Slow Steaming as a Sustainable Measure for Low-Carbon Maritime Transport

The paper addresses the urgent need to reduce greenhouse gas emissions, in line with the Paris Agreement targets and the IMO Initial Strategy goals to reduce CO2 emissions per transport work by at least 40% by 2030 and 70% by 2050 relative to 2008. Within this context, maritime transport accounts for 2.89% of global anthropogenic emissions, and speed reduction (slow steaming) is highlighted as a short-term operational measure to curb emissions. The research question is whether slow steaming can sustainably reduce CO2 emissions and fuel consumption while understanding its technical and operational impacts on ship performance when operating off design conditions. The work investigates a representative post-Panamax container ship to quantify emissions reductions, assess the validity of the commonly used cubic law relating fuel consumption to speed, and analyze hydrodynamic performance at reduced speeds.

The study situates slow steaming among technological and operational measures for maritime decarbonization, alongside alternative fuels. Prior works report slow steaming adoption driven by fuel cost reductions and market conditions, with observed fuel and CO2 savings ranging from 20% to 40% and even over 60% depending on speed reduction. Economic analyses indicate strong incentives under high fuel prices and low charter rates. Speed reduction affects voyage frequency and may necessitate additional ships to maintain constant transport work; relationships between speed, fleet size, routing, and bunkering under uncertainty have been modeled. Concerns include weakened just-in-time delivery, potential modal shifts to higher-emission transport, and capital investment due to fleet expansion. Hydrodynamic and CFD literature shows speed and scale effects on resistance and propulsion characteristics, including changes in form factor at low Froude numbers, trim optimization benefits, and biofilm impacts that vary with speed. Prior studies have questioned the validity of the cubic law for speed–fuel relationships and emphasized the need to use accurate speed–power curves tailored to ship type and operating profile.

A full-scale post-Panamax container ship, the Duisburg Test Case (DTC), was used as the case study. The ship is a typical single-screw 14,000 TEU containership with Lpp = 355 m, BWL = 51 m, T = 14.5 m, displacement volume V = 173,467 m³, wetted surface S = 22,032 m², block coefficient CB = 0.661, and design speed Vd = 25 kn. A five-bladed propeller (D = 8.911 m, P/D = 0.959, AE/A0 = 0.8, c = 3.208 m, θeff = 31.97°, d/D = 0.176) was used. Numerical simulations comprised resistance tests (RT), open water tests (OWT), and self-propulsion tests (SPT) using STAR-CCM+ (2020.1.1). RANS with k-ω SST turbulence model and wall functions solved the averaged continuity and momentum equations. Free surface tracking employed the VOF method with HRIC; second-order upwind convective discretization and first-order temporal scheme were used. RT and SPT used a rectangular domain; OWT used a cylindrical domain, with appropriate velocity inlet, pressure outlet, slip/no-slip walls, and VOF wave damping to prevent reflections. OWT applied an MRF in the propeller region at constant n = 1.5 rps across different J. SPT modeled propeller effects via a body force method informed by RT and OWT results, using Goldstein optimal radial force distribution on a virtual disk. Meshes were unstructured hexahedral with refinements near hull, bow/stern, free surface, Kelvin wake, propeller, and prism layers targeting y+ ≈ 50. Verification used three grid levels and three time steps to compute grid and temporal uncertainties via the GCI method; numerical uncertainties for CT, PD, n, J, KT, and KQ were found to be ≤ ~1.25%. Validation was performed at Fn = 0.174 and 0.218 using ITTC 1978 extrapolation of towing tank data; OWT propeller characteristics were compared to experiments following published methodology. Fuel oil consumption (FOC) was computed from brake power PB times specific fuel oil consumption (SFOC) based on MAN B&W project guide data under ISO reference conditions, including effects of engine load. CO2 emissions were derived using the carbon conversion factor CF = 3.114 gCO2/gHFO. Engine selection was based on the operating point at 25 kn using margins (SM = 10%, EM = 10%), yielding MAN B&W 12 K98MC-C7 with MCR 72,240 kW at 104 rpm and PSMCR = 70,602 kW at nSMCR ≈ 102.61 rpm; light running margin LRM in 0.04–0.07 range was considered for propeller rpm. Propulsive efficiencies were calculated: ηP = ηH ηO ηR ηM, with ηM = 0.98, ηH = (1 − t)/(1 − w′), ηO = (J / 2π) (KTO/KQO), and ηR = KQO/KQ. The Admiralty coefficient CADM = Δ^(2/3) V^3 / PB was used to assess speed–economy. Simulations assessed speeds of 25, 22.5, 20, and 17.5 kn in calm water, and emissions reductions were further estimated for constant yearly transport work on a 5141 nm route including median port time (0.71 days).

