Berthing Energy and Fender Selection

Berthing energy is the kinetic energy a ship carries into a fender at the moment it contacts a berth, calculated as half the ship’s displacement mass times the square of its velocity normal to the berth, corrected by four coefficients for added mass, eccentricity, berth configuration and hull softness. The result is expressed in kilojoules, and it is the quantity every fender system on every commercial quay in the world is sized to absorb.

The governing document changed recently and comprehensively. PIANC MarCom Working Group 211, published as the PIANC Fender Guidelines 2024 in March 2024, completely supersedes WG 33 of 2002, in PIANC’s own words, and the transition period that allowed suppliers to reorganise their catalogues ended on 1 May 2026. Any fender specification still written to WG 33 is now written to a withdrawn method. This article carries the calculation, the coefficient derivations, the velocity evidence, the fender selection and the hull pressure check, on the current basis, with the WG 33 method shown where it explains what changed. It is the berthing-and-fendering leaf of the ports and terminals overview cluster; the load on the ship once it is lying alongside belongs to mooring forces and station keeping .

What berthing energy is, and what it is not

Berthing energy is a transient impact absorbed over a few seconds as the ship makes contact. It is not the mooring load, which is the sustained force wind, current and passing ships impose while the ship lies alongside. The two are sized separately and the distinction matters commercially: a fender chosen only on impact energy can still be the wrong fender, because a lean-on load lasting a tide can transmit higher forces into the same rubber than the berthing did.

The chain of consequence runs one way. Approach velocity sets the energy. Energy sets the deflection at which the chosen fender absorbs it. Deflection sets the reaction force. Reaction force sets the hull contact pressure, the chain loads, the panel scantlings and the pile group under the breasting dolphin . Every number downstream is hostage to the velocity at the top, which is why the velocity is the part of the calculation that is measured rather than assumed at any well-run berth.

Get the sizing wrong in the low direction and the fender bottoms out, the reaction spikes, and the shell plating dents or the structure cracks. Get it wrong in the high direction and the berth costs more than it needed to and the fender is too stiff for the small ships that also call there.

The guideline changed in 2024, and the change is not cosmetic

PIANC WG 211 replaced WG 33 in March 2024. PIANC states plainly that “users cannot simply change WG 33 into WG 211 in their own fender specifications”, and that doing so “will lead [to] significant cost increases due [to the] different design approach adopted in WG 211”. Four changes carry most of that difference.

WG 211 describes the berthing process with higher velocities, lower berthing angles and multiple fender contact than WG 33 assumed. It puts the safety margin into the rubber rather than into the supporting structure, so the marine structure has to be designed as a whole rather than as a fender bolted to a dolphin designed separately. It replaces the single abnormal berthing factor with partial factors derived by reliability methods. And it strongly recommends site-specific velocity data, treating its own tabulated velocities as the fallback for a berth with no records: “Only use presented velocities when no knowledge on, data [for the] site is available. Always talk [to] pilots, harbourmasters [and] tug masters if present.”

The scope has boundaries worth knowing before reaching for it. WG 211 addresses fender systems for seagoing vessels that are tug assisted or thruster equipped, usually pilot supervised, executing controlled berthing manoeuvres. It does not cover collision protection structures such as bridge piers, it gives no guidance on lock chambers, and it does not cover inland vessels and barges, which frequently moor against plain concrete.

WG 33 remains citable as the historical method and as the source of tables that later documents reproduce, notably the hull pressure guide. It is no longer the design basis.

The berthing energy equation

Characteristic berthing energy is the product of the ship’s kinetic energy and four dimensionless coefficients:

WG 211 names the output of that expression the characteristic berthing energy, and obtains the design berthing energy by multiplying it by a partial energy factor. The older WG 33 nomenclature called the same two quantities the normal and the abnormal berthing energy.

Displacement, not deadweight, is the mass term. Owners and terminals specify a design vessel in deadweight because that is how ships are traded, so the displacement has to be derived first, either from published ship-data tables or from the hull dimensions:

with seawater density \(\rho\) taken as 1.025 t/m\(^3\). The characteristic displacement is the largest operational displacement of the design vessel, not its average.

Added mass, Cm

The added mass coefficient accounts for the water the hull drags with it as it moves sideways. That water has to be decelerated along with the ship, so it adds to the effective mass arriving at the fender.

