IIoT and Industry 4.0 initiatives are rapidly adding to the amount of available data, driving the need for advanced analytics applications to create insights.

Byline: Katie Pintar, Seeq

In the age of the industrial internet of things (IIoT) and Industry 4.0, the sheer amount and complexity of data has greatly increased. Add the emergence of artificial intelligence (AI) and machine learning (ML), and the process industries have the potential to uncover more meaningful insights than ever before.

However, for process manufacturers, the journey from raw data to meaningful insight is still disjointed. To meet these companies where they are today, leading issues such as data access and connectivity, the lack of time-series specific analytical tools, and collaboration limitations must be addressed.

Traditional tools lacking

For most process manufacturers, numerous data sources exist, from equipment and process data to quality and inventory data, but this information is stored in a variety of different databases. Using standard spreadsheet-based tools to collect, cleanse, and align this type of process data from a variety of sources is time intensive for subject matter experts (SMEs), typically process experts and engineers.

These inefficiencies are compounded by a lack of live data connectivity, which leaves analyses perpetually out-of-date. These hurdles make it difficult for SMEs to gather the data, let alone prepare it for meaningful analysis. Traditional solutions also make sharing data and analyses across teams difficult or impossible, limiting the power of collaboration and knowledge transfer.

Traditional spreadsheet-based tools are not optimized for time-series data analysis and are decoupled from real-time data visualizations, which makes quick, iterative data analysis prohibitive. While SMEs bring a wealth of process knowledge and insight, they have long been underserved when it comes to effective and efficient data analytics tools, but better solutions are now available.

Advanced analytics address issues

Advanced analytics applications make handling disparate data sources much easier by connecting them to a single cloud-based or on-premises application. These types of applications, such as Seeq, can be used to cleanse and contextualize data, and to perform time stamp alignment in the background, enabling SMEs to quickly derive meaningful insights across all available data. Equipped with live data connections, these applications allow users to apply analyses to near real-time data.

With these data access barriers removed, SMEs are empowered to leverage the application’s purpose-built, time-series, analytical tools—provided in a no-code or low-code point-and-click format—to discover transformational data insights. These tools are coupled with trending and data visualization, empowering SMEs to visualize the impact of their data analysis in real-time, and allowing for quick, iterative analyses.

In addition to empowering SMEs with rapid analysis, advanced analytics applications enable more streamlined collaboration across teams and sites. For example, in a cloud-based application, an SME and a data scientist can access the same data simultaneously, working together to combine process knowledge with AI and ML expertise. Algorithms developed by data scientists can be operationalized within the application, providing seamless adoption and validation by process manufacturing teams. Additionally, analyses can be easily shared and scaled across plants or product portfolios, providing maximum impact.

Advanced analytic applications enable process manufacturing organizations to place the right data in the right hands at the right time, providing SMEs with the tools they need to leverage their process knowledge and experience to improve production outcomes.

Use Cases

Golden profile operationalized across plants

A chemical company uses advanced analytics to improve product quality by monitoring the temperature profile during the reaction stage. The product’s dependency on temperature control is well-known by engineers and chemists, but the ability to track the temperature profile through different stages of a reaction has been difficult to visualize and communicate.

Using Seeq, the engineers identified golden batches and built a golden profile with plus or minus three standard deviations using point-and-click tools to help identify temperature limits and optimize product quality (Figure 1).

Figure 1: A golden profile analysis was created in Seeq to monitor the real-time temperature profile and keep the product within specification.

This visualization enables operators to see how the real-time temperature profile is performing against the model, and to make changes to keep the product within specification. Because the analysis is browser-based and has constant connectivity to the data source, this analysis can be shared directly with SMEs at different plants, scaled across similar assets, and replicated to different products.

Remote truck fleet monitoring

A mining company leverages IIoT to monitor their truck fleet using condition-based monitoring. These trucks often run long hours in harsh conditions and require frequent maintenance. By accessing each truck’s telemetry signals using Seeq, the maintenance team can quickly identify trucks that need maintenance by combining excessive travel time identification with profile monitoring to assess truck health. This enables the team to monitor trucks remotely and view pertinent data on truck health across the entire fleet to perform predictive maintenance (Figure 2).

Figure 2: A treemap created in Seeq displays truck health across the fleet to determine when predictive maintenance should be scheduled.

Greenhouse gas emission reduction

A super-major oil and gas company was able to reduce NOx emissions using Seeq by creating an algorithm operationalized as a point-and-click tool. To create the algorithm, the centralized data science team worked with the site engineers to develop a model of the upstream, downstream, and auxiliary processes, accounting for residence times and transitions, so sensors could be properly shifted to account for process lags.

The company’s data scientists leveraged open-source ML libraries in Seeq to create the algorithm, and then operationalized it as an Add-on in Seeq. This point-and-click capability empowered site engineers to make proactive process adjustments, reducing overall greenhouse gas emissions and improving site environmental performance.

Without a comprehensive solution to connect to disparate data sources, provide intuitive tools for engineers, and enable effective collaboration, the path from raw data to meaningful insight will remain disjointed and prohibitive. Advanced analytics applications can be used to address these and other issues, making them an essential tool for process manufacturing companies to realize the full potential of Industry 4.0 and IIoT.

Katie Pintar is an Analytics Engineer at Seeq Corporation, where she helps companies maximize value from their data. She has a process engineering background with a B.S. in chemical engineering from Montana State University. Pintar has over 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. She has expertise in batch and continuous processing for a wide range of chemistries.

All figures courtesy of Seeq