Problem:
Quality defects in the carmaker’s painting line – especially on Mondays. No obvious explanation for variations in quality since production parameters like spray pressure and speed were constant at all times.
IIoT analytics:
Correlation of wide range of measured values and quality level
Result: Paint quality depends on airborne dust volume and therefore the ventilation system
Solution: Ventilation system now also runs outside business hours on weekends and public holidays
Big Data:
- Environment data: Temperature and air humidity/pressure/flow rates and quality such as dust content
- Process data from machinery and production robots: Spray pressure, quantity, speed, color code, etc.
- Quality level: Good, poor, imperfections, etc.
Problem:
Quality defects in manufacturing as soon as more than one production line is operated in parallel. Each line produces zero defects on its own. Process input values offer no obvious root cause(s) or conflicts.
IIoT analytics:
Correlation of measured values from all production lines
Result: During parallel production, pressure increases in the supply lines.
Solution: Understanding interdependencies in the production process enables comparison, adjustment, and optimization of manufacturing parameters.
Big Data:
- Process status: Running, in preparation, shutdown, etc.
- Process parameters: Pressures, temperatures, volume flow rates, etc.
- Quality attributes
Problem:
Ongoing quality defects in a steel stamping works cannot be explained by the few manufacturing parameters available.
IIoT analytics:
Multi-week analysis correlating measured values with production and process parameters
Result: Quality defects resulting from drop in stamp pressure.
Solution: Optimize manufacturing parameters to increase quality of stamp operation, resulting in less scrapand reduced production costs.
Big Data:
- Process and machine data: Pressures, oil flows, stamp speed, temperatures, etc.
- Quantity and quality of stamped parts
Problems:
- Variations in the quality of LED panels.
- Maintenance has to be synced with production, since completion of a batch must not be interrupted (as it would cause quality issues).
IIoT analytics:
Second-by-second correlation of measured values with metrics; several GB of data a day can be processed.
Result: Product quality depends on ventilation temperature.
Solution:
- Continuous ventilation.
- Rolling trend analysis provides early notification of maintenance/repair requirements, so maintenance can be matched to production.
Big Data:
- Environment data: Air pressures/flows, temperatures, etc.
- Machine data from fans: RPM, vibrations, valve settings, etc.