What is BetterFleet?

BetterFleet is an online information portal that provides free, transparent ship operational efficiency data to help those using the shipping industry as part of their supply chain make more informed decisions when it comes to how they ship their goods.

Why has BetterFleet been developed?

BetterFleet is based on the idea that improving efficiency in the shipping fleet requires greater data transparency and a broader perspective than focus on technical efficiency alone. The end goal of efforts to drive greater ship efficiency is really to drive better freight efficiency. BetterFleet information seeks to expand the conversation about how to move goods from one place to another more efficiently, by bringing charterers, shippers, cargo owners further into the conversation. A consistent baseline of universal data is crucial to understanding and ultimately finding opportunities to drive greater efficiency and lower emissions in the broader system.

What is the difference between BetterFleet and other sources of operational efficiency information?

Universal, free, transparent, consistent – these are all words that set BetterFleet apart from other sources of operational efficiency data. BetterFleet uses state of the art models, similar to those used to evaluate fleet greenhouse gas emissions for the IMO, to generate the best possible information about ship operational efficiency, short of getting data directly from ships themselves.

What are the benefits of BetterFleet data?

BetterFleet will deliver greater understanding to any stakeholder (owner, charterer, shipper, cargo owner, ship manager, equipment manufacturer, technology developer) of the operational efficiency and operational efficiency variability of ships and fleets of similar ships. It will allow them to identify potential inefficiencies, find opportunities to optimise operations, and reduce costs for everyone. It will also empower owners to further conversations with their charterers and shippers about how to improve their operations as well as enabling them to improve their commercial competitiveness by showing strong efficiency.

What is the relationship between Rocky Mountain Institute’s ShippingEfficiency.org website, exactEarth and UMAS?

The partnership between UMAS, exactEarth, and CWR is based on a common vision that the future of shipping will be increasingly data dependent and that better data and analysis are fundamental to the low-carbon future of the industry. CWR’s ShippingEfficiency.org site serves as a host to BetterFleet information as well as a gateway to potential future in-depth analysis by UMAS. BetterFleet data are produced using the market leading AIS data (terrestrial and satellite) supplied by exactEarth and a series of algorithms for processing and modelling using AIS data, developed by UMAS.


Where does the BetterFleet data come from?

BetterFleet data is derived from Automated Information System (AIS) information that all ships in the world transmit continuously. This data covers very basic attributes of ship operation including speed, location, and IMO number. This basic data is commonly available from other providers, but BetterFleet combines these with fleet technical specifications, engineering and statistical models developed by UMAS to transform the raw data into estimates of operational efficiency. This is done using software that has been gradually refined over five years and heavily tested against actual ship and voyage data. You can read the full method behind BetterFleet’s results on our Method page.

What are the individual and peer graphs, and how do I read them?

BetterFleet provides two main types of information in the form of histograms. The first histogram shows the average normalised estimated efficiency range and frequency for all ships that are of a similar size and type. This is intended to show how the individual vessel operates relative to similar vessels.

The second histogram shows how frequently an individual vessel is operated at different levels of estimated operational efficiency, measured in grams CO2 per tonne mile of freight, over the course of a year. The more often a ship is operated at low CO2 per tonne-miles of freight figures, the more efficiently it is operating. This data shows how much a ship’s operational efficiency varies over the course of year, in effect comparing the ship to itself over time.

In both histograms, the “zero” value of the x-axis is pinned to the median EEOI value of the comparison group (ship’s of similar size and type). The range of the x-axes is in positive and negative increments away from that median value, expressed in grams of CO2 emitted per ton-nautical mile.

What is the scale and how is the BetterFleet score calculated?

The BetterFleet score is a 1-10 value that is intended to simplify the comparison between the estimated annual operational efficiency of an individual ship and that of its comparison group. The 1-10 scale represents the range between the maximum and minimum values of estimated operational efficiency achieved by that comparison group. The BetterFleet score can change over time as fleet and individual operational change. Future improvement to BetterFleet may include the ability to choose the fleet comparison that is most meaningful to a given user, in which case the Scale value could change even more.

Is it meaningless to compare ships’ operational efficiency (EEOI) values?

The comparison that BetterFleet offers is deliberately not a ranking. It is intended as a tool for stakeholders responsible for making decisions around the operations of individual vessels. We believe these decisions are enhanced to favour efficiency with the ability to see how an individual vessel's efficiency varies over time and performs in the context of its peers.

Why is it useful to compare individual ship efficiency against peers?

