2.1 Definition of the measures of helideck motion and wind severity by reference to the ROS

Part A provided a complete framework by which the ROS of an unconstrained helicopter on a moving helideck can be calculated based on the most salient governing parameters.

To recap, the ROS has been defined in terms of the ratio of destabilising (M_D) versus restoring (M_R) moments:

(1) \begin{align}{\rm{ROS}} = 1 - \frac{{( - {{\rm{M}}_{\rm{D}}})}}{{{{\rm{M}}_{\rm{R}}}}}\end{align}

It is equal to zero at the point of failure (tipping and/or sliding) and equal to 1 (or equivalently, 100%) when no destabilising forces act on the helicopter.

It has been proven analytically that the reduction in ROS caused by helideck motion is a function of the total helideck accelerations in all three directions, the helicopter undercarriage dimensions and the location of the helicopter’s centre of gravity.

It is useful to group these parameters into two main governing measures — the Measure of Motion Severity (MMS), and a helicopter orientation factor, O_f.

The MMS has been defined as the ratio of total acceleration (gravitational and inertial) component resolved in the plane of the helideck, a_h, divided by the component normal to the deck, a_z:

(2) \begin{align}{\rm{MMS = }}\frac{{{{\rm{a}}_{\rm{h}}}}}{{{{\rm{a}}_{\rm{z}}}}} = \frac{{\sqrt {{{\left( {{{\rm{g}}_{\rm{x}}} - {{\rm{x}}_{{\rm{ddot}}}}} \right)}^2} + {{\left( {{{\rm{g}}_{\rm{y}}} - {{\rm{y}}_{{\rm{ddot}}}}} \right)}^2}} }}{{{{\rm{g}}_{\rm{z}}} - {{\rm{z}}_{{\rm{ddot}}}}}}\end{align}

where g_x, g_y and g_z are the components of the gravitational acceleration g in the plane of the helideck (longitudinal, x, lateral, y) and normal to the deck, z (axis pointing downwards). The ddot suffix denotes the respective inertial accelerations due to the motion of the helideck (surge, sway and heave).

Another very useful way of interpreting the MMS is as the tangent of the effective helideck inclination; this is the static helideck inclination that would produce the same destabilising effect as the moving helideck. The effective inclination is larger than the actual inclination of a moving helideck measured relative to earth axes, since it also includes the effect of the inertial accelerations acting on the helicopter due to the motion of the helideck. Conversely, in the absence of any helideck motion, the effective inclination is equal to the actual inclination.

The orientation factor, O_f, represents the relative destabilising effect of the longitudinal and lateral helideck accelerations as experienced by the helicopter with respect to each particular failure axis (tipping and/or sliding). This depends on the orientation of the helicopter axes relative to the helideck axes (angle Θ_h), and the magnitude of the orientation factor is different for each failure mode.

In addition to helideck motion, a number of external forces, F_i, and moments, M_j act to destabilise the helicopter. It was shown in Part A that the ROS for tipping failure can then be expressed simply in the following general form:

(3) \begin{align}{\rm{ROS = 1}} - {\rm{MMS}} \cdot {{\rm{O}}_{\rm{f}}}({{\rm{\uptheta }}_{\rm{h}}}) \cdot {{\rm{f}}_{{\rm{grav}}}} - \sum\limits_{\rm{i}} {\left( {\frac{{{{{{\vec {\rm F}}}}_{\rm{i}}} \cdot {{\rm{f}}_{\rm{i}}}}}{{{\rm{m}} \cdot {{\rm{a}}_{\rm{z}}}}} \cdot {{\rm{f}}_{{\rm{grav}}}}} \right)} - \sum\limits_{\rm{j}} {\left( {\frac{{{{{{\vec {\rm M}}}}_{\rm{j}}} \cdot {{\rm{f}}_{\rm{m}}}}}{{{\rm{m}} \cdot {{\rm{a}}_{\rm{z}}}}} \cdot {{\rm{f}}_{{\rm{grav}}}}} \right)}\end{align}

where m is the mass of the helicopter, a_z is the total vertical acceleration. f_grav and matrices f_i and fmcontain purely geometrical helicopter factors reflecting the undercarriage geometry and the point of action of each force/moment at any given location, CFX (longitudinal distance from front wheel), CFY (lateral offset relative to centreline), and CFZ (height above the helideck).

The above expression makes it evident that the destabilising effect of any additional forces and moment depends on the product of m. a_z, which is the effective weight of the helicopter, which acts to stabilise the helicopter. Logically, it then follows that a_z is an additional helideck-dependent parameter that needs to be quantified. Although a_z is part of the MMS (=a_h/a_z), knowing only the MMS is not sufficient; for example, it is possible to have two helidecks with identical MMS but with one that has a higher a_z than the other. Although the destabilising effect of the helideck motion per se will be the same for both, the destabilising effect of any additional external forces will be less for the helideck with the higher a_z (since it corresponds to a larger, more stabilizing, effective weight).

The wind is another important helideck environmental factor that destabilises a helicopter since it generates aerodynamic forces in addition to the gravitational and inertial forces, namely fuselage drag and main and tail rotor forces. These forces are mainly governed by the time-varying, instantaneous wind speed, which can be decomposed into a mean and time varying (turbulence) component uʹ:

(4) \begin{align}{\rm{U(t)}} = {{\rm{U}}_{{\rm{mean}}}}({\rm{t}}) + {\rm{u'}}({\rm{t}})\end{align}

The wind direction relative to the helicopter is also a significant factor; the fuselage drag is sensitive to the wind direction relative to the fuselage, β, and the main rotor lift is sensitive to the wind angle-of-attack relative to the rotor disc, α_s.

2.2 Definition of the limits curve

Using Equation 3 and by expressing the aerodynamic forces acting on the helicopter as a function of U, it is possible to calculate the ROS analytically as a function of MMS and U. These analytical expressions allow the time-series of ROS(t) to be calculated for any given helicopter, based on MMS(t) and U(t). They also allow the equation of the limiting curve to be derived, i.e. the threshold separating safe from unsafe operating conditions.

The safe operational envelope can be defined in terms of the main two parameters, the MMS and the instantaneous wind speed U, as illustrated in Fig. 1.

Figure 1.

Illustration of a limiting curve as a function of MMS (t) and U(t).

Fundamentally, the safe operational envelope represents the conditions for which the helicopter ROS is greater than zero (or greater than some minimum acceptable safety factor, ROS_min), across all possible modes of failure (tipping and/or sliding).

