Since 1 January 2023, every cargo ship, RoPax and cruise vessel above 5,000 GT trading internationally has been required to calculate an annual operational Carbon Intensity Indicator and receive a rating from A to E under Regulation 28 of MARPOL Annex VI.
For the first three years, the tightening was gentle: 5% in 2023, 7% in 2024, 9% in 2025. Many ships coasted along on a C rating without changing much.
2026 is different, for three reasons:
This guide walks through the calculation the way a fleet manager or DPA actually has to do it — step by step, with the real tables you need, and a complete worked example for a Kamsarmax bulk carrier that ends up with a D rating and needs a plan.
CII is an operational efficiency measure, not a design one. It answers a single question:
How many grams of CO₂ does this ship emit to move one tonne of its capacity one nautical mile, over a full calendar year?
That distinction matters. EEXI (Energy Efficiency Existing Ship Index) is a one-off, design-based calculation — you do it once, you pass or you fail, and it never changes unless you modify the ship. CII is recalculated every single year from real operational data, and your rating can move up or down depending on how you traded.
The metric used for almost all ship types is the AER (Annual Efficiency Ratio), which uses deadweight as the capacity term. Note that AER uses design capacity, not cargo actually carried — so a ship that sails in ballast or part-laden is penalised exactly as if it were full. This is one of the most criticised features of the framework, and one of the things Phase 2 of the review is examining.
Key consequence: speed and time at sea dominate your result. Two sister ships with identical machinery can land two grades apart purely because of trade patterns, waiting time and charterer instructions.
CII under MARPOL Annex VI Reg. 28 applies to ships of 5,000 GT and above engaged in international voyages, in the following categories:
| Applies | Does not apply |
|---|---|
| Bulk carriers | Ships below 5,000 GT |
| Tankers | Offshore support vessels, tugs, dredgers, cable layers |
| Container ships | Fishing vessels |
| Gas carriers and LNG carriers | Pleasure yachts, naval vessels |
| General cargo ships | Platforms, FPSOs, FSUs |
| Refrigerated cargo carriers | Category A ships as defined in the Polar Code (exempt from Reg. 28) |
| Combination carriers | Ships not propelled by mechanical means |
| Ro-ro cargo ships (incl. vehicle carriers) | |
| Ro-ro passenger ships (incl. high-speed craft) | |
| Cruise passenger ships with non-conventional propulsion |
If a ship type has no reference line defined in the guidelines, it has no required CII and therefore no rating — though it may still be subject to IMO DCS reporting.
Before calculating anything, know which document governs which step. Auditors will ask.
| Guideline | Resolution | What it gives you |
|---|---|---|
| G1 — CII Guidelines & calculation methods | MEPC.352(78) | The formula, CO₂ conversion factors (C_F), definitions of distance and fuel consumption |
| G2 — Reference Lines | MEPC.353(78) | The a and c parameters for the 2019 reference line, per ship type and size band |
| G3 — Reduction Factors | MEPC.338(76), as amended by MEPC.400(83) | The annual Z factor, 2023 through 2030 |
| G4 — Rating Guidelines | MEPC.354(78) | The dd vectors that set the A/B/C/D/E boundaries |
| G5 — Correction Factors & Voyage Adjustments | MEPC.355(78) (interim) | Legitimate deductions: reefer power, cargo heating, shuttle tanker DP, ice navigation, etc. |
| SEEMP Guidelines | MEPC.346(78), amended by MEPC.401(83) | SEEMP Part III content; revised definition of “underway” |
| Verification & Company Audits | MEPC.347(78) | How the Administration/RO verifies your Part III |
| DCS reporting format | MEPC.395(82) | The standardised electronic reporting format |
A note on “underway”: MEPC.401(83) amended the SEEMP guidelines so that “underway” is defined as the period between Full Away on Passage (FAOP) and End of Sea Passage (EOSP). Make sure your noon reporting and voyage logging systems use this definition consistently — mismatched definitions between your onboard software and your verifier’s is one of the most common sources of rejected submissions.
