Food Manufacturing IoT solution: real-time monitoring and algorithms to reduce production line waste and giveaway.
The solution
The CEO of a cheese repackaging company was after a fast and affordable Food Manufacturing IoT solution to enable him to see key performance metrics across multiple facilities in real-time.
His greatest concerns were an increase in waste and his factories’ inability to reach their daily target of 20,000 units per facility.
We focused on creating a solution (built on our world-leading Dynamix integrated data, AI + IoT platform) that could find the answers to those questions quickly, while providing a long-term path for performance measurement – our philosophy is always start small, show a clear ROI fast, and then grow your solution over time when you’re ready.
Defining what data to collect is critical
We started by identifying the key metrics necessary to achieve the desired business outcomes. In our Food Manufacturing IoT solution, these included:
- Production targets
- Machine status
- Downtime duration
- Downtime reasons
- Unit weights
- Target weights
- Production rate
- Production schedule.
Initial goals of the Food Manufacturing IoT solution
The first goal of the project was to connect to an Ishida checkweigher using a RS232-WiFi module installed in a cold / wet environment and a 4G gateway outside the cold area, which would then send the data in real-time to the Dynamix platform. Through this set-up, we could collect the right data in a consistent and reliable manner, whilst ensuring data quality and integrity, and provide the following features:
- Monitoring of critical operational data received from the equipment via a web-based and mobile system
- Ensuring that device management monitoring for the Ishida check-weigher sensors was set in place
- Defining business logic to the application in-order to meet the CEO’s business objectives
- Providing notifications via email or SMS when there was downtime, or throughput was below the hourly target
- Collecting of historical data to enable the ability to test different logic to diagnose reasons for downtime
- Providing real-time throughput and waste metrics on the mobile phone for senior management and line managers
- Connecting the ERP system, to collect daily target, and package weight and sizes
- Testing of the application, making sure all of the above goals are met, based on the below solution architecture.
Before setting the solution live, we tested four critical aspects of the Food Manufacturing IoT solution:
Security
The Rayven Dynamix data, AI + IoT platform is built with security as a top priority and our proprietary security architecture ensures that data is secure at all points of the environment.
The Food Manufacturing IoT solution includes data encryption in transit from device-to-cloud, as well as device authentication; security (Bearer) tokens; SSL, AES and RSA encryption; as well as additional device security checks done via automated polling.
Connectivity
Once securely connected, an Ethernet cable connection from the cold room to a 4G Gateway proved to be the best solution. It required a small level of investment with cabling and gateway connection setup, and then ongoing minimal costs to maintain the connection in the form of a monthly 4G data sim fee. The combination of a secure and encrypted transmission path together with a dedicated, direct connection (that eliminates connecting into the factories network) meant a fast and secure connection with out needing to involve IT.
Data integrity
The Food Manufacturing IoT solution is only as good and reliable as the data collected, so ensuring continuous connectivity means data is always up-to-date and that it can be relied on to make critical business decisions in real-time. It was critical for us to test the data we were seeing on the factory floor matched 100% to what we was being seen in the dashboards and build in capability to back-fill data in sequence in the event that communication went down.
Industrial data science
The objective of exploratory data analysis was to observe trends in the data and compare them with what was happening on the factory floor, which included:
- Calculating throughput, IE packs of cheese per minute, hour, day and week
- Real-time forecasting using a pitch chart of daily packs of cheese based on real time production throughput
- Amount of cheese kg waste per minute, hour, day and week
- Cost of cheese kg waste per minute, hour, day and week
- Amount of cheese kg give away per minute, hour, day and week
- Check weigher weight bell curve.
What’s next?
After providing evidence of the amount of giveaway in dollar terms, which amounted to millions across all facilities over 12 months, and providing insights on how to reduce the give away, we are now focusing on the next saving opportunity, which is around reducing energy consumption and improving OEE.
