MIT research shows ocean transit reliability isn't as problematic as landside delays.
By Eric Johnson
and Chris Dupin
Ocean shippers have long cried out for more schedule reliability from liner carriers but have had few ways to precisely measure the true performance of their ocean service providers.
That’s changing with the development of a handful of new schedule reliability indexes and tools from the shipping industry’s more recognizable trade software developers.
But as ocean carrier schedule reliability becomes more transparent, new research from the Massachusetts Institute of Technology (MIT) shows that shippers probably shouldn’t fixate too heavily on schedule variability of the ocean leg of goods transit.
More specific, the MIT research found lack of schedule variability was far greater in the ports at either end of the ocean leg.
As Chris Caplice, executive director of MIT’s Center for Transportation & Logistics, put it to American Shipper
: “You need to follow the container, not the boat.”
Caplice and his team tracked the inbound and outbound schedule reliability of shipments from a single company over four continents. They studied the pick-up at origin, arrival at origin port, ocean transit, arrival at destination port, dwell time, and journey from destination port to the inland drop-off point.
“The ocean leg wasn’t that bad,” Caplice said. “If you’re worried about reliability, it’s at the land ends, the handoffs. We haven’t seen a company where the real variability problem is port-to-port.”
For instance, on a shipment from the Pacific Rim to North America, the average total transit time was 29.9 days, with 14.4 days taken up by the ocean transit. The coefficient of variation (essentially a measure of variability) for the ocean transit was 5.1.
Meanwhile, the average dwell time at the origin port was 3.2 days with a coefficient of variation of three. The average dwell time at the destination port was 4.1 days, with a coefficient of variation of four.
executive director, Center for Transportation & Logistics, Massachusetts Institute of Technology
|“You need to follow the container, not the boat.”
What this says is that while there’s more variation in terms of actual days on the ocean leg than at the ports, in terms of what that means to the length of the ocean transit, it matters less. Or to put it another way, if a 3.2 hour flight is delayed by three hours, it’s more of a problem than if a 14.4 hour flight is delayed by 5.1 hours.
“Five days is five days, but the absolute matters less than the relative,” Caplice said.
Especially since on the landside transits on both ends, the coefficient of variation was even higher relative to the average transit time.
Put it all together, and ocean carriers have probably been blamed a little too much for lack of reliability, when in reality, no one — neither ports nor landside carriers — are performing all that well in terms of schedule variability.
Caplice’s research turned up another interesting insight — that there is often no comparison between the transit times shippers were actually experiencing and what they were promised in their service contracts.
“There’s a gap between what they procured for and what they’re getting,” he said.
Caplice also said his preliminary research found that major, one-off incidents tend to taint management’s perception of reliability.
“What we found was that there’s a lot of management by anecdote,” he said. “The ‘bad thing’ happened three years ago, and they apply that to everything. Discretionary (cargo rolls) are the ones that really piss off senior management, but we looked at data, and they were very rare. What we’re trying to do is apply some data to show how infrequent these things are.”
According to MIT’s research, the number of interim stops made by a vessel between a shipper’s origin and destination port can have a negative influence on ocean leg reliability. Meanwhile, there’s a small positive correlation when carriers use their own-operated terminals.
. While reliability on the ocean leg may not be as problematic as its perceived to be, it still is important, judging by the sheer number of measuring tools that have emerged in recent months. Entering this year, London-based consultant Drewry was the only entity to offer a regular, publicly available index of carrier on-time performance.
Drewry reviews, on a quarterly basis, on-time performance of the top 20 carriers at 10 benchmark global ports.
Aside from a couple individual carriers — namely APL and MOL — releasing quarterly performance updates on a trade-wide basis, there was little help for shippers to delineate which carriers consistently showed up on time.
Then, in September, the Denmark-based maritime consultancy SeaIntel started its own Global Liner Schedule Reliability report, which on a monthly basis breaks down on-time performance for 44 carriers by trade lane and individual service.
And more recently, two of the online trade services software developers, which became known primarily for providing Web-based ocean freight bookings, have released functions that allow customers to track ocean carrier schedule reliability.
