Reading EEOI without being misled
Why the largest emitter on a list is rarely the least efficient ship on it.
Absolute emissions answer a simple question: how much carbon did this ship put into the atmosphere last year? It is a real number and it matters. It is also almost useless for judging a vessel, because it mostly measures how much work the ship did.
A container ship on continuous liner service will out-emit a handysize bulker that spent four months idle. That tells you about trade patterns, not about engineering or seamanship.
The efficiency ratio
EEOI, the Energy Efficiency Operational Indicator, divides emissions by transport work: grams of CO₂ per tonne of cargo carried one nautical mile. A lower figure means more freight moved per unit of carbon. Because it is a ratio, it lets a 300-metre vessel be compared with a 100-metre one, provided both are the same type of ship doing the same kind of work.
Across types the comparison collapses. A cruise ship's transport work is measured in passengers, a car carrier's in vehicle units, a tanker's in tonnes of cargo. Those units do not convert. Every ranking on this site is therefore calculated within a ship type and within a single reporting year, never across them.
The outlier problem
EEOI is a division, and the source data contains returns where reported transport work is close to zero: a ship that spent the year in lay-up, or one that carried almost nothing while still burning fuel to keep systems running. Dividing by a near-zero denominator produces figures in the millions.
Measured across the published dataset, the median EEOI is around 12 and the 99.9th percentile is roughly 4,100. Values above 100,000 are therefore withheld on this site rather than displayed. They are not evidence of an extraordinarily dirty ship. They are evidence that the ratio has stopped meaning anything for that return.
Design index versus operation
Many vessels also report a technical efficiency figure, an EEDI or EIV. That is fixed at construction and describes the hull and machinery as designed. It says nothing about how the ship is actually run. A vessel can hold an excellent design index and a poor EEOI by sailing fast, waiting at anchor, or running half empty. Read the two together.
Median EEOI by ship type
| Ship type | Vessels | Median EEOI | CO₂, latest (Mt) |
|---|---|---|---|
| Bulk carrier | 7,439 | 8.0 | 25.2 |
| Oil tanker | 3,308 | 8.6 | 26.9 |
| Container ship | 3,249 | 17.9 | 62.4 |
| General cargo ship | 2,768 | 30.2 | 9.3 |
| Chemical tanker | 2,202 | 15.7 | 11.5 |
| Vehicle carrier | 644 | 60.2 | 5.1 |
| Gas carrier | 616 | 32.9 | 4.1 |
| LNG carrier | 528 | 6.5 | 12.4 |
| Ro-pax ship | 492 | 757.2 | 14.6 |
| Other ship types | 454 | 87.2 | 2.5 |
| Other ship types (Offshore) | 327 | 1,367.1 | 2.0 |
| Ro-ro ship | 301 | 114.0 | 6.0 |
Each vessel's most recent return. The spread between types is the reason rankings on this site never cross a type boundary.
How many vessels can be ranked
| Year | Ranked | Not rankable | Share ranked |
|---|---|---|---|
| 2018 | 8,960 | 998 | 90 % |
| 2019 | 9,222 | 1,040 | 90 % |
| 2020 | 8,884 | 970 | 90 % |
| 2021 | 9,162 | 958 | 91 % |
| 2022 | 9,924 | 1,123 | 90 % |
| 2023 | 9,196 | 1,102 | 89 % |
| 2024 | 8,622 | 1,163 | 88 % |
| 2025 | 9,512 | 1,473 | 87 % |
A return is not rankable when transport work is missing or implausible, when the type has too few returns to compare, or when the ratio exceeds the cut-off. Unranked vessels lists the reasons.
Two ends of one scale: bulk carriers in 2025
Among the 3,241 bulk carriers ranked for 2025, MARYAM D reported the lowest EEOI, 1.1 g CO₂ per tonne-mile on 3,913 t of CO₂, while KMAX EMPEROR reported the highest, 78.5, on 6,310 t. The absolute figures do not order the same way as the ratios: the ship that emitted less is not the one that carried its cargo more efficiently.
Questions readers ask
What exactly is EEOI?
The Energy Efficiency Operational Indicator: CO₂ emitted divided by transport work, expressed as grams of CO₂ per tonne of cargo moved one nautical mile. It is calculated from the fuel, cargo and distance figures in each MRV return.
What is a good EEOI?
It depends entirely on ship type. In this register the median is 8.0 for bulk carriers and 114.0 for ro-ro ships; across every 2025 return the median is 12.2. Compare a vessel with the median for its own type and year, never with the fleet.
Why can EEOI not be compared across ship types?
Transport work is measured in different units: tonnes of cargo for bulk carriers and tankers, passengers for cruise ships, vehicle units for car carriers. A gram per passenger-mile and a gram per tonne-mile are not the same quantity.
Why are some ships not ranked?
Because the return lacks the transport work needed for the ratio, the type has too few returns in that year, or the reported ratio is an artefact of a near-zero denominator. For 2025, 9,512 vessels could be ranked and 1,473 could not.
Is EEOI the same as the IMO's CII rating?
No. The Carbon Intensity Indicator uses the ship's design capacity (deadweight or gross tonnage) as the denominator and grades vessels A to E against a yearly reference line. EEOI uses the cargo actually carried, so a ship sailing half empty scores worse on EEOI than on CII.
Why are values above 100,000 hidden?
Measured across the published dataset the median EEOI is around 12 and the 99.9th percentile roughly 4,100. A value in the hundreds of thousands means the transport work in the return was close to zero, typically a ship in lay-up, and the ratio has stopped describing efficiency.
Does a low EEOI mean a clean ship?
It means the ship moved a lot of cargo per tonne of CO₂. A large vessel on a busy liner route can have a low EEOI and still be among the largest absolute emitters in its type. Read the ratio next to the absolute figure on each vessel page.