Using the analytical equations derived in Part A as a function of MMS and U and setting ROS to zero (or more generally, to ROS_min), allows the (minimum) MMS value at which failure will occur (MMS_crit) to be calculated for any given wind speed and for any given set of assumptions for all other parameters (helicopter and helideck specific, that correspond to any given on-deck operation).

Separate equations will apply to each mode of failure (whether tipping or sliding), and the MMS_crit value at which failure will occur will be different for each mode; the mode with the lowest MMS_crit will be the mode of failure that will occur in practice. This is illustrated in Fig. 2. The analytical equations for the limits curves for all modes of failure have been derived and are presented in Section 3.

Figure 2.

Illustration of the overall limiting line for a single set of helicopter/helideck operational conditions, as a function of MMS (t) and U(t).

Also, different limiting curves can be calculated based on different sets of helicopter and helideck specific parameters, e.g. depending on helicopter type, mass and centre of gravity (CoG) location, updraft characteristics at each individual helideck.

The on-deck time history of any given landing can be plotted on the MMS vs wind speed plot, from the time of landing t_land onwards, as shown in Fig. 2. If the trace remains below the most onerous limiting curve (overall limiting line), the helicopter operation remains safe.

3.1 Tipping failure limits equation

Using the analytical expression for calculating the ROS for tipping failure relative to axis NS (rollover towards starboard side) derived in Part A and setting ROS = 0 (or to a minimum safety margin of ROS_min), the expression for the minimum MMS for failure, MMS_crit, is equal to:

(5) \begin{align}{\rm{MM}}{{\rm{S}}_{{\rm{crit}}}} = {\rm{MM}}{{\rm{S}}_{{\rm{max}}}} - \frac{1}{{{{\rm{O}}_{\rm{fTIP}}}}}\sum\limits_{\rm{i}} {\left( {\frac{{{{\vec{\rm F}}_{\rm{i}}} \cdot {{\rm{f}}_{\rm{i}}}}}{{{\rm{m}} \cdot {{\rm{a}}_{\rm{z}}}}}} \right)} - \frac{1}{{{{\rm{O}}_{\rm{fTIP}}}}}\sum\limits_{\rm{j}} {\left( {\frac{{{{{{\vec {\rm M}}}}_{\rm{j}}} \cdot {{\rm{f}}_{\rm{m}}}}}{{{\rm{m}} \cdot {{\rm{a}}_{\rm{z}}}}}} \right)}\end{align}

where MMS_max is equal to:

(6) \begin{align}{\rm{MM}}{{\rm{S}}_{{\rm{max}}}} = \frac{{1 - {\rm{RO}}{{\rm{S}}_{{\rm{min}}}}}}{{{{\rm{O}}_{{\rm{fTIP}}}}}} \cdot \left( {\frac{{{\rm{LY}}}}{{\rm{L}}} \cdot \frac{{{\rm{CGX}}}}{{{\rm{CGZ}}}} - \frac{{{\rm{FR}}}}{{\rm{L}}} \cdot \frac{{{\rm{CGY}}}}{{{\rm{CGZ}}}}} \right)\end{align}

MMS_max is the maximum possible helideck motion that a helicopter can withstand (i.e. when no other external destabilising forces act to reduce the ROS). Any additional force/moment in a destabilising direction will reduce the limiting value of MMS_crit, as represented by the minus sign of the external force and moment terms.

3.2 The orientation factor

The orientation factor is equal to:

(7) \begin{align}{{\rm{O}}_{{\rm{fTIP}}}} = \frac{{{\rm{FR}}}}{{\rm{L}}} \cdot {\rm{cos}}\,{{\rm{\uptheta }}_{\rm{h}}} + \frac{{{\rm{LY}}}}{{\rm{L}}} \cdot \sin\,{{\rm{\uptheta }}_{\rm{h}}}\end{align}

and the helicopter geometry matrix factors are:

(8) \begin{align}{{\rm{f}}_{\rm{i}}} = \left( {\begin{array}{c}{\frac{{{\rm{CF}}{{\rm{Z}}_{\rm{i}}}}}{{{\rm{CGZ}}}} \cdot \frac{{{\rm{LY}}}}{{\rm{L}}}}\\ \\[-5pt] {\frac{{{\rm{CF}}{{\rm{Z}}_{\rm{i}}}}}{{{\rm{CGZ}}}} \cdot \frac{{{\rm{FR}}}}{{\rm{L}}}}\\ \\[-5pt] {\frac{{{\rm{FR}}}}{{\rm{L}}} \cdot \frac{{{\rm{CF}}{{\rm{Y}}_{\rm{i}}}}}{{{\rm{CGZ}}}} - \frac{{{\rm{LY}}}}{{\rm{L}}}\frac{{{\rm{CF}}{{\rm{X}}_{\rm{i}}}}}{{{\rm{CGF}}}}}\end{array}} \right)\quad \quad {{\rm{f}}_{\rm{m}}} = \left( {\begin{array}{c}{\frac{{{\rm{FR}}}}{{\rm{L}}} \cdot \frac{1}{{{\rm{CGZ}}}}}\\ \\[-5pt] {\frac{{ - {\rm{LY}}}}{{\rm{L}}} \cdot \frac{1}{{{\rm{CGZ}}}}}\\ \\[-5pt] {\rm{0}}\end{array}} \right)\end{align}

as derived previously for the ROS expressions in Part A.

3.3 Sliding failure limits equation

The ROS equations for the sliding failure modes are much more complex than those for tipping failure because the restoring moment of the frictional forces is implicitly related to the destabilising forces and moments. Once ROS is set to 0 (or to a minimum safety margin of ROS_min) and after some algebraic manipulation, it is possible to reduce the equations into the same general form as that for the tipping failure mode (Equation 5), with terms corresponding to the MMS_max, O_f, and geometric factors f_i and f_m, all of which also include the coefficient of helideck surface friction, µ.

These factors are different for each of the sliding failure modes (about the nose wheel, N, and about either of the main wheels, S or P), as given below. The failure mode that will occur first will be that with the lowest MMS_crit.

