Why Use Manufacturing Analytics?

A look at how automating the collection and analysis of data provides greater process control and reduces downtime, maintenance costs, and the negative impacts of staff turnover.

Luigi De Bernardini, chief executive officer at CSIA-Certified Autoware
Luigi De Bernardini, chief executive officer at CSIA-Certified Autoware

I recently attended three different conferences. One of the common themes that emerged is the value that statistical data processing can bring to the manufacturing process. This is referred to as manufacturing analytics. I met several technology providers, system integrators and end users from different industrial environments. In all cases, the consensus was that the ability to analyze manufacturing data with complex mathematical algorithms has the potential to provide competitive advantage through increased efficiency and reduced risk of being heavily dependent on staff experience.

Most manufacturing companies began methodically gathering process data some years ago. It started in the regulated industries, such as pharmaceuticals and life sciences. Others soon followed, motivated by the idea that knowing what happens in the production process can be a useful element to diagnosing and troubleshooting problems. This has led to the storage of significant volumes of data. Statistical analysis tools, previously applied primarily in the study of demographics and macro-economics, are now being used to find patterns and correlations between independently collected data for manufacturing purposes.

Use of manufacturing analytics gives companies three main advantages:

  1. Highlighting the correlation between data relating to different process variables. While data may not seem related to each other, IT tools can be used to implement sophisticated algorithms and discover hidden relationships, in the data. The discovery of a relationship that is not immediately apparent, particularly when the variables belong to distant stages of the production process, generates a new understanding and awareness of the process itself. The revealed information can then be used both for more accurate and responsive production management, as well as for the optimization and improvement of the process.
  2. Using analysis tools for process control. Once a pattern is identified by analyzing data collected over time, the analysis tools used to identify the pattern can be used to process data collected in real time as a process control. This has a significant impact on both product quality and production efficiency. Early detection of condition changes related to quality problems or plant failures allows for timely action, often avoiding the actual occurrence of problems or at least limiting the impact on productivity. This has an immediate economic return in terms of reduced failure times, maintenance costs and waste.
  3. The ability to leverage employees' personal knowledge. This third benefit is less immediate or measurable, but still highly relevant. Often the information generated by the analysis tools is simply an explicit formalization and measurable representation of what skilled staff already know through experience. This has historically been part of the value that experienced operators have compared to newer and younger ones. Being able to formalize, represent mathematically and automate that expertise enables companies to reduce the impact of and the risk related to staff turn over. This is going to be increasingly important considering the greater mobility of the new generation of workers. Companies are now able to transform knowledge gained from personal experience and use it to govern processes independently and without staff involvement, to reduce the impact on process quality and efficiency when experienced operators depart, and to have useful targeted information to train new staff more quickly.

To reap the benefits of manufacturing analytics, however, companies must invest heavily in hardware and software systems to process the data and recognize patterns, first on a historical basis and then in real time. They must also invest heavily in new professionals. These people, who will be in charge of implementing and managing these systems, can either be part of a modified internal structure, or external system integrators who develop the appropriate skills working with different customers and can, in turn, offer this experience to the company.

Luigi De Bernardini is chief executive officer at Autoware, a Certified Control System Integrators Association member based in Vicenza, Italy. For more information about Autoware, visit the Autoware profile on the Industrial Automation Exchange.

More in Data