The development and application of artificial intelligence tools to improve construction risk management

Over the last two decades the construction sector has increasingly been seen as a part of the wider set of manufacturing industries. More recently, there has been an increasing use of digitally based technologies to help manufacturing, including for example the development of Artificial Intelligence (AI) techniques to improve manufacturing productivity, partly by reducing risk and uncertainty in both future thinking and manufacturing processes. This research project will harness AI activity to date in 'traditional' manufacturing industries and proceed to develop and apply a suite of techniques and tools for the sector of National Economic Infrastructure. This will not merely be a repurposing of existing methods - Infrastructure has very different characteristics from traditional manufacturing and understanding this very different and complex context will be key to developing practical tools of lasting value, with a strong focus of using previous experience to reduce risk in both bidding and undertaking capital infrastructure projects.
The infrastructure sector is a significant contributor to the world economy, with an annual spend of approximately $10 trillion on construction-related products and services. However, productivity growth and development within the sector, estimated at 1%, is below the total world economy, at 2.8%, and far below comparable sectors, such as manufacturing, estimated at 3.6%. This can be attributed to several factors, including risk culture, project size and tight profit margins.
Infrastructure projects will often find that information can be presented in unstructured formats, such as with existing drawings, surveys and technical data. This information can often be filled with uncertainty and subjectivity, particularly when related to old existing sites where record keeping may be unsatisfactory. These issues can make it difficult to identify and rationalise the magnitude and probability of risks. Therefore, time and resource are required to extract accurate and relevant information that may be better used in other, more productive value generating areas of the business.
The aim of this research project is to develop tools that can support the management of infrastructure firms' risks. This will primarily focus on learning to identify risks, but also to support the analysis and management of risk associated with a company's activities, i.e. capital projects and asset management services. The tools will need to integrate with the company's existing procedures to ensure effective utilisation to support teams in the delivery of business activities. The tools should help to increase efficiency and accuracy of risk identification and analysis to allow teams to redistribute resource to areas of greater risk more appropriately.