One of a Kind
Can standardization make shipping more efficient?
(Article originally published in July/Aug 2019 edition.)
The phrase, “Every ship is a prototype,” is common in the maritime industry. Thanks to the way vessels and even sister ships are built, each one represents a unique configuration of equipment linked together in a unique way to account for unique characteristics.
This creates a headache for shipowners and managers that is specific to maritime. It’s not present in the aviation industry, for example, where all equipment is standardized and thus all issues have a straightforward and replicable solution. It’s a challenge for shipowners trying to improve efficiency.
Fortunately, there are a host of software companies dedicated to solving this problem. Matt Heider, CEO of one such company, Nautilus Labs, claims that programming tools – specifically, his own – can improve fuel efficiency by up to 30 percent.
“To start, there are simple examples of fuel waste that constitute the low-hanging fruit of conservation and optimization,” he states. “This ranges from unplanned stoppages to unoptimized generator configuration to RPM modulation. Noon reports, the traditional means of getting this information, are inadequate. They don’t provide intra-day visibility into performance.”
“Voyage optimization” is the operative concept, designed to achieve both efficiency and economy. “Voyage optimization looks at the P&L, future spot rates, bunker prices, weather, routes and other factors,” he says, “and helps advise the crew on the next best action for the ship. This can be an intelligent RPM or speed instruction, an optimal trim recommendation, and the like.”
But the software can do more, Heider says: “As we extend the scope beyond active voyage support, Nautilus helps its clients maximize TCE pre-fixture and over the course of multiple voyages while improving description accuracy and claims mitigation for time charters. This builds into an entire matrix of fleet decisions around bunkering, service, repair, hardware improvements and beyond that can drastically reduce overall fuel consumption in a fleet by upwards of 30 percent.”
Another recent breakthrough is machine learning, a technique which only recently became available to software developers but offers theoretically limitless capacity for improvement in just about every arena. “The more data we can feed into our models, the more accurate predictions we can generate,” says Heider.
“Recently, we’ve been doing some work to assess precisely how effective we’ve become at predictive analysis, to help us quantify the impact of our software on our clients’ businesses,” he explains. “We looked at a voyage where we used our TCE-based speed optimization to predict the financial outcome of a leg. The captain followed our RPM instructions, which were provided by an algorithm that based its calculation on the ship’s performance profile, weather forecasts, routing details and commercial market inputs. By using Nautilus’ recommendation, our client was able to achieve a 13 percent improvement in TCE on that one voyage. Moreover, our platform was almost 70 percent more accurate at predicting the financial results from the voyage than their existing tools.”
These results are a persuasive endorsement of digitalization, machine learning and the benefits these technologies can bring. “Given the skepticism when we talk about the total potential for fuel optimization to be in the 20-30 percent range,” Heider adds, “data points like these help bring those numbers into starker relief.”
With disparate systems on board a vessel, there’s much to be gained from finding a way to link them together. But because of the way ships are built, It’s difficult to find a solution that works in every case. Heider says there are three basic ways to aggregate data – directly from the sensor, via an aggregation point on the vessel or via software from a shoreside system.
“The third is our favorite,” he says, “and it’s increasingly common. An owner has a shoreside software system that already has the vessel data available from another provider. In those scenarios we use structured file transfers, API integrations or other methods to access the data. And overall, any of the three varieties has worked out well in terms of getting our clients what they need.”
Meanwhile, competitor Alpha Ori Technologies (AOT) has also undertaken to wire these disparate systems together with SMARTShip, which has been installed on 30 vessels so far. The platform recently won a vote of confidence in the form of type-approval from ClassNK aboard mid-sized gas carrier Hourai Maru, delivered in March.
The system works by collecting data from various ship systems in order to algorithmically construct a complete platform, which can then be used to improve ship performance. It can draw on as many as 5,000 data points at once. “The main challenge is the non-standardization of the data communication protocol within maritime OEMs,” says Sanjeev Namath, Chief Business Officer. “Alpha Ori has tackled this challenge by developing interfaces with most of the major OEMs.”
