Meet our customer hero
Huntsman’s journey towards analytics maturity and operational excellence has been groundbreaking all the way. Historically, the company’s process engineers would use their extensive experience and process knowledge to try to understand sensor-generated data. This approach was regarded as experience-driven. But the company is on a journey to fully leverage its captured time-series data by transforming towards a data-driven work approach. And it needed tools to enable this.
Process experts know about production processes and think in terms of trends. Data scientists think in terms of algorithms and statistical models and do not necessarily have production knowledge. Huntsman needed to overcome the disconnect between these two central problem-solving groups. And it had heard that TrendMiner’s self-service analytics would allow process experts to do the data analytics themselves. “TrendMiner democratizes the analytics allowing process experts to analyze, monitor and predict production processes themselves, without the help of data scientists,” says Jasper Rutten, Advanced Analytics Manager. “It is a powerful tool using a combination of daily data, new sensor data and lab-analysis data, enabling us to scale our operations through global collaboration. The uses cases are compelling when it comes to showing the value we’re getting out.”
Use Case 1: Setting soft sensors for product quality monitoring
For years, a Huntsman continuous isocyanate plant had been collecting daily process and offline-created lab analysis data, both of which were stored in the historical database. With TrendMiner the company’s teams can now build soft sensors on operating conditions to predict product quality for certain Isocyanates. The process experts then use these to make micro-adjustments to process setpoints to proactively minimize impurity levels. One of the monitors, for example, predicts hydrolysable chloride levels in the final product, and by tweaking vacuum pressure conditions, Huntsman has mitigated impact on product quality. In addition, monitors have been set up to send out early warnings to tell the operators not to load trucks, preventing off-spec material from going to a customer.
TrendMiner has also made it possible to have 24/7 quality control compared to the situation before using lab analyses only available during work hours. With trucks being sent out 7 days per week, the soft sensors have eliminated 75% of the expensive off-spec transportation cases which used to occur on weekends. In addition, a significant positive impact on lead time has been achieved as unnecessary wait hours for in-spec products has been eliminated—with the average lead time reduced by several hours. Finally, the extra insights on product quality have reduced the demand on lab resources by as much as 10%.
Use Case 2: Fingerprinting batch processes to check product quality
Before TrendMiner the team would check batch profiles using Excel. This approach required a lot of work, time, and expertise. Now they can create “fingerprints” to check batch quality against specifications. In one of the polyols processes, distinct pressure and temperature profiles are required to consistently create high-quality material. To ensure this quality, time-series patterns from known good batches are grouped and saved as fingerprints. The golden batch fingerprint is then used as a real-time monitor to continuously check the process for deviations. This means, batches don’t require checking for abnormalities afterwards because the monitors give early warnings for unexpected heat input allowing operators to take appropriate action in time. This new approach to batch analysis and monitoring has led to a significant reduction in off-spec batches and an increase in product quality.
Use Case 3: Quality improvement for DMAIC
In one of the advanced materials plants, a lot of batches in a wiped-film evaporator were exceeding the solvent specification limit, resulting in off-spec products. In fact this same plant had observed a multi-year drift in quality according to the QA lab. Process experts suspected that this issue was due to a change in testing methods. Using TrendMiner capabilities, a complete Six-Sigma DMAIC analysis was performed. The analysis included value-based searches, layer comparisons, statistical comparison tables, scatterplots, filtering, and the recommendation engine.
By observing scatter plots, process experts can now easily track batch performances to see which ones are inside and which ones are outside the operating zone. That’s because TrendMiner allows them to complete a faster root cause analysis on a much larger data set. It’s also helped them identify subtle differences in pressure readings resulting in quality improvements in a matter of just days.
It’s quite significant what Huntsman has achieved in a relatively short space of time. The company’s digital evolution and journey to process excellence has received a serious boost through self-service analytics. Greater insights into processes have improved operational efficiency and quality control, decreased losses and downtime, and streamlined continuous improvement cycles. Before TrendMiner, Huntsman’s data used to be applied in silos making collaboration very difficult: It was only possible to run pure diagnostics—to show what had happened. Now the company’s process experts around the world can analyze larger data sets and say why something has happened—and prevent negative outcomes. This is releasing large value for Huntsman in terms of efficiency, quality and employee empowerment.