How to Turn Raw Data into Actionable Data for Manufacturing
Data, data everywhere: With factory digitization and advancements in sensor and transmitter technology, there is no shortage of raw data available from the plant floor for industrial manufacturers:
- Transmitters can now provide you with multiple variable values along with their health and other diagnostic information;
- There are large amounts of time-series data locked up within process historians, along with processing event data;
- Manufacturing operations also generate terabytes of data that needs to be collected and stored for either regulatory compliance or process analytics.
The Downside of Data: Limited Analysis & Availability
Unfortunately, a lot of these data are never properly analyzed and the results are not available until the following day. Data are stored in silos making them difficult to retrieve and perform process forensics.
This is one of the biggest challenges in manufacturing operations management: How to take raw data and transform it into actionable data for making informed decisions.
As manufacturing engineering consultants, we find that it’s very important to provide some context to these raw data as they are collected from the plant floor to make them actionable.
Take food and beverage manufacturing for example. If you can add the product being made, the recipe being utilized, and the operator who was executing the production order to the time-series data, you have a much more robust set of criteria to compare results between multiple runs of a similar product and identify discrepancies. You can then act on these to streamline the variables outside the normal.
This same principal can apply to almost any industrial process.
Providing Different Data for Different Departments
Another issue of manufacturing process control is that different users across the facility need different information.
The same data collected from the plant floor can be analyzed very differently by various departments. The maintenance department may be more focused on evaluating downtime reasons while the production department may be looking at schedule adherence and production quantities. Providing context to the raw data enables us to analyze the data in multiple ways applicable to the various roles in the organization so action can be taken.
The Crucial Role of Contextual Data
Whether it’s downtime root cause analytics or Overall Equipment Effectiveness (OEE) calculations, contextual data plays a critical role. The data is also being utilized to provide electronic batch records, increase production throughput by performing constraint analysis, and identifying production waste and non-value add activities.
More and more emphasis also is being given to automated data collection. Manual data collection is known to be inconsistent and completely dependent on the operators. To get a better understanding of data, we advise manufacturers to automate data collection and be sure it is accurately collected with event timestamp as well as the actual duration of the event.
Why Big Data = Big Opportunities for Manufacturers
With the advent of big data analytics, the opportunity to get a real-time view of your production and ascertain its health has increased tremendously:
- Data is enabling the creation of mathematical models that can help predict asset performance and condition;
- Data is being used in a number of ways to identify Key Performance Indicators (KPI);
- And in some cases data is even being updated to the cloud to create a digital twin to perform offline diagnostics and operator training.
With the increase in plant floor automation and connectivity to enterprise network, we anticipate a wider acceptance of these capabilities within the manufacturing community. We are moving definitively towards an enterprise that is connected from corporate headquarters to the plant floors.
Matrix Technologies is one of the largest independent process design, industrial automation engineering, and manufacturing operations management companies in North America. To learn more about our manufacturing operations management capabilities and manufacturing process control solutions, contact John Lee, Strategic Manager of Manufacturing Intelligence.
This schematic space planning helped the manufacturer and design team to be on the same page during the design phase. Once the process layout space planning was approved, the design team was able to quickly move to detailed design of major packages which required earlier installation dates.
Reflecting back on my earlier days spent collecting field data, I vividly remember performing topographic surveys with a three-person team. Two members of the team were responsible for the chain (or linear tape measure generally 100’ long), while a third operated the theodolite or instrument used to measure horizontal angles and vertical planes.