The PIANC method makes the coefficient a function of under-keel clearance, not of hull fullness:

Source: PIANC WG 33 (2002), carried into WG 211 section 5.7.

Shallow water raises the coefficient. Water beneath a ship in a shallow berth cannot escape downward and must be pushed around the hull, so more of it is entrained. That is the opposite of the intuition that a confined berth somehow cushions the arrival, and getting the direction wrong understates the design energy by up to 20 percent.

Two alternative expressions are in legitimate use. Vasco Costa’s 1964 form, \(C_m = 1 + 2D/B\), is the one BS 6349 applies. Shigeru Ueda’s 1981 form, \(C_m = 1 + \pi D / (2 B C_B)\), is widely used in Japan and returns similar or slightly lower values. Neither is the PIANC method, and the choice is not free: if the PIANC added mass coefficient is used, the berth configuration factor is 1.0 by definition, because the under-keel effect is already inside Cm. Applying a Vasco Costa added mass and a separate configuration factor below 1.0 counts the same physics twice.

Eccentricity, Ce

The eccentricity factor is the fraction of kinetic energy that actually reaches the fender. Contact away from the centre of mass makes the ship yaw, and rotational energy is energy the fender never sees.

Here \(K\) is the radius of gyration, \(R\) the distance from the centre of mass to the contact point, and \(\gamma\) the angle between \(R\) and the velocity vector, given by \(\gamma = 90 - \alpha - \arcsin(B/2R)\) with \(\alpha\) the berthing angle. The radius of gyration follows from the block coefficient:

which gives about \(0.224,L_{BP}\) for a fine hull at \(C_B = 0.6\) and about \(0.262,L_{BP}\) for a full one at 0.8.

The resulting values run counter to a common misreading:

A midships contact is the worst case, not the mildest. The velocity vector passes through the centre of mass, the ship does not rotate, and essentially the whole kinetic energy is delivered into the fender. A quarter-point contact halves it. Any design that assumes a midships landing is generous is generous in the wrong direction.

Berth configuration, Cc

The configuration factor covers the cushion of water squeezed between the hull and a solid structure as the ship closes the last metre.

Source: PIANC WG 33 (2002).

“Solid” here means a continuous quay wall, which is the commonest berth type in the world, not a dry dock. And the whole table collapses to 1.0 once the berthing angle exceeds 5 degrees, because displaced water then escapes toward the bow or the stern instead of being trapped.

Softness, Cs

The softness factor accounts for energy absorbed by deflection of the ship’s own shell plating rather than by the fender. The criterion is fender deflection, not fender family: 0.9 or less where fender deflection is 0.15 m or less, and 1.0 or less above that. Modern fenders deflect several hundred millimetres, so the hull takes no meaningful share and Cs = 1.0 is the normal assumption.

Worked example

Take a laden 100,000 DWT tanker at a displacement of 125,000 t, \(L_{BP}\) 236 m, beam 43.0 m, draught 15.1 m, \(C_B\) 0.796, berthing at a third-point contact at 5 degrees against a solid quay with an under-keel clearance ratio of 0.3, at 0.10 m/s.

The radius of gyration is \((0.19 \times 0.796 + 0.11) \times 236 = 61.7\) m. With \(R\) 44.8 m and \(\gamma\) 56.3 degrees, the eccentricity factor is 0.761. The added mass coefficient at Kc/D 0.3 is \(1.875 - 0.75 \times 0.3 = 1.65\), and because the PIANC method supplies it, Cc is 1.0. Cs is 1.0.

Halve the eccentricity by moving contact to the quarter point and the energy falls to about 516 kJ. Raise the velocity from 0.10 to 0.15 m/s and it rises to about 1,766 kJ. The velocity term dominates everything else in the expression, which is the practical argument for spending money on berthing aids rather than on rubber.

Design berthing velocity

Berthing velocity is the asset owner’s input, not the fender supplier’s. WG 211 requires it to come from site-specific information wherever any exists: berthing records, past performance, the approach speed limits the port applies, and the judgement of the pilots , harbour masters and tug masters who work the berth. Its own tabulated values are explicitly a fallback for a berth with no data.

Where a published figure is needed, the Port of Rotterdam design impact speeds are traceable and dated:

Source: Broos, Hoebee, Peeperkorn and Sibbes, PIANC World Congress Panama 2018, Table 4. The second column uses an IACS 5 mm permanent set as the hull damage criterion.

