The client is a leading global manufacturing group with more than 60 operative subsidiaries and production facilities in 12 countries.
PRODUCTS & SERVICES
Azure Stream Analytics
Azure Event Hub
Azure SQL Data Warehouse
Power BI Embedded
- An increasing need to monitor manufacturing equipment on customer sites.
- Recurring machine problems.
- Predict manufacturing equipment problems before it impacts production.
- Real-time monitoring of equipment and units.
- Automate troubleshooting.
- Alerting field executives immediately after equipment issues are discovered.
SNP’s predictive analytics solution offers the simplicity and self-service the client needed while meeting the governance and automation requirements of their IT teams.
The solution was offered in 2 phases:
Phase 1: Source Analysis & Requirement Gathering
- Analyze the overall platform for a full-blown predictive analytics implementation.
- Implement the Azure-based solution to prove the concepts of the cloud and predictive analytics.
- Create a preventive maintenance dashboard from the log files.
- Create an Azure channel to extract data files from various machines.
- Implement an ETL tool and import it into the data model to create requisite dashboards.
- Track activity by location.
Phase 2: Roadmap for Predictive Analytics Solutions
- Set up real-time alerts and data monitoring across complex manufacturing environments.
- Analyze high volume streaming data to get real-time insight into the Power BI analytics dashboard.
- Enable data correlation to give clarity on how variability in one area will impact another.
- Detect and resolve any errors or unruly patterns among departments, machines, processes or individuals.
- Set up predictive maintenance during streaming forecasting on data and alert when thresholds are surpassed.
- Use streaming data from sensors and devices to recognize warning signs (e.g., predict equipment failure) and perform maintenance before equipment breakdown occurs.
- Send notifications to field executives or customers when errors or warnings occur.
- Set up advanced preventive analytics for periodic machine health check and historical analysis.
- Establish customer satisfaction and SLA monitoring.
- Develop exception reporting for failure and error analytics.
- Real-time monitoring of performance.
- Detect equipment failures before they happen and fix them using smart sensors and real-time data.
- Remote diagnosis by gathering and transforming data from sensors and systems.
- Replace legacy ETL technique with modern IoT workflows that enable event hubs for real-time analytics and notifications.