The attained annual operational CII is:
Σ (FCⱼ × C_Fⱼ)
Attained CII = ─────────────────────
Capacity × Distance
Where:
The result is expressed in gCO₂ per DWT·nm (or gCO₂ per GT·nm).
The required annual operational CII is:
Required CII = CII_ref × (1 − Z/100)
where CII_ref = a × Capacity^(−c)
And the rating is determined by the ratio:
Rating ratio = Attained CII / Required CII
Lower is better. A ratio of exactly 1.00 sits comfortably inside band C.
Sounds trivial; it isn’t. Errors here invalidate everything downstream.
You are not creating new data; you are reusing what you already report under the IMO Data Collection System. You need, for the full calendar year:
Include everything. Fuel burned at anchor, in port, while waiting for a berth, during drifting, and while shifting all counts toward the numerator. Only distance over ground counts toward the denominator — so 20 days at anchor off a congested port is pure CII damage with zero offsetting transport work. This is the single largest hidden risk in the framework, and it is why Phase 2 of the review is specifically examining port and idle emissions.
Fuel consumed by shore-supplied electricity is not counted, since it is not consumed by the ship.
Multiply each fuel’s annual consumption by its C_F factor from G1:
| Fuel type | Reference | C_F (gCO₂ / g fuel) |
|---|---|---|
| Heavy Fuel Oil (HFO) | ISO 8217 RME–RMK | 3.114 |
| Light Fuel Oil (LFO) | ISO 8217 RMA–RMD | 3.151 |
| Diesel / Gas Oil (MDO, MGO) | ISO 8217 DMX–DMB | 3.206 |
| LPG (Propane) | — | 3.000 |
| LPG (Butane) | — | 3.030 |
| LNG | — | 2.750 |
| Methanol | — | 1.375 |
| Ethanol | — | 1.913 |
Important: these are tank-to-wake factors based on carbon content. They are not well-to-wake, and biofuels are not automatically zero-rated here — treatment of alternative fuels for CII purposes follows separate IMO guidance and typically requires certification. Do not confuse these factors with the life-cycle factors under the proposed Net-Zero Framework or with the well-to-wake intensities used under FuelEU Maritime. Three different regimes, three different numbers.
Sum across all fuel types to get total annual CO₂ in grams.
G5 (MEPC.355(78)) permits certain fuel consumption to be deducted. These are interim guidelines and are being reviewed, but as of 2026 the main categories are:
Correction factors (deduct fuel used for a specific non-propulsion purpose):
Voyage adjustments (deduct fuel consumed during specific voyage periods):
Two rules that catch people out:
If none apply — which is the case for most dry bulk operators — skip this step. Your uncorrected figure is your attained CII.
Divide corrected total CO₂ (grams) by (Capacity × Distance). Watch the units: the numerator must be in grams, so multiply tonnes of CO₂ by 10⁶.
