Advanced design software for mastering ship geometry
Why is / are the Bwl / Lwl calculations sometimes erroneous ?            

MAAT Hydro's Bwl and Lwl are calculated by scanning model's current lines immersion (including face's outlines), face's inner areas being, nevertheless, ignored. 

This simplification usually has no incidence on Lwl / Max Draft calculation, as ship's silhouette (usually present for allowing windage calculations) and / or faces outlines allow an accurate calculation. 

Nevertheless, the Bwl calculation can be strongly affected when no line or station is located in the midship area, Bwl being then only calculated with the present lines (and, therfore, sometimes only with stem and transom). 

Including a midship section in the model therefore allows solving simply the problem. 

Moreover, it is also important to recall that the "Max Hull Dimensions" selector located on [Ship] tab's [Hydro] page allows selecting the Lwl / Bwl calculation mode as follows: 

  • "/All Immersed Hull": All the immersed line parts are scanned for the calculation ( immersed bulb's length will therefore be fully taken into account). 
  • "/ Floatation Only": Only the intersections of present lines with the current floatation plane are taken into account for the calculation (immersed bulb's length will therefore not be taken into account).
What is the incidence of "Computation Accuracy" on the calculated results ?            

The "Computation Accuracy" selector, located on [Ship] tab's "Hydro" page, allows setting the best speed / accuracy compromise for a given model. 

The "Smart xxx" settings correspond to a constant surface tesselation, which means that face's patches will be represented by a variable number of facets, whose area will therefore be almost constant. 

The "Const xxx" settings correspond to a constant parametric tesselation, which means that face's patches will be represented by a constant number of facets, whose area will therefore be variable (large patches will therefore be tesselated like the small ones). 

When a model is made of numerous solids (large & highly compartmented vessel for example), the "Smart / Very Low Resolution" setting generally provides the best accuracy / response time ratio. 

Conversely, when a model is only made of few simple solids, defined by few simple faces (a single parallelepipedic box for example), the "Const xxx" settings may cause inaccurate results (few large patches cause a poor tesselation density and, therfore, inaccurate calculations ) . In such cases, the "Smart Medium Resolution" / "Smart Very High Resolution" setting usually provide the best results. 

Nevertheless, as "Smart xxx"'s optimization algorithm may sometimes fail on certain cases, the "Const Medium Resolution" setting (usually slower for a similar accuracy), has been retained as the default setting, being the safest compromise in terms of precision, as soon as the number of patches is sufficient (which is the most common case). 

The only case on which this default setting may be unadapted is the case of very simple models (patches are not enough numerous); the "Smart Medium Resolution" / "Smart Very High Resolution" settings then provide the best results. 

It is therefore a good habit to open the hydrostatic viewport as soon as projects solid(s) is/are loaded or created, in order to find its best accuracy / speed compromise by trial and error.

A model is imported but no hydrostatic data is displayed ?            

The imported faces have probably not yet been converted into solid(s) and no buoyant object can therefore "float" in MAAT Hydro's virtual tank. 

The various ways to create solids from imported faces are detailed in "Part II: MAAT Hydro's Quick Hydrostatic Modelling Guide" and further details are available in"Part IV: MAAT Hydro's advanced Data Management". 

Nevertheless, when the imported data only contain a single half hull surface, the corresponding symmetrized solid can be generated immediately thanks to menu bar's "/Solid/Mirror Surface" function. 

Moreover, if the original model was a solid in your initial 3D modeler, the appropriate IGES export options will probably allow to export it as a solid, which will be directly recognized by MAAT Hydro. 

In addition to this need of including at least one buoyant solid, it is also necessary to create at least one realistic mass allowing to calculate ship's weight / buoyancy balance. 

At last, several complementary lines may also be necessary, depending on the calculations you want to perform and the associated criteria (see manuals for further details).

Why are MAAT Hydro’s calculations sometimes slower            

As MAAT Hydro directly operates on solid boundaries without needing any station or section, computation speed directly depends on the amount of faces to integrate. 

Model’s size being extremely variable, setting the "Computation accuracy" displayed on [Ship] tab’s “Hydro” page allows finding the best speed / accuracy compromise: 

A “light” model may require a high computation accuracy in order to get accurate results without slowing down the program and, conversely, a heavy model may be accurately and quickly calculated with low computation accuracy, as the amount of data is naturally high. You can immediately evaluate the incidence of this setting by controlling the values displayed in the hydrostatic window for various accuracy settings. 

When the initial hull model becomes divided into compartments, the amount of data grows and it is generally possible to decrease the computation accuracy accordingly. 