He says SMARTShip brings transparency across the value chain – charterers, shipowners, ship managers, classification societies, insurance – and addresses the fundamental flaws that exist in the sector and the resultant losses: “Transparency, in itself, has great potential in changing the business dynamics. Further, the data analytics and IoT solutions facilitate value creation.”
Giampero Soncini, CEO of Italian software firm Influencing Business, also calls for data standardization to enable the integration of systems throughout the vessel. He believes digitalization ought to be implemented not as an end in itself but with bigger goals in mind: cutting back on crew expenses.
The mathematics are compelling. Shipowners often bemoan the cost of providing data for crews, which is regarded as essential for retention. Meanwhile, wages, ship agent fees and the space and expenditure dedicated to keeping an accommodation block running are all concerns. “Things are changing and, unusual for the maritime sector, very quickly,” Soncini says. “For such changes to happen, two things must absolutely take place.”
The first is the use of diesel electric propulsion, combined with azipods, as the main driver for crewless vessels. “The obsolete one main engine, one shaft, one propeller scheme must go,” he says. The other is standardized data interfaces: “Today, all manufacturers write their technical handbooks in any way, form or shape they want. Instead, data must be standardized or efforts to digitalize a fleet will continue to stumble over the issue of manual data entry.”
The Stand-Alone Alternative
It’s worth noting that not everyone is on board with the notion of a fully integrated, interlinked ship. Satcom provider KVH has an altogether different vision for how the future of ship efficiency will pan out.
“Shipowners are looking for answers, not data,” says Executive Vice President for Mobile Connectivity Mark Woodhead. “Our eureka moment was when we considered ourselves to be an equipment manufacturer rather than a service provider.”
KVH Watch uses a separate antenna and software from that of the main channels of crew and operational communications. This distinct channel, unconnected to other systems, serves as a dedicated link from shore to the vessel’s equipment such as gensets and ballast water systems – one that OEMs and not shipowners pay for.
“They can use as much data for remote monitoring as they like,” Woodhead says, “and troubleshooting can be implemented remotely. Because they don’t have to send engineers around the world, the benefit is so great they’re willing to pay for it. This more closely follows the aviation model – engines as a service.”
KVH’s software allows each OEM to administer its own equipment, gathering data, sending updates and fixes and even upgrading it in the field without requiring it to be connected to equipment from other manufacturers on the vessel. With manufacturers running their own kit, all shipowners need to worry about is day-to-day management.
To Pool or Not to Pool
But other aspects of ship operations can benefit from machine learning.
Signal Group’s Signal Ocean platform applies the concept of machine-learning to owner dealings on the spot market, allowing them to marshal information about their vessels in one place and inform the negotiation of new spot charters. COO Dimitris Tsapoulis explains, “If you have multiple sources of information – market gossip, contradictory market information about one specific ship – how do you resolve this conflict? Fuse it into one specific report.”
The system is used to make individual negotiations more beneficial. With a great deal of penetration in the tanker industry – some 50 percent of vessels in the crude spot market, he claims – Tsapoulis sees Signal Ocean serving as a key enabler for the pooling of vessels in the future, helping spot market operations become much more profitable.
“Pools have not been the main choice,” Tsapoulis says. “Only 20 percent of vessels in this market are in a pool. The problem is that some ships in a pool get a premium based on attractive characteristics, such as ice class. But for obvious reasons, this is not very valuable when there is no ice. But pool points are usually not dynamic, and when vessels are getting a 10 percent premium despite there being no ice, other owners are going to complain. So we are working to improve the main points that hinder the industry joining pools.”
Despite calls from his peers for greater transparency and standardization, Tsapoulis believes this must not come at the cost of competitive advantage. “It’s important to remember that markets do not necessarily want full transparency,” he says. “In fact, I think most markets in the world don’t want it. It’s unrealistic to try and democratize the market. If we set out to democratize the information, we would never be given it in the first place.”
The opinions expressed herein are the author's and not necessarily those of The Maritime Executive.