Food & Beverage Can Do Root Cause Analysis Better, Simpler & Faster 

Avoid Bitter Beer Face with Proper Tooling

If you’re part of the Food & Beverage Industry, you most likely deal with batch and/or continuous processes and all of the issues and concerns that go with these. We’re talking quality assurance, energy efficiency, regulatory compliance, and much more. You may also have to deal with asset analytics where you need to monitor your equipment and operating zones, keeping both in check. And for sure you need to get a handle on solving any process issue that arises.

Recently we discussed how self-service analytics can help the industry maintain precise process records by integrating time-series data and process context data. In this blog, we want to focus specifically on how food and beverage professionals can use it to perform better, simpler, and faster Root Cause Analysis (RCA).

Improving Operations with Accessible & Fast RCA

Performing RCA is a critical part of any operational process. It saves operating and maintenance costs, ensures asset health and reliability, aids in energy management, and improve performance.

Knowing this, it’s important to have an accessible and fast approach to root cause analysis. Consider this robust RCA checklist:

  • You’ll need to perform pattern recognition and value-based searches.
  • You should have monitoring capabilities that can alert personnel when deviations occur, so you can act before major events happen (Btw: This process “heads up” saves substantial time and money.)
  • You’d want a tool that can give correlating signals for recommended insights so that you can easily navigate problems. 
  • You’d want a top-notch, user-friendly trend viewer that allows you to quickly navigation through multiple years of data, tag and sensor groupings. Multiple visualization modes don’t hurt, either.
  • You should be able to tie in your contextual data, so things like quality batches, shift log information, and so on, are visible on top of your trends and available for analysis as well.

Now, you’re probably asking yourself: “Great, where can I find such a tool?” No worries – self-service analytics has these capabilities.

And as use cases are always a great way to visualize and understand the value of a tool and get a feel for how your can use self-service analytics, let’s look at one for a beer brewing anomaly.

Use Case: RCA & Quality Control in Beer Brewing 

In this use case, the brewing team was dealing with inconsistent low-quality batches that were causing off specification beer parameters, delays in production, and rising costs. The team needed to find the root cause and fast. Using self-service analytics, they were able to find the cause and prevent it from happening again.

1. Create an overview.

First, one engineer created an overview of all the equipment involved in the brewing process. The time-series data from each brewing asset was plotted in its own swim lane, and the respective context items for each were displayed in a chart using self-service analytics (Figure 1).

This context information included the following: 

  • Mash mill information which showed when the malt mill was actively milling. 
  • Mash ton information which showed production information like alarms for pH limits, process anomalies, and planned maintenance. 
  • Fermenter information including lab samples, batch ID’s and beer measurements like alcohol percentages, bitterness and pH.  

Figure 1. Time-series data plotted in its own swim lanes with the respective context data displayed in a chart.

2. Look at the data.

There were a lot of lab samples for the beer, so he took at an overview of all the quality samples in the process. Viewing these in a Gantt chart, the engineer could see the attached context data (Figure 2). 

Next, he checked the key beer parameter – the bitterness. To do this, he performed a value search for batches with off-spec bitterness. Once he found these, he saw that these batches also had an off-spec pH, determining the root cause of the issue.  

Figure 2. Brewing data plotted in Gantt charts.

3. Prevent reoccurrence. 

To prevent this bitterness anomaly from happening again, the engineer set a monitor to alert personnel via email about a potential problem with pH and recommended actions to take.

Recommendations like this one can be set up to influence last minute changes or even detect upcoming failures in the process, so personnel can correct the process to ensure a successful batch.  

Wanna see this use case in action? 

Check out our free webinar on demand “Advanced Analytics for the Food & Beverage Industry”

A Smarter Route to Finding the Cause

As Food and Beverage process experts, you have an extremely difficult job so having a tool that can help is great, especially when it comes to root cause analysis and quality control.

Self-service analytics is the smart route to finding root causes and ensuring a clear road to operational excellence.