As a last resort, important speed improvements can also be obtained by simplifying float’s descriptor and removing small unnecessary faces directly in your 3D modeler.

How to Use the new 'Turbo' option ?

Thanks to a high optimization of its hydrostatic engine, MAAT Hydro provides a new 'Turbo' option  since Rev. 7.3, allowing to speed up the calculations by a x4 to x50 ratio (average between x7 to x10).

Following to this implementation, new 'MAAT Hydro Turbo +' and 'MAAT Hydro Turbo ++' licenses are now proposed on the license purchase page.

Nevertheless, it is important to notice that operating MAAT Hydro in the 'Turbo' mode with a 'non Turbo' license will overwrite 'For Demo Only / No Commercial Use'  labels on the resulting reports.

The 'Turbo' mode must therfore be switched off by clicking on the corresponding button at the bottom of the window in order to be able to use your 'non Turbo' license normally.

 How to parallelize long probabilistic stability calculations when extra computersare available ?

As the probabilistic stability calculation may concern tens of thousands cases, 1, 2 or more days of calculation may sometime be necessary for getting the results (thousands of report pages !), so that it may be interesting to process damage system's branches simultaneously on distinct computers when possible.

The procedure is the following:

- Install MAATHydro on each of the computers to be used for the calculation (no MAAT Hydro license is needed for these 'extra' computer). 

- Generate the probabilistic damage system on the licensed MAAT Hydro version thanks to the '/Tools/Probabilistic Stability/Make Damage System' function and store the resulting project in distinct files corresponding to each of the 'extra' computers.

- Open each of these duplicated files on a distinct computer with its dedicated MAAT Hydro version and select the branch to calculate according to computer's power: Heaviest branches will be preferably calculated on the most powerful hardware and vice versa. .

- Launch the computation of each of these branches on each computer thanks to the '/Tools/Probabilistic Stability/Calculate Damage System' function. The computation speed will then be multiplied by the number of computers running simultaneously (typically x4 for 4 computers running simultaneously on each branch of a 4 zones damage system. A 2 days calculation may therefore be achieved in about 12 hours !).

- When a branch is calculated, its partial 2D reports can be scratched, as well as the other project data, so that only the calculated branch will remain (this is all we need for reconstructing the calculated complete damage system in the original file).  

- When all these branches are calculated and stored in distinct files, they just have to be imported, one by one, in the initial project, in order to transform the initial non-calculated damage system (blank damages) into a calculated one (colored damages) by replacing the initial branches by the imported ones (i.e deleting the old branches and dragging the imported one to their place). Special attention must be paid to conserve exactly damage system's structure

- When the calculated damage system is completely (and exactly) reconstructed, just select the '/Tools/Probabilistic Stability/Calculate Damage System' function for the final calculation and click on damage system's root: The global results will then be obtained very quickly, as all the cases are already calculated (nevertheless, this must be done with a licensed MAAT Hydro++ version in order to get usable reports).

What are MAAT Hydro's Swell Parameters ?

In several functions, MAAT Hydro allows taking an optional sinusoidal swell into account in order to calculate its (quasi-static) incidence, like, for example, in a longitudinal strength calculation.  

The height h(x) of this optional swell, according to x,  is calculated as follows

   h(x) = a.sin(2.pi.(x+x0)/wvl)


   a: Swell’s amplitude (‘Amplitude’, mnemonic :  WVA)

   wvl : Swell’s wave length (‘Wave Length’, mnemonic: WVL)

   x0: Swell’s crest position (‘Crest @’, mnemonic : WVP)

   x : Current longitudinal position.

Therefore, MAAT Hydro’s swell amplitude must be clearly discriminated from its total height, which is twice greater.

What is the incidence of the 'Variable (real) FSM' / 'Constant FSM' option on 'Transverse Stability' calculations ?

The '/Tools/Transverse Stability' dialog box provides a 'Variable (real) FSM' / 'Constant FSM' option allowing to control the liquid effect process during the stability calculations:

  • The 'Variable (real) FSM' allows calculating tank's real free surfaces at each stability iteration. This calculation mode provides the most realistic results, but with a performance loss depending on the number of filled tanks to process.
  • The 'Constant FSM' mode 'freezes' tank's contents at 0° heel and only applies the user defined FSM corrections(FSMt / FSMl) to this 'frozen' center of gravity for calculating ship's transverse stability. This calculation mode is the fastest, although it is less realistic and, usually, more pessimistic. 

The default mode is 'Real (variable) FSM', which is, now the most commonly used, but the 'Constant FSM' mode may sometimes be useful, in particular when ancient generation stability software results must be reached as closely as possible or, simply, when old regulations require it.

