Chemical processors are harnessing the powerful fusion of advanced analytics and increasing artificial intelligence capabilities to optimize processes and enhance operational productivity
As analytical technology matures throughout the process industries, a new generation of chemical process industries (CPI) engineers are examining their jobs differently. In the not-so-distant past, upwards of 90% of the time spent analyzing data was dedicated to simply collecting, cleansing, contextualizing or otherwise preparing information for investigation. After intense preparation, the analyst applied the final approximately 10% of total time to actually analyze the dataset, identifying anomalies, patterns and areas for optimization.
However, the meteoric rise of automated analytics software and now, artificial intelligence (AI), is transforming the CPI and beyond by significantly reducing data preparatory time, and even generating novel operational insights across complex datasets (Figure 1). In fact, a Deloitte study found that 94% of organizational leaders surveyed said AI is or will be critical to the ongoing success of their companies [1].
FIGURE 1. Artificial intelligence (AI), is transforming the CPI and beyond by significantly reducing data preparatory time, and even generating novel operational insights across complex datasets (photo credit: Shutterstock)
Evaluating AI
AI is showing up in unique ways within the CPI and spanning several areas. Its reach includes innovative research and creation with new molecules, in addition to accelerating insight generation for improving productivity, quality and other key performance indicators (KPIs) in manufacturing.
There are different ways of approaching AI innovations. While the healthcare, retail, supply-chain and logistics industries are eager to adopt AI and uncover exactly how it impacts them, the chemicals and several other process industries are taking a more cautious approach. As a result, growth may not be as instantaneous as witnessed in some consumer sectors, but safety, quality and compliance cannot be compromised in the chemicals space.
Furthermore, skepticism and concern about security remain, and as the industry prepares to embark on an AI-driven excursion, it must construct a foundation of safeguards to ensure success.
AI ranges from a narrow scope to self-awareness. Most U.S. consumers are familiar with self-driving cars, an example of narrow AI, which must identify and follow a path to deliver a payload from point A to point B. While variables such as traffic, construction, weather or event closures may exist and cause changes mid-execution, the program scheme is predictable and definitively clear for the most part.
On the other side of the spectrum, AI applications capable of thinking and adapting with more human-like responses are arising. The CPI is most comfortable at this time focusing on generative AI (GenAI), which is largely predictable, given the input information used for creating results, but it also includes machine learning (ML) elements and evolution over time.
Like ML, GenAI relies on computer systems capable of learning and adapting without following explicit instructions, using algorithms and statistical models to analyze and draw inferences from patterns in data. These platforms are primarily being used in the CPI to improve reliability and facilitate predictive maintenance.
Many phrases can quickly become misunderstood or misapplied in the technology sectors, and right now, “AI” is that top buzzword. Past examples include “new normal,” “cloud computing,” “metaverse,” “big data” and “information superhighway,” some of which have stuck around and gained clarification, while others have tailed off within a year or two. The fate of AI has yet to be seen, but with the undeniable impact it has already wrought on the world, it appears not to be going away soon.
AI’s influence in the CPI
Chemical processors are trying to identify AI’s ideal roles in manufacturing environments. As part of the industry’s digital transformation, AI is being leveraged to improve productivity, enable preventive maintenance and improve quality.
Regardless of an AI engine’s sophistication, the clearest way to achieve success is to ensure the model can access the right data. But therein lies another challenge — many manufacturing environments house either too much or not enough data, or the data are spread across so many various locations, that the data first needs to be centralized. With this in mind, the essential first step is clearing the data availability and quality hurdle to drive accuracy and uncover key insights for improving productivity, quality and other KPIs.
The global GenAI market is forecast to grow at a compound annual growth rate (CAGR) of 27.8% over the next 8 years in chemicals, with its primary potential in identifying new molecules [2]. Secondarily, the industry must ensure the ability to reliably produce these molecules. Within these efforts, AI is expected to increasingly serve as a learning platform, helping train technical personnel and bring them up to speed within a company.
This is especially crucial today because the workforce has sharply changed in recent years, with a drop in experience and subject matter expertise. Additionally, knowledge transfer and retention are especially critical to eliminate valuable information losses. Leveraging AI helps CPI manufacturers upskill their labor force more quickly, preparing them to identify and tackle critical operational issues at a faster pace.
