Improve Your Production Processes With Real-Time Data Analytics Monitoring
Using advanced data analytics models in real time opens up a whole new world of possibilities for improving your production processes. Not only does real-time process monitoring provide a level of confidence in your process performance, it can also help improve the overall quality of your production output.
This article is posted on our Science Snippets Blog.
Real-time monitoring of multivariate data analytics on shop floor
A real-time multivariate data analytics monitoring system can help your production floor staff to know whether a process is performing optimally or not.
But to make it happen, you need data. Data representing good process behaviour. Data from a lot of difference sources. And maybe from sources that are not connected to each either. You need to monitor the data in real-time using advanced multivariate analytics models.
Your models must be proven statistically accurate enough to determine when a production process is deviating from any normal operating condition, and perhaps even predict when a current process might start deviating from accepted conditions. And you probably don’t want to have to know how to do multivariate data analytics calculations in order to manage it.
That’s why a multivariate data analytics tool for real-time process monitoring is essential. Especially if the tool allows your production floor staff to gain assurances that a process is performing optimally or not (and even to see alerts when something is not performing at expected levels).
Why a Multivariate Data Analytics Tool?
Using multivariate analytics application for real-time monitoring turns all the process data you are collecting into a benefit for operations. It’s useful if you want to:
- Minimize operational costs.
- Increase confidence in your process performance.
- Ensure more consistent product quality.
- Maximize efficiency throughout operations.
- Monitor processes in real time so you can quickly take corrective actions if needed.
How a Multivariate Data Analytics Tool Works in Real-Time
A data analytics solution such as SIMCA-online utilizes data regression models to summarize all of the individual parameters from various operations into multivariate models so they can be monitored in real time. This becomes very efficient in the control room because instead of looking at a large number of individual parameters or signals, you have a small set of summary parameters that let you monitor all the variables at the same time.
Another advantage of using a multivariate data monitoring technology is that you can use correlations between variables to your advantage. You can increase the sensitivity of your system to pick up changes in the process as they happen.
Being able to tie all the data from various systems together into one single monitoring system is a great advantage.
A multivariate data analytics tool like SIMCA-online works in real time to pull data from various data sources together and allows people in different locations to view the process. It can help you increase the awareness and visualization of the process throughout the company.
With SIMCA-online, you will have at your fingertips functionality such as:
- remote monitoring
- multivariate predictive monitoring
- fault detection
- root cause analysis (whenever a fault is detected)
- automatic corrective recommendation (available now with Control Advisor in SIMCA-online)
What Can SIMCA-online Do?
With SIMCA-online you can:
Monitor in real time. You can create an ideal model of your process and then compare your actual data for the process to the model in real time. This works for both batch processes and continuous processes.
Predict with confidence. The multivariate analysis model provides a basis for predicting quality parameters over time using regression analysis. Using this tool, you can predict the final critical quality attributes with a high degree of confidence.
Control at a glance. SIMCA-online improves your overall understanding of process and equipment because you are getting real time data and updates about what is happening in the process right now. Plus, with real-time drill downs you can pinpoint issues and detect problems with equipment as they happen.
SIMCA-online is proven in a wide range of industries. For example:
- A multinational paper company reduced costs, achieved a more consistent product quality, gained a deeper understanding of their data.
- A major food processing company saved over 3 million USD each year in production costs.
- A pharmaceutical company paid for their investment several-fold in recovered batches alone.
- SIMCA-online multivariate data analytics in real time
A real-time view of multivariate data using SIMCA-online lets production managers easily see how each batch in a process is doing.
System Design for Multivariate Data Monitoring
The basic infrastructure of SIMCA-online is set up to fit within data collection areas of a process, such as a laboratory or production plant. SIMCA-online is a server client application installed as a service on a computer, usually a server computer, and it connects to a process database such as OSI PI.
The database pulls data at specific intervals from the process components into the SIMCA-online server. The SIMCA-online server also has loaded within it a USP file built from SIMCA, which is then used to compare the current values from the process to the SIMCA model. The results are delivered to a number of SIMCA-online clients that have been installed at various locations throughout the company: in the control room, in development, technology staff, as well as the management level.
The communication happens through OPC/ODBC from the SIMCA-online server into the clients. The connection between SIMCA-online and the process database takes place through a SIMAPI, which is an interface between SIMCA-online and a particular database. There are a number of different SIMAPIs available, and you can have multiple SIMAPIs running at the same time so you can pull data from many different data sources.
This chart shows the various APIs or interfaces available for SIMCA-online.
Basic Operating Principle of the Monitoring System
The basic operating principle for SIMCA-online uses the concept of a traffic light with red, green and yellow color codes to provide an easy visual overview of what is happening with processes.
▪ Green – indicates a process is operating as it should (normal condition)
▪ Yellow – indicates a process is starting to deviate (warning condition)
▪ Red – indicates a process is out-of-control (critical condition)
In SIMCA-online, you have dashboards with overviews that list units and batches and when everything is operating as it should, they will appear in green. Alarms are configured on the system to show when things are going wrong. When you have batches or processes that start to go beyond warning limits, the colors turn to yellow, and when you have a true deviation, they turn to red.
The visual panels for SIMCA-online are grouped in a way that allows drill-down from a higher level view (unit overview) to investigation (for process upset) to viewing a contribution plot (which shows culprit variables) and a further drill down into the trend plot of the original data.
When an alarm is triggered, you can double-click on that point to see an interpretation of why there is a process upset.
From there, you can see the culprit variables. And double-click on any interesting variable to drill-down to the raw data.
The trend plot of the variable shows the upset in the original data, and the time point at which the alarm was triggered is highlighted.
A Powerful Tool for Monitoring Process Data in Real-Time
SIMCA-online is very powerful because it is user-friendly and suitable for implementation on the shop floor. With SIMCA-online you get an off-the-shelf software that has been on market for a long time and is fully validated. With detailed but intuitive graphics, it’s suitable for a GMP environment and can make a big difference in a company’s process quality output.
Want to Know More?
Find out more about how SIMCA-online works, see the video on how to analyze batch process data, a step-by-step guide