The following pictures illustrate the different FSM correction modes:

1: Real (variable) FSM:

* G0 is the 'frozen' solid+liquid upright center of gravity.

* G is the current 'frozen' solid+liquid center of gravity.

* Gc is the current corrected solid+liquid center of gravity, according to current FSM (in this case, GZ = GcZc).



2: Constant FSM:

 * G0 is the 'frozen' solid+liquid upright center of gravity.

* G0c is the upright 'frozen' solid+liquid center of gravity constantly corrected normally to current floatation (in this case, GZ = G0Z = G0cZc).



3: Fixed solid load (no liquid effect, used for Max KG calculation):

* KG0c is the Max KG calculated with solid loads (no liquid effect).

* KG0 is the 'frozen' Max KG deducted from KG0c according to a given FSM correction.




What can I do when a "No balance can be found" message appears?            

As float’s balance depends on several physical factors, common sense usually allows understanding what happens: 

  • Load / CG / buoyancy mismatch (ship sinks or capsizes). 
  • Wrong plate thickness causing a disproportionate plate thickness correction (i.e. thickness entered in mm instead of m). 
  • Error(s) in float / compartment permeabilities. 
  • Erroneous water density. 
  • Erroneous face pressure orientation or missing faces. 
  • Wrong calculation initialization: Initial floatation being located at “Dwl Height” (check [Ship] tab’s “Hydro” page “Dwl Height” field), should always intercept ship’s hull, otherwise ship balance calculation may sometines fail.
Why does the selected criterion sometimes come to nothing?            

Depending on script’s content and safety tests set to process exceptions, its execution may fail, warning messages then provide information on problem’s cause.

A perfectly debugged script should never fail, but perfection is unfortunately only a goal ! 

Fortunately, this kind of script error is gentle and must not be taken for a system crash. 

One of the most common causes of such problems is an inadequate heel range; it is therefore a good habit to enter generous heel limits. 

Another common cause is the lack of necessary data, i.e. a silhouette descriptor for windage calculations, sheer line(s) for freeboard calculations, opening(s) for downflooding calculations… 

Finally, non-protected divisions in the script may cause classical “division by zero” errors.

Why does the IMO A749 (18) criterium calculation stop on theta1 angle calculation ?               

IMO A 749 (18)'s theta1 calculation depends on the roll angle, which is itself calculated according to ship's mean moulded draft. A theta1 calculation problem usually denotes an erroneous midship draft value, generally resulting from a wrong K point location..

In order to solve this problem, it is then usually necessary to correct K point's height on [Ship] tab's [Hydro] page ('Keel Point Height (Z)' field), which must correspond to midship's keel height.

 How to Calculate the 'Maximum Steady Heel Angle Curves' ?

The 'Maximum Steady Heel Angle Curves' diagram, requested by par certain regulations like the 'MCA  Intact Sail (Mono)', allows evaluating sailing ship's vulnerability to gusts and squalls according to their strength, depending on the current apparent wind velocity / corresponding heel.

In order to get this diagram, ship's transverse stability must be calculated first and the selected criterion must be designed for such calculations (i.e contain an initialization line like 'SetMaxSteadyHeelData(hdfl, 1.3, "Downflooding")').

When these condition  requirements are fulfilled, the 'Maximum Steady Heel Angle Curves' tick becomes active in the stability results dialog box, allowing to generate the diagram co rresponding to the current case (the created document has a CSV content, which can be exported to EXCEL for any external use).

See  the  'STC  scripting  manual' for further details on the use of the 'SetMaxSteadyHeelData' statement.


How to Calculte Dredge’s Stability ?            

Dredge’s stability calculation is based on 3 simultaneous phenomenons : 

  • Cargo shifting, whose free surface is not parallel to floatation (cargo’s “shifting law” according to heel and density is defined in criterion script’s header). 
  • Cargo’s progressive spilling out of hopper, depending on heel and hopper’s spilling / inflow line. 
  • Hopper’s partial inflow, depending on cargo’s free surface, permeability and hopper’s spilling / inflow line. 