3.3.1 Sliding about nose wheel (N)

The MMS_crit terms (as per Equation 5) for sliding failure about the nose wheel (N) are:

(9) \begin{align}{\rm{MM}}{{\rm{S}}_{{\rm{max}}{\rm{.N}}}} = \frac{1}{{{{\rm{O}}_{{\rm{fN}}}}}} \cdot \frac{{{\rm{CGX}}}}{{{\rm{CGZ}}}} \cdot {\rm{\mu }}\end{align}

where O_fN the orientation factor for sliding about wheel N is equal to:

(10) \begin{align}{{\rm{O}}_{{\rm{fN}}}} = \frac{{{\rm{CGX}}}}{{{\rm{CGZ}}}} \cdot \frac{{{\rm{FR}}}}{{\rm{L}}} \cdot \cos \,{{\rm{\theta }}_{\rm{h}}} + \left( {{\rm{\mu + }}\frac{{{\rm{CGY}}}}{{{\rm{CGZ}}}} \cdot \frac{{{\rm{FR}}}}{{\rm{L}}}} \right) \cdot \sin \,{{\rm{\theta }}_{\rm{h}}}\end{align}

and the corresponding helicopter geometry matrix factors are:

(11) \begin{align}{{\rm{f}}_{{\rm{Ni}}}} = \left( {\begin{array}{c}{{\rm{\mu }} \cdot \frac{{{\rm{CF}}{{\rm{Z}}_{\rm{i}}}}}{{{\rm{CGZ}}}} + \frac{{{\rm{CF}}{{\rm{Y}}_{\rm{i}}}}}{{{\rm{CGZ}}}} \cdot \frac{{{\rm{FR}}}}{{\rm{L}}}}\\ \\[-5pt] {\frac{{{\rm{CF}}{{\rm{X}}_{\rm{i}}}}}{{{\rm{CGZ}}}} \cdot \frac{{{\rm{FR}}}}{{\rm{L}}}}\\ \\[-5pt] { - {\rm{\mu }} \cdot \frac{{{\rm{CF}}{{\rm{X}}_{\rm{i}}}}}{{{\rm{CG}}{{\rm{Z}}_{\rm{i}}}}}}\end{array}} \right)\quad \quad {\rm{f}}{{\rm{m}}_{\rm{N}}} = \left( {\begin{array}{c}0\\ \\[-5pt] { - {\rm{\mu }} \cdot \frac{{\rm{1}}}{{{\rm{CGZ}}}}}\\ \\[-5pt] { - \frac{{{\rm{FR}}}}{{\rm{L}}}\frac{1}{{{\rm{CGZ}}}}}\end{array}} \right)\end{align}

This is for an anticlockwise rotation, consistent with forces causing tipping relative to NS.

To include the effect of a safety margin ROS_min, the above expressions can be adapted simply by replacing μ by a reduced effective value of μ. (1-ROS_min).

3.3.2 Sliding about main wheel S

To include the special case that the nose wheel is free to castor, the coefficient of friction at the nose wheel has been assumed equal to α·µ, i.e. a fraction α of the coefficient at the two other wheels. When the wheel is unlocked and free to castor α = 0; otherwise, if locked, it is assumed that α = 1.

The MMS_crit terms (Equation 5) for sliding about the nose wheel (S) are:

(12) \begin{align}{\rm{MM}}{{\rm{S}}_{{\rm{max}}{\rm{.S}}}} = \frac{1}{{{{\rm{O}}_{{\rm{fS}}}}}} \cdot \left[ {\left[ {\left( {\alpha \cdot \frac{{{\rm{FR}} - {\rm{CGX}}}}{{{\rm{CGZ}}}} + \frac{{{\rm{LY}}}}{{\rm{L}}} \cdot \frac{{{\rm{CGX}}}}{{{\rm{CGZ}}}}} \right) - \frac{{{\rm{FR}}}}{{\rm{L}}} \cdot \frac{{{\rm{CGY}}}}{{{\rm{CGZ}}}}} \right] \cdot {\rm{\mu }}} \right]\end{align}

where O_fS is an orientation factor equal to:

(13) \begin{align}\frac{{{\rm{FR}}}}{{\rm{L}}} \cdot \left( {\frac{{{\rm{FR}} - {\rm{CGX}}}}{{{\rm{CGZ}}}} + {\rm{\mu }}} \right) \cdot \cos \,{{\rm{\theta }}_{\rm{h}}} + \left[ {\frac{{{\rm{FR}}}}{{\rm{L}}} \cdot \frac{{({\rm{LY}} - {\rm{CGY}})}}{{{\rm{CGZ}}}} - {\rm{\mu }} \cdot \left( {\alpha - \frac{{{\rm{LY}}}}{{\rm{L}}}} \right)} \right] \cdot \sin \,{{\rm{\theta }}_{\rm{h}}}\end{align}

The geometrical matrix factors for a clockwise rotation (consistent with forces causing tipping relative to NS), are:

(14) \begin{align}{{\rm{f}}_{{\rm{Si}}}} = \left[ {\begin{array}{c}{\frac{{{\rm{FR}}}}{{\rm{L}}} \cdot \frac{{({\rm{LY}} - {\rm{CF}}{{\rm{Y}}_{\rm{i}}})}}{{{\rm{CGZ}}}} - {\rm{\mu }} \cdot \left( {\alpha - \frac{{{\rm{LY}}}}{{\rm{L}}}} \right) \cdot \frac{{{\rm{CF}}{{\rm{Z}}_{\rm{i}}}}}{{{\rm{CGZ}}}}}\\ \\[-5pt] {\frac{{{\rm{FR}}}}{{\rm{L}}} \cdot \left[ {\frac{{({\rm{FR}} - {\rm{CF}}{{\rm{X}}_{\rm{i}}})}}{{{\rm{CGZ}}}} + {\rm{\mu }} \cdot \frac{{{\rm{CF}}{{\rm{Z}}_{\rm{i}}}}}{{{\rm{CGZ}}}}} \right]}\\ \\[-5pt] { - {\rm{\mu }} \cdot \left[ {\alpha \cdot \frac{{{\rm{(FR}} - {\rm{CF}}{{\rm{X}}_{\rm{i}}}{\rm{)}}}}{{{\rm{CGZ}}}} + \frac{{{\rm{LY}}}}{{\rm{L}}} \cdot \frac{{{\rm{CF}}{{\rm{X}}_{\rm{i}}}}}{{{\rm{CGZ}}}}} \right] + {\rm{\mu }} \cdot \frac{{{\rm{CF}}{{\rm{Y}}_{\rm{i}}}}}{{{\rm{CGZ}}}} \cdot \frac{{{\rm{FR}}}}{{\rm{L}}}}\end{array}} \right]\quad \quad {\rm{f}}{{\rm{m}}_{\rm{S}}} = \left[ {\begin{array}{c}{{\rm{\mu }} \cdot \frac{{{\rm{FR}}}}{{\rm{L}}}\frac{1}{{{\rm{CGZ}}}}}\\ \\[-5pt] {\left( {\alpha - \frac{{{\rm{LY}}}}{{\rm{L}}}} \right) \cdot {\rm{\mu }} \cdot \frac{1}{{{\rm{CGZ}}}}}\\ \\[-5pt] {\frac{{{\rm{FR}}}}{{\rm{L}}} \cdot \frac{1}{{{\rm{CGZ}}}}}\end{array}} \right]\end{align}

To include the effect of a safety margin ROS_min, the above expressions can, again, be adapted simply by replacing μ with a reduced effective value of μ. (1-ROS_min).