The gap between the two columns is the real margin the port is working with, and it is narrow. For a VLCC the damage threshold sits at 0.10 m/s against a design value of 0.08 m/s.

The older Brolsma curves, which WG 33 and BS 6349-4:2014 Figure 9 both reproduce, are keyed to displacement rather than deadweight and are read off the published figure rather than tabulated by ship class. Their provenance is often given wrongly: they come from Brolsma, Hirs and Langeveld, Paper on Fender Design and Berthing Velocities, presented to the PIANC 24th International Navigation Congress at Leningrad in 1977, with data from shore-based docking systems at three Rotterdam berths and one in Scotland. The lineage runs back through Saurin to Baker’s field observations of 1953.

What field measurement changed about berthing velocity

The assumption that berthing velocity falls with ship size, which shaped every velocity curve from 1953 onward, does not survive measurement. A campaign covering 555 berthings in the Port of Rotterdam concluded that “the collected data do not confirm the historical assumption that berthing velocities are strongly related to ship dimensions of large seagoing vessels.”

The numbers behind that finding are worth stating plainly. Mean velocities by vessel class fell between 0.03 and 0.05 m/s, and the highest single value across all 555 berthings was 0.13 m/s. There was no evidence that laden vessels berth more slowly than ballasted ones, no correlation with berth type or fender type, and no correlation with wind speed in the sheltered Rotterdam basins.

What did correlate was berthing policy: pilot experience, tug assistance, the presence of a berthing aid system, and whether a target velocity had been stated at all. Velocities of 0.25 and 0.26 m/s have been recorded for the largest container ships, so the tail is real and it is long, but the tail is a function of how the berth is run rather than of how big the ship is.

That has a direct consequence for the safety factor. WG 211 removed vessel type and size as inputs to the partial energy factor precisely because the correlation the old table assumed was not found in the data.

From characteristic to design energy

WG 33 applied one factor, once. The abnormal berthing factor from its Table 4.2.5 ran inversely with vessel size:

Source: PIANC WG 33 (2002) Table 4.2.5. BS 6349-4:2014 calls the same quantity an energy factor.

There was never a separate “normal” factor of 1.5 in WG 33, a widely repeated error: one factor was applied to the calculated energy, and the value came from that table.

WG 211 abolished the table. In its place the partial energy factor is assembled from a reference value for 100 berthings per year, corrected for the actual berthing frequency, for whether the berthing is pilot assisted, and for correlation between the design variables. The inputs are the consequence class, the navigation conditions, the variability in displacement, and whether contact falls on a single fender or several.

The consequence class is the new organising idea and it is worth understanding because it, not the ship, now drives the margin:

• Class A: the fender sits on a structure with functional redundancy and few people at risk, and its failure does not close the berth. A continuous earth-retaining quay wall, or a dolphin berth with more than two redundant breasting dolphins.
• Class B: no functional redundancy, so failure most likely closes the berth. A single berth with two breasting dolphins.
• Class C: failure is likely to endanger public lives or close the berth with significant economic loss. Floating storage and regasification units, critical floating power plants.
• Class D: failure is likely to cause significant socio-economic disruption, progressive damage or cascading effects.

Classes A and B cover most commercial marine structures. Navigation conditions moved from WG 145’s two categories to three in WG 211: favourable, moderate and unfavourable.

Fender types and where each is used

Selection turns on the energy demand, the reaction force the structure and the hull can take, the tidal range, the spread of ship sizes, and whether the unit is fixed to a structure or deployed operationally.

WG 211 additionally covers element and leg fenders, parallel motion systems, pile and pivot fenders, and rolling fenders, and tabulates typical sizes with their energy and reaction values. Published catalogue energy and reaction figures are product-specific and change between suppliers and rubber grades, so a design is run against the supplier’s own type-approved data sheet rather than against a generic range.

Base, characteristic and design performance

A fender catalogue figure is not a design figure, and WG 211 formalises the gap between them in a four-step ladder:

1. Base performance, determined by constant-velocity compression under the standard conditions.
2. Characteristic performance, the base performance multiplied by four correction factors: velocity, temperature, angular, and multiple fender contact.
3. Design performance, the characteristic performance divided by partial resistance factors covering the single fender and, separately, multiple fender contact.
4. Verification of the design performance against the design berthing energy.

The correction factors are compound-specific and product-specific. WG 211 defines the standard conditions under which all four equal 1.0 and sets out the test protocols for establishing the velocity, temperature and angle factors, which is what allows one supplier’s factors to be compared with another’s.

