Maple Leaf Foods Achieves the “Perfect Batch” via Data Analysis

Maple Leaf Foods is a leading prepared meats and poultry producer in Canada and one of the top 10 pork producers in North America. Speaking at AVEVA World in San Francisco in October, Andrew Thorne, Maple Leaf’s IS solutions architect, described the company’s move to modern manufacturing, noting that, in 2019, Maple Leaf Foods launched an initiative to change the way the company works. The goal: optimize operations using an existing manufacturing execution system (MES) as the center of digital transformation.

The digital transformation initiative, called the "IoT Project", focused on the deli section of the Heritage facility, the company's largest prepared meat plant, and leveraged AVEVA MES for data collection and analysis. The objective was to add sensors across the production line for advanced analytics, aiming to improve yield and reduce waste and product loss.

Blair Hembruff, president, Cygnus Consulting, speaking at AVEVA World 2023 in San Francisco.

The project, conducted in collaboration with systems integration partner Cygnus Consulting and analytics provider Braincube, involved applying technology to eight applications. One such application uses a standard set of processes to ensure deli meat logs move along the line with consistent length, width, and shape. Here’s where IoT comes in, by using a vision system to measure the log dimensions so that operators can make adjustments, if necessary.

With the meat logs in a consistent shape, they cook at the same amount of time, and go through the slice halls presented in exactly the same way. Plus, by looking at the temperature data and oven data, the optimum temperature profile can be determined to ensure bacteria is killed, but the meat is not losing moisture due to overcooking. 

But perhaps the biggest benefit of the IoT project was achieved in the slicing area. Given the scale of production, slicers are complicated machines that use recipes, which, in the past were based on production over the life of the machine.

Now, the MES dashboards on the shop floor show yield in real time to operators so they know when they’re running off spec. That data is pulled from the slicers into Braincube, which uses data science and digital twin technology to adjust the recipes accordingly.

“[Braincube] iteratively pulls data up every 10 minutes and keeps refining it,” said Thorne. “Eventually, it comes out with the perfect parameters for that day and that machine on that line. Once we started setting machines up this way, we got the perfect batch almost every single time.”

Overall, the results were impressive, with significant improvements in time savings, waste reduction, and power savings. The implementation led to a 10% to 12% increase in gross profit for the relevant production lines, surpassing initial return-on-investment (ROI) predictions. Project IoT achieved an ROI in just three months.