6a. Find the 2019 reference line. Look up a and c from G2:
| Ship type | Capacity band | Capacity used | a | c |
|---|---|---|---|---|
| Bulk carrier | ≥ 279,000 DWT | 279,000 | 4,745 | 0.622 |
| Bulk carrier | < 279,000 DWT | DWT | 4,745 | 0.622 |
| Self-unloading bulk carrier | — | DWT | 5,498 | 0.621 |
| Gas carrier | ≥ 65,000 DWT | DWT | 14405 × 10⁷ | 2.071 |
| Gas carrier | < 65,000 DWT | DWT | 8,104 | 0.639 |
| Tanker | — | DWT | 5,247 | 0.610 |
| Container ship | — | DWT | 1,984 | 0.489 |
| General cargo ship | ≥ 20,000 DWT | DWT | 31,948 | 0.792 |
| General cargo ship | < 20,000 DWT | DWT | 588 | 0.3885 |
| Refrigerated cargo carrier | — | DWT | 4,600 | 0.557 |
| Combination carrier | — | DWT | 5,119 | 0.622 |
| LNG carrier | ≥ 100,000 DWT | DWT | 9.827 | 0.000 |
| LNG carrier | < 100,000 DWT | DWT (min. 65,000) | 14479 × 10¹⁰ | 2.673 |
| Ro-ro cargo ship (vehicle carrier) | ≥ 57,700 GT | 57,700 | 3,627 | 0.590 |
| Ro-ro cargo ship (vehicle carrier) | < 57,700 GT | GT | 3,627 / 330 * | 0.590 / 0.329 * |
| Ro-ro cargo ship | — | GT | 1,967 | 0.485 |
| Ro-ro passenger ship | — | GT | 2,023 | 0.460 |
| Ro-ro passenger ship (high-speed craft) | — | GT | 4,196 | 0.460 |
| Cruise passenger ship | — | GT | 930 | 0.383 |
* Vehicle carriers have a further break at 30,000 GT. Always verify the exact band against the current text of MEPC.353(78) as amended, since the reference lines are within scope of the ongoing review.
6b. Apply the reduction factor Z. From MEPC.400(83):
| Year | Z (reduction vs. 2019 reference line) |
|---|---|
| 2023 | 5% |
| 2024 | 7% |
| 2025 | 9% |
| 2026 | 11% |
| 2027 | 13.625% |
| 2028 | 16.250% |
| 2029 | 18.875% |
| 2030 | 21.500% |
The 21.5% endpoint for 2030 was chosen to align with the 2023 IMO GHG Strategy target of cutting CO₂ per transport work by at least 40% by 2030 compared with 2008, which is why the annual increment rises from 2% to 2.625%.
Compute Attained / Required, then compare against the dd vectors from G4:
| Ship type | Capacity band | exp(d1) | exp(d2) | exp(d3) | exp(d4) |
|---|---|---|---|---|---|
| Bulk carrier | DWT | 0.86 | 0.94 | 1.06 | 1.18 |
| Gas carrier | ≥ 65,000 DWT | 0.81 | 0.91 | 1.12 | 1.44 |
| Gas carrier | < 65,000 DWT | 0.85 | 0.95 | 1.06 | 1.25 |
| Tanker | DWT | 0.82 | 0.93 | 1.08 | 1.28 |
| Container ship | DWT | 0.83 | 0.94 | 1.07 | 1.19 |
| General cargo ship | DWT | 0.83 | 0.94 | 1.06 | 1.19 |
| Refrigerated cargo carrier | DWT | 0.78 | 0.91 | 1.07 | 1.20 |
| Combination carrier | DWT | 0.87 | 0.96 | 1.06 | 1.14 |
| LNG carrier | ≥ 100,000 DWT | 0.89 | 0.98 | 1.06 | 1.13 |
| LNG carrier | < 100,000 DWT | 0.78 | 0.92 | 1.10 | 1.37 |
| Ro-ro cargo ship (vehicle carrier) | GT | 0.86 | 0.94 | 1.06 | 1.16 |
| Ro-ro cargo ship | GT | 0.76 | 0.89 | 1.08 | 1.27 |
| Ro-ro passenger ship | GT | 0.76 | 0.92 | 1.14 | 1.30 |
| Cruise passenger ship | GT | 0.87 | 0.95 | 1.06 | 1.16 |
Reading the bands:
| Ratio falls | Rating | Meaning |
|---|---|---|
| Below exp(d1) | A | Major superior |
| exp(d1) to exp(d2) | B | Minor superior |
| exp(d2) to exp(d3) | C | Moderate |
| exp(d3) to exp(d4) | D | Minor inferior |
| Above exp(d4) | E | Inferior |
Within three months of the end of the calendar year (i.e. by 31 March), report the attained annual operational CII and the resulting rating to your flag Administration or an authorised RO, electronically, using the standardised format in MEPC.395(82). Once verified, the rating is recorded and a Statement of Compliance is issued. The ratings for the previous three years must be recorded in SEEMP Part III and kept onboard.