WG 33’s equivalent requirement expressed the same idea as a single inequality against the catalogue rated performance data:

Working the other way, from an energy demand to the reaction force it produces on a fender of known rated performance, the approximation below is the one used for a first pass before the supplier’s own curve is interpolated:

It is a screening tool only. The relationship between energy and reaction is set by the shape of the individual fender’s curve, and a buckling unit’s curve is not well described by a square root over its whole range, so the published curve governs any figure that reaches a specification.

The test basis moved as well. WG 33 rated performance came from a decreasing-velocity compression starting at 0.15 m/s and finishing at 0.05 m/s or less, at 23 degrees C. WG 211 bases catalogue performance on a constant-velocity compression with the unit thermally stabilised at 23 plus or minus 5 degrees C. Two catalogues written to the two bases are not directly comparable, which is a large part of why the transition period existed at all.

Two practical points sit inside this section and are missed often enough to be worth stating. A new fender can return a reaction force 30 to 40 percent above catalogue on its first compression, so WG 211 makes a break-in compression compulsory for buckling rubber elements above 1,000 kN reaction and for load-sensitive structures such as dolphins. And there is no such thing as a PIANC-certified fender: PIANC is not a certifying body, and it says so in the guideline because suppliers had been claiming otherwise. Type approval means the fundamental testing was observed or confirmed by a qualified independent third party.

Batch acceptance at manufacture works on a sample. A unit passes if it reaches the required base energy less the manufacturing tolerance without exceeding the reaction value plus the tolerance at any deflection up to maximum design deflection. Testing starts at a 10 percent sample; one failure escalates to 20 percent, and a failure there means the whole batch is tested.

Temperature and velocity effects on rated performance

Rubber stiffens as it cools and as the strain rate rises. Both effects raise the reaction force at a given deflection, which is the direction that matters, because reaction force is what the hull and the structure have to survive.

The Japanese Technical Standards apply a temperature factor between 0.95 and 1.25 against the standard 23 degrees C. Rubber conducts heat poorly: a fender takes on the order of 1.2 days to stabilise at 15 cm depth, so the design input is the daily average maximum and minimum rather than an instantaneous air temperature, and applying a spot reading overstates the correction.

At extreme low temperature a buckling fender can stop buckling altogether. The characteristic reaction peak disappears, the curve rises monotonically, and the unit no longer limits force the way its design intends. The design reaction is then taken as the maximum reaction at which the fender still absorbs the energy it would have absorbed at 23 degrees C. That behaviour, rather than a percentage, is the reason a berth in the northern Baltic or the Arctic gets a different specification from the same berth in the tropics.

Allowable hull pressure and panel sizing

The hull pressure check is the point at which the fender design meets the ship’s structure. Reaction force divided by panel face area gives contact pressure, and that pressure has to stay inside what the shell plating and framing of the design vessel can take.

Source: PIANC WG 33 (2002) Table 4.4.1, the hull pressure guide, reproduced at BS 6349-4:2014 clause 4.6.2 Table 3.

Three qualifications travel with that table and none of them is optional.

These are guidance values, not certified limits. No IACS Unified Requirement governs allowable hull pressure from fender contact, and the criteria in the national codes trace back to PIANC work of 1984 rather than to ship structural rules. The governing document for a particular ship is the owner’s specification derived from that ship’s own structural analysis.

The national codes disagree, widely. For a fifth or sixth generation container ship, PIANC WG 33 and BS 6349-4 both give 200 kPa, the Spanish ROM gives 250, the German EAU gives 150 and the Japanese standards give 200 to 290. A design accepted under one code can fail under another for the same ship, so the specification has to name the code it is written to.

The flat-panel pressure model represents cylindrical fenders poorly. Port of Rotterdam design impact speeds, which cause no vessel damage in service, produce calculated hull pressures of 415 to 817 kPa on cylindrical fenders, far above the guidance, with no owner or terminal complaint recorded across 25 years of operation. Most of the load goes into web frames and stiffeners rather than unsupported plating, which the model does not capture.

Panel sizing itself is arithmetic. A fender producing 2,500 kN at its design deflection, against an allowable 200 kPa, needs at least 12.5 m\(^2\) of flat face. Peak pressure then has to be separated from average: pressure is distributed evenly only where the reaction into the panel is symmetrical, and common practice arranges the units so that maximum hull pressure is no more than twice the average.

