KEY POINTS
- There is a particular focus in the construction industry on improving scheduling, cost estimation and management, and claim management.
- In particular, the use of 4D and 5D Building Information Modelling (BIM) is seen as one of the major advances in improving efficiency. It is currently being used as a planning tool for both scheduling, sequencing of works and early detection of clashes between various trades or construction components.
- In construction disputes, this technology would allow for early analysis of delays, cost overruns and for evidence-based mitigation to be put in place.
- With the addition of AI, 4D & 5D modelling may become as advanced as to predict problems in advance of them occurring
Introduction
The construction industry faces a myriad of challenges as the global population and world connectivity increases, creating growing demand for infrastructure in a variety of sectors such as housing, transportation, energy, education, healthcare and defence. The challenge is exacerbated by the need to reduce global carbon emissions, with the construction sector contributing 37% to global emissions according to the United Nations (UN), largely froin m carbon intensive materials such as concrete, steel, plastics and aluminium.
In addition, shortages in skilled labour are further reducing the ability for the industry to meet demand. Governments are making efforts to rectify this by launching training schemes and incentives for those joining the sector, but they seem to be having little effect in the immediate term.
These complexities are compounded by an increase both in the volume of projects and in the complexity of infrastructure programmes, with so-called 'mega-projects' becoming more common. Despite contributing around 8% to global GDP, according to industry estimates, these projects are plagued with problems such as delays and cost overruns, damaging public confidence in the industry and ultimately the utility of the project.
To counter all of these issues, the sector must become more efficient and effective in its working methods to reduce delays and manage costs. A key method to improve efficacy is the integration of AI with 4D and 5D BIM.
AI has the potential to increase the efficiency during construction of major infrastructure projects, enhancing analysis of records and schedules post construction, and providing an effective tool for infrastructure managers throughout the building's lifetime.
In this article we discuss the AI developments in schedule analysis, the use of 4D and 5D BIM, and the effects on expert testimony.
The Use of AI in 4D & 5D BIM
The potential of BIM in the construction industry and claims management
The use of BIM in the construction industry has been expanding globally, with most major projects now using BIM in some form. Use cases span from visualisation tools, computer aided design (CAD) tools, clash detection, change management and document control throughout the building lifecycle.
While there isn't a universal definition of what constitutes 4D BIM, several researchers have identified it as a visualisation of time and space. Gledson and Greenwood write that 4D planning "involves linking a time schedule to a 3D-model to improve construction planning techniques". Elecosoft, a manufacturer of construction software, defines 4D BIM as "the process of combining 3D models with time and schedule-related information such as programmes and logistic models to create a virtual construction sequence."
Similarly, 5D BIM lacks a clear definition, however it is generally considered to contain cost information within a 3D model. Access to a common data environment supports the design and construction process, enabling asset owners, consultants and contractors to collaborate and effectively manage changes, version control and innovation as they progress through each design phase. This approach also simplifies management of new legislation such as the UK's Building Safety Act 2022, which requires a "Golden Thread" of information throughout a building's lifetime. The use of BIM software allows change control to be observed and integrated. In addition, Operations & Maintenance (O&M) data, industry specifications and standards can be monitored in real-time to form relevant, timely and succinct recommendations for building operators.
The UK Government Infrastructure & Projects Authority (IPA) has released reports on its ambitions to utilise AI in tackling major UK infrastructure programmes. It highlights three areas pertinent to the construction sector: data quality and availability, evidence-based decision-making and data collaboration. Furthermore, Transforming Infrastructure Performance: Roadmap to 2030 highlights the necessity for constructors to use the UK BIM Framework and the advancements in 4D and 5D BIM.
Wider adoption in the UK and globally is likely to create common operating standards for the use of 4D and 5D BIM across platforms. Several AI based BIM models have been used by constructors and experts to determine delay factors, cost estimation, contractual arrangements, variations, claim management and safety risks. Productivity surveys have found that traditional project management principles and 4D BIM have led to improved workforce productivity, increased the output of individual technical staff reduced waste and improved knowledge sharing. Pictorial and time-based visualisation have also allowed greater understanding of the impacts of client variations and unforeseen technical challenges; which are common foundations of delay and quantum dispute claims. The adoption of AI-enabled 4D and 5D BIM can therefore reduce the volume of claims by tackling the root causes.
Developments in AI in the construction industry
Further AI developments outside the immediate construction realm have the potential to improve infrastructure development efficiency. For example, the use of predictive weather models to positively identify inclement weather would, when integrated with 4D/5D AI BIM, allow smart rescheduling of projects, early understanding and mitigation of potential extension of time claims and consequent variations in productivity and cost.