In early November, INTTRA said it will offer its OceanMetrics module to its entire network of customers in the first quarter of 2012. It has offered the product through a pilot program since February to some of its largest ocean carrier customers, and more than 40 shippers and freight forwarders.
“OceanMetrics goes way beyond measuring performance based on schedule reliability,” said Michael Roberts, INTTRA’s director of marketing. “We can measure the service reliability against what the carrier promised the shipper on the booking confirmations. If it is door-to-door, that is what will be measured — not just vessel arrival or transport time.
“What is great is we are providing the same information to both the carrier and the shipper so they can all speak from the same page,” he added. “We are giving them both common information so they can have a collaborative and productive discussion based on a common set of data about how do we improve service performance.”
Roberts said in the third quarter of 2011, using “several million TEUs across a significant range of carriers,” INTTRA found that 81 percent of vessels arrived on time and 57 percent of containers arrived on time. On time, for this exercise, was a three-day window — either the day of the booking or the day before or after. Looking only at the containers delivered on the same day as the booking confirmation, the on-time percentage dropped from 57 percent to 26 percent.
“This data shows that a carrier with high schedule reliability might not be the carrier that delivers the containers on time,” Roberts said.
He said the product also allows carriers to measure their shippers, such as the number of booking cancellations they make.
. But the tool was developed foremost with shippers in mind.
“On-time performance, as we have taken a view of it within OceanMetrics, is all about looking at it from a shipper’s perspective,” said Cindy Bergelt, senior product manager for OceanMetrics. “From the shipper’s perspective, what they care most about is whatever they put in the booking. That is what they want to be able to see. So if I booked something that was a door-to-door, then I want to know for that carrier what is the on-time performance door-to-door.”
Bergelt said OceanMetrics doesn’t actually break out the components of a specific move.
“We actually just give a performance measurement for on-time performance,” she said. “So it can go across those different types of moves. We are looking at a three-month rolling average. It is not specific to one particular shipment. We are looking across their shipments in that three-month period to say ‘well, did this carrier meet this carrier’s expectations according to a confirmed booking that was confirmed by the carrier.’
| “Importers and exporters are managing their cash more tightly. They are trying to conserve their cash and reduce their inventory. With that the reliability of transit time is really critical.”
“This is all about summarized performance measurements. It is not about specific performance measurements. This is not track and trace, though we certainly leverage some of the same information that would be used in a track and trace product.”
For example, a shipper can ask to gauge a carrier’s performance on whether shipments arrived on the day promised, or within a three-day window. Bergelt noted that some commodities, like fruit for example, would be much more sensitive to delivery time. The idea of the three-month rolling average is so that shippers can gauge whether their carriers’ performance is getting better or worse.
A shipper with multiple destinations for its goods could look across a summarized view of all its locations. So, if one carrier is delivering containers to two warehouses — one in Chicago and one in Houston — the shipper can see how delivery compares at the two locations.
As of now, OceanMetrics allows INTTRA customers to compare a carrier’s performance from one country versus another, not by trade lane. So, a shipper would be able to compare Brazil versus China, but not Shanghai versus Yantian.
“Trade lane we are looking at as a future enhancement,” Bergelt said. “We feel this is revolutionary, we feel this will start to drive the kind of conversations that we have heard, as we have talked to carriers and shippers — primarily shippers — that they wished they could have these kind of conversations with carriers based on real data, common measurements. This is something that we feel by giving it to both parties helps to foster that conversation.”
. Meanwhile, another software developer that began life as an ocean freight booking portal, CargoSmart, on Nov. 8 released a schedule reliability application of its own.
The company’s Schedule Reliability product allows users to compare actual vessel arrival times with those shown on schedules for 14,000 vessels at 800 ports.
CargoSmart said it launched the product because shippers and logistics service providers are facing frequent vessel schedule changes due to increased slow steaming, shifting port coverage, vessel diversions and service changes.
“Customers want better quality schedules,” said Kim Le, director of CargoSmart North America. Since the 2008-2009 market decline, “importers and exporters are managing their cash more tightly. They are trying to conserve their cash and reduce their inventory. With that the reliability of transit time is really critical.”
