While the infrastructure industry is getting to grips with AI capabilities, its true potential is still to be fully unlocked, according to a new report from FIDIC and EY.
A new report from international engineering federation FIDIC (International Federation of Consulting Engineers) and EY highlights how AI has the potential to transform the infrastructure industry, but only if the sector speeds up its adoption to unlock new efficiencies and opportunities across the asset lifecycle.
The infrastructure industry faces major challenges over the coming decades, including climate change, urban expansion, economic uncertainty, labour shortages and supply chain disruptions amongst others. To address these issues, the report, How artificial intelligence can unlock a new future for infrastructure, calls for new ways to deliver projects faster, more cost-effectively and more sustainably. “AI will certainly be a crucial tool in this effort, but while the industry is getting to grips with AI capabilities, its true potential is still to be fully unlocked,” says the report.
To better understand the industry’s current perspectives and AI adoption, EY surveyed FIDIC Global Leadership Forum (GLF) members – representing the key players and leaders in the infrastructure industry – and ran a workshop at the FIDIC GLF meeting in April 2024, focusing on the following areas.
- Their current use of AI.
- Barriers to AI adoption in infrastructure.
- What is needed for wider AI adoption.
- The potential benefits of AI.
The survey results from 44 senior leaders and industry observations showed a mixed picture. The respondents said that they are investing in AI to varying degrees, in line with findings from other industry reports. Some organisations are investing conservatively (less than 2% of their revenue) while others are taking bolder steps (of up to 10%). In addition, startup funding for AI-driven businesses in the built environment is outstripping that of AI-enabled fintech solutions. However, this investment is resulting in spot solutions, mainly focused on the early project stages.
The report says that this is understandable when AI solutions are trying to operate in the traditional, siloed delivery models that define infrastructure delivery today. Other barriers to AI adoption include unclear returns on investment, cultural resistance, skill gaps and technological fragmentation.
Given those barriers, it’s little wonder that AI adoption is low and that the industry is yet to realise its true value. To address this and to help move the industry to an informed sector that understands the benefits of AI and deploys it to the best effect, the FIDIC/EY report seeks to bring greater clarity of the different types of AI in the market, how they can be used in different stages of an asset lifecycle and the benefits and outcomes they deliver to the sector and society more broadly. It also explores how AI can unlock new opportunities and value across the infrastructure ecosystem and for the many stakeholders who work within it.
Although the report concedes that the sector has made a good start into the world of AI, it says that to harness its potential to improve decision-making, build resilience, increase productivity and enhance outcomes throughout an asset’s life, the sector needs to accelerate its adoption of AI tools.
The report advocates looking at AI through the lens of a ‘systems thinking’ approach to infrastructure development. With its immense processing power and advanced abilities in data analysis, relationship discovery and outcome prediction, AI can handle complexities that traditional linear approaches cannot. AI can analyse vast amounts of data to uncover hidden patterns, predict potential issues, and best use resources across all project phases. This would break down barriers between stakeholders, reduce costs and expedite delivery.
Five guiding principles that frame the traditional asset lifecycle can support the ‘systems thinking’ approach and illuminate AI’s potential, the report outlines. By building the connections of various factors across all five guiding principles, better decision-making can be achieved.
- Determine purpose − AI improves asset planning and selection.
- Plan end-to-end delivery − AI enables more comprehensive insights for robust business cases.
- Confirm operating model − Data-centric digital twins facilitate AI adoption.
- Integrate ways of working − AI enhances controls and performance oversight.
- Operate responsive assets − AI improves operations and sustainability.
The report says that accomplishing the systematic change will require targeted cross-stakeholder collaboration across eight key groups – asset owners, operators, consultants, contractors, government, academia, technology providers and capital investors – all of whom must align on shared ‘ambitions’ to use AI to improve infrastructure.