4.1 Background

Currently in the UK, the 20-minute maximum values of helideck motion parameters such as static roll (R) and pitch (P), total inclination (INC), and heave rate (HR) are used as the limiting helideck motion parameters for offshore operations.

The limiting R/P/INC/HR values are jointly set by the helicopter operators via the Helideck Certification Agency (HCA). They are dependent on vessel type as well as helicopter type, and on a number of other factors, as published in the Helideck Limitations List (HLL)⁽¹⁾.

The intention is to ensure that conditions at the time of touchdown are such that they, a) allow pilots to land safely and, b) do not compromise on-deck stability during the entire landing duration. The limiting values are based on service experience and pilot judgement and have generally proven to be satisfactory in terms of the touchdown itself.

However, the governing parameters affecting touchdown and on-deck stability are fundamentally different, and it is much more difficult for pilots to assess on-deck stability than it is to judge conditions at touchdown.

While the maximum helideck angles of R, P and INC are correlated with the MMS, none of them capture directly the dynamic effect of the helideck accelerations. As mentioned above, the MMS is equal to the tangent of the total ‘effective’ helideck inclination, which includes the static component (represented by INC) but also the dynamic component due to the acceleration of the helideck. The helideck accelerations are correlated with the maxima of the helideck angles and the heave rate, but also depend on the period, T, of the helideck motion, i.e. how fast the helideck motion alternates between maxima. The dynamic component of the MMS is significant, and typically of the same order as INC. It is evident, therefore, that the MMS is more suited as an on-deck stability limiting parameter than any of the measures currently used to regulate operations.

In addition, wind is not currently a factor limiting the decision to land. Based on the analysis and the evidence presented in Part A (including data from helicopter trials, and the West Navion/GBKZE accident investigation findings), this is a significant omission in the current scheme. In fact, even in the absence of any motion (i.e. for fixed helidecks), the wind alone can destabilise a helicopter. In addition to wind speed, wind direction relative to the helicopter can have a large effect on helicopter stability (as demonstrated in Part A), and therefore needs to be monitored and controlled.

For any given helideck, the MMS will depend on the response of the vessel to the waves, currents, and the wind. More onerous helideck motions are expected to co-exist with windier conditions, but this correlation will clearly not be the same in all situations and for all vessels as the response of vessels to a sea state and the wind will vary.

4.2 Definition of the MSI, WSI and wind direction limiting parameters

The MMS and instantaneous wind speed, U, are continuously time-varying so, as for the R/P/INC/HR limits, it is necessary to, a) define a single number for each of these parameters to characterise the expected conditions on the helideck over the 20 minute on-deck duration and, b) base this on measurements taken prior to landing. These parameters are the Motion Severity Index (MSI) and the Wind Severity Index (WSI).

There are a number of different ways in which these indices can be defined. For the MSI, one approach is to simply use the MMS as the basis and, with reference to the existing convention used for R/P/INC/HR adopt the 20-minute maximum measured prior to landing. However, as a forecasting measure, this is not necessarily the most representative measure of conditions over the following 20 minutes, especially if the sea state and wind conditions are changing and cannot be assumed to be statistically stationary. This is discussed further in Section 5.1

In terms of the definition of the WSI, the maximum instantaneous wind speed can be represented by the mean plus the likely maximum fluctuation (or gust). A forecasting element could also be added to represent changing trends in the wind. This is discussed further in Section 5.2.

In addition to the MSI/WSI, the wind direction also needs to be taken into account. This presents some additional difficulties since there are two main governing parameters to consider: a) the wind direction relative to the helicopter in the plane of the helideck (angle β), which is the main parameter affecting wind drag; and b) the angle-of-attack of the wind relative to the main rotor disc (α_s, as defined in Part A), since this has a significant effect on the main rotor lift generated at Minimum Pitch On Ground (MPOG).

Angle β is a wind parameter defined relative to the helicopter (rather than relative to earth axes or relative to the helideck/vessel). It was therefore named the Relative Wind Direction (RWD) and was identified as an additional important parameter affecting on-deck safety. It is equal to zero for a head-on wind direction, and equal to ±90° for a beam-on wind direction (the sign representing a wind towards starboard or portside, respectively). Importantly, the RWD depends on the orientation of the helicopter on the helideck after touchdown, so this cannot be forecast prior to landing.

The orientation of the helicopter at the point of landing will usually be chosen by the pilot with reference to the wind direction at the time of landing, but not always; an out-of-wind orientation may be chosen in some situations, e.g. to avoid placing the tail rotor over or close to a helideck access point. However, after touchdown the orientation of the helicopter relative to the helideck will remain fixed unless it takes off to reposition or, unintentionally, if it slides relative to the helideck.

The RWD will nonetheless vary continually during the time the helicopter remains on a helideck due to the variability in the ambient wind direction. This variation will be more marked during certain meteorological conditions such as the passing of a squall. Furthermore, any variations in the vessel heading after touchdown will also affect the RWD. These are expected to be small for vessels that use dynamic positioning (DP) systems, provided that no intentional changes to heading are allowed during helicopter operations. However, for vessels that weathervane with the wind the vessel heading will be more variable, although this will usually be correlated with the oncoming wind direction and self-correct to some degree.

In terms of the wind angle-of-attack α_s, this is not only a function of the incident wind direction, but also of a) helideck motion in roll and pitch, b) any updrafts or downdrafts generated by the interaction of the wind with the vessel structures surrounding the helideck, and c) the response of the main rotor disc to the wind. Quantifying the wind angle-of-attack α_s is very difficult, even during dedicated helicopter and helideck instrumented trials. Instead, the destabilising effect of α_s has to be estimated indirectly, based on other parameters that can be measured (e.g. RWD, R, P) and/or tailored for each helideck environment (e.g. updraft/downdrafts expected at each helideck as a function of wind direction).

