“Today’s refineries are dealing with many challenges including operational agility, higher margins, different crudes, corrosion and reliability, as well as how to use the Industrial Internet of Things,” said Ed Schodowski, refining director, Emerson, to attendees of the Refining and Petrochemical Industry Forum at the 2019 Emerson Global Users Exchange in Nashville, Tennessee.
With the guidance of live audience polls, Schadowski moderated a panel of four industry experts who discussed six topics that brought out a range of views and useful insights for automation engineers at refining and petrochemical companies. They dove in by asking the audience what types of operational excellence programs they’ve used to improve Solomon benchmarks, and how automation was involved. The audience chose mechanical availability (73%), turnaround index (35%), net cash margin, USD/bbl net input (19%), energy intensity index (15%), and personnel efficiency index (13%).
“Solomon is a double-edged sword,” said panelist Bruce Taylor, director, digital transformation, Sinclair Oil Co. “It can lead management to slash and cut budgets and head counts, so you have to produce information that explains maintenance efficiencies—for example, sometimes it’s less expensive to just replace a relief valve than to remove it, test it and put it back.” Taylor recommended that managers make the Solomon criteria more granular, so plant personnel could recognize their influence. He said, “We gave them metrics directly related to them so they know their contribution.”
Panelist Michael Barham, principal engineer and control systems technologist, Marathon, said, “Looking at the metrics, our mechanical availability is high, but we haven’t done much using automation. We could do much more to increase that number, but how can we quantify it? We measure things falling apart, not holding together.” Barham emphasized that turnarounds are a quick way for plants to lose money. “They’re often 20% over budget,” he said. “Good turnarounds equal good maintenance. Do turnarounds well, and the rest will fall out.”
Panelist Richard Marcantel, instrument engineering manager, Citgo, said, “We’re an old plant with analog signals and few smart devices. We need to have as much availability as possible. We’re trying to leverage HART and get information to people who can use it.”
Panelist Dustin Beebe, vice president, control and operator performance, Emerson, pointed out that “Procedural automation can transform the way people work. Don’t just deploy technology, lift and shift operators. Automation is one way to do it.”
Improve equipment reliability
Attendees were then asked where they were on the journey toward best practices in Integrity Operating Windows (IOW) as described in API 584, and improving fixed equipment reliability. The poll showed 48% “have started and are on the journey,” while 28% are “talking about it and creating plans,” 21% have “only discussed it,” and 3% have programs that are “fully developed and comprehensive.”
Emerson’s Beebe observed that, “With millennials coming in, most people are on the journey.”
Marcantel at Citgo said, “We started, and are early in the journey, asking how to get data—information—to the computers on their desks to make better decisions.” It’s important because, “We make measurements on rounds, but what happens between rounds?” he said.
“The bottom line is money,” said Marathon’s Barham. “We look at the IOWs when it’s important, but mostly pull data before meetings, and that’s shortening downtime.”
Taylor at Sinclair warned that, “Many people don’t know what an IOW is versus a limit or target, so we usually give information in terms operators can appreciate,” such as dollar signs or value symbols. Further, “When we see step changes from shift to shift, we try to learn from those anomalies and get everyone on the same page,” he said.
Upskill for more data
Asked, “What have been the best upskilling methods and procedures to utilize additional information about the process and assets?” the audience poll showed attendees use formal onsite training (69%), formal vendor offsite classroom training (48%), lunch-and-learns (52%), webinars (16%) and universities/technical schools (8%).
Training is good, but “It’s hard to get change into the operator rhythm, to continually reinforce training so it becomes part of the operating norm,” Taylor said. “Training is valuable, but it has to be followed up.”
At Marathon, “A chemical company customer requires our operators to have four-year technical degrees, so we sponsor and follow that,” said Barham. He says the problem with formal training is that it’s a one-time thing. “Webinars are very valuable because they’re always coming up,” he said. Repetition is important because, “Our models that we train them on don’t replace the models in their heads—they compete with them.”