Integrating advanced analytics
Further addressing these and other challenges, AI, advanced analytics and ML software platforms significantly ease data integration, contextualization and predictive modeling procedures. This helps produce accurate predictions that empower teams to strategically schedule maintenance when required to prevent downtime. These software platforms leverage analytics to improve operational reliability.
Leading predictive models can forecast failure events and other process issues based on complex patterns within historical and real-time data. These tools are also effective for identifying anomalies in data that can precede numerous types of process upsets.
Upon implementation, these platforms collect data from field sensors, process historians, data lakes, enterprise-resource-planning and asset-management systems, and other sources, then cleanse and organize the information. The results are combined to create a comprehensive dataset, helping fill in operational and maintenance gaps by leveraging AI.
At this stage, the platform also automatically flags outliers and inconsistencies. Users can interact with the contextualized data using graphs, charts and heat maps to visualize trends, correlations and patterns, and to help piece together a well-understood overall plant story.
Beyond these tools for amalgamating information, understanding process interactions and generating insights, modern advanced analytics platforms leverage ML models that self-improve over time to produce failure probabilities and communicate optimal maintenance intervals. These steps help minimize upkeep costs, facilitating just-in-time maintenance to stave off failure while preventing unnecessary downtime.
The required complex multivariate calculations execute in the shadow of a no-code front end to help plant personnel gain clear insights regardless of their analytic or programming skill levels, enhancing operational decision-making. Additionally, these platforms provide user interfaces designed to streamline workflows around process performance.
Identifying causal relationships
A global specialty chemicals company leveraged a no-code ML advanced analytics platform to predict fouling in its distillation towers and improve operational efficiency. The company’s subject matter experts (SMEs) used the platform’s multivariate modeling capability to analyze numerous operating parameters and conditions to identify and predict fouling instances, including composition, temperature, pressure, flow data, and more.
Armed with this information, the ML model produced nearly immediate insights covering multiple contributing factors, in addition to discernment regarding these factors’ interactions. The tool also provided plant personnel with high-level summary metrics, alongside trend views and recommended maintenance actions.
Using the software tools, the team began by identifying baseline operating conditions, along with target periods for analysis. SMEs used the platform’s signal selection tool to identify and remove low-variance signals, along with occurrences of high intercorrelations among multiple independent variables. They then ranked signal importance. These collective efforts emphasized the variables contributing most to fouling, and they highlighted causal relationships for comparing dynamics in the stages before, during and after fouling as a method of determining root causes of these unfavorable outcomes (Figure 2).
FIGURE 2. A specialty chemical manufacturer leveraged an ML advanced analytics platform to establish an operational baseline model for its distillation columns. It then implemented predictive maintenance by comparing live operational data to the standard to reduce fouling (photo credit: Seeq
The manufacturer optimized the model over time to the point where it now identifies conditions that lead to fouling two months before serious anomalies appear, providing plenty of time to act and perform required upkeep steps. These predictions communicate proper downtime planning as part of the company’s overall maintenance schedule.
Additionally, plant personnel are now aware of the key contributors to fouling, which is primarily caused by condenser temperature variances. These findings have prompted operational changes that extend runtime between cleanings, increasing profitability through both enhanced efficiency and reduced downtime that was previously prompted by more frequent routine maintenance.
AI-enhanced insights
Advanced analytics and AI platforms empower SMEs to solve some of the toughest process problems efficiently, implement effective process improvements, transition from reactive to predictive maintenance strategies, increase productivity and uptime, and operate more profitably.
These gains empower chemical manufacturers to achieve new levels of efficiency and remain competitive in the fast-paced market. ■
Edited by Dorothy Lozowski
References
- Deloitte Consulting LLP, Fueling the AI transformation: Four key actions powering widespread value from AI, right now, 2022.
- McKinsey & Company, How AI Enables New Possibilities in Chemicals, Nov. 20., 2024.
Author
Janelle Armstead-English is the industry principal for chemicals at Seeq Corp. (113 Cherry St., PMB 78762, Seattle, WA 98104; Email: [email protected]; Phone: 206–801–9339). She has an engineering, market research, sales and product management background with a dual B.S. degree in chemical engineering and mathematics from the University of Pittsburgh. Armstead-English has nearly two decades of experience working with various chemical manufacturers like Honeywell UOP and Praxair (now Linde). In her current role, she enjoys analyzing the ever-changing chemicals market and understanding the challenges around digital transformation for chemicals customers.