The solutions developed for monitoring the RWD and quantifying its destabilising effect, as well as that of α_s, are discussed in more detail later in Section 5.2.

5.1 Deck motion (MMS) variability

The current consensus in the industry is that forward predictions of helideck motion time series are impossible to perform accurately for more than a few minutes, even when state-of-the-art technology is used.

As discussed above, the MMS can simply be defined as the 20-minute maximum measured prior to landing, assuming that the maximum that will occur in the next 20 minutes will be the same. However, the maximum occurring during any given duration, even if ambient conditions remain exactly the same, will vary randomly and could vary substantially from one 20-minute record to the next.

It can be shown theoretically that for narrow-banded waves, and assuming a linear helideck motion response to the waves, helideck motion maxima will follow the statistical distribution described below, expressed as multiples of the Root Mean Square (RMS) of the signal:

(15) \begin{align}{\rm{Max = }}\zeta \cdot {\rm{RMS}}\end{align}

with the multiplier ζ following the probability distribution:

(16) \begin{align}{P_N}(\zeta ) = {\left( {1 - {e^{ - {\zeta ^2}/2}}} \right)^N}\end{align}

where P_N is the probability that the maximum over a number of cycles N will be less than or equal to ζ·RMS. The number of cycles can be calculated using N = Δt/T_z, where Δt is the observation window duration over which the maximum is taken (i.e. 20 minutes), and T_z is the mean zero up-crossing period of the vessel motion.

This is a standard, widely used theoretical result, e.g. as described in Ref. (Reference Baltrop and ADAMS2). In-service measurements of linear helideck motions such as roll, pitch, heave rate and accelerations on helidecks on large Floating Production Storage and Offloading vessels (FPSOs) as well as smaller support vessels fit this correlation very well, provided that an appropriate value is used for T_z, and any average offset relative to zero (e.g. due to the vessel listing) is subtracted.

It is therefore possible to use this approach to predict the maximum roll, pitch, or heave rate in the next 20 minutes, based on the RMS value measured previously, for any required level of probability, P_N. The average measured previously is also needed to account for any constant offsets from zero. This approach has in fact been applied to a new definition of the heave rate as a Significant Heave Rate (SHR) replacing the so-called Norwegian Method.

The MMS, however, is not a linear helideck motion parameter and it does not vary either side of a mean value. It is therefore not possible to apply the linear motion theory meaningfully to the MMS. Comparing MMS maxima against the RMS of an MMS time series has shown clearly that the distribution of the ratio ζ does not follow the distribution of Equation 16 and is more extreme, i.e. the MMS maxima are larger multiples of the RMS at any given probability than all other motion parameters. There are also marked differences between vessels, as shown in Fig. 3.

Figure 3.

MMS distributions of ζ for two vessel types, compared to those for linear motions.

Therefore, it is not clear which probability distribution might best describe the variability in the MMS, and whether the RMS or any other statistical measure of the signal represents the most appropriate basis. Any non-stationarity of the helideck motion, i.e. changes in the mean between the measurement period and the subsequent 20-minute on-deck period, presents additional difficulties.

Given these difficulties, it has not been possible to improve on the approach of using the 20-minute maximum measured prior to landing, and therefore this has been retained for the definition of the MSI.

Figure 4.

Illustration of the expected shape of the distribution for the variance factor R.

The variance between the previously measured (expected) value and the actual maximum values can be assessed probabilistically by calculating the ratio R of these values:

(17) \begin{align}{\rm{R = }}\frac{{{\rm{Valu}}{{\rm{e}}_{{\rm{actual}}}}}}{{{\rm{Valu}}{{\rm{e}}_{{\rm{expected}}}}}}\end{align}

and plotting its Probability Distribution Function (PDF) and/or Cumulative Distribution Function (CDF).

As illustrated in Fig. 4, since maxima over the same duration of 20 minutes are compared, the distribution of R should be equal to 1 on average (50 percentile), i.e. the on-deck maximum will on average be the same as that measured prior to landing (for statistically stationary conditions) and the spread of values either side of 1 would likely be symmetrical. The narrower the spread, the smaller the uncertainty and the better the MSI is as a predictive measure. Plotting the CDF allows the R value (or error in the prediction) to be read out for any required probability/certainty level. It is not obvious how the CDF of the R value might vary for any given vessel during the year, across different vessel types, and for the same vessel stationed at different sites; this is an area meriting further study.

5.2 Mean wind and turbulence variability

The detail of the variability of the wind during a landing event cannot be predicted in advance of a landing, but it is possible to represent the variability of the mean wind and the associated gusts in a probabilistic way.

The variability of the wind occurs over several timescales, as described by Van der Hoven^{(Reference Van Der Hoven3)}. The main variations are due to the passage of weather systems (synoptic, of the order of 5 days), the daily variation between and day and night (diurnal variation) and the variability due to turbulence. There is little variability in the wind for timescales between about 2 hours and 10 minutes; this is known as the spectral gap. Wind speed averaged over 10 minutes to 1 hour will therefore be relatively stable and will provide a good basis for describing the wind conditions.

In this work, CDFs of the variability of the mean wind speed and wind direction between consecutive measurement records have been quantified probabilistically using long-term historical data gathered at the Norwegian coastal site of Frøya, in Ref. (Reference Andersen and Løsveth4).

Real-time forecasting information could also, in principle, be used during operations to improve on the uncertainty between successive mean wind records and, for example, to warn pilots prior to landing when extreme events such as squalls may be passing. Another approach is to monitor the wind continuously after landing and to warn pilots to take mitigating action as appropriate; this has in fact been adopted for the mean direction relative to the helicopter (angle β or RWD), as discussed further in Section 7.

The fluctuations of wind speed and direction relative to the mean also need to be considered. In order to model their destabilising effect, it is not sufficient to predict the maximum gust value (e.g. with the use of simple gust factors). Instead, it is necessary to take into account the variability of each of the individual turbulence components, uʹ, vʹ and wʹ. The magnitudes of each of these components are different (turbulence in the atmospheric boundary layer is anisotropic) and each of the turbulence components affect the main rotor lift and fuselage drag in different ways.

