Model seamanship: Using data to manage trans-oceanic supply chains

Supply chain management professor Dan Brooks does research and consulting about decision making in uncertainty and environmental issues. Also director of the W. P. Carey MBA Custom Corporate program, Brooks works closely with oil companies on models that use the collective intelligence of marine professionals to calculate the risks associated with various routes -- or, as the companies say, to make high quality decisions about shipping. Recently, knowSCM talked with Brooks about his work with the petro-chemical industry.

Supply chain management professor Dan Brooks does research and consulting about decision making in uncertainty and environmental issues. Also, director of the W. P. Carey MBA Custom Corporate program, Brooks works closely with oil companies on models that use the collective intelligence of marine professionals to calculate the risks associated with various routes -- or, as the companies say, to make high quality decisions about shipping. Recently, knowSCM talked with Brooks about his work with the petro-chemical industry.

Transcript:

KnowSCM: Supplies of oil are located a great distance from the population centers that use it, and for efficiency, oil is shipped in huge quantities. This combination of great distance and bulk cargo creates challenges for the petrochemical industry. For example, in fall 2008, the Saudi super tanker Sirius Star was hijacked 500 miles off the coast of Kenya. The brand new vessel loaded with 100 million in cargo was eventually ransomed for $3 million. Supply chain management professor Dan Brooks is Director of the W.P. Carey MBA Custom Corporate program. His research and consulting are in the area of decision making in uncertainty and environmental issues, and he works closely with the oil companies. Brooks says the hijacking of the Sirius Star sounded an alarm bell in an industry that otherwise manages engineering and supply chain challenges of global magnitude.

Brooks: Even if you’re big, even if you’re new, even if you’re working for a global company, you’re still vulnerable to seven people in a skiff with a grenade launcher and the equivalent of World War II weaponry. [The Sirius Star] really highlighted the fact that the risks are real from this particular challenge, and the degree to which they can extract really significant impact was something that was of global concern not just one of the costs of doing business.

KnowSCM: But piracy is not the only challenge facing the industry. Routing tankers around outlaw waters sometimes opens up a sea of other threats.

Brooks: One of the choices that oil companies can make in shipping petrochemicals is they can take a different route. They can try to stay away from the areas where piracy is the greatest, but this exposes them to other types of risks. There’s a risk from the weather; there’s a risk of being late; there’s a risk of being over cost. So there are many ways in which running a supply chain has to be a pretty delicate balancing act of the various aspects of getting the product to the customer in a way that is cost effective and a way that is safe and a way that is timely.

KnowSCM: So the petrochemical companies are in the business of managing risk, but Brooks says that the industry doesn’t look at it that way. In fact, the companies don’t even use the word risk when analyzing their options.

Brooks: Rather than focus on that risk in a sort of passive sense and say that we’re trying to manage risk, instead companies have adopted the point of view that we are taking proactive steps to assure the quality of this supply chain. The vernacular would call this risk management; in fact, it’s viewed as marine quality assurance. We are mastering those events which we believe could adversely impact something that we value.

KnowSCM: So from the point of view of the oil industry, says Brooks -- from the point of view of any supply chain manager -- every marine quality assurance decision made concerning those shipments is a bet that they can navigate around trouble and deliver their cargo intact.

Brooks: In the case of shipping petrochemicals, the bet is hugely lopsided. If they’re right, that the vessel with its cargo can make it from point A to point B successfully, they make some money. If they’re wrong, they lose a huge amount of money. On the upside, they can make a small percentage of one year’s revenue stream on shipping. On the downside, if this ship is lost, if there’s a major release, if there are other problems with the shipment that is catastrophic, they can lose five to 10 years’ worth of revenue stream. So there’s a tiny upside and a very large downside. That means that for them to make this bet, they have to have extremely good odds before they’re able to feel good about going ahead and making that shipment.

KnowSCM: Companies analyze data in an effort to improve the odds that their bets are successful, but one of the problems is the amount of available information. Industries are capable of collecting virtually numberless data points about their supply chains.

Brooks: When you have lots and lots of information, and you need to make a decision now about this vessel, this cargo, this voyage, how is that you use the most relevant data, use it real time and use it in a way that supports your decision making? So you can think of it as a card counter might. A card counter in a betting environment says when it is good to make a bet and when it is not in your best interests to make a bet. So in this case, the industry has modeling that uses information in the same analogical sense that a card counter would use to estimate which of the vessels, which of the cargoes, which of the voyages are most preferable to use for this particular engagement.

KnowSCM: But who are the card counters? Brooks says that the oil companies tap the expertise of professionals who have been to sea -- many of them former captains and who have also managed their supply chains from a desk onshore.

Brooks: The decision makers are marine professionals with a lot of experience on the sea, a lot of experience with tankers, and a lot of experience with shipping. The marine professionals shape what data is collected and how it’s organized. So the decision makers drive the data collection and the data organization as opposed to ‘bring me everything you’ve copied and measure, and I’ll figure out a way to combine it once I get it.’ Instead of data driving the decision makers, decision makers are driving the data collection in the way that is linked to the decisions that they have to make.

KnowSCM: These marine professionals engage in what’s called a vetting process, a function Brooks says is separate from the revenue decisions of the company.

Brooks: They’re really looking at the quality of the supply chain. When a voyage is proposed, when we need to get this cargo from point A to point B, they look at the vessel that’s available, the integrity of the vessel, the challenges of that voyage, and they decide whether or not that is a voyage that should be engaged. It has nothing to do with how much money could we make, how much would it cost. It’s not a commercial decision at all. It’s truly a focus on the quality of that link in the supply chain and the advisability of making that delivery.

KnowSCM: In a sense, this is knowledge management: capturing all of the relevant experience of the vetters into a model.

Brooks: The modeling that’s taken place has come thanks to those marine professionals, the vetters, who have the experience on the sea, have the experience managing, and have worked with the data sources to capture their experience in the form of a model. What is the insight that you bring that allows us to take those particular pieces of data and combine them in a way that gives us insight about this vessel, this voyage, and this cargo? So the modeling in that sense is a way of encoding and preserving the experience of these marine professionals and linking it to decision making for the future so that we have a real base for leveraging that experience and being able to use it over and over again.

KnowSCM: These models are used to make hundreds of thousands of decisions in a year.

Brooks: How can you make that many decisions and make them all high-quality decisions? Well, one way is to base each of those decisions on a model that incorporates the insights of marine professionals so that the relevant data for that particular shipment is grabbed, is combined in a way that provides insight and is supportive of the vetter’s decision on whether to go forward with that voyage or not.

KnowSCM: As time goes by, conditions, ships, shipping agencies, maintenance practices, all of these factors change, and the models must change as well.

Brooks: About every year to two years, every year and a half on average, the models are reevaluated. For example, 15 years ago piracy was not a part of the vetting process. It wasn’t formally included in risk assessment when you were making shipments. The last 10 years have changed that. Now piracy is incorporated explicitly in evaluating a particular voyage.

KnowSCM: Although each shipper runs its own models, information is shared across the industry.

Brooks: There are organizations that collect data and make it available to everyone. No one profits at all by having competitors in shipping that are not very safe because it reflects on all of the shipping when there are incidents. So there is collaboration making sure that people are aware of the information that’s available, effective ways to combine it, and effective ways to use that for the vetting process in keeping a supply chain running safely and efficiently.