Challenges

One of the global FMCG manufacturers faced a number of manufacturing challenges, including quality inconsistencies, downtime due to equipment malfunctions, and suboptimal production scalability. There was a lot of variability in product quality due to inconsistent temperature and mixing parameters across plants and lines. Unscheduled maintenance caused by a lack of real-time monitoring of critical machinery also led to huge losses in addition to the high energy costs from non-optimized usage of HVAC and refrigeration systems.

Solution Proposed:

Phase 1: Sensorization and Edge Computing

We first initiated the client IIoT journey by deploying a number of sensors:

• Thermocouples and RTDs (Resistance Temperature Detectors): To monitor the pasteurization process and ensure precise temperature control.
• Pressure Transducers: To monitor the homogenization process and ensure product consistency.

These sensors were connected to devices adjacent to the machinery, sending it to the cloud, and reducing latency and bandwidth requirements.

Phase 2: Data Aggregation and Real-Time Analytics

The data from these devices were processed and aggregated in a cloud-based Industrial Data Lake. A data analytics engine equipped with machine learning algorithms assessed the data in real time, generating actionable insights. These included predictive analytics for equipment failure, quality control parameters, and even demand forecasting.

Phase 3: PLCs and HMI Integration

Next, the actionable insights were integrated into Programmable Logic Controllers (PLCs). The PLCs were equipped with Human-Machine Interfaces (HMIs), allowing for both automated and manual controls. For instance, if a deviation in pasteurization temperature was detected, the PLCs would auto-correct it, while simultaneously flagging the event on the HMI for manual review.

Phase 4: Energy Management and MES Integration

We also used the IIoT data to integrate with their Manufacturing Execution System (MES) and dynamically manage energy consumption based on real-time production schedules, thereby optimizing utility costs.

At Levintutu, you will discover a variety of solutions meticulously crafted to meet the needs of diverse industries. From precision-guided robotics that optimize your manufacturing processes to data analytics that unveil hidden patterns, we offer services designed to increase your efficiency and redefine your success: https://levintutu.com/industrial-iot/

Results and Impact

1. OEE and Maintenance

Predictive maintenance algorithms, built on regression analysis and machine learning, reduced unscheduled downtimes by 15%, contributing to a 30% improvement in OEE.

2. Cost-Effectiveness

Dynamic energy management resulted in a 20% reduction in energy costs.

3. Scalability and Demand Matching

Real-time production data was used for more accurate demand forecasting through time-series analysis and seasonal decomposition, enabling better scalability of operations.

To read about Resolving Train Door Failures through Predictive Maintenance, visit our case study on Resolving Train Door Failures through Predictive Maintenance.

Conclusion

The factory is now a seamless blend of cyber-physical systems, enabling higher productivity, better quality, and significant cost savings while providing a blueprint for other manufacturers in the food industry seeking to upgrade their operations.

Tech@Levintutu is the technical team at @Levintutu who writes on the latest technology and its applications in industrial automation. To read more about us, visit our blogs at https://levintutu.com/blog/.

We are really excited about writing more on the application of technology, Artificial Intelligence, industrial automation, AMRs, AGV, ASRS, Industrial IOT, etc.

Follow us on https://www.linkedin.com/company/levintutu-automation/.