As discussed in Part A, the main rotor lift can be assumed to vary linearly with wind speed, the slope depending on α_s:

(18) \begin{align}{\rm{LIFT}}({\rm{U}},{\alpha _{\rm{S}}}) = ({\rm{a}} + {\rm{b}} \cdot {\alpha _{\rm{S}}}) \cdot {\rm{U}} + {\rm{LIF}}{{\rm{T}}_{\rm{0}}}\end{align}

where a and b are constants, and LIFT₀ is the lift in zero wind.

The fuselage wind drag is assumed to be of the form:

(19) \begin{align}{{\rm{k}}_{\rm{w}}}({\rm{\upbeta }}) \cdot {{\rm{U}}^2}\end{align}

where k_w is a constant of proportionality which depends on the relative wind direction, β.

Using a linear decomposition of the mean and gust component (U = U_mean + uʹ) and taking into account small changes in both α_s and β due to the turbulent fluctuations (Δα_s = wʹ/U and Δβ = vʹ/U), separate analytical expressions for the destabilising terms due to the mean wind and gusts have been derived. A numerical model has been developed that calculates the turbulence in uʹ, vʹ and wʹ as a function of mean wind speed, and then calculates the probability function of the destabilising effect (expressed as a reduction in MMS_crit).

Thus, the Wind Severity Index (WSI) has been simply defined as the mean wind speed, using 10 minutes as a typical averaging time (in line with METAR data averages, rather than the 20 minutes used for the MSI). An appropriate correction is also included in the definition of the WSI, to refer the wind speed from that typically measured at the top of a tall mast, to that applicable to the height of the helicopter’s main rotor when landed on the helideck. The effect of wind variability and gusts can then be modelled probabilistically as a function of WSI and included in the calculation of limit curves.

6.1 Defining the variability of each input parameter

The first step in the probabilistic modelling approach has been to quantify the variability or uncertainty in each input parameter by defining a probability distribution for each of the main parameters. For the helideck motion and wind, this process has already been discussed above in Section 5. The same process applies to all other input parameters, whether helicopter-dependent or helideck-dependent.

One example is the minimum weight of the helicopter while on-deck. This occurs during the period in the on-deck turnaround when the helicopter is emptied of passengers and cargo, and prior to any re-fuelling. Thus, the variability in the minimum weight is primarily driven by the variability of the weight of fuel at the time of landing. A distribution was derived based on data gathered by helicopter operators. An example of such data (for the Sikorsky S-76 helicopter derived from a UK helicopter operator’s database) is presented in Fig. 5, which clearly illustrates the spread in the values. Although 90% of helicopters landed with more than 300kg of fuel, there were some operations that were close to the limit of minimum allowable fuel.

Figure 5.

Example of a probability distribution used to represent an operationally variable input.

For input parameters for which in-service data were more difficult to obtain, e.g. for the helicopter heading relative to the wind at the time of landing, expert estimates obtained from experienced technical pilots of a realistic worst case were used to reconstruct a probability distribution. As discussed by Bradbury et al^{(Reference Bradbury and Gill5)}, expert estimates of a realistic (rather than the absolute) worst case tend to identify the 2σ value of normal random variables (corresponding to a one-sided 2.5% probability of occurring).

For other parameters, a worst case or representative constant value had to be chosen. This was applied, for example, to the maximum expected vertical acceleration a_z (which was linked to the HR limit via assumed values for the vessel pitch period, T, for two main classes of vessels, large FPSOs and smaller support vessels). For the orientation factor, O_f, it was possible to use the maximum theoretical value, without introducing undue conservatism into the overall result.

6.2 Monte Carlo simulations

Having defined the probability distributions for all the variable input parameters, Monte Carlo simulations have been used to combine these probability distributions, and to calculate limits curves to any required level of probability. With this approach, an ensemble of possible on-deck scenarios is created, each input variable following its assigned probability distribution. Interactions between variables are also included as appropriate (e.g. dependence of mass and CoG location on minimum fuel at landing). All wind-related variables (i.e. variations in the mean following the measurement time, and the effects of turbulence) are generated by reference to a given mean wind speed, or WSI.

For each of the landing scenarios in the input ensemble, a corresponding ensemble of MMS_crit values is calculated deterministically using the analytical limits equations of MMS_crit versus U. From the ensemble values of MMS_crit the centile corresponding to any required level of probability for failure is calculated; this gives the limiting value of MMS_crit for each given WSI. This is repeated for the range of wind speeds from zero upwards, and for each failure mode, to build up a combined overall limiting curve. This process is shown schematically in Fig. 6.

Figure 6.

Building the limits curve using Monte Carlo simulations.

An example of calculated limits curves is shown in Fig. 7. Three curves are compared:

1. a) worst-case curve, assuming all parameters take their realistic worst-case values simultaneously

2. b) Monte Carlo result calculated for a 2.5% probability of failure (2σ level of probability, corresponding to an overall realistic worst case)

3. c) a mean or 50%ile curve, with all parameters set at their mean values

Figure 7.

Example limits curves calculated for various probabilities (worst case input parameters), realistic worst case (2.5 percentile) and 50% probability of failure.

The 2.5 percentile curve calculated with the Monte Carlo model is clearly much less restrictive than the worst-case curve, demonstrating the conservatism in assuming that all parameters take their worst-case value simultaneously. At the other extreme, the 50 percentile (mean value) curve would clearly be too risky to adopt as an operational limit; the 2.5 percentile curve lies in-between the worst case and 50 percentile assumptions.

The probability modelled in the Monte Carlo simulation represents the probability of failure for a helicopter operating in conditions on the limit curve. Due to the current restrictions imposed by the HCA roll and pitch limits, current operations only cross the 2.5% line at higher wind speeds, which will occur relatively infrequently compared to most operations.

Another important point to make is that, for each of the ensemble of events considered in the Monte Carlo simulation, a single value is assigned to each of the governing input parameters to represent the entire on-deck period. In effect, it is assumed that the 20-minute motion and gust maxima will always coincide, and that they will occur during the time the helicopter is at its lightest. This adds a certain conservatism to the modelling given that the period during which the helicopter is lightest is much shorter than the assumed 20 minutes. The magnitude of this uncertainty has not yet been quantified, although it would be informative to estimate it based on historical helideck observations of how helideck motion and wind variability might be correlated.

In the analytical expressions failure is assumed to occur when the ROS reduces to zero, or to a minimum acceptable level ROS_min. As discussed previously, there is uncertainty in the accuracy with which the ROS can be modelled (mainly due to possible additional lateral drag forces currently not accounted for in the model), and it is therefore prudent to add a safety margin in the Monte Carlo simulation in the form of ROS_min to cover the modelling uncertainty.

