The COVID-19 pandemic has highlighted a need for food manufacturers to automate and streamline their processes, primarily due to workforce and material shortages. Many food manufacturers look to robotic applications and equipment modernization to ensure that production keeps pace with demand.
Manufacturers are beginning to focus their attention on possible improvements found in the digital transformation of their process controls, specifically with regard to process measurements such as temperature, flow, level, and humidity.
A process is typically monitored and manipulated using feedback from instrumentation located within the process stream. For instance, level is used to control the volume of liquids in tanks, temperature is used to control the heat applied to a product, relative humidity is used to control the moisture content of a product before it’s fried, and so on. Instruments like these have historically been used to control a closed loop process, offering only limited data to operations on how the process is performing. The feedback from the instrumentation is used more for information on how that loop is being controlled and not necessarily on the quality of the product. To fully understand how a process affects product quality, tests are performed on the product ex situ as a quality control (QC) method.
QC typically performs a set of tests on a product after it has completed the production process to ensure that it meets manufacturing specifications. In food processing, a representative sample of a product is retrieved at certain stages of production (generally after packaging) and tested for quality. The inherent issue with measuring end-of-line product is the delay from when a problem arises to when it is detected in QC. For some processes, this can range from dozens of minutes to hours. If a problem is determined to exist and the source is identified, say, 20 minutes upstream in a process, this results in more than 20 minutes of product potentially going to waste. If you consider the time it takes to correct the issue and verify the effectiveness of the correction again at the QC checkpoint, losses can be doubled or even exponentially increased. With supply chain delays and issues facing manufacturers, this can result in tremendous losses beyond waste product.
The typical way to combat this problem is by adding more QC checkpoints in the process and retrieving more samples; however, this leads to added labor requirements and increased processes to monitor the quality of a product. Through modern technology and the digitization of process controls, manufacturers can migrate from a QC model to a quality assurance method. The conversion or addition of digital instrumentation gives more accurate, real-time data that operations can act on to control their process. Modernizing process controls can provide the ability to keep a process closer to baseline while optimizing it. This moves the QC in situ with the process to assure that product quality is maintained at multiple locations along the process and corrections can be made earlier and with more precision.
With the digital transformation of processes, production-aided controls such as model predictive control can reduce labor force requirements and increase production output by analyzing real-time data and manipulating control variables to achieve optimal performance.
Where to Start
A smart place to begin with a digital transformation is in temperature control and process monitoring. Temperature affects many attributes of product quality and is used to ensure that a product, particularly in the food and beverage industry, adheres to food safety and regulatory requirements. Temperature is used as a means of determining many things in the production process. In frozen potato products, temperature measurement is used at almost every stage of the process: from raw potato storage, to monitoring the temperature of batter to determine the potential of bacterial growth, to the most obvious of measurements—the temperature of the cooking oil.