Data in CargoSmart’s product can be parsed by service or port pair. Le said that functionality can aid shippers in considering routes they hadn’t in the past. She said the tool could prove especially beneficial to non-vessel-operating common carriers and freight forwarders.
“There are more 3PLs and NVOs chasing customers and it is very competitive right now, and with that the 3PLs and NVOs have to provide more reliable service and higher quality service to retain their customer base and to grow,” she said. “This tool allows them to better identify carriers, service providers, and also offer better quotes by peer pricing for their customers.”
| “The key for companies is to cast as wide a net as possible across the broader supply chain, to go beyond a single transportation mode, even beyond the transportation segment itself, in order to get the full chain-of-custody picture.”
executive vice president and co-founder,
CargoSmart said shippers want to measure carriers’ performance so that they can select the best carriers and routes and make informed decisions for their supply chain planning. They also can use the information about reliability to decide who they want to invite to respond to requests for proposal (RFPs).
The product bases arrival time on both information it receives from carriers and automatic identification systems from onboard transponders that tell when ships arrive in a harbor. CargoSmart then integrates that with data from carrier’s pro-forma schedules and the Automated Identification System (AIS), which tracks vessels.
Shippers can set tolerances — say 12, 36 or 48 hours from the scheduled arrival day — to decide if a vessel’s arrival is timely for their purposes.
The information could underlie a tiered pricing system for carriers and NVOs, enabling them to potentially obtain a premium for services that are both speedy and reliable, Le suggested.
“The current situation in the market is that any time a customer has a late shipment, they tend to go back and look at a carrier and ask ‘is this because of the vessel or the service,’” Le said. “They may complain to the carrier and the carrier may say ‘try this other service, it is better.’”
If the shipper tries the other service, it may or may not be better, and if they encounter the same issue they may switch again to yet another service, or switch to another carrier, but they don’t know the reliability of the new carrier either.
Le said it is a trial and error situation, a “vicious cycle where they are constantly trying different vessels and services, but they really don’t know the reliability of the carrier. This is the dilemma in the marketplace today that shippers and carriers are complaining to us about.”
CargoSmart’s Schedule Reliability currently covers the top 20 or so carriers and is available online and through downloadable reports. Carrier reliability data is also available on the latest version of CargoSmart’s sailing schedule application for the iPhone.
Le said CargoSmart has previously offered customers insight into non-ocean leg components. The company’s Carrier Performance Reports identify delays before the cargo is loaded on a vessel, after cargo is discharged, or during the inland transit.
. The other big fish in trade software development, GT Nexus, told American Shipper
that “supply chain data is like gold to companies because it can expose huge opportunities for improvement.”
“The key for companies is to cast as wide a net as possible across the broader supply chain, to go beyond a single transportation mode, even beyond the transportation segment itself, in order to get the full chain-of-custody picture,” said Greg Johnsen, executive vice president and co-founder of GT Nexus. “We’ve been providing powerful analytic solutions for years so our customers can tap into the insights their big data sets provide on GT Nexus.”
Caplice, meanwhile, held a roundtable discussion in October with shippers and a handful of carriers and truckers to determine where his research should go next.
“They’d love to find a better way to tie the value of reliability to their overall system,” he said. “Where are the causes of variability? The big takeaway is: what’s the business case. Then I’ll decide how much I care. What’s the business case for the value of reliability? How much effort should I put into it?”
Caplice said the follow-up research will have to tie the business case to inventory levels. He expects to have more data by spring.
“People shipping waste have very different metrics than inbound retail shippers,” he said. “It’s less important to them. Everyone knows reliability is important. But there comes a point where the cost of becoming more reliable outweighs the benefits of being more reliable.”
| Shipper takeaways
- Take advantage of new tools from trade software developers that allow shippers to gauge carrier performance.
- Shippers shouldn’t fixate on variability of the ocean leg of their container shipments, but rather look at the reliability of a container from end-to-end.
- New research by MIT may help shippers determine necessary resources for improving end-to-end reliability.