By making factories self-aware, Singapore-based AI startup FireVisor Systems is kickstarting a new field in manufacturing called cognitive automation.
From flashy cars to life-saving drugs, items that improve our everyday lives were all brought to life on a manufacturing line somewhere. Indeed, the manufacturing sector is a significant contributor to the global economy, accounting for approximately 16 percent of global GDP and 14 percent of employment, according to a 2012 report by McKinsey.
While these industries all rely heavily on global innovation, today’s factories are, ironically, themselves not all that clever: human inspectors pace the factory floor to make sure no fault escapes the production line, while process control engineers constantly monitor the machines to ensure they don’t stall or break down.
Now imagine a self-aware, thinking factory that has enough cognition to learn its own processes and improve its own productivity. This is the mission of Singapore-based AI software solutions startup FireVisor Systems, founded in 2018 by Surbhi Krishna Singh and Long Hoang. Together, they have built a machine-learning enabled analytics platform that uses data from the manufacturing line to perform engineering failure analysis in real-time.
FireVisor is on a mission to bring the power of data science to the manufacturing floor, and in the hands of humans with a few clicks.
Finding Meaning from a Deluge of Data
Before co-founding FireVisor, Singh built automation systems for semiconductor and memory chip makers, saving them thousands of unproductive work hours. “I worked in manufacturing before, and I saw first-hand how useful information was getting lost because all of this data is inexplorable,” she shared.
When Singh met co-founder Long Hoang, who now serves as the company’s chief technology officer, they originally planned on improving the machine vision systems used in factory lines by automating and improving visual checks. But after speaking to industry experts and line personnel, they realised that they were not fixing the root of the problem—how to best handle the deluge of data produced by factories.
“Factories today are extensively using ‘if-else’ type of automation. This creates many islands of intelligent machinery inside the manufacturing line, but none of them are talking to each other,” Singh explained. “So if something goes wrong with machine X, which is step one of the manufacturing process, machine Y at step three of the process would know nothing about it and continue processing defective material.”
This realisation led to Singh and Long on a journey towards pioneering a new field called cognitive automation, which is to give learning and predictive capabilities to manufacturing lines across any industry.
There are two parts to FireVisor’s value proposition: first, its visual defect detection system allows for defect detection at a level that is above the accuracy of human operators, and requiring as few as 200 defect images to get started. Second, its proprietary deep learning algorithm uses the data to predict why those defects are happening, reducing unscheduled downtime and maintenance.
“Our strength is in dealing with data. We are able to extract the data, fill it in the missing information with machine learning, and then synthesise meaning and find value from it,” Singh said.
In 2019, FireVisor raised close to S$1 million in seed funding in a round co-led by the 500 Startups Durians II Fund and Acequia Capital. SGInnovate and Entrepreneur First (EF) also participated in the round. Prior to the seed round, the company received grants and support from NVIDIA, NUS Enterprise and the Vizag Million Dollar Challenge.
FireVisor currently follows a subscription model for its existing products, plus a fixed one-time charge per machine. Moving forward, Singh plans to switch to a model that allows FireVisor’s clients to pay according to the data volume the company handles for them.
The young company has an impressive client base in Southeast Asia, India and China, including manufacturers such as REC Solar, a European company brand of solar panels. Accordingly, the company claims that its defect detection and predictive analytics systems can help its customers save manufacturing value equivalent to 15 to 30 percent of their revenue.
Besides manufacturing clients, Singh also plans on extending the company’s platform to startups and the R&D community. “Our defect detection system can also be used across industries for all types of visual defect detection and analysis. Although our products are tailor-made for the manufacturing line, they can also be used by research labs and institutes that engage in new technology transfer to factories.”
A Happy Challenge
If there is any bigger challenge than endowing factory floors with artificial intelligence, it would be scaling up the management team and expanding FireVisor’s client base, Singh said, calling it “a happy challenge”.
“Building a company is a new challenge every day, but tackling these challenges is what makes the process very fulfilling as well. For us, handling various client projects while expanding the team and scaling up is the obvious challenge. Also, thankfully for us, we have a great team, helpful advisors and supportive investors on our side.”
Just like how their industry clients have benefitted from FireVisor’s analytics programme, Singh said founding team is also looking to receive feedback on how best to improve their products. “In general, we value feedback a lot, and we always act upon it to assess the best possible way of doing a particular thing,” she said.
As founders from the third cohort of EF, a platform which brings together individuals to build startups from scratch, Singh and Long developed and launched FireVisor during EF’s intensive six-month programme at SGInnovate’s office.
“SGInnovate has been immensely helpful to us not only for navigating the local ecosystem, but also because they go above and beyond to connect us with their global network of partners. They have even helped us with regulatory matters in Singapore,” said Singh. “Most importantly, they have a brilliant portfolio of Deep Tech companies and founders, all of them one call away and a huge resource to tap into.”
At SGInnovate, we work with entrepreneurial scientists to build and scale their companies. These companies develop solutions to solve globally relevant challenges.
FireVisor is one of our portfolio companies. Read more about our other investments here.
Share this with your network!