Under Construction: A Week In Tech & Innovation
Amazon, AI, and Knowledge
What I’m consuming
I’m constantly scanning the news, social media, and industry for case studies, innovations, and generally interesting construction related content. Here’s what I’ve been consuming recently:
Bouygues and Amazon Partner for Construction Digital Transformation
<link>
The story: Bouygues Construction and Amazon Web Services (AWS) have announced a multi-year collaboration to accelerate the digital transformation of the construction industry. The partnership aims to leverage AWS's cloud technologies, including generative AI, to improve project efficiency, sustainability, and safety. This collaboration will focus on developing and deploying innovative solutions for areas like building information modelling (BIM), project management, and on-site operations, ultimately streamlining processes and enhancing decision-making across the construction lifecycle.
Why it stood out: For years now I’ve had this niggling thought that once a tech juggernaut (or a group of software engineers from said cohort) figure out the state of tech in the construction industry that we are ripe for disruption. Amazon linking up with one of the biggest construction firms on the planet gave me pause to reconsider this. No doubt this will give Bouygues a strategic and tactical leg up to look at their processes and rethink them with a tech giant.
I look forward to seeing how deep they go in rebuilding the backend of a such a business to modernise and leverage emergent technology such as AI. Hopefully, for the sake of the industry, we see success and it solidifies the need to ‘build in, not bolt on’ technology.
However, this industry is fragmented with deep rooted path dependence, and an established incentive structure. This has tempered my prediction of impending disruptions in recent months. Still, very interesting.
AWS Leverages AI for Construction Document Analysis
<link>
The story: Staying in the Bezos empire, Amazon Web Services (AWS) has detailed how its AI services, including computer vision and large language models (LLMs), can be used to analyse construction documents. The solution enables automated extraction of key information from various document formats, such as drawings, contracts, and specifications. This automated analysis can improve efficiency, reduce errors, and accelerate processes like cost estimation, risk assessment, and compliance checks, leading to better project outcomes.
Why it stood out: This is extracting real knowledge from AI. There have been a few attempts at this from start-ups since Machine Learning and AI became accessible a few years ago. The striking thing here is that a giant like Amazon - with its vast resources, scale and distribution - is getting into this area. Whilst I’m a fan of start-ups and SMEs it’s super interesting, and perhaps exciting, to see a juggernaut take on construction problems.
Survey Reveals Emerging Tech Adoption in Construction
<link>
The story: A recent survey has highlighted the growing adoption of emerging technologies within the construction industry. The survey reveals that technologies like IoT, 5G, and AI-powered tools are gaining traction as companies seek to improve productivity, collaboration, and sustainability. The findings indicate a shift towards data-driven decision-making and a greater focus on integrating technology throughout the construction process.
What stood out: The shift and acceleration toward data-driven decisions and integration is super encouraging. Though I prefer the term data-informed - leaving rightful space for intuition and experience - there is still a hesitance and skill gap for construction to struggle through to really exploit the data and knowledge we capture every day.
Genia Secures Funding for AI Structural Planning Tool
<link>
The story: Genia, a startup developing AI-powered tools for the construction industry, has secured $3 million in funding. Their flagship product, Structural Copilot, uses AI to generate physics-validated structural plans. This technology aims to automate and accelerate the structural design process, allowing engineers to explore different options quickly and optimise building designs for cost, performance, and sustainability.
What stood out: Speed. I think this is the underlying value here. If Genia is successful a tool which can offer compliant and sound structural options based on inputs provides a huge acceleration to the design and optioneering process. That said they aren’t here yet.
It’s great to see great minds putting effort toward construction pains and venture capitalists putting their money behind them.
Replit Agent App
<link>
The story: Replit launched its app, which uses AI to… make apps. The "Replit: Make an app for that” ad shows users using simple text prompts to develop their own apps.
What stood out: I used it. It is mind blowing how far we’ve come in a few years. I can’t help but imagine a future where user interface generation is unrecognisable in the next few years. One possible future is that construction companies manage and provide the trusted data sources and pipelines but the user defines how they interact with that data based on the specifics of the project, task, and personal preference. The mind boggles. What’s for sure is that data management will become more critical as citizen development gets more and more accessible.
Substack is a great resource for knowledge. Here’s a recent post which stood out:
When do you need a Knowledge Graph? by Kurt Cagle
The article: A fantastic resource for understanding in which situations a knowledge graph can add value.
Quote: “Knowledge graphs should be considered an integral part of your organization’s data life cycle - to facilitate integration, establish consistent terminology, store complex data structures, improve search and transformation, and ground artificial intelligence systems. They should not be seen as just another database but as a data hub to connect various heterogeneous data systems and applications and, as such, to ultimately power the applications and data processes that run your organization.”
On my mind
I’ve been thinking a lot about the interconnectedness of projects and construction. And the absurdity of managing interconnectedness in a fragmented industry.
It feels like those who can find a solution to managing these two realities can produce better results.
There is lots of concurrent and siloed working in construction. But at the end of the day that fragmented industry must come together to deliver the same assets in the built environment.
In a project, each ‘silo’ has an impact on others. Design decisions impact procurements and methodology. Methodology decisions impact budget, safety, and environmental. Logistics decisions impact design. And so on, in an endless complex loop.
I picture this as a network. An interdependent and complex network.
We have tried to tackle this. As a leader in digital technologies my mind goes initially to BIM. Aiming to mobilse, share, and coordinate information - done will it fosters collaboration toward the end goal.
There is also Planning and Scheduling. Contract and Scope. And on, but it strikes me that the problem of network complexity and the dynamic impact has never truly been solved.
Our current tools often look forward in forecasting and look back after the event/decision, but rarely tackle the immediate and dynamic nature of events and decisions being made.
We have entities (people, teams, assets, documents, etc…) and we have relationships (how these entitles relate to each other).
This takes me back to Knowledge Graphs as a potential solutions. This will be the focus of my next article.