Dredge’s float must be modelized like for an OMI A749(18) transversal stability calculation (including silhouette openings and sheer descriptors)), except for hopper, which must comply with 4 constraints, due to its special behavior in this calculation:

  • Hopper must be a MAAT Hydro Tank, directly located in the calculated directory, and therefore not affiliated to any parent compartment. Hopper’s hole must therefore be modelized in dredge’s float in order to avoid accounting its volume twice. 
  • Hopper must be set in the desired cargo layer, corresponding to its density (densities and permeabilities can be specified on [Ship] tab’s “Materials” page). 
  • At least 1 inflow and spilling line set in the 'Overflow' layer must be included in hopper tank, in order to allow spilling / inflow calculations to be done automatically. 
  • Similarly, inflow (without spilling) lines must also be included and set in the 'Downfloodable Opening 1' layer if necessary (i.e hopper's bottom when not watertight). Cargo's permeabilty can be specified in the 'Permeability' column of [Ship] tab's [Material] page. 

When dredge’s model is ready:

  • Select «/Tools/Dredge Stability» in the menu bar. 
  • Select a dredge stability criterion (i.e. «D» ). 
  • Select the cargo layer, which will allow MAAT Hydro to identify the hopper tank(s). 
  • Define a reasonable heel range (consider the negative roll angle before setting the min. heel. Moreover, due to cargo spilling, it is not recommended to exceed a 90° max. heel, 80°/85° being the best limit. 
  • Tick the expected outputs before starting the calculation. 

Depending on hopper’s geometry and cargo’s density, the free surface effect may sometime be considerable, occasionally avoiding calculation to result (locally unstable ship).In such cases, it may be helpful to increase the 'Downflooding Range' value on [Ship] tab's Hydro page in order to make hopper's downflooding more progressive and, therefore, ship's balance easier to calculate if possible.

   See function's help for further details


How to Calculate Water on Deck Stability (Torremolinos convention) ?

The /Tools/Dredge Stability function presented above also allows calculating the transverse stability with water on deck, in the Torremolinos convention sense.

Like for modelling the hopper in a dredge calculation, a fictitious tank representing the deck + bulwark space must be added to ship's model, in order to contain the water on deck and adjust its level according heel and trim.

In order to respect the Torremolinos specifications, this tank must:

- Be set to the layer  selected as  'Dredge Cargo Layer' on [Ship] tab's [Hydro] page (normally used for defining cargo's density and identifying the hopper in a dredge calculation) and filled to 100% (WoD's level will be automatically adjusted according to heel and trim as soon as overflowing lines will be included. See 'Advanced buoyancy Management' for further details on this mechanism).

- Be affiliated to an intact compartment in order to ignore 'Deck - Bulwark' tank's buoyancy (otherwise ship's righting arm would be overvalued as soon as this fictitious tank would be sumerged).

- Contain one or more overflowing lines (i.e. bulwark's upper edges, identified by the [Ship] tab's [Hydro] page's 'Overflow Layer') in order to allow WoD's level to be automatically adjusted according to heel and trim.

Moreover, as water on deck's weight must not be directly taken into account in a Torremolinos stability calculation, a minimum density material must be associated to the 'Dredge Cargo Layer', by example a 'void' material with a 0.001 density, associated to a 'Torremolinos Water on Deck' layer. In this way, WoD's volume and center of gravity will be automatically calculated according to heel and trim without taking its mass, thus neglectable, into account in ship's balance calculation (anyway, the '' script will finally restore WoD's real weight according to the real water / fictitious Wod density ratio...). 

For further details on this topic, see the script comments in '', inwhich you will also notice that the specific dredge header equals WoD's density / heel to floation's values, contrary to dredge stability scripts, where they are generally distinguished.

When ship's Water on Deck is modelled like this:

- Select  '/Tools/Dredge Stability" in the menu bar
- Select the '' criterion
- Enter a realistic calculation range and select the layer identifying the 'Deck-Bulwark' tank
- Select the awaited reports and validate the calculation


 How to analyze the Incidence of Progressive Damages on Ship's Stability ?            

When damaged compartment's inflow(s) cannot flow freely outside, the incidence ot such progressive floodings must be taken into account and ship’s stability must be analyzed for various intermediate flooding stages.

Since MAAT Hydro Rev. 7.9, menu bar's '/Tools/Progressive Flooding' provides 2 complementary approaches, depending on model's content:

A) Progressive Flooding (Basic):  

This approach provides a quick way for analyzing the incidence of a progressive damage on ship's stability but the progressive damage is assumed to be known and modellized by identifying the ‘progressively flooded compartments’ on the [Data] page by:

   - Their layer, which must correspond to [Ship] Tab / [Hydro] page’s 'Intermediate Flooding Layer'.

  - Their damage percentage which, in this case, represent their current inflow / final inflow ratio. For obvious reasons, intact compartments (0%) are ignored, even when located in the ‘Intermediate Flooding Layer’.

