Jan-2020

Where transparency-driven value comes from

Safe and sustainable operation is a key factor for the profitability of every processing business, such as a petroleum refinery.

Andrey Kostyukov, Alexey Kostyukov and Sergey Boichenko
DYNAMICS Scientific Production Center USA Inc

Article Summary

The US petroleum refiners have a chance to increase their profit at least 20% by using contemporary cutting-edge digital technology which provides safe resource-saving operation and maintenance of machinery. To be able to provide safe, reliable and effective operation, the personnel of the refinery must have precise, focused, and timely information on machinery health (“Make The Future Visible”, n.d.). By making machinery health visible, using the unbiased AI prescriptions in real-time becomes a reality, which leads to effective reliability management which dramatically reduces losses that come from: incidents and unplanned shutdowns due to machinery breakdowns, short uptime and extended turnaround outage, overestimated maintenance costs.

Any petroleum refinery is a hazardous processing plant that constantly experiences incidents due to machinery breakdown. According to F. Buder, CEO at Pruftechnik: “Avoiding failures that impact production is time-sensitive and critical, and yet there were 2200 unscheduled shutdowns in the US alone between 2009 and 2013, an average of 1.3 incidents a day, costing roughly $20 billion per year, according to Deloitte Insights. It’s no wonder that 54% of owners said their spending should be on predictive maintenance” (Petrochemical Update, 2018).

According to another trustworthy source (HPI Market Data, 2018a), the US refining industry comprises 135 refineries with total capacity of approximately 18.6 million barrels of crude oil per day. Almost every incident came with a shutdown that led to refinery losses in the average amount of $36.4 million dollars ($20 billion dollars divided by 2,200 unscheduled shutdowns and divided by four years). Moreover, following the statistics above, an average refinery encounters around four incidents annually. By installing the AI system which transforms sudden defect in machinery into the gradual ones, refiners can avoid most of the incidents and corresponding losses. Even if the AI system prevents only 75% of the incidents and shutdowns, the total annual savings at one average refinery can be to $110 million dollars.

The AI system can bring more additional value if it provides extended uptime and decreases turnaround’s outage. According to the lobby interviews during 2018 AFPM Annual meeting which was held in New Orleans, LA, it was pointed out that the average gross margin in US petroleum refining in 2017 was $17 dollars per barrel. Also, it was mentioned that an average annual availability rate of refinery uptime was revealed as 91.2%. This information means that in 2017 an average refinery lost as much as $76 million dollars (365 days in a year multiplied by the difference between 100% and 91.2% of availability rate, than multiplied by $17 gross margin per barrel, than multiplied by the average specific capacity of US refinery, 140,000 barrels a day). Knowing machinery health and eliminating destructive forces at the early stages of machinery degradation extended uptime twice and cut turnaround’s outage by 10-15 days, which totally brings over $50 million dollars of additional profit annually.

Last but not least, the source of transparency-driven value are overestimated maintenance costs. “US refiners increased planned maintenance spending by an estimated 38.5% to $1.26 billion dollars in 2018… (HPI Market Data, 2018b).” This means that an average US refinery spends over $9 million dollars on the maintenance annually. It’s well-known that the later a defect is identified, the greater the costs and losses become. By using AI system and identifying defects timely, refiners can reduce the maintenance costs up to 40% (HPI Market Data, 2018b), which is approximately $4 million dollars annually.

Finally, by summarising all value sources mentioned above, the total amount of additional profit that comes from the AI system implementation at an average US refinery could be as much as $164 million dollars annually. The estimated AI system’s lifespan is at least 10 years, which allows the refinery to aspire the total net additional income in the amount of 1.6 billion dollars.