An Intellyx Brain Candy Brief
Industrial operations is an industry primed to take advantage of advances in machine learning and predictive analytics. But in spite of the stories extolling this future, the current reality is a bit less utopian. Applying machine learning and identifying early predictors of equipment failure or quality issues is more difficult than it sounds. The reason is that these indicators are not one-dimensional and require many different data points from multiple channels (called signals) over time to generate reliable insights.
Discovering meaningful patterns from multivariate (coming from multiple signals), time-series data is the thorny problem that startup Falkonry is solving. The company’s platform can analyze historical and real-time operational data, deciphering the relationships within the data across multiple signal sources and over periods of time. Moreover, understanding the nature of industrial systems, the platform can work with irregular data sampling, data gaps, frequency content and other factors that are common in industrial operations.
But while being able to identify these complex patterns is critical, it is also meaningless if organizations cannot convert that information into actionable insights. The company, therefore, designed the platform to enable practitioners to use the system without requiring a large team of data scientists to make it work. Once practitioners connect the platform to data sources and allow it to ingest operational data, the system uses a mix of unsupervised and semi-supervised learning models to rapidly identify insights and give practitioners ‘knobs to turn’ that allow them to improve and deploy models without data scientists on staff.
Copyright © Intellyx LLC. Intellyx publishes the Agile Digital Transformation Roadmap poster, advises companies on their digital transformation initiatives, and helps vendors communicate their agility stories. As of the time of writing, none of the organizations mentioned in this article are Intellyx customers. To be considered for a Brain Candy article, email us at pr@intellyx.com.
