the design of a bespoke museum roof; which in turn meant that a crane was required to be in a location for an extended period, in order to adequately price and plan for that change the contractor would be required to assess all possible outcomes both internally and externally. The contractor would have to consider the effects of the prolonged presence of the crane on associated work packages / interfaces and would be required to collaborate with subcontractors; who in turn would have to assess implications with suppliers and so forth. It may seem farfetched, but the prolonged presence of the crane could result in lower building works being delayed, which in turn delays flooring works, and consequently results in marble suppliers being unable to meet demands when flooring works are due to commence — therefore delaying the project.
The above scenario retrospectively demonstrates that inherent order does exist; however, the contractor’s argument may be that chaos trumps simple cause and effect on the grounds that issues in relation to flooring works were not reasonably foreseeable. It could perhaps argue that contractual time constraints deemed it impossible to query works and associated supply chains beyond those of the lower buildings, therefore resulting in the requirement to make assumptions for the likes of the flooring; which upon being incorrect, ultimately provoke dispute.
As a claims consultant I regularly see such scenarios, typically associated with design change. Having retrospectively collated claims for extension of time on large airport projects, I have witnessed the extensive evolution of project designs from which complex networks of change events have developed. In isolation it can be difficult to distinguish relationships between vast amounts of change events; however, I find that by plotting their connections graphically, it helps to decipher the chaos.
Such ‘change event graphs’ allow labelling and processing of multiple events, involving multiple parties in one single system. Events within a graph may have multiple connections, loops and cyclical links and as a result allow parties to visually and transparently track change on projects. Collating such graphs on live projects encourages parties to agree on causation (or settle on the reasonable probability of causation) at or closer to the time of event occurrence; ultimately before the task of analysing it becomes too time consuming and expensive.
In absence of such tools contractors and subcontractors continually battle to manage change on projects. I often find that money claims are captured for compensable events with distinct and direct cause and effect; yet events of a more complex nature, with less distinguishable cause and effect, are left on the table or are subject to dispute. As a result, chaotic change events are often unaccounted for within interim claims yet tend to be significant contributors to contractors overspending on the likes of plant and labour. Consequently, and in attempt to recover such losses, contractors retrospectively collate global claims. Most claims for disruption and productivity losses are for example dealt with on this basis. The issue however is that such claims often come under scrutiny, as even retrospectively it is a huge and ultimately expensive task to clearly link cause and effect. This is largely due to the amount of data analysis that is required to undertake the task.
When it comes to substantiating claims, the amount of project data available for mining and analysis is only increasing. The gradual adoption of Building Information Modelling (BIM) within the industry brings with it the requirement for projects to integrate large volumes of information into common shared databases, concerning all components of the building lifecycle. The evolving term for such large volumes of data is referred to as “big data”.
The ability to find patterns and associations within large structured and unstructured data sets allows systems to analyse, learn and predict. The utilisation of tools that permit such smarter analysis of project data will assist parties in establishing more accurate effects of change, both prospectively and retrospectively. It will also help to increase the speed of undertaking such analysis.
It is for this reason that the industry ought to acknowledge that projects are generating ‘big data’, the analysis of which requires the utilisation of smart new tools and processes. There is an increasing necessity to engage specialists like Driver Trett on projects to ensure that data is utilised to its full potential, as efficiently as possible. Similarly, there is a growing need for tools that visually simplify relationships between change events and demystify causation complexities.