Fender panels, chains and facing pads

The rubber is the cheapest part of the system to specify and the least likely to fail. The steelwork and the accessories are where fender systems actually give trouble.

Panels. WG 211 sets minimum steel thicknesses of 12 mm for plates exposed on two surfaces, 9 to 10 mm for plates exposed on one surface and 8 mm for internal members. Panel weights run about 200 to 300 kg/m\(^2\) for standard duty, 300 to 400 for heavy duty and above 400 for extreme duty. Edges are chamfered so that a vertical or horizontal load cannot land on the panel perimeter and put destructive shear into a rubber unit that was never designed to take it.

Chains do three separate jobs. Weight chains carry the panel’s dead load, shear chains resist the vertical movement that friction against a rising or falling hull imposes, and tension chains control the panel’s angle through the compression stroke. The working load per assembly follows from the panel weight, the friction coefficient and the fender reaction:

Two things surprise people about chain design. The highest chain loads often occur near half the rated deflection, where the buckling fender is at peak reaction, not at full compression. And a chain assembly should include a cheap, easily replaced weak link, so an overload breaks a shackle that can be changed from the quay rather than tearing a bracket out of the panel. About 2 percent slack in chain length produces about 9 percent droop at the centre, so a specification demanding zero slack is asking for something that is neither achievable nor necessary.

Facing pads are the wearing surface. Ultra-high molecular weight polyethylene sheets bolted to the panel face let the hull slide against plastic rather than steel. Standard pad thicknesses of 30, 40 and 50 mm carry wear allowances of 5, 10 and 15 mm respectively, and the limit is set by the fixing rather than by an absolute residual thickness: the plastic under the washer has to retain enough material to stop the bolt head pulling through. Design friction values are 0.2 or more for polyethylene on steel, 0.3 or more for high-density polyethylene, 0.6 or more for timber and 0.8 or more for rubber.

Fender spacing, bow geometry and multiple contact

A ship’s bow is curved, so the gap between two fender units has to be small enough that the stem cannot reach the structure between them. Bow radius follows from the beam and the bow curvature length:

with \(x/L_{OA}\) taken as about 0.3 for a fine hull below \(C_B\) 0.6, about 0.25 between 0.6 and 0.8, and about 0.2 above 0.8. The spacing limit is then

with \(h\) the compressed fender height and \(C\) the clearance to the quay face. BS 6349-4 adds a separate cap of \(S \leq 0.15 L_S\), where \(L_S\) is the overall length of the shortest ship using the berth. Clearance between hull and structure is usually 5 to 15 percent of the uncompressed fender projection, including the panel and any spacer spools.

Multiple fender contact is a distinct design case under WG 211 rather than a conservative bonus. When two or more units are engaged the energy is shared, but unevenly and unpredictably, because each unit sits at a different point on the hull and compresses by a different amount. WG 211 gives multiple contact its own correction factor, its own partial resistance factor and its own table of reference partial energy factors. Ships with a short parallel mid-body, which describes most modern container ships , and ships with pronounced bow flare both complicate the contact geometry further.

Belted ships are their own case. Ro-ro vessels, ropax ferries and some small tankers carry a belting that takes the load, which is why the hull pressure guide reads “not applicable” for them: the force goes into the belt, not the plating. An arch fender whose maximum deflection is smaller than the belting height will contact the hull twice, above and below the belt, and produce line loads the flat-panel model does not describe.

Breasting and mooring dolphins

A breasting dolphin is a piled structure standing clear of the shore that carries the fender panel, gives the ship something to push against, and transfers the horizontal load into a pile group. The design load is the fender reaction, so a system delivering 3,500 kN at full deflection sets 3,500 kN of horizontal shear into the piles.

Pile groups are usually battered in opposing pairs so that the horizontal component is carried mostly as axial compression in one set and axial tension in the other, rather than as bending. A vertical-only group would need very large diameters to resist the same load in bending alone. Deck elevation is set from the loaded ship’s waterline plus the panel height plus a working margin, with the light ship riding higher but arriving with less energy.

Mooring dolphins are separate, smaller structures placed forward and aft of the breasting line. They carry the bollards or quick release hooks and are not designed for berthing impact. Separating the two functions means the berthing shock does not pass through the mooring hardware, and each structure can be positioned where its own job requires: the fender where the parallel mid-body will be, the lines where the lead angles work.

