Data and Analytics are transforming the construction industry as companies are able to increase productivity, reduce risks, increase safety and make more informed decisions. I caught up with Adinda Savitri, Founder, Nucon one of Singapore’s hottest ConTech start-ups looking to revolutionize the construction industry.
ND: In BWT tradition, please start by giving me the Nucon elevator pitch.
AS: Nucon is the AI Quality Manager for construction. It automates the analysis of inspection reports and gives early warnings for project threats. It assists directors of mega projects in decision making. Nucon’s knowledge graph learns concepts like experts do, and creates unique recommendations for each project in order to ensure completion on time and on budget.
ND: The construction industry has been notoriously slow in the adoption of innovative technologies and digital transformation. Why do you think this is?
AS: Yes, this is because margins are so thin. Construction businesses are treading water. There is almost no time to think of innovation when you’re constantly fighting fires. A McKinsey research report recently found that the construction industry on average spends less than 1% of their revenue on R&D. In comparison, the aero sector spends up to 4.5% of revenue on R&D.
ND: Is this trend changing?
AS: Definitely. One factor is the push and support from the government. Singapore is doing a good job at enabling traditional businesses to apply technologies with both mandates and grants for BIM adoption and integrated development and delivery (IDD). Labor crunch means there is a push for automation. I think people feel that innovation is inevitable for survival, so better adapt or lose. The statistics from OECD found that there is a 50% increase in spending on R&D by construction players in Singapore from 2010 to 2013.
It will be great when the industry stops being dependant on grants for innovation. It takes a certain level of maturity and growth culture for companies to overcome the activation energy to breed innovation. ROI of innovation itself is very hard to quantify and justify. So innovation projects often get prematurely killed before they get the chance to show benefits.
A technology will never fit a company like a glove immediately. It takes collaboration and patience from both technology adopters and providers to fine tune the ideal workflow. Construction businesses get pitched thousands of new tech each year — hundreds of which get trialed, but only one or two reach stable adoption. Not because of failure in technology, but because the customers have a hard time incorporating the technology into their natural workflow.
ND: What are the major benefits for construction companies in the utilisation of machine learning and data and analytics technologies?
AS: We have seen excellent applications of machine learning in combination with laser scanner or drones at replacing visual inspection onsite. These technologies are quite mature and can be quick wins. Though the area of data science that I am most passionate about is all about the fundamentals.
Traditional businesses are rich with data. The most valuable application of data science is for internal understanding of actors and drivers within a business. This is the type of knowledge that you typically have to rely on external consultants to get answers for. Which is ironic, because chances are you already have the answers in your internal data all along; just not visible to the naked eye. Even simple analytics can be a huge help.
For example, at Nucon, we discovered that 20% of construction defects contribute to 80% of the costs. This is a crucial piece of knowledge because strategic prioritization and allocation of resources can push project productivity in late-stage construction, translating directly to bottom-line. And we achieved it from analyzing otherwise-boring paper data that businesses already have aplenty.
At Nucon, we are now able to surface unique characteristics of each contractor in a construction project: how efficient they are; their areas for improvement; top problems and challenges; etc. And we’re only getting started!
We like to remind our clients that our insights should be used with the right mindset: it’s all about improvement and not blame.
ND: How was Nucon founded?
AS: It all started with my love for Sichuan food. In university, I have a group of friends, mostly civil engineers, who gather regularly for spicy food in Chinatown. We always talk about site problems over dinner. Some were incredibly scary, and some were comically hilarious. Most site problems are funny when you don’t have to pay for them.
My background is in computer science, so I always pointed out something like, “Hey this scheduling problem has been solved in CS a long time ago! Why can’t we do the same in construction?” Late nights at Chinatown was how I got to understand the industry so well and all of its quirks. At that time I worked as a tech consultant, and that gave me an edge with construction clients looking for digital transformation projects.
The Chinatown gang told me if I wanted to start somewhere it’s gotta be on defects. That’s what sparked my interest in defects research. A year later, Lendlease and JLL started the first Propell Asia Accelerator. I pitched Nucon.io as a concept, and it was one of the five startups selected from 100+ other applications from the region. The people at Lendlease were pivotal at helping us find our problem-solution fit. We have made many strategic pivots since Propell, and here we are!
ND: What's been your funding model so far, and are you currently raising capital?
AS: We are completely bootstrapped and raising seed fund later this year. We have a strong preference for strategic investors who understand construction and are as passionate as we are at transforming the sector. The future of construction is data-driven, and Nucon wants to be the Bloomberg for construction.
ND: What's the next big step in terms of growth?
AS: Nucon is unique because we have the machine intelligence that has been trained on a diverse set of production data in SEA. Our AI is robust enough to handle residential, industrial, and life science projects.
We started getting interests from oil and gas, engineering, and infrastructure projects due to the high-stake nature of such projects. I don’t think there is anyone who has seen inspection data from those projects all in one place and try to analyze it. So it’s an exciting exploration for us, which will definitely open up a new adjacent market for Nucon.
Our goal in the next few years is to conquer the Singapore building market. We are in category creation phase with our AI, so we’re constantly socializing our vision to forward-thinking businesses. We’re lucky that we’re in Singapore. We have access to many quality-driven MNCs with regional HQs here. Nucon is very well positioned for regional expansion in the coming years.
Nucon was a Featured Technology of Built World Technology Asia
Interview conducted by Neall de Beer, Director, Built World Technology Alliance.