It should be noted that despite increasing popularity in other industries, uptake of AI in construction remains slow. Moreover, the lack of common standards between organisations and countries prevents its effective deployment on large multi-national projects. Further collaboration between major players in the global construction arena is needed to form common processes and a business-as-usual approach for AI-enabled 4D and 5D BIM.
AI in Construction Delay & Quantum Dispute Analysis
How can 4D & 5D BIM predict the most common delays
Construction claims and disputes are one of the key blockers to improving efficiency in the construction industry. The identification of claims, preparation of supporting materials and subsequent analysis can prove a lengthy process, especially when a claim goes through arbitration, adjudication and litigation proceedings. After discussing how the use of AI in the BIM environment may reduce the need for a claim, we investigate how AI can support dispute experts in their analysis and forming their opinion and testimony.
A common problem faced by in-house claims experts or external consultants is the speed with which it is necessary to analyse documentation. Documents control is often lacking, while the complexity of detailed technical documents makes it challenging to understand the design philosophy and establish critical paths and cost models. Contractual obligations tend to force clients, legal professionals and consultants to act quickly, considering the US FIDIC standard form contract, which determines that claims must be raised within 28 days of the alleged delay event. These constraints reinforce the need for effective information management and collaboration between parties during design and construction, but also the requirement for disputes experts to be able to efficiently analyse data and form comprehensive recommendations.
There has been considerable research in recent years into the development of predictive AI models for claims management. While different approaches have been used taking into consideration different AI tools and the merits of individual technologies, there is increased evidence that AI is capable of consistently producing the same delay and quantum conclusions as human experts. It is also apparent that the traditional, human expert analysis approach will continue to be enhanced by advanced AI analysis.
Research papers have found that delays and cost disputes can be accurately predicted using AI. These models considered several typical factors causing construction disputes. One study produced three models which achieved an accurate understanding of the criticality of delay, with an accuracy of 89% to 98%. A similar study, used to predict and identify causes of delay based on real world data, achieved an accuracy of 91.97%.
Another study analysed the outcomes of ten construction dispute cases to attempt to establish whether AI could accurately predict the outcome compared to the traditional approach. This study achieved an accuracy of 93% and accounted for a number of dependant and independent causation factors which were established using surveys and literature reviews.
The conventional vs the AI approach
The use of an AI-based approach in analysing construction disputes presents significant potential value. However, current research underscores the importance of employing such tools as supportive tools to guide analysis, rather than be fully relied upon. In many dispute scenarios, the reliability of data and contemporaneous records is questionable, and expert reports must be caveated with how appropriate the evidence is when drawing conclusions. Little evidence has been found to suggest that AI could evaluate the quality of critical data such as contractor site reports, programmes, purchases orders and communications.
The speed at which AI is able to operate may allow for multiple different types of delay and quantum analysis to be used in each case, as opposed to the single type of analysis commonly used by humans.
Retrospective and prospective analyses could be made readily available due to the adaptability of the tools. This would allow effective comparison and comprehensive reviews of expert claims to come to a more informed opinion. At present, a conventional approach should be maintained, however guided by AI tools, to allow for efficient analysis. Whilst case studies have demonstrated the potential of AI in predicting case outcomes, it is not yet a replacement for human litigators.
Conclusion
So to what extent can experts and legal professionals across the construction industry rely upon AI for expert analysis?
With some AI experts predicting that computers will have human-level intelligence by 2029, does this mean that construction practitioners have four years to look for alternative employment? Are the Courts and arbitrators going to have a solution where they press a button rather than deploy expert testimony?
It has been shown that AI can accurately predict the outcomes of various construction scheduling scenarios and, to some extent, there is a possibility this can be used to predict the outcome of cases.
However, can we rely upon the source data? When researching the background to the circumstances of a case, the quality of the data made available on the internet and other industry sources is mixed. It may be challenging to retrace the sources of the data to ascertain the quality of the analysis.
When reviewing industry norms, productivity circumstances will vary according not only to the sources, but to human factors, weather, geography, external influences and impacts – all of which will require some form of expert input to determine.
Typically, most construction experts provide opinion based on their experience and interpretation of the facts presented to them. We expect that, rather than AI taking over entirely, the more likely scenario is that experts will become more adept at using AI data to enhance their analyses ad complement their experience.
In summary, AI will act as a key tool in improving the efficiency and cost of construction projects through the use of BIM and other software. There is a strong argument for the use of AI in conjunction with BIM, scheduling and costing in the sector. These enhancements should be embraced, whilst recognising that AI is a tool and not a replacement for expert analysis.
Originally published by 13 May, 2024
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.