WG 211’s structural doctrine matters here. By putting the safety margin into the rubber rather than the supporting structure, it requires the marine structure to be designed as a whole, so the dolphin, the fender and the accessories are one calculation rather than three.

Tidal range and contact geometry

A fender panel has to stay against the hull across the full tidal range. At a berth with a 6 m spring range the contact point on the hull moves 6 m over about 12.4 hours, and a panel only 3 m tall will be out of contact at some stage of the tide, concentrating load on the panel frame instead of the rubber.

The design responses are a tall panel spanning the range, stacked units with overlapping vertical coverage, or floating fenders that ride the tide. Contact geometry is checked at the extreme water levels the site produces rather than at mean level only, and it is checked against the ship’s parallel mid-body at each of them, because a panel that is on the flat of the hull at high water can be on the curve at low.

Pneumatic and foam units solve the problem differently, by floating. That is their main advantage over fixed rubber at a high-range berth, and it is bought with a maintenance obligation.

What OCIMF MEG4 governs, and what it does not

MEG4 is a mooring publication. The Mooring Equipment Guidelines, Fourth Edition, published by OCIMF in June 2018, covers mooring forces and environmental criteria, mooring arrangements and layouts, lines and tails, fittings, berth design and fittings, alternative mooring technology, and the ship-shore interface. It sets no fender specification, no berthing velocity and no berth tier by deadweight. Fender energy and approach velocity belong to PIANC and BS 6349-4, and a claim attributed to MEG4 in either of those areas is misattributed.

What MEG4 does supply to a berth designer is the mooring side of the same structure. Its standard environmental criteria are 60 knots of wind from any direction combined with either 3 knots of current from ahead or astern or 0.75 knots on the beam, assessed across current directions from 0 to 360 degrees and in both loading conditions, for vessels of 16,000 DWT and above. They are deliberately arbitrary benchmarks against which a ship’s mooring system is sized, not conditions measured anywhere.

The asymmetry between 3 knots ahead and 0.75 knots abeam is not arbitrary. Transverse current acts on the length between perpendiculars times the draught, which for a laden VLCC is around 7,000 m\(^2\), so a beam current produces thousands of kilonewtons where the same speed from ahead produces a fraction of that. The full force calculation, the drag coefficients and the line-load arithmetic belong to mooring forces and station keeping and are not repeated here.

Berth environmental operating limits are established per berth, from the mooring evaluation, the fender capacity, tug availability and the loading arm envelope. There is no industry-wide Beaufort cut-off, and a limit quoted as a Beaufort number should be read against the actual scale: force 6 is 10.8 to 13.8 m/s and force 8 is 17.2 to 20.7 m/s.

Two pieces of long-standing berth lore are now wrong and worth correcting explicitly. Emergency towing-off pennants are not recommended by OCIMF. A study found no documented case of one being used since 1967, against roughly 1,700 injuries incurred handling the heavy wires, and a 2007 risk assessment concluded they should no longer be recommended; ISGOTT requires a terminal that still wants them to justify it by risk assessment. Where one is rigged it goes on the ship’s offshore side, made fast to the ship’s own bitts with the eye just above the water, and it is not secured to a shore hook. It is also a different thing from a SOLAS emergency towing arrangement . And snap-back zones are no longer painted on deck: MEG4 and the Code of Safe Working Practices both dissuade it, because a parted line’s trajectory cannot be bounded reliably and paint creates false confidence outside the marked area. The whole mooring deck is the danger area, and IMO now frames it that way.

The regulatory hook on the ship side is SOLAS regulation II-1/3-8 , replaced by resolution MSC.474(102) adopted on 11 November 2020 and in force from 1 January 2024. Shipboard fittings on ships constructed on or after that date are designed for 1.15 times the ship design minimum breaking load under MSC.1/Circ.1175/Rev.1; ships constructed on or after 1 January 2007 and before 1 January 2024 carry the earlier 1.25 times the line breaking strength. The ship design minimum breaking load and the equipment sized against it are covered in marine mooring equipment and winches .

The berthing manoeuvre

The calculation assumes a controlled arrival, and the controlled arrival is a piece of work in its own right.

The pilot boards at the port’s designated boarding ground and the master-pilot exchange settles the plan: the side to be alongside, the swing, the tug plan and where each tug makes fast, the mooring pattern the terminal expects, the abort point, and what happens if a tug or the main engine fails. That exchange is also the document a casualty investigation reads first.