• Validation against extrapolated towing tank data showed satisfactory agreement; for Fn = 0.174, relative deviations included CT: −0.896%, PB: −4.492%, n: −2.129%, J: 0.330%, ηO: 6.091%, ηP: 6.481%. For Fn = 0.218, CT: −0.918%, PB: −2.498%, n: −1.439%, J: 0.527%, ηO: 6.349%, ηP: 4.695%. • CO2 emissions per hour decreased with speed reduction: at 22.5 kn (0.9 Vd), CO2 ≈ 20.57 t/h (−32.163%); at 20 kn (0.8 Vd), ≈ 15.379 t/h (−49.282%); at 17.5 kn (0.7 Vd), ≈ 11.292 t/h (−62.759%). SFOC increased as speed decreased due to lower engine load (168.568 to 183.962 g/kWh from 25 to 17.5 kn), yielding FOC reductions from 9.737 to 3.626 t/h. • With constant annual transport work on a representative 5141 nm route and median port time, yearly CO2 emissions reductions were −16.89% (10% speed reduction), −21.97% (20%), and −25.74% (30%). • The classical cubic law (FOC ∝ V^3) is not valid for the investigated ship in calm water; speed exponent n1 for FOC was < 3 and decreased with speed: n1 = 2.683 (22.5 kn), 2.042 (20 kn), 1.769 (17.5 kn). Brake power exponent n2 was > 3 and also decreased with speed: n2 = 3.868, 3.278, 3.014, reflecting higher SFOC at lower engine loads. • Hydrodynamic characteristics varied with speed (from Table 9): CT·10^3 was lowest at 22.5 kn (1.629) and highest at 17.5 kn (1.814); ηP ranged 0.736–0.767; ηO slightly increased near reduced speeds (max 0.676 at 22.5 kn); ηH increased with speed reduction (max 1.154 at 17.5 kn); wake fraction (1 − w) and relative rotative efficiency (ηR) remained nearly constant across speeds. • The Admiralty coefficient CADM indicated a hydrodynamic optimum at 0.9 Vd (22.5 kn): CADM = 746.263 (higher than at 25 kn: 681.021; 20 kn: 724.660; 17.5 kn: 684.527).

The findings confirm that slow steaming can substantially reduce fuel consumption and CO2 emissions in calm water, but the benefits depend on operational context. The demonstration that the cubic law overestimates savings underscores the need for ship-specific speed–power curves when assessing emission reduction potential. While hydrodynamic performance parameters generally improve or remain acceptable at reduced speeds, the hydrodynamic optimum for the studied ship lies at 90% of design speed based on CADM and CT/ηP, rather than at the lowest investigated speed, indicating trade-offs between efficiency and speed reduction. Operationally, slower speeds entail longer voyage times, fewer annual voyages, and potential weakening of just-in-time services. For cold chain logistics, extended transit can increase energy use for refrigeration and risk product quality degradation. Market responses may include modal shifts to air or road transport, which could negate environmental gains. Maintaining constant transport work under slow steaming often requires more ships, increasing capital and operational costs. From an engine perspective, off-design low-load operation necessitates technical measures (e.g., blower activation) to avoid thermal stress, and higher SFOC at low loads diminishes the proportional FOC savings. Policy-wise, while mandatory speed reduction would cut sectoral emissions, it may slow innovation in emission-reducing technologies; voluntary, market-based approaches and ship-specific optimization are suggested. Overall, slow steaming is a viable short-term measure within a broader strategy, provided that hydrodynamic performance, operational constraints, and economic variables are jointly considered.

This study evaluated slow steaming as a short-term operational measure for maritime decarbonization using a full-scale post-Panamax container ship in calm water. CFD-based RT, OWT, and SPT simulations, verified and validated against extrapolated towing tank data, quantified hydrodynamic characteristics, FOC, and CO2 emissions across speeds. The main contributions are: (1) CO2 emissions per hour decreased by 32–63% as speed reduced from 25 to 17.5 kn; with constant transport work, yearly reductions were −16.89% (10%), −21.97% (20%), and −25.74% (30%). (2) The classical cubic law for fuel–speed dependence is invalid for the investigated conditions; FOC speed exponent is < 3 and brake power exponent > 3, both decreasing with speed. (3) Most hydrodynamic characteristics are speed-dependent; the hydrodynamic optimum for this ship is 90% of design speed, with the highest Admiralty coefficient and lowest total resistance coefficient. (4) Broader implementation should account for operational, market, and technical factors to avoid adverse effects. Future work will extend the analysis to other ship types, consider performance in waves, and conduct detailed evaluations of engine operation under off-design conditions, including efficiency losses and wear, to provide comprehensive guidance for operators on slow steaming across diverse scenarios.

Results are limited to a single post-Panamax container ship (DTC) operating in calm water and may not generalize to other ship types or wave conditions. Emissions and fuel consumption estimates rely on SFOC data under standard tuning and ISO reference conditions and assume HFO use. The yearly emissions analysis assumes constant transport work on a specific route length and median port time, without a full economic optimization of fleet size, charter rates, or fuel prices. Engine off-design behavior is discussed but not modeled in detail; impacts such as efficiency losses and wear require further study. Validation comparisons use extrapolated towing tank data rather than full-scale trials.