Despite the limitations identified above, the Monte Carlo approach provides a clear and rational way to calculate limits curves and has, so far, been implemented successfully to calculate limits for two helicopter types, the Airbus Helicopters AS332 Super Puma and the Sikorsky S-76.

It is also noted that while the limits calculation takes into account the variability of the wind for a given WSI value, it does not take into account the variability in the MMS for a given MSI value — it assumes that the maximum MMS that actually occurs during the on-deck period is always equal to the MSI (and, as discussed above, that this occurs during the time the helicopter is at its lightest, and that this coincides with the maximum gust). Including this variability in the existing Monte Carlo modelling method not only makes the calculation much more complex, but also requires a probability distribution for the MMS maxima to be defined. This, as discussed previously in Section 5.1, is very difficult to quantify for any one particular vessel, let alone for all different types of vessels in the UK.

The simplest way to account for the uncertainty in using the MSI to predict the actual maximum on-deck MMS value is to multiply the measured MSI value by an uncertainty/safety factor, i.e. a chosen R value, as discussed previously in Section 5.1. This should be vessel-dependent and could be refined based on real-time data.

The Monte Carlo approach was chosen for its robustness and simplicity. Other statistical approaches could also be used, e.g. Receiver Operator Characteristics (ROC) methods to help to find the best balance between preventing genuinely dangerous situations (true positives), while minimising false alarms (false positives).

6.3 Sensitivity studies

In addition to the Monte Carlo simulations, sensitivity studies have been carried out to assess and rank the effect of each of the input parameters.

Whether rotors are on or off is the most important destabilising factor, and the effect of rotor lift increases linearly with wind speed; this is dependent also on the rotor disc angle-of-attack α_s relative to the wind. For a Super Puma in a (mean) wind of just over 20 knots, the reduction of tipping failure ROS_TIP due to the (mean) lift is about the same as that due to helideck motion within ±3° limits of maximum roll and pitch, together reducing ROS_TIP to about 50%.

The contribution of fuselage drag also becomes more important with increasing wind speed. At low winds its contribution is relatively small, but because the drag depends on the square of the wind speed its destabilising effect grows faster than that of the lift. The drag term is also very sensitive on the relative wind direction RWD; there is an increase in the mean drag by about a factor of 3 for a change in RWD from 10° to 30°.

Non-centralised main rotor cyclic and tail rotor pedal inputs are another important factor, irrespective of wind speed; e.g. for the Super Puma, every 10% misalignment in either lateral cyclic or pedal causes about an additional 10% reduction in ROS_TIP.

The relative importance of all the input parameters to the probabilistic limits model depends not only on their mean values, but also on their range of variation and/or uncertainty relative to the mean. The Propagated Uncertainty approach described by Bradbury et al.^{(Reference Bradbury and Gill5)} has been used to derive rough estimates of the probabilistic limits curve algebraically, by approximating all probability distributions as Gaussian (normal) distributions relative to their mean values and combining their standard deviations. This has helped explore sensitivities without having to carry out full Monte Carlo simulations and as a high-level check of the Monte Carlo simulated limits curves.

8.1 New HMS system requirements

For the MSI/WSI/RWD limits to be implemented in practice, current HMS have to be upgraded to measure, calculate and monitor these parameters. Most vessels are equipped with Motion Reference Units (MRUs) to measure current helideck motion parameters, and these can easily be adapted to calculate helideck accelerations and the MSI.

Wind speed and direction are also routinely monitored on most vessels; however, this is not necessarily linked into the HMS, and some modification to the hardware and data logging arrangements may be needed. Graphical user interfaces used to display HMS data also have to be updated accordingly, and the opportunity is being taken to apply a degree of standardisation to the displays.

To enable the RWD to be calculated, new HMS operational procedures require the system operator on the vessel to input the heading of the helicopter at landing into the system, allowing the HMS to calculate and monitor the RWD automatically over the entire on-deck period.

A detailed new HMS specification has been developed by the authors in consultation with stakeholders. Prototypes built to this specification have been tested in-service during extended trials on the Maersk GP III FPSO, and the Chevron Captain FPSO and Alba FSU.

The MSI/WSI/RWD scheme is now incorporated in CAP 437⁽⁶⁾, and the new HMS standard was published on the HCA website in April 2018⁽⁷⁾. All moving helidecks operating on the UK will be required to upgrade their HMS to comply to this new standard by 31 March 2021.

8.2 Preliminary limits line

The new scheme is introduced with a preliminary, generic lower bound MSI/WSI limit, designed to cover all helicopter types. The limits curve has been simplified in the form of a straight line, linking the maximum MSI point (at zero wind) to the maximum WSI point (at zero MSI). Initially, this will be applied as advisory only, invoking consideration of revised on-deck handling procedures in order to reduce the risk of failure (as discussed below in Section 8.3).

The preliminary limit line is shown in Fig. 10 where the MSI is presented as 10 tan⁻¹(MMS) in order to generate a number that is meaningful (an obvious multiple of the effective helideck inclination), and distinct from R/P/INC/HR values. Note that the WSI presentation has been modified from that shown in Fig. 10 to display it as a percentage of the maximum WSI for the helicopter type selected; this is in order to avoid confusion between the maximum wind speed limit currently used by pilots which includes gusts and the WSI, which does not.

Figure 10.

MSI/WSI limits line compared to estimated values for selected examples.

To put this into context, the estimated MSI/WSI values that would have been recorded by the new HMS are plotted for selected examples: a) Super Puma GBKZE/West Navion accident in Nov 2001; b) the S-76 Foinaven trials in Nov 1999; c) a Super Puma G-TIGZ near-miss incident at the Tartan fixed offshore platform in Jan 2005. Examples a) and b) are discussed in Part A.

1. a) G-BKZE/West Navion accident: the wind speed quoted at the time by the West Navion was 26kt, but the Met Office provided an estimated mean value of 32kt (at standard surface met height of 10 m) gusting to 42kt. Nearby vessel Faroe Connector (within 500 m) recorded a mean of 37kt, as did Foinaven FPSO (37kt gusting to 45kt), though the height of measurement for both these vessels was not known. The helicopter commander announced to the passengers over the public address that the West Navion helideck was moving around and it was quite windy, quoting (gust) wind speeds of between 40kt to 45kt. He asked the passengers to disembark/embark two at a time to maintain the weight of the aircraft for these reasons. Using the Met Office value as the reference, the mean wind speed corrected to the West Navion helideck height is WSI = 38kt. This illustrates the difficulties in obtaining reliable wind measurements when these are not part of the HMS, and how important it is to use a consistent definition for the WSI.

2. b) For the Foinaven trials, the wind speeds plotted are the actual means as measured at the helideck; the wind changed during the trials, and the WSI that would have been measured prior to some of these trials would have been less that the actual mean values (so the WSI values shown are probably overestimated). The Foinaven trials were carried out in conditions close or just above the limits. The aircraft was deliberately placed at large angles relative to the wind (between 35° and 60°), and the minimum recorded reserve of stability for the helicopter was indeed low, at about 25% for the rotors turning tests.

3. c) The last example is for a fixed, rather than a moving helideck. The MSI in this case is effectively zero, assuming that the helideck is perfectly horizontal and completely flat — in reality, fixed decks will be inclined slightly to allow for drainage. The fuel weight was 900kg (corresponding to a 40%ile value of assumed North Sea operations, i.e. slightly lighter than average). The mean wind speed was of the order of WSI~40kt, which then increased to 50kt as a result of a passing squall (a change in the mean wind speed from 40kt to 50kt corresponds to a 1 in 4 event in the probabilistic limits model). The gust was quoted as 63kt, which corresponds to a gust factor of 1.26 for a mean of 50kt. The relative wind direction also changed, from β = 0° (head-on) to 25°, which is at the limit of triggering an RWD warning.

It is intended that helicopter type-specific MSI/WSI limits calculated and certified by the helicopter manufacturers will subsequently be introduced; these are expected to be less conservative but will likely be mandatory do-not-land limits, especially if the current R/P/INC/HR limits are relaxed. It is also strongly recommended that more in-service measurements of helicopter ROS are gathered, as well as more data on the variability of the MSI/WSI and corresponding actual maxima, to support the calculation and setting of future limiting curves.

8.3 New operational procedures

New operational procedures have been developed in consultation with helicopter operators and vessel Offshore Installation Managers, using the Hazard Identification (HAZID) approach used in the oil and gas industry to identify potential hazards and operability issues and to recommend preventive actions.

The new operational procedures cover pilot and helideck crew actions prior to landing and during the on-deck period. It also sets out suggested modified operational procedures in the event of MSI/WSI exceedences and/or RWD warnings.

The modified helideck handling procedures to be considered when operating to an offshore installation in steady amber conditions (see Section 8.5) include:

1. • taking particular care to align the aircraft with the wind (to minimise destabilising forces due to wind).

2. • both pilots remaining at the controls during re-fuelling, embarking or disembarking of passengers, bags and freight (to minimise reaction time to deteriorating conditions).

3. • swapping embarking and disembarking passengers one or two at a time (to maintain helicopter weight as high as possible).

4. • if necessary, re-fuelling with passengers on board (to maintain helicopter weight as high as possible).

5. • carrying out one operation at a time (to minimise reaction time to deteriorating conditions).

Any number of the above may be selected by the pilot depending on the prevailing conditions, but the pilot should make clear to the helideck crew in advance exactly what course of action is to be taken to prevent any confusion.

In the event of an RWD warning (flashing red) immediately after touchdown, the pilot should take-off and re-align the helicopter with the wind. At any time after touchdown, an RWD caution (flashing amber) will normally occur to provide advance notification of an impending RWD warning (flashing red). In the event of an RWD caution, the pilots and the helideck crew should investigate the cause and agree appropriate mitigating action with the pilot such as:

1. • if due to vessel heading change, helideck crew to take action to correct vessel heading.

2. • if due to wind speed increase and/or direction change, pilot to consider taking off and re-aligning the helicopter with the wind.

If the RWD caution develops into a warning, the pilots should prepare to take-off and re-land on the helideck oriented into wind. If this occurs towards the end of the on-deck period while passengers/cargo are embarking, the pilots should prepare for take-off as per normal but without delay.

8.4 Effect on operability

The new MSI/WSI limits are to be used in combination with current HLL landing limits. Initially, the MSI/WSI limits are to be introduced as advisory only, ensuring minimal loss of operability while still providing important safety information to the flight and helideck crews. The MSI/WSI limits will cover on-deck stability with no loss of safety guaranteed while the current HLL limits are retained.

If and when the MSI/WSI limits become mandatory, maintaining the current HLL landing limiting values likely would lead to a reduction of operability at higher wind speeds. However, the HLL landing limits would then only be required for the touchdown and could potentially be relaxed. This provides scope for operability gains, especially at lower wind speeds as illustrated in Fig. 11 where the current R/P/HR limits are presented schematically as a line of constant MSI.

Figure 11.

An illustration of the effect of new MSI/WSI limits on operability.

The effect of the new MSI/WSI limits on operability will vary from vessel to vessel. For example, for the same value of static roll and pitch, MSI values for smaller vessels will be larger than those for larger vessels, since helideck accelerations on the former will be larger due to their shorter motion periods, T.

8.5 New helideck motion status repeater lights

A helideck mounted repeater light system has been added to the new HMS in response to a request by helicopter operators. The main purpose is to provide helideck status information directly to the pilots and helideck crew, instead of relying on the system operator to relay this information.

Prior to landing, the helideck status is indicated by steady burning signalling lights as follows:

1. • BLUE status: all parameters within limits, safe to land.

2. • AMBER status: MSI/WSI only limit exceedance, land with caution/consider employing modified helideck handling procedures.

3. • RED status: R/P/INC/HR exceedance, do not land.

   (NB: BLUE is used instead of GREEN, to avoid confusion with the green helideck perimeter lighting.)

An analysis of near misses in the CAA Mandatory Occurrence Reporting Scheme database has shown that about 30% of occurrences were caused by incorrect reporting of helideck motion prior to landing. It is therefore expected that the new helideck motion status lights should help to eliminate such incidents, improving landing safety significantly by this simple measure alone.

In addition, the repeater lights are required for the new RWD functionality to ensure that timely warnings are provided to the pilot. After landing, signalled to the HMS by entry of the helicopter heading immediately after landing, the lights are used to relay the RWD status of the helideck as follows:

1. • BLUE status (slow flash): relative wind direction within limits;

2. • AMBER status (fast flash): impending relative wind limit exceedance (investigate cause and identify appropriate mitigating action required);

3. • RED status (fast flash): relative wind limit exceeded (take appropriate mitigating action).