Thursday Thoughts: Are you really having a conversation on Schedule Reliability for the last 10 years+ ?

For over a decade BCO / NVOCC have engaged Carriers with sometimes confidence, sometimes unsureness, on a report said to measure Schedule Reliability “globally” - but does it, really???

Through this article I will walk you through the mist, by factually explain the nuances, demystify the concepts, and head-knocking the limitations.??

Afterwards, let me know what your perspectives are, and if it changed after reading this.?

?

1. Reliability, one word, three concepts.?

1. “Global Liner Performance”, classic or Jurassic??

1. “Schedule Reliability Scorecard”, a Gen-y (genie) or a Gen-Z??

1. A conclusion?

?

Gear, secondo !        

?

?

Reliability, one word, three concepts.?

?

About 93% of the Container ships sail as per a Proforma Schedule, a baseline set for a “Service” connecting ports at (most often) weekly intervals, design to meet customers' requirements on speed, cost, carbon emissions one day.?

This Proforma schedule is often presented as a flyer, a smooth and slick document for Sales to build upon, and within Carriers’ systems it is the core element onto which the ships deployment will be planned.?

The first concept is born from the very nature of the Liner business: be in the port every week, on the day you designed to be – this is Schedule Reliability.?        

?

?While calling on the day you planned to is ideal, most of the supply chains can accommodate delays – as long as there is a call every week.

However, Carriers will omit ports as a common lever to recover and declare blank sailing / slide when the impact on the schedule is too big, or demand too low.??

Thus, the second concept is about how consistently the ports are being called over a period – this is Schedule Stability.?        

?

?For we cross oceans, follow demand into congested ports, compete for the same windows, the plan doesn’t always come true. However, when placing bookings customers should be given an accurate expectation of when their containers will depart and arrive in ports.?

That’s the third concept, the ability to forecast precisely a schedule weeks in advance – this is Schedule Accuracy.?        

?

?

“Global Liner Performance”, classic or Jurassic??

?

Sea Intelligence started more than a decade ago, and produces a monthly “GLP” report that is said to be about Schedule Reliability – but is it really ??

?

“Our standard schedule reliability definition is actual vessel arrival within +/- 1 calendar day of scheduled arrival, so a vessel scheduled to arrive on the 15th would be counted as on time if it arrived on the 14th, 15th, or the 16th.”?

The methodology measures the actual date of arrival compared to a forecasted date of arrival, which is truly a Schedule Accuracy measurement.?

Also, this approach of calendar day inherently tweaks the perception: a vessel arriving 23 hours late on next calendar day can be considered “on time”, which by right should considered “late”; and a vessel arriving 1hr early on the day prior (13th 23:00) would be considered a failure: “early”.?

??

“At present, our methodology requires that both a scheduled arrival and actual arrival be recorded, in order for a call to be included in the schedule reliability calculation.”?

Secondly, this highlights that Sea Intelligence does not measure the Schedule Stability, and it leads to inflated Schedule Accuracy scores especially in regions with frequent blank sailings or omitted calls.?

?

“Depending on the carrier schedule used, we pre-select one source carrier per service, that we source the schedules from. The choice of source carrier is based on many factors, but generally reflect our expectation of which carrier will have the most updated schedules”?

The methodology relies on a selected “source carrier”, which introduces a bias though is initially intended to ensure consistency:

1. The reader has no guarantee that the choice is not based on perceived data quality, rather than objective & measured one ;
2. To put one carrier in charge of a service that has multiple carriers operating, assumes that all carriers share far-out schedules to one another, and do so at regular interval, or that the “source carrier” can forecast schedules better than its partners can on their own vessels.?

?

“An example, if we are collecting schedules on the 5th of December, we will ignore any schedules collected on the 5th for the period between the 5th and the 19th, and only consider schedules from December 20th on-wards. The schedules for December 5th to 19th would then have been collected in our previous schedule collection process, roughly 14 days earlier.”?

The methodology only considers data beyond two weeks, imposing a “quiet period”, which negates carriers efforts to come out with contingency plans to remedy schedules, plans that very often can be decided closer to the date.

Imagine: the vessel is meant to call Qingdao, Busan, Shanghai, Ningbo, Nansha and Singapore – suddenly, Qingdao close because of inclement weather and delay you by 2 days. Today you share data to Sea Intelligence:

• if you do nothing, you will be late versus proforma in Singapore but rated “on time”;
• if you do omit Shanghai and speed up, you will be on time in Singapore as per Proforma (Reliable) but fail Sea Intelligence and rated “early”.?

?What would you prefer ?

“It is important to note that calculating carriers’ scores across all the different trade lanes does not add up to that specific carriers’ overall scores. This is because of how we define regions and trade lanes, as the carrier scores count each distinct vessel arrivals only once, but one vessel arrival may be counted on multiple trade lanes.”?