| Parameter | Value |
|---|---|
| Ship type | Bulk carrier |
| Deadweight (design) | 82,000 DWT |
| Reporting year | 2026 |
| HFO consumed | 4,600 t |
| MGO consumed | 320 t |
| Distance over ground | 45,000 nm |
| Applicable G5 corrections | None |
HFO: 4,600 t × 3.114 = 14,324.4 t CO₂
MGO: 320 t × 3.206 = 1,025.9 t CO₂
─────────────
Total = 15,350.3 t CO₂
= 1.53503 × 10¹⁰ g CO₂
Capacity × Distance = 82,000 DWT × 45,000 nm
= 3.69 × 10⁹ DWT·nm
Attained CII = 1.53503 × 10¹⁰ / 3.69 × 10⁹
= 4.160 gCO₂ / DWT·nm
Bulk carrier, below 279,000 DWT → a = 4,745, c = 0.622, Capacity = 82,000
CII_ref = 4,745 × 82,000^(−0.622)
= 4,745 / 1,138.66
= 4.167 gCO₂ / DWT·nm
Z = 11%
Required CII = 4.167 × (1 − 0.11)
= 3.709 gCO₂ / DWT·nm
Ratio = 4.160 / 3.709 = 1.122
Bulk carrier dd vectors: 0.86 / 0.94 / 1.06 / 1.18. A ratio of 1.122 falls between exp(d3) = 1.06 and exp(d4) = 1.18.
Result: Rating D
Expressed as absolute boundaries for this ship in 2026:
| Rating | Attained CII range (gCO₂/DWT·nm) |
|---|---|
| A | below 3.190 |
| B | 3.190 – 3.486 |
| C | 3.486 – 3.931 |
| D | 3.931 – 4.376 ← this ship at 4.160 |
| E | above 4.376 |
To reach the top of band C, attained CII must come down to 3.931, which means:
Required improvement = (4.160 − 3.931) / 4.160 = 5.5%
In absolute terms, that is roughly 844 tonnes of CO₂, or about 271 tonnes of HFO over the year — assuming the same distance sailed.
What does 5.5% look like operationally? Because fuel consumption scales roughly with the cube of speed, fuel per nautical mile scales roughly with the square. A 5.5% cut in fuel per mile therefore needs only about a 2.8% speed reduction — from, say, 12.5 knots down to 12.15 knots. About 0.35 of a knot. That is the good news.
Hold attained CII flat at 4.160 and let the Z factor do its work:
| Year | Z | Required CII | Ratio | Rating |
|---|---|---|---|---|
| 2026 | 11% | 3.709 | 1.12 | D |
| 2027 | 13.625% | 3.599 | 1.16 | D |
| 2028 | 16.250% | 3.490 | 1.19 | E |
| 2029 | 18.875% | 3.381 | 1.23 | E |
| 2030 | 21.500% | 3.271 | 1.27 | E |
Three consecutive D ratings (2026–2028) would already trigger a corrective action plan requirement. But in this trajectory the ship hits E in 2028 anyway — and a single E triggers the requirement immediately.
This is the core lesson of the worked example: a D rating in 2026 is not a stable position. Under the MEPC.400(83) trajectory, standing still means sliding down a grade roughly every two years.
Under Reg. 28 of MARPOL Annex VI, corrective action is mandatory if a ship is:
In either case, the company must develop a plan of corrective actions showing how the required index rating of C or above will be achieved, and include it in a revised SEEMP Part III, which must be submitted to and approved by the Administration or RO before 1 April of the following year.
