Breaking down the data silos for production monitoring insights
A leading Print media publishing company with multiple plants across the country has a high level of automation at various stages of its workflow. This has created silos of data systems and a lack of an aggregated and correlated source for applying data analytics that drives business value enhancement at the factory and organisational levels.
A leading Print media publishing company with multiple plants across the country, has high level of automation at various stages of its workflow. This has created silos of data systems and lack of an aggregated and correlated source for applying data analytics that drive business value enhancement at factory level and at organizational level.
Data needs to be aggregated from over 15 disjoint systems and sensors to create an aggregated and correlated database system.
The aggregated and correlated data to facilitate the development of deep insights at a plant level as well as at the organization level to drive business benefits such as:
- Operational excellence
- Cost optimization
- Cost of downtime
- Budget monitoring
- Waste Identification and elimination
- Real time status monitoring of ongoing operations
- Automatic Analytical Dashboards with decision-making insights.
- Predictive and Prescriptive Maintenance of equipments
- Predictive insights for unplanned downtime
- Data to be obtained from multiple sources such as ODBC databases (about 10), EMS server, Newsprint loading system server, CTP Servers, Mail Room Data, and Line monitors.
- Drive Systems, Security Systems like – CCTV, Attendance System, SAP, Standalone sensors, PLCs so on
Data Capture Method
- Using ODBC connections
- Through network access
- Direct data collection from sensors through edge gateway
- Manual data entry
Break this data project down into phases
- Aggregation of data from existing sources
- Using simple analytical algorithms
- Computation of KPIs and Dashboards
- Manual entry for certain segments of data
- Building of quality aggregated historical data
Next Phase (s)
- Integration of additional data sources
- Incrementally replace manual data entry with automated interfaces and data collection
- Implementation of sophisticated analytical methods for predictive insights
- Continue building quality historical data
- Incrementally automating the production monitoring and management.
- Building of ML/DL models for analyzing data
- Prescriptive models for optimized outcomes like:
- Action recommendations
- Resource optimization so on
Download this Success Story