Consider Cranberry Sauce Data Before Talking Turkey

You might say it’s the trip to grandma’s, the day off work, or the afternoon football games, but there’s only one real reason Americans look forward to Thursday: the Thanksgiving feast.

Forget the diet. People look forward to indulging in their own Thanksgiving staples each year, but there are some constants. Most Turkey Day meals will include:

  • Turkey
  • Stuffing
  • Potatoes
  • Gravy
  • Canned cranberry sauce

If you are like half of Americans, you are not a cranberry sauce fan. But there it is, sitting on the dining table, waiting for its first victim to take a bite of the bittersweet concoction. In fact, it’s hard to imagine a Thanksgiving feast without cranberry sauce there. And yet, it could happen. So in the spirit of Thanksgiving, we thought we would take a look at how data analytics makes the holiday’s most polarizing side dish possible.

Canning It

Cranberry sauceSure, cranberries grow in a bog. They are harvested, packed, and sent off to a plant. When they arrive, they are kept frozen before the next process.

But cranberry sauce is a different story. Unlike raw cranberries, making canned cranberry sauce is a manufacturing process.

There’s not much that goes into canned cranberry sauce. The cranberries, of course, and water. They also contain some kind of sweetener and acid.

Cranberries go through a sorting and cleaning process before they are cooked. Cranberry sauce manufacturers know that this is when they can sweeten the pot before they cook the cranberries again. Finally, the hot cranberry mixture is poured into cans, sealed, and allowed to cool.

Just like any manufacturing process, cranberry sauce has a golden fingerprint. But the process also can have anomalies, deviations, or other problems during production that could lead to inferior sauce sitting in that dish on Thanksgiving Day.

The solution?

Using advanced analytics to solve trends in production deviations.

Drilling into Data

A major food manufacturer already was using OSIsoft PI to collect their time-series data, but it needed a deeper dive into a problem that was causing their batches to deviate from their golden fingerprint (their ideal operating zone) every time.

The company used TrendMiner to take a closer look at the data they already had. It compared the visual analysis of a “perfect” run with one that didn’t come out just right. Then, the company used TrendMiner’s layers feature to compare the two runs.

As it turns out, one of the valves with a safety shut-off feature would, inadvertently, close during a run. The company still had to repair the valve, but TrendMiner made it possible for the company to understand where its problems were coming from.

In fact, since it adopted TrendMiner, the food manufacturer has saved $1 million in operating costs.

Turkey Time

Enough about cranberry sauce. We know most Americans are looking to feast on turkey this Thanksgiving, not the red berries that grow in a bog. Still, the cranberry sauce might not be the only processed food on the table.

Instant mashed potatoes? Store-bought rolls? Box-based stuffing? While some still make Thanksgiving meals the old-fashioned way, the time-saving methods are attractive alternatives to cooking all day.

Just remember that many of the things you eat this holiday season may have come from a manufacturing process. Using TrendMiner to solve anomalies in the Food and Beverage industry just might be the reason your holiday treats taste so good.

Discover how TrendMiner helped food manufacturer CP Kelco reach a savings of $1 million in operating costs in this free Customer Success Story>