Deming helps users run and deploy industrial models at the Edge thereby allowing them to bring the intelligence and insights closer to their workflow and processes. Because industrial data comes from various sources, each data stream might represent data about the same operation or equipment, but in different contexts. Once the data is put into Deming, the platform extracts a common context around the machine—whether it is the machine’s condition, its operation, or the performance history. Using digital twins, the platform then generates timelines for each machine whereby users can monitor the happenings, related to throughput, product quality, or how much downtime or availability it has. In addition, the AI platform integrates with the powerful open-source rules engine, Drools that enables users to apply business rules to streaming data so that the insights get directly ingested into their work processes. With Deming, manufacturers can reduce the number of unnecessary safety shutdowns thereby achieving a more reliable but also a safer operation that is highly flexible.
Quartic.ai heavily invests in developing Edge gateways software to connect to any type of old legacy system, regardless of the protocols they follow. “Our focus is to make legacy infrastructure smarter as we enable manufacturers to retain their existing OT infrastructure but add intelligence to it through Deming,” Anand adds. Also, once the manufacturers build their intelligence through Deming, they can consume it within their existing infrastructure— manufacturing execution systems, client plant control systems, SCADA, and more.
We want to accelerate the espousal of AI in the manufacturing domain
Elaborating on the solution, the CEO notes that while it brings innovative approaches such as reinforcement learning to enhance accuracy for industrial models, their focus is not strictly confined to accuracy. The team emphasizes more on the speed and ease of deployment along with empowering customers to understand and interpret these models in their existing domain language.
In context, Anand refers to a particular instance where a client was greatly losing out on productivity due to the high-maintenance of their manufacturing equipment. This is when Quartic.ai swung into action and embedded AI into the client’s system, enabling it to tell them which particular components of the machine are starting to behave abnormally. The insights were an aid for them to address only those specific components instead of shutting down the machine for 7-8 days. “Our solution offered operational certainty to the client thereby enabling them to manufacture a batch of quality products on time, on spec,” remarks Anand.
In coming times, Quartic.ai’s team will continue making its platform scalable and have as many people use the tool on small-scale projects as possible in order to build their confidence and move towards doing bigger projects.