The inability to connect a carrier score, to its Trade Lane score, and the Alliance it is part of, is confusing for the reader and may produce very sterile discussions. Nowadays, data and storytelling are necessary tools to convince your audience, and one must be able to switch from macro to micro without losing the thread.?

?

“Any fixed schedule container liner service that makes scheduled port calls in more than one continental region AND has an average schedule roundtrip duration of more than 14 days”?

Though the term “Global” may have been a marketing decision, the above clarifies what is and isn’t included in the report; with Gemini and the multiplication of short shuttles between hub & outports, this may significantly mis-represent Maersk/Hapag performance as better than it really is.

?

Sub-conclusion:?

To put things in context, Sea Intelligence started in an age when Carriers were reluctant to share data, were not applying any standard on their schedules, and AI was only a threat in Schwarzenegger's movies.?

It only stands to reason that they had to take positions, to aggregate data, to fix rules and methodology points, to produce the best version of the truth they could.?

Above should not be read as a critic, as I merely quote their own words and propose ways they could follow to improve their product – for a product that does not evolve, is doomed – ask BlackBerry.

print ( So, Classic or Jurassic ? )??        

?

“Schedule Reliability Scorecard”, a Gen-y (genie) or a Gen-Z??

?

There is now a contender to Sea Intelligence’s hegemony: eeSea, taking radically different approaches, be it on the content and the shape.?

?

“Our primary measurement is the average delay in days ? Proforma vs actual time of the vessel event ? For example: 5h45m = 5.75 hrs = 0.24 days late ? A delayed vessel is expressed with a negative number. ? A positive number indicates an early arrival”?

?

“Our secondary measurement is the on-time percentage ? We mark < 12 hrs delay as an on-time arrival ? This variable can be adjusted to fit your use case in our data ? A port event < 12 hrs late gets 100%, > 12 hrs late gets 0%. The aggregate percentage of vessels on-time is used throughout”?

?

From this we understand that eeSea SRS is really dealing with Schedule Reliability, holding carriers to their advertised Proforma Schedule.?

This is also how carriers measure internally their performance, and what they can present to their customers, therefore there is an advantage for all parties to exchange on this report.?

?

?

“? Event-based: port arrival, berth arrival, berth departure and port departure ? Primarily from un-biased, geo-fence-based AIS events”?

?

The choice to use the AIS (imagine a GPS transmitter onboard the ships continuously sending their positions) has a few benefits, one to remove the hassle for carriers to manually share their schedules every second week.??

However, it is important to highlight that AIS signals can be disrupted, or even turned off (some ships crossing the Red Sea have done so), and Geofencing quality depends considerably on the quality of the fencing.?

The alternative for eeSea is then to default to Carriers website schedules.?

??

“All can then be aggregated and analyzed through several lenses ? Trade lane – last load & first discharge ? Service & alliance ? Port, country, region ? Vessel operating carrier ? VSA partner ? Berth/ port arrival/ departure ? Terminal, terminal operator"?

“And always – Each visualization is accompanied by an explanation of measures and filters used.”?

?Transparency seems at the core of how eeSea build their platform and reporting, which enables the user to better understand carriers' performance, but also to identify ports becoming bottleneck and take proactive measures.?

?

On the downside, the SRS is only published quarterly while GLP is monthly, which limit the impact of eeSea in current discussions Carriers and Customers have.?

But that’s notwithstanding that eeSea proposes a comprehensive website with numerous capabilities that could take a whole article to explain, and a Tableau to access “live” data, another testimony to their keenness on transparency and technology.?

?

?

Sub-conclusion:?

?Key challenges for eeSea include:?

• To convince a market that has been dominated by 1 provider for more than a decade.

• To maintain the right balance between digestible insights and a whole universe of data points & measurements.?

?

A conclusion?

?

Honestly, I thought of not writing a conclusion, partly because I want to read your genuine ones, and partly because I am who I am: a Gen-Y that is passionate about data, an Operations leader strongly advocating for holisticness, and forever in quest of becoming a better version of me.??

But here I go:??

Sea Intelligence deserves much credit for pioneering in a mined field and offering some visibility to the industry – though, in my opinion, that is not so humble (paraphrasing from Dumbledore, Harry Potter), they don’t present Schedule Reliability but Schedule Accuracy – and in ways that can be debated in 2024.?

It is impossible to have a constructive discussion, internally or with customers, based on their Global Liner Performance report because of the scoping, methodology, and inability to link the high-level numbers to the very granular ones.?

There are signs that the new kid in the block, eeSea, can produced comprehensive benchmarking of carriers and ports, using technology and placing transparency at the core of their endeavor.?

?

?

Disclaimer:?        

The views and opinions expressed in this article are solely my own and do not reflect the official policy, position, or opinions of my employer or any other organization I am affiliated with. Furthermore, I do not have any financial interest, participation, or involvement with the companies mentioned, and the content provided is for informational purposes only.?        

?