My car has predictive analytics, why shouldn’t my refinery?
When I bought my first car, I thought (naively) ‘perfect, pay for capital cost and done’. What I didn’t realize was how much work (and $$$) went into making sure it didn’t fail! I followed dealer best practices with oil changes, tire rotation, brakes and rotors, etc. Sure, as it aged, there was no stopping some failures. But the reason I kept up with regular maintenance was to get more time out of my asset. It’s expensive to buy a new car; it had to last as long as my monthly payments, and hopefully, longer.
Refineries and chemicals plants are cars on steroids! Much more complex with a much greater potential for failure. In addition to lost profit from downtime, is the possibility of a safety incident with injury to property and personnel. The U.S. Department of Energy reports one U.S. oil/chemical incident per day incurs on average an estimated cost over $2MM USD per incident. Or in terms of my first car, 100 Chevrolet Cavaliers!
Predictive Asset Analytics can provide early warning notification and diagnosis of equipment issues days, weeks or months before failure. Predictive asset analytics helps asset-intensive organizations reduce equipment downtime, increase reliability and improve performance while reducing operations and maintenance expenditures.
As the world leader in gases, technologies and services for industry and health, Air Liquide serves more than 3.5 million customers and patients by delivering the essential molecules for life, matter and energy—oxygen, nitrogen and hydrogen. From an operational point of view, Air Liquide has many plants across the US. Some are too small to have personnel onsite. They are run remotely using the enterprise data coming into a centralized site from each facility. From there, predictive analytics is used to compare current performance against historical trends, rules based logic is applied and alarms will sound when a process or equipment is outside the normal range of operation or compliance. Back to my car analogy, it’s like when my dashboard tells me I have low pressure in a rear tire without me having to manually check the pressure of each tire hourly, and then comparing it to normal operation, applying a range-based rule and alerting the driver if necessary. I’m going to fix that low pressure before something drastic and costly happens.
Having a Predictive Maintenance strategy can help asset intensive organizations increase asset performance, reliability and availability to ultimately maximize economic return. Watch here to learn more on how Air Liquide derives real value (to the tune of more than $500K) to detect equipment failures before they occur and prevent unscheduled system downtime.
Livia Wiley is the Sr. Product Marketing Manager for SimSci software at AVEVA. She is primarily responsible for expanding SimSci’s brand awareness and marketing of its design, simulation & training, and optimization software. She has more than 20 years of experience in process simulation; assisting clients model, troubleshoot, and optimize their processes through technical and economic studies. Prior to joining Schneider Electric in 2014, she worked for suppliers of process simulation and automation technology, including Honeywell, Aspen Technology, and legacy SimSci. Livia holds a B.Sc. in Chemical Engineering from Queen’s University, and a M.Eng in Chemical Engineering from the University of Houston.