Blog

Quenching the Thirst for Sustainability in The Food & Beverage Industry

Engineers hungry for information on how to improve sustainability can find it in operational data

Consumers are increasingly concerned about sustainability in the food and beverage industry and are actively seeking eco-friendly and socially responsible products. Companies across the sector, from ingredient manufacturers to market-ready product creators, are facing growing demands to adopt sustainable practices that extend from the farm to the factory floor.

Embracing these changes is not only critical for meeting global food demands but also for complying with strict government regulations and maintaining a positive reputation with consumers and investors. Many are investing heavily in technology to meet the growing need. One of the ways the food and beverage industry can embrace sustainability is to explore operational data. Generated through sensors and captured by technology, this time-series data presents a gold mine of information for food and beverage manufacturers.

Sustainability Clues Found in Data

Manufacturers can leverage operational data to track their effect on the environment and optimize production processes to achieve sustainability goals. Through advanced industrial analytics, engineers gain valuable insights that enable them to:

  • Optimize Maintenance Schedules: By monitoring process behavior and performance, companies can determine the best time for maintenance, which prevents asset damage and unplanned shutdowns.
  • Reduce Energy Consumption: Identifying energy-intensive processes allows manufacturers to discover ways to reduce energy usage and greenhouse gas emissions.
  • Optimize Water Usage: Manufacturers can save fresh water and explore possibilities for recycling process water for future use.
  • Reduce Carbon Footprint: By measuring and analyzing carbon footprint data, companies can implement strategies to control and reduce their environmental effect.
  • Minimize Waste: With data-driven insights, production waste can be minimized or eliminated, which results in significant cost savings.

Preventing Shutdowns: A Step-by-Step Approach

A new pasteurization unit that consists of two lines is shutting down whenever Line 1 is running, and Line 2 starts up. There are some challenges:

  1. There’s a lack of process knowledge about the new unit.
  2. The root cause of the sudden and unexpected shutdowns is unknown.
  3. A dashboard showing specific KPIs must be created so performance can be monitored.

Learn how to optimize your pasteurization unit by using TrendMiner`s self-service analytics. Study the 10-steps process or watch the video.

Here is the main dashboard that is going to be both the starting point for the analysis and the monitor to follow up on process performance once the anomaly has been solved.

At left, the box with Line 1 shows the main variables of interest and the limits that must be overcome. Below that, the dashboard shows events for Line 1:

  • 5 anomalies have occurred,
  • 3 corrective maintenance requests have been performed, and
  • 2 preventive maintenance requests have been executed.

At right, the same information can be shown for Line 2. Also notice that the dashboard shows that Line 1 is running while Line 2 is stopped. Between that are a list of recent events, and above that list is a calculation performed in TrendMiner. In this case, it shows the total yield of the pasteurization unit, which is at 37.1%.

The most notable KPIs on the dashboard are the 5 anomalies in Line 1. By clicking on the title Anomalies, a table opens that shows the shutdowns.

Clicking on each shutdown provides further information. In one case, an operator has suggested that pressure could be causing the line to shut down. Click the Options box, and a menu will appear. Select Visualize.

Click Add tags/attributes to navigate through the asset Framework.

To create the visualization:

  • Select Flowrate, Pressure, and Status for Line 1.
  • Select Status for Line 2 to show any correlations.

At left, select Tags & Attributes. A new view will appear. Note the orange bubble on the Flowrate of Line 1. This shows the anomaly.

Drag the cursor to create a box that selects a time of interest. A new view will appear that magnifies this period.

The new view reveals that there’s an expected pressure drop in Line 1 when Line 2 starts (due to the pipes connection – see PFD). However, Line 1 cannot return to its previous state and the shutdown occurs.

In the menu at left, select the magnifying glass. Then select Similarity Search.

With the cursor, select the most important area (pressure drop itself) to give it an extra weight in the algorithm. Then, change the Minimum Similarity Score from 70 to 60 to find other pressure drops with different behavior (slightly different profile).

The patterns appear to be similar during periods of good and bad behavior. However, a more closely correlating event showed that the shutdown occurs when the pressure drops below 1.5 bars on Line 1. This report was shared with the maintenance team. When the pressure drops below 1.5 bars for more than a few minutes, a safety measurement automatically shuts it down.

A pressure drop is causing the shutdown. Further analysis of the same chart reveals that the pressure drops too low when it starts out too low.

In the menu at left, select the magnifying glass. Then select Value Based Search. This is where the conditions that need to be monitored are defined. In this case, it will watch for periods when the pressure falls below 3.5 bars.

Select the Monitoring button on the right side of the ribbon. On the slide out menu, enable a new monitor based on the conditions created during the previous step. Save the monitor.

Navigate back to the dashboard, select Action, and then Add new tile. Add a title (in this case, Risk of Shutdown), then add values to show on the dashboard (Risk/No Risk). Save the tile.

The new title will reveal the default normal status until conditions are met that initiates an alert. Once the alert is received, the pressure can be adjusted to prevent the shutdown.

Conclusion

As the demand for global food production rises, it is imperative that the food and beverage industry embraces sustainability to create a better future for all. By exploring operational data, companies can turn hidden clues into actionable insights that drive efficiency, reduce waste, and save valuable resources such as energy and water. Manufacturers can become environmental stewards while meeting the ever-increasing needs of consumers and the planet.

From our Blog

Low-code and No-code AI & Analytics Solutions Yield High-Value Use Industrial Use Cases

TrendMiner's Industrial AI & Analytics platform demonstrates how user-friendly solutions empower engineers to make more informed decisions about operational performance.
Data Fabric featured image

Building a Foundation for AI & Analytics with an Industrial Data Fabric

The Enhanced Data Layer indexes, structures, and contextualizes time-series, asset, and event data for rapid visualization in a universal production client.
The Organizational Gap

Bridging the Organizational Gap in Pursuit of an Augmented Factory

By closing the Organizational Gap, manufacturers can do more to improve operations with machine learning models and AI-powered systems.