The approach runs through water that is behaving differently from open sea. Bank effect pushes the bow away from and sucks the stern toward a channel side, squat reduces the water under the keel as a function of speed, and the under-keel clearance in the berth pocket may be less than in the channel that led to it. Under-keel clearance feeds straight back into the added mass coefficient, so the depth at the berth is a fendering input and not just a navigational one.

Control authority changes as speed comes off. A bow thruster loses most of its useful side force with headway, because the transverse jet is deflected aft along the hull and the pressure field it induces over the after body cancels much of the thrust; above roughly 3 to 4 knots through the water the ship depends on tugs for lateral control. Tug type matters at that point: a tractor tug , a rotortug or an azimuth-drive unit can apply force in directions a conventional screw tug cannot, and the assisted ship’s speed is the variable that determines whether the tug is working safely or is at risk of girting .

Lines then go out in an order that has a reason behind it: ordinarily a spring first to stop movement along the berth, then breast lines to hold the ship in, then head and stern lines to control the ends, adjusted for the direction of the set. At an oil or gas terminal the ship-shore safety checklist and the pre-transfer conference follow before anything is connected.

Berthing aids and monitored berths

A berthing aid system measures the distance from the berth face and the closing velocity at two points along the ship, usually by laser or Doppler, and displays them to the pilot on the bridge and to the terminal on a board. It replaces a judgement of closing speed made by eye at distances where the eye is unreliable.

The measured effect is specific and it is not what most operators expect. Establishing a target berthing velocity decreases extreme berthing events but does not necessarily decrease velocity during normal berthing. The system cuts the tail, not the median. That is exactly the right outcome for a fender designer, because the tail is what the design case is drawn from, and WG 211 gives it a concrete consequence: a separate and lower reference partial energy factor for monitored berths.

Smart fender instrumentation extends the same idea to the rubber itself, capturing berthing speed, compression percentage, over-compression events and the number of compressions, which turns the inspection regime from a calendar exercise into a condition-based one and feeds the next berth’s design data.

Inspection, wear limits and replacement

Fender systems fail through fatigue cracking from repeated compression, ozone attack on exposed surfaces, corrosion of the steelwork and accessories, and mechanical damage from hard contacts and ice.

Source: PIANC WG 211 section 11.3 for the inspection regime; ISO 17357-1:2014 for pneumatic manufacture and testing.

Acceptance limits for cracking and wear are left to the supplier, and WG 211 catalogues the common failure modes rather than fixing numerical criteria, because they differ by compound and geometry.

Two counterintuitive points close this out. Holding spare rubber units is not recommended, because rubber ages in storage and a unit held for years may not meet its certificate when fitted; the recommended practice at a facility with many systems is to replace a damaged unit with an end unit, and to hold spares only for the things that wear or get knocked off. And fatigue can govern instead of impact. Where constant swell or frequent gusty wind works the moored ship against the fender, WG 211 recommends keeping continuous load to 40 to 50 percent of the design load, equivalent to 5 to 10 percent deflection, and keeping cyclic deflection below the buckling limit at the peak of the performance curve. In a protected port that analysis is usually not required; at an exposed berth or a high-frequency ferry berth it can be the binding constraint.

Ship-to-ship transfer and specialised berths

For a transfer between two free-floating hulls the mass term uses both vessels and the velocity is a closing or relative approach velocity rather than a speed toward a fixed structure. Fendering is pneumatic, suspended between the hulls, sized on the combined mass. The division of labour between the guidance documents is clean: OCIMF covers offshore ship-to-ship, and WG 211 covers nearshore transfers and floating storage and regasification units. The regulatory frame for the operation itself is MARPOL Annex I regulation 41 , which requires an approved STS Operations Plan on board.

Gas carrier jetties add their own constraints. Manifold position drives the fender line, the emergency shutdown system may release the mooring hooks as well as stopping cargo, and the loading arm envelope sets a tighter limit on ship movement than the mooring system alone would.

Ferry and ro-ro berths invert the usual design driver. A ferry berth may take several thousand contacts a year against a crude oil jetty’s few dozen, so fatigue rather than single-contact energy governs, end berthing puts the eccentricity factor back at 1.00, and the linkspan geometry constrains the approach. Ice-affected berths change the fender type decision, and floating berths and pontoons carry their own arrangements.

