Analytics Provide Process Insights
EP Editorial Staff | June 6, 2022
Advanced-analytics tools make it possible for engineers and other SMEs to greatly increase productivity and overall plant reliability.
By Katie Pintar, Seeq
In the process-manufacturing industries, there has been a significant increase in the amount, complexity, and accessibility of operational and equipment data. Engineering teams now have visibility into historical and near-real-time data, from local and remote locations. Add machine learning and artificial intelligence and these industries have the potential to discover more valuable insights than ever before.
However, process manufacturers still face a long list of challenges on the journey from raw data to valuable insight. Fortunately, they can now leverage advanced-analytics applications to address the leading challenges along this journey, such as lack of data access, connectivity, time-series specific analytical tools, and collaboration capabilities.
From equipment and process data to quality and inventory data, process manufacturers have a variety of existing data sources. This information is often stored in different databases, e.g., process historians and asset-management systems, either on-premises or in the cloud.
For decades, organizations have used standard spreadsheet-based tools to collect, cleanse, and align time-series data from these existing sources. These tasks can be very time-intensive for SMEs (subject-matter experts), such as process experts, engineers, and data scientists. In fact, a 2016 CrowdFlower (CrowdFlower Inc., Kirkland, WA, crowdflower.com) study found that SMEs spend nearly 80% of their data-analytics time collecting and wrangling data into a suitable format for analysis, leaving only about 20% to create insights.
When these inefficiencies are combined with a lack of live data connectivity, SMEs are left with analyses that are perpetually outdated, making it challenging to create meaningful insights. Additionally, this type of spreadsheet-based analysis is siloed and error-prone, limiting the ability of a team to collaborate and communicate analyses with the broader organization.
Spreadsheet applications are not optimized for time-series data analysis and have poor visualization functionalities, prohibiting quick and iterative analysis. Although SMEs bring invaluable process knowledge and insight to the table, traditional tools lack the effective and efficient data-analytics capabilities they need to improve production outcomes. Better solutions are now available.
Manufacturing organizations can leverage advanced-analytics applications to connect disparate data sources to a single cloud-based or on-premises application, immediately alleviating the challenges of live data connectivity. These types of applications provide simplified data-cleansing tools and contextualization, including time-stamp alignment, empowering SMEs to quickly derive meaningful and reliable insights across all available data. Equipped with live data connections, SMEs can apply their analyses to near-real-time data.
By removing these data-access barriers, SMEs can leverage an application’s purpose-built, point-and-click interface for descriptive, diagnostic, predictive, and prescriptive analytics to improve performance based on transformational data insights. These tools incorporate visualization into the data-analysis process, empowering SMEs to immediately visualize the impact of their data analysis, identify missteps and successes in real time, and iterate and innovate faster.
Advanced-analytics applications also enable organizations to maximize the effectiveness of SMEs—who may work from different sites or countries—by enabling streamlined collaboration, knowledge capture, and reporting.
For example, product assets frequently change hands in the process industries, so it is common to find multiple sites within the same organization struggling with the same issues over time. Advanced-analytics applications enable these organizations to implement enterprise-wide analytics strategies that promote and enable cross-site collaboration, such as sharing best practices for predicting and preventing common failure modes among assets. These analyses can also be shared and scaled across plants or product portfolios and then used to train new personnel.
With advanced-analytics applications, organizations can provide SMEs with solutions they need to improve production outcomes by placing the right data in the right hands at the right time, as illustrated in the following use cases.
When producing a biopharmaceutical compound, a membrane-filtration system is used to separate the desired molecules from other species. As portions of the material being produced begin to coat the membrane, a clean-in-place procedure is conducted between batches to remove the buildup.
A major pharmaceutical manufacturer suspected that the clean-in-place procedures were becoming less effective. They needed a way to predict when maintenance would be required to prevent unplanned downtime. Over time, membrane fouling, due to ineffective cleaning or degradation with usage, reduces overall quality and performance, requiring a membrane replacement.
The manufacturer’s engineering team applied Darcy’s Law to determine the filter-membrane resistance based on pressure- and flow-sensor data, as well as known values of surface area and fluid viscosity. With the number of variables of interest reduced, the team could clearly visualize the decaying-membrane performance in the advanced-analytics application.
A regression analysis was performed to model the degradation rate and the model extrapolated into the future to predict when maintenance triggers would be exceeded. Maintenance activities were scheduled proactively to maximize membrane lifespan, with minimal impact on operations, leading to improved cycle times, yield, and batch quality.
Historical loss analysis
To maximize future production capacity, process manufacturers must understand the source of historical losses, their warning signs, and past mitigation strategies. Most of these companies are already tracking and categorizing periods of capacity loss to identify bad actors, justify improvement projects, and perform cross-site benchmarking. But each analysis iteration in this process is tedious and time-intensive for SMEs who must identify losses, perform root-cause investigation, and document the events leading up to the loss and the ensuing actions.
Using an advanced-analytics application, teams identified performance losses by comparing actual production to theoretical capacity, and by creating conditions when operation is constrained. Losses were categorized by breaking this broader event or condition down into multiple sub-categories of similar events, either manually or logically, based on configured thresholds.
Summary visualizations created in the application represent losses in graphic and tabular formats for presentation to different stakeholders. Summary views can be assembled in a report, with scheduled date ranges used to auto-update reports, providing the required collaboration and sharing capabilities.
Automatically generated reports for periodic analyses, such as these capacity-loss investigations, can save as many as five days per month of valuable process engineering time. Easy filtering and aggregation of historical loss data enables SMEs to spend more time adding value to improvement projects and less time developing cost justifications.
Advanced-analytics applications are paving a smooth path from raw data to meaningful insight for process manufacturers by connecting disparate data sources, providing intuitive tools for SMEs, and enabling seamless collaboration. These applications are an essential tool for organizations to realize the full potential of their growing data repositories by improving production outcomes. EP
Katie Pintar is an Analytics Engineer at Seeq Corp., Seattle, WA (seeq.com), where she helps companies maximize data value. She has a process engineering background with a B.S. in chemical engineering from Montana State Univ. Pintar has more than five years of experience working for chemical manufacturers to optimize existing processes and develop processes for new materials, scaling them from the lab to pilot plant to full-scale manufacturing.