A food manufacturer requires a complete understanding of their manufacturing process as well as its management’s responsibility to drive a culture, which ensures that the facility, its production processes, and its employees understand expectations throughout the production chain. The company should analyze and document what automation and information systems exist (if they do not already know), generate a risk-assessment plan, and prioritize next steps. Cost of implementation is always a part of the equation, so the fact that increased knowledge can also increase product quality and improve the production methodology— hence reducing the production costs—is an important argument to get buy-in from management.
Integrated modular automation and information systems, when designed and configured properly, address product quality, compliance, productivity, information accuracy, process repeatability, human error, human safety, product traceability, downtime, operational efficiency, and more.
So how do you collect process, batch, and quality data today? Kjell Francois, industry software delivery manager, Siemens, comments, “It is not just a matter of combining your process and batch data, it’s also the move from sample-driven lab data to process-driven online measurements, which will increase the product quality monitoring abilities and process knowledge. The idea is to move QA from the lab to the line. This is an approach that is now already applied, for example, in some dairy companies.”
Within your manufacturing process, do you know your key production critical-to-process and critical-to-quality data points that, when out of spec, can adversely affect final product quality? Do you collect all data or just key data, and what are you doing with it? Are you just storing it? A high rate of quality data has been collected in food companies in LIMS environments over the years. Typically, trend charts are made and key performance indicators (KPIs) are followed over time, often based on sample data from samples collected in the production lines. Integrated data—process, batch, and quality—would enrich the (near-) real-time dashboard information displayed. Is this a practice you are following today?
How do you retrieve this data if it is requested? Are you recording data automatically, and/or on paper? Are you still using chart recorders in certain areas? Are you doing any statistical analysis? If yes, is it off-line and after-the-fact, or is it on-line, at-line, in-line, and proactively making real-time corrections to your manufacturing process? Can you predict final product quality by proactively controlling around these key data points? Technology exists today that enables you to meet these on-line/at-line/in-line sampling goals, allowing you to more accurately predict your final product quality.
Finally, how do you handle an investigation? When it comes to your final batch report, is it a hodgepodge of paper and electronic data that requires too much time to assemble, then analyze? Is your batch report retrievable in a timely manner? In today’s investigations, it might take a significant amount of time to find and extract the necessary data, but it doesn’t have to. FSMA makes it clear that the FDA would ideally like the information in under 24 hours, and emerging technology will allow you to meet this goal.
Putting It Into Practice
Today’s modular automation and information systems, including LIMS, collect a lot of raw data. Are you simply collecting this data (rendering it basically useless), or are you merging collected data to allow full contextualization? Once combined or merged, this information can be made available to the employees in real time, enabling them to make timely decisions. KPIs can be generated and utilized by all process food manufacturers, and these metrics can be managed and measured. These KPIs are then able to address many common challenges that companies face, as well as uncover points in the manufacturing process that could be risky. In addition, this information can be made available to the entire business for deeper analysis, and the results are then used to help your company make better decisions when addressing potential risk.