Highlights
  • Customer aimed to eliminate data silos to share intelligence from edge to cloud
  • Deployed Litmus Edge via virtual machine and started collecting data in just days
  • Leveraged and shared ready-to-use analytics across the business for shop floor visibility
  • Integrated data with maintenance system to optimize scheduling
  • Sent data to Azure for analytics and deployed machine learning models at the edge

A bottled water manufacturer with bottling plants around the world set out to adopt an Industrial IoT platform to achieve visibility into manufacturing operations. They had several types of machines and data acquisition systems creating data silos and sought a solution that would allow the IT team to unify factory data and share it across the business.

The customer chose the Litmus platform to connect to all their disparate devices, then collect, normalize and analyze the data. They deployed via virtual machine in each plant data center and connected and collected data in just a matter of days. They quickly enabled visualizations and pre-built KPIs to share key data intelligence across both OT and IT teams, eliminating the data silos and creating a steady stream of good, complete manufacturing data and analytics for use across the company.

Data visualization and analytics quickly led to real business value including a reduction in downtime and cost savings. The company quickly began to add use cases like anomaly detection to improve quality and integration with their Oracle Maintenance System to create automatic work orders.

The IT team was already using Azure as their cloud analytics platform of choice, which Litmus Edge integrates to natively for rapid data sharing and further processing. They created statistical models to predict machine failures, and then deployed those models back at the edge for quality, efficiency and maintenance improvements. The customer started small with edge analytics and visualizations, and quickly expanded the scale to multiple plants and more advanced use cases based on observing machine behavior.