When the ship or the berth is damaged

Hard contacts happen, and the consequences are structural, regulatory and contractual at once.

On the ship, excessive local pressure at the panel dents or buckles shell plating. That is damage affecting classification, so it goes to the class society , gets measured, and may need thickness measurement and repair before the ship trades on. Underwater components of the fender system are inspected in the same way a hull is, and an in-water survey can cover part of it.

On the commercial side, liability depends on the governing law and the charter form. Under United States federal maritime law the safe-berth clause in the ASBATANKVOY form is a warranty of safety, imposing liability without fault on the charterer, decided in CITGO Asphalt Refining Co. v. Frescati Shipping Co. on 30 March 2020 by 7 votes to 2. English law treats the safe-port undertaking as a duty of due diligence in the nomination instead. The two positions are set out in safe port and safe berth warranties and the United States safe berth warranty , and the applicable law clause decides which one a claim runs under. Damage done alongside by shore equipment rather than by the ship is a different route again, through the stevedore damage provisions.

Prevention is cheaper than any of that, and it happens before arrival, in the terminal questionnaire and berth fit exchange where the caller’s dimensions, manifold position and hull pressure tolerance are checked against what the berth actually has.

Limitations

The kinetic energy method is an engineering tool, not a simulation of an impact, and several of its assumptions are known to be approximations.

It treats the ship as a rigid body and the fender as a lumped spring. Hull flexibility is not modelled, and neither is the distribution of load into frames and stiffeners, which is why calculated hull pressures on cylindrical fenders read far above the guidance while service records show no damage.

The hull pressure criteria are not derived from ship structural rules. No IACS Unified Requirement governs allowable hull pressure from fender contact, and the values in every national code trace back to PIANC work of 1984. They are guidance carried forward by convention, and the codes disagree with each other by up to a factor of two for the same ship. Meanwhile ships have grown and the published allowable pressures have moved downward, so the margin implied by the tables has changed without the tables saying so.

Velocity is the dominant uncertainty and the historical basis for it has been undermined. The size-velocity relationship that the classic curves encode was not confirmed by 555 measured berthings in Rotterdam, and what did correlate was berthing policy. A berth with no measurement and no stated target velocity is being designed against a distribution nobody has observed at that berth.

The eccentricity factor assumes a contact geometry that an oblique approach breaks. Where a ship comes in at 15 to 20 degrees, the geometry of the contact point relative to the centre of mass changes during the approach itself and the simple expression becomes approximate. WG 211 recommends dynamic simulation where oblique approaches are common.

Single-contact sizing does not describe an exposed berth. In moderate swell a ship may contact, rebound and contact again within tens of seconds, and each cycle adds strain to the rubber. Fatigue life rather than single-contact capacity can then be the governing parameter, and the guidance on fatigue limits is explicitly provisional in the absence of manufacturer data.

Some of the WG 211 numerical tables are not reproduced here. The report is a paid publication, and the numerical contents of its hull pressure and reaction-force table, its typical fender types table, and its partial factor tables were not read in preparing this article. The structure of each is described and the report cited; no value is stated from them. Where a design depends on those cells, the report itself is the source, not a secondary summary.

Related Articles

• Ports and Terminals Overview : the cluster hub that maps the full port call, from arrival through berthing to cargo and departure.
• Mooring Forces and Station Keeping : wind, current and passing-ship loads on the moored ship, the OCIMF coefficient method and the line-tension arithmetic.
• Marine Mooring Equipment and Winches : the ship-side hardware, the ship design minimum breaking load and the winch brake render setting.
• Pilotage Operations : the pilot’s conduct of the approach, the boarding ground and the master-pilot exchange.
• Tug Operations and Bollard Pull : the tug force that controls the transverse approach velocity at the berth.
• Terminal Questionnaire and Berth Fit : the compatibility check that settles whether a caller suits the berth before it arrives.
• World Port Profiles : the per-port channel depth, tidal range and maximum design vessel that fender systems are sized against.
• Container and Bulk Terminals : the continuous quay and the fender line that runs along it.
• Port Dues and Disbursements : the towage, pilotage and berth charges the call attracts on the port bill.
• Safe Port and Safe Berth Warranties : the charterer’s undertaking and what makes a berth unsafe in fact.
• Canals and Straits : the chokepoints and approach waterways that feed the berths these fenders protect.
• Vessel Traffic Services : the traffic organization that schedules the approach and the berthing window.