The comparison between a ship and comparative vessels of a similar size and type can indicate where the operational efficiency of a vessel can be improved. Many times operational requirements are specified in charter party agreements. This comparison illustrates how those operational requirements result in varying efficiency over a vessel’s annual voyages. This comparison will be most meaningful for those who are closest to decisions about how that vessel is operated.

Why are relative efficiency values shown and not absolute efficiency values?

BetterFleet is not intended to be a tool for calculating carbon footprints or ranking ships. A relative scale for operational efficiency allows a comparison of an individual ship to itself or to its peers but does not allow for more in-depth information to be extrapolated. If additional output information is needed, UMAS is working to develop more detailed reports on individual operational efficiency.

How are the “comparison vessels” referenced in the comparison scale grouped – how are the groups defined?

The method groups ships of similar type and size together, consistent with the categories used in the Third IMO GHG Study. There are many ways to define groups of similar ships, and this one is simple and transparent, however alternative groupings may be considered in the future.

What do the maps depict?

BetterFleet’s maps show the track of an individual ship over the course of the past 12 months (at launch, for simply 2015), corresponding to the hourly-average values shown in the individual ship histogram. The “track” is represented by dots for each data point that are colour coded to correspond to the relative operational efficiency achieved by the vessel. In effect, the map offers a visualisation of the locations and routes where vessels operated more and less efficiently.


How do the results account for ballast legs?

Fleet average values of utilisation are calculated for each ship type and size category, and applied consistently for all ships within that fleet. Utilisation takes into account both the distance travelled on loaded and ballast voyages, and the average payload carried relative to the ship’s capacity (dwt), when loaded. For further detail, see the Method document.

This method removes the variability in operational efficiency of different ships caused by differences in their utilisation (for example, a ship optimised with fewer ballast legs relative to similar ships in the fleet), and will be reviewed in the future as methods to define ship specific utilisation are refined.

How does the model account for weather and hull and propeller condition?

Both adverse weather and hull and propeller fouling can increase fuel consumption, and therefore impact the operational efficiency of a ship. Sailing with or against currents is a further source of variability in operational efficiency. The method currently assumes an annual average increase in fuel consumption related to the weather (varied according to whether the ship is coastal or ocean going), and hull and propeller fouling. This application of annual average fuel consumption increases will cause some inaccuracy in the histogram showing the variability in a ship’s operational efficiency over the course of a year, and should be considered when interpreting the results.

Furthermore, this use of ‘fleet average’ assumptions means ships with high performance hull coatings, or high frequency hull and propeller cleaning, will have operational efficiency improvements that are not currently being captured by the BetterFleet data.

AIS derived models that include the weather’s impacts (metocean including currents) on fuel consumption have been developed, but require testing to ensure that uncertainties in any metocean input data do not adversely impact the overall accuracy relative to a more robust and transparent application of annual average fuel consumption impacts.

How complete is AIS data collected by exactEarth?

AIS data’s coverage is dependent on both the ship’s AIS transponder being on and within range of a receiving station in exactEarth or Vesseltracker’s network. The approximate maximum range between the ship and a receiving station is 50nm, so coastal activity coverage is often excellent, but the deep-sea coverage is dependent on an orbiting satellite being within range. exactEarth currently operate a constellation of 10 orbiting satellites, which ensure several satellite overflights of every point on the earth’s surface per day. This ensures that even when on the high seas, ship’s activity can be observed with coverage gaps normally not exceeding approximately 6 hours. exactEarth’s planned next phase of deployment will increase the constellation to ~60, and provide effectively continuous coverage of the earth’s surface.

Ship’s captains have the authority to turn off AIS transponders at their discretion, if they consider they are creating safety risks. Examples where they can impair safety include when operating near piracy zones, or undertaking operations on board for which electromagnetic induced sparks could create a hazard. For this reason, even with excellent satellite and shore-based coverage with receivers, ship’s AIS can sometimes not be observed.

Both for shortages in receiver coverage and for when transponders are turned off, supplementary data and algorithms can help to estimate the activity.

See a further discussion and coverage statistics, see the full method behind BetterFleet’s results in the Method document.

For container vessels, what is the quality of cargo information without which you cannot determine cargo tonne/mile?

Container vessel’s operational efficiency is often indicated specific to the number of containers carried, for example in terms of gCO2/TEUnm. In this release of BetterFleet data, only cargo mass is used so the operational efficiency is presented as gCO2/tnm. In this calculation, a containership’s cargo mass includes both the mass of the containers carried and the mass of their contents, and is obtained as for tankers and bulk carriers through the estimation of the fleet average utilisation. Please see the Method document for more detail.

If wanting to estimate efficiency and emissions in terms of containers moved, data such as CCWG’s trade lane specific emissions factors may provide an alternative perspective.