Marcantel at Citgo said, “We do all of these. We have a really good technical school in our area, and companies in the area require a degree from that school.” He prefers off-site training because people tend to get pulled out of on-site training for emergencies. But the most important thing is to put training to use immediately, he says. “If you don’t use it, you lose it.”
Beebe agreed. “We can’t do enough formal training, so we use more online webinars,” he said. “Operators can see a webinar on, for instance, alarm management or an instrument procedure, then they can go do it.”
Asked to comment on lessons they’re learned about wireless justification, implementation, maintenance, etc., the audience poll revealed that 51% need a better vision for use cases, 44% need better financial justification, 28% need management buy-in, 23% need a total wireless IT plan, and 21% have had trouble tackling cybersecurity issues.
“We put in a comprehensive Wireless HART and Wi-Fi system,” said Taylor at Sinclair. ”The infrastructure was five times the cost of the devices for a single application, so we canvassed for all the applications and created a vision—no one use case justifies the cost, but together, the return was huge. We cut two to three days out of a turnaround.” But he cautioned, “When you build it, who owns it? Who changes the batteries? What about monitoring and cyber? And you must understand the protocols—Wi-Fi is not industrial.”
Marcantel at Citgo pointed out that, “Management doesn’t understand all the terminology. Dumb it down so they can understand.” He agreed that it’s important to make decisions about ownership and maintenance upfront.
“We designed a wireless temperature sensor innovation program with great Emerson sensors, but it’s not moving forward due to cost priorities,” lamented Barham.
Beebe replied, “Think about what the project means to your company. Then challenge the vendors on cybersecurity and understanding your use case.”
Get a grip on mass balance
Asked about their successes and best practices in improving mass balances, the audience poll found that 80% have improved data quality for unit or plantwide optimization/modeling, 13% have improved losses, 5% have reduced financial risk, and 3% have reduced theft.
“Our biggest challenge is people,” said Barham. “It’s so basic, it’s not sexy, and people don’t care.” There are great automated tools that use statistics to pinpoint where the problems are, “But then you have to get the fix done,” he said, adding, “You must have mass balance for advanced process control (APC) projects and model-based control.”
Beebe noted that available systems “can pinpoint the instruments causing loss, tell you where to put instruments, where to improve accuracy. A refinery runs on barrels, which equal mass, and you really need to understand where it’s going.”
Marcantel said, “We have a lot of orifice plates, not mass flowmeters. New processes and new crews change things up all the time, and when you start with bad data, you end up with bad results.”
Taylor observed, “It has the same problems as APC. Who keeps it evergreen? A new engineer gets the job for their first 18 months, then what? It’s a great application for artificial intelligence (AI).”
Connect islands of automation
Finally, the audience was asked to fill in a blank: “Where has your plant removed islands of automation, such as stranded PLCs?” Responses were many and varied.
“First, what’s the value? It’s a capital cost, it has to have ROI,” said Barham.
Marcantel replied, “The costs include maintenance of stranded PLCs, doing maintenance and diagnostics out in the plant. Technicians could be doing something else more productive. We’ve integrated a lot of PLCs to our DCS—a DeltaV can integrate them very easily. Or run a Modbus TCP connection and bring it back. Use an OPC connection if there’s no DCS to get it to operators,” adding, “It really helps you on the maintenance side.”
Beebe said, “It’s also for optimization, by bringing data together and into the cloud. It helps you coordinate cybersecurity, deploy patches, control attack surfaces.”
Taylor pointed out that, “When islands are not integrated, they run on their own for functionality as well as data. It’s hard to know what’s going on.” Furthermore, “AI is coming, and AI totally depends on cross-functional views. For example, a pump’s usage, history, environment—we need all that to optimize it.”
This is the official online community site of the Emerson Global Users Exchange, a forum for the free exchange of non-proprietary information among the global user community of all Emerson Automation Solution's products and services. Our goal is to improve the efficiency and use of automation systems and solutions employed at members’ facilities by sharing our knowledge, experiences, and application information.
User Groups |
World Areas |
Community Guidelines |
Legal Information |
Contact Community Manager
Website translation provided by
© 2015-2019 Emerson Global Users Exchange. All rights reserved.