In his 1950 paper ‘Computing Machinery and Intelligence’, the godfather of computer science, Alan Turing, posed the question: can machines think? He cannot possibly have known then how far technology would come in the 70 years that followed.
What is artificial intelligence?
In a lecture in 1947, Turing is recorded as saying “What we want is a machine that can learn from experience,” and that the “Possibility of letting the machine alter its own instructions provides the mechanism for this.” Arguably, this would represent the beginning of the meaningful exploration of artificial intelligence (AI).
In its infancy, programming mostly focused on puzzles and games. Today, AI has implications for almost every sector and every aspect of our lives, and construction is no exception.
AI for project delivery
Projects in all their forms are vital for growth and economic prosperity. Yet, according to research by Professor Bent Flyvbjerg and Alexander Budzier of the University of Oxford, the probability of delivering a major project within the sanctioned cost, time and benefits envelope is only 0.5%. AI has the potential to change all that.
The primary benefit of AI in project delivery lies in its ability to process enormous quantities of data. By analysing past and current project data, AI can enable more effective decision-making around project delivery – better predicting the future performance of a project by considering all the variables.
Will AI in project delivery render humans obsolete? I would argue that there will always be a place for us as decision-makers – and thanks to AI we will have considerably more information at our fingertips.
AI for the real world
So how is AI manifesting itself on today’s projects? Rapid advancements in capabilities of technology mean we’re seeing a huge increase in the number of tools and services coming into the project delivery space that utilise some element of AI or machine learning.
Tools like nPlan use machine learning to review the realism of project delivery plans with a view to identifying any unforeseen risks. To give one example of its impact, on a major liquefied natural gas (LNG) project with a project outturn of US$50bn, nPlan correctly forecast a 15-month delay.
Had nPlan existed before this project was commissioned, it predicted the owner could have saved US$1bn through optimal project selection, greater certainty and cost savings from avoidance of overruns.
Hackathons: a practical approach to AI in project delivery
If you’ve never heard of Project:Hack, you’d be forgiven if you’re imagining teams of computer geeks trying to break into the Pentagon from their basements – but you’d be wrong. Hackathons are essentially events where a group of people come together to find a digital solution to a given challenge.
From leveraging blockchain technology for smart construction contracts to revolutionising the recruitment process and exploring the potential of natural language extraction, the creativity and ingenuity on display is truly inspiring. These hackathons are a great way of helping project professionals to better understand AI and learn how it can be applied to project delivery.
We’re sponsoring another event in October, and I’d highly recommend getting involved.
Educate and inform
There are also a number of bodies entering the space seeking to educate and inform. The Construction Data Trust is a not-for-profit that brings together data from across the industry, collaboratively and securely. Its aim is to reform how projects are managed and to provide industry professionals with the confidence to deliver them successfully.
The advancement of technology is happening faster than we as an industry can adapt, but what is clear is that digital skills should be a core competency for all professionals entering construction. At Gleeds, we have taken proactive steps to prepare for this by launching our own data academy called Upskill.
The future of AI in project delivery
The big paradigm shift is not just in doing things digitally, but in totally reinventing how we do those things. AI is important because it is allowing us to think outside the box.
Going forward, instead of taking a process that exists and asking how we can do it digitally, what we need to ask is: why are we doing it like this in the first place? What would it look like if we were to reinvent it from scratch?
I often wonder what Turing would make of today’s advances in AI. Since he first observed that all tech development is exponential, I don’t believe he would see it as a big leap from his Enigma machines to ChatGPT. He may just be surprised it’s taken so long.
James recently shared his first ‘Global AI in Construction and Property Round-up’ on LinkedIn, providing an overview of six of the biggest AI stories from the past few weeks. If you missed it, you can take a look here.
You may also be interested in: