Database and AI specialist Splice Machine has launched Livewire, an open source Operational AI platform for industrial IoT use cases. Livewire is a combined data and AI platform designed to alert plant operators of likely outages or performance degradation soon enough for them to take remedial action to avoid outages and improve performance. 

The Splice Machine Livewire platform is designed to allow teams of data engineers, operators, and data scientists to work together with speed and agility. By using an integrated platform these teams can deploy machine learning models faster and with less staff than conventional alternatives.

The platform brings together: tools and APIs to integrate and ingest data from DCSs, SCADAs, historians, ERP systems, MES systems and other data sources; visualisation and alerting tools to surface temporal data, static data and predictions; and an end-to-end Machine Learning platform for developing, experimenting, and deploying machine learning models.

According to Splice Machine, Livewire has been designed to help oil and gas, utilities and process manufacturing companies move ML projects into production more rapidly and reduce infrastructure complexity for a more agile ML lifecycle.

The new system aims to make feature and model creation easier by giving data science and machine learning teams access to real-time data. It also provides MLOps tools for managing and deploying models to production faster.

For companies in asset-heavy industries like oil and gas, desalination, process manufacturers, energy, utilities, and telecom, Livewire is designed to augment native DCS/SCADA systems.

Livewire is built upon the Splice Machine SQL RDBMS with built-in machine learning. The system is elastic and designed to be deployable anywhere. If the plant is isolated from external networks or restricted, the Livewire Kubernetes edition allows them to implement the same cloud-based solution on a private cloud or Kubernetes cluster on premises.

Splice Machine works in common IIoT use cases that follow the OODA structure (Observe, Orient, Decide, Act) by ingesting sensor data and suggesting actions. Use cases where Splice Machine is being used or evaluated include outage avoidance, load forecasting, predictive maintenance, threat detection, and spare parts planning.

“Many AI and ML projects start out with great ideas and intentions, but go nowhere,” said Monte Zweben, co-founder and CEO, Splice Machine. “We are here to help those projects move out of the ‘ivory towers’ and into production applications. With Livewire, industrial enterprises can capitalise on investments made in sensors, AI and staff and drive a digital transformation that can generate superior business outcomes on an ongoing basis.” 

“Splice Machine’s unique combination of operational, analytical and machine learning capabilities has been proven at scale in highly regulated industries like financial services and healthcare,” said Zweben. “Now, with Livewire, we’re aligning IT and operations technology to deliver operational AI for industrial companies around the world.”

“Accenture and Splice Machine are taking AI out of the lab and into the plant with the new Livewire solution,” said Andreas Braun, Managing Director, EMEA Lead, Data Business Group and Applied Intelligence, Accenture. “This merging of databases and machine learning enables companies to achieve better business outcomes faster, but better yet, it allows them to quickly iterate and continuously improve the predictive accuracy of models leading to even better outcomes such as fewer outages and better performance.”