Hear from CIOs, CTOs, and other C-level and senior execs on data and AI strategies at the Future of Work Summit this January 12, 2022. Learn more
While it might not be the first example that comes to mind when thinking of AI applications, AI systems are increasingly being used in the manufacturing sector. In industrial factories and warehouses, AI has the potential to improve equipment efficiency and production yields as well as uptime and consistency. According to a 2021 survey from The Manufacturer, 65% of leaders in the manufacturing sector are working to pilot AI. Implementation in warehouses alone is expected to hit a 57.2% compound annual growth rate over the next five years.
Many barriers stand in the way of successful AI manufacturing deployments, however. Both hiring and retaining AI technologists remain difficult for businesses, in addition to addressing the technological issues associated with AI systems. For example, in a recent report, 82% of data executives told Precisely that poor data quality was jeopardizing data-driven projects in the enterprise — including AI projects.
This is a modal window.
No compatible source was found for this media.
In recent years, platforms designed to abstract away the complexity of AI applied to manufacturing have emerged as awareness of the technology grows. One of these is Elementary, which uses AI to enable customers to inspect manufactured goods down to the individual parts and assemblies. The company claims that interest in its solution in particular has climbed at an accelerated pace as labor shortages worsen. As many as 2.1 million manufacturing jobs could go unfilled through 2030, according to a study published by Deloitte and The Manufacturing Institute.
Elementary AI
Elementary was founded in 2017 by Arye Barnehama, who previously launched and sold wearable technology company Melon to Daqri, an industrial augmented reality startup. Elementary’s no-code platform and hardware allows customers to create routines and train AI models to inspect products for quality assurance by labeling data through a dashboard.
Elementary describes its product as a “full stack” robotic technology, with everything from motor controls to an API that keeps human inspectors in the loop to trace and train the models over time. The company’s robots can learn to perform monotonous tasks and leverage RGB cameras, depth sensors, and AI to perceive the world, allowing them to learn from processes they observe.
Elementary partners with companies like Rapid Robotics, a startup providing out-of-the-box automation products for manufacturers, to deliver turnkey automation solutions to manufacturers. Barnehama asserts that the combination of Elementary’s and Rapid’s products lets customers achieve greater levels of autonomy without sacrificing quality.
Above: One of Elementary’s robotic inspection products.
“Elementary performs use cases from cosmetic inspections (making sure finished goods are acceptable for the end consumer) to defect detection (making sure no critical issues are present in a product) to foreign material detection (making sure no foreign material or objects are present) to label verification (making sure the right label is on the right product),” Barnehama explained to VentureBeat via email. “[M]anufacturers can use the platform for a global view at their production yields, their most common defects, and full reporting to drive insights and improvements to the production line.”
Growth in automation
Elementary’s success — the company today raised $30 million in a series B funding round led by Tiger Global — reflects the surging demand for AI technologies in physical industries. Barnehama estimates that more than 10% of all open roles in manufacturing are quality- or inspection-related, making it among the hardest kinds of positions to fill.
Among other startups, Landing AI is developing computer vision-based technologies for various types of manufacturing automation. Cogniac and Seebo are other recent entrants in the field, as well as tech giants like Google, which offers a visual inspection product that spots — and aims to correct — defects before products ship.
The no-code nature of Elementary’s platform dovetails with another trend: the growth of tools that allow non-developers to create software through visual dashboards instead of traditional programming. An OutSystems report shows that 41% of organizations were using a low- or no-code tool in 2019/2020, up from 34% in 2018/2019. And if the current trend holds, the market for low- and no-code could climb from between $13.3 billion and $17.7 billion in 2021 to between $58.8 billion and $125.4 billion in 2027.
“During the pandemic, manufacturing and logistics have undergone major labor shortages … As companies look to continue to automate without having to rely on expensive and hard-to-find engineering talent, our business has scaled because we’re able to provide them with no-code AI solutions,” Barnehama said. “Not only do we enable them to automate a task that they cannot find enough labor for — quality assurance — but we make our system easy to use, removing the need for machine vision experts that are even harder to find today.”
Fika Ventures, Fathom Capital, Riot VC, and Toyota Ventures also participated in 50-person Elementary’s series B. It brings the company’s total raised to over $47.5 million.