There is no direct financial penalty from the IMO. The pressure is commercial and administrative:
Worth knowing: a rating pattern of C, C, D, D is fully compliant with no consequence defined by the IMO, because there are not three consecutive Ds. Compliance and good performance are not the same thing, and building a strategy around scraping past the trigger is a fragile plan given the Z-factor trajectory above.
Ranked roughly by cost-effectiveness for a typical dry bulk or tanker operator:
Operational (low cost, fast payback)
Technical (capital cost, longer payback)
Fuel switching
Phase 2 of the short-term measure review. Phase 1 concluded at MEPC 83. Phase 2 began on 1 January 2026 and runs through intersessional and correspondence groups to spring 2028. On the table: the treatment of port and idle time (a persistent complaint, since waiting time is largely outside the shipowner’s control), possible revisions to the reference lines and the CII metric itself, the correction factors and voyage adjustments in G5 — which remain formally interim four years after adoption — and the interaction between CII and the mid-term measures.
The IMO Net-Zero Framework. The NZF’s adoption was postponed at an extraordinary session in October 2025. MEPC 84 (27 April – 1 May 2026) agreed to preserve the NZF as the primary basis for continued negotiation, with two intersessional working group meetings scheduled before MEPC 85 (30 November – 3 December 2026), followed by a resumed Extraordinary Session on 4 December 2026 at which adoption is scheduled. Support appears to have consolidated since 2025, but the outcome remains genuinely uncertain. If adopted, the NZF would introduce a global fuel standard and an economic mechanism operating alongside — not replacing — the CII framework.
IMO DCS transparency. MEPC 83 approved draft amendments to Reg. 27 of MARPOL Annex VI to make DCS data more accessible: fully non-anonymised to MARPOL Annex VI Parties, and available in anonymised form to the public. These amendments were expected to be taken up at MEPC 84. Once in force, ship-level fuel data becomes considerably harder to keep private.
Regional overlays. EU ETS (fully phased in for shipping from 2026) and FuelEU Maritime already impose separate, and differently calculated, obligations on ships calling at EEA ports. A ship can be CII-compliant and still face substantial EU ETS surrender obligations. Model all three regimes together, not sequentially.
Is CII calculated per ship or per fleet? Per ship, per calendar year. There is no fleet averaging or pooling mechanism under the IMO framework — unlike FuelEU Maritime, which does allow pooling.
What happens if a ship changes owner mid-year? The CII is calculated for the full calendar year regardless of ownership changes. Data must be transferred at change of ownership, and a partial-period DCS report is required at the point of transfer.
Does a ship in lay-up get a rating? A ship with no voyages has no distance travelled, which makes the calculation undefined. Handling of laid-up periods and drydocking should be agreed with your flag Administration in advance rather than assumed.
Can a good CII rating offset a poor EEXI? No. They are independent requirements. EEXI is a pass/fail design threshold that must be met; CII is an annual operational rating. Failing EEXI is a certification problem, not something a good operational year can cure.
How does CII relate to the EEOI? The EEOI (Energy Efficiency Operational Indicator, MEPC.1/Circ.684) uses actual cargo carried rather than design capacity, which makes it a more honest measure of real transport efficiency. It remains voluntary. Its cargo-based logic is one of the alternatives raised in discussions about revising the CII metric.
Do I need special software? No — the calculation is arithmetic and can be done in a spreadsheet, as this guide demonstrates. What software buys you is in-year visibility: a rating you only discover in March is a rating you can no longer influence. Class societies offer platforms that project year-end CII from live DCS data, which is where the actual value lies.
This article is intended as general guidance and reflects the regulatory position as of July 2026. It is not legal or compliance advice. CII is an area of active regulatory development — Phase 2 of the short-term measure review runs to 2028, and the G5 correction factor guidelines remain interim. Always verify parameters against the current text of the applicable IMO resolutions and confirm your specific calculation methodology with your flag Administration or Recognized Organization before submission.
© 2025 Matasys. All rights reserved.