Such progressive damages, with or without ‘progressively flooded compartments’, can be registered once, by creating a ‘Damage Condition’, which will allow restoring this damage directly, for the automated calculation loops or, simply, by a double click.

When a transverse stability calculation is carried out with such ‘progressively flooded compartments’:

   -  An equilibrium floatation is first calculated buoyancy loss in the damaged compartments.

   - The ‘progressively flooded compartments’ are then automatically transformed into tanks containing the specified percentage of the final inflow, calculated by equalizing the free surfaces at equilibrium.

   - Transverse stability is then calculated on this transformed model.

   - This calculation model is finally restored into the original one, as soon as the stability calculation is finished.?

When ship's current model contains such damaged compartments, the 'Tools/Progressive Flooding' function allows repeating this calculation automatically from 0% to 100% of inflow, by 10% steps (see online help); Nevertheless, it will be refused when the current model doesn’t contain ainy ‘progressively flooded compartment’.

When such intermediate flooding calculations must be repeated for numerous damage / loading cases or case combinations, a selector allows repeating this function automatically for all the present damage and / or loading conditions. These damage and loading conditions must therefore be preliminarily created before starting the ‘Intermediate Flooding Stages’ calculation, in order to allow repeating it automatically on any case / combination.

At last, when progressive floodings must be included in the probabilistic damage system, the corresponding intermediate damages will be automatically created as soon as ‘progressively flooded compartments’ will be damaged, depending on [Ship] tab / [Hydro] page’s ‘Interm. Flood. Stages Number’. Nevertheless, it is recommended to keep this number of intermediate stages (2 by default) as small as possible in order to avoid increasing significantly damage system’s number of cases to process.

 B) Progressive Flooding (Advanced):  

This new approach provides an accurate way for analyzing the incidence of a progressive damage on ship's stability, as MAAT Hydro can now calculate the damage extension scenario automatically, according to the properties, immersions and status of ship model's pipes and openings.

The 'Tools/Progressive Flooding' function automatically selects this advanced process (see online help) when:

   - Opened Pipe(s) and / or Opening(s) are present in the current model.

  - Ship's current status allows the current pipes and openings to cause progressive inflow(s) because:

- They are opened 'outboard'

- They are connected to a damaged space.

Of course, no progressive flooding calculation is possible when ship's status doesn't cause any progressive inflow, especially when it is in the intact state, without any outboard pipe / opening, or when no damaged space is connected to them.

At last, it is recalled that, although pipes and openings are also used for adding the probabilistic damage systems,  



How to calculate Ship's Longitudinal Strength for a given Load and Swell ?            
  • The first longitudinal strength calculation step consists in defining a realistic weight curve by creating a set of masses whose Xmin / Xmax fields must be realistically defined (don't forget that mass' Gx must remain confined beween xmin+(xmax-xmin)/3 and xmax-(xmax-xmin)/3).
  • The hydrostatic tab must then be opened and displays the red weight curve resulting from the defined masses. Swell's wave length (WVL), phase (WVP) and amplitude (WVA) can then be specified at the bottom of the right grid.
  • Hydrostatic  menu's  "Show Longitudinal Strength" option must then be selected in order to turn ship balance calculation from the 'normal' default mode to the 'transverse' mode, as the summed strength must be only be located in the transverse plane (i.e. such calculation is only valid for small trims).
  • You can then calculate ship's balance for this load and swell in this appropriate 'transverse' mode in hydrostatic tab's menu. The displayed shear forces and bending moment curves will then be updated and closed at their extremities.
  • Detailed outputs can then be produced for this floatation by selecting 'Tools/Longitudinal Strength' in the menu bar.
  • Ship's strength can then also be displayed dynamically for increasing swell phases by selecting 'Start Swell Motion' in hydro tab's menu: After calculating all the intermediary balances, the views will start animating in real time; animation can be stopped by pressing [Esc] key or selecting 'Stop Swell Motion' in the menu.

        See function's help for further details


How to Calculate RoRo's damage stability according to Stockholm regulations            

RoRo's damage stability calculation adds an extra liquid load on the RoRo deck, depending on its freeboard and the average height of swell in the service area (Stockholm).

The RoRo spaces are represented by compartments identified by a dedicated layer whose density correspond's to the current floatation (during calculation, they will be automatically transformed into tanks whose filling will be automatically updated depending on the scripted statements).
The calculation procedure is the following: 

  • Ensure that the RoRo compartment(s) is/are identified by a dedicated layer whose density corresponds to the current floatation, and that the current damage involves at least one RoRo compartment. 
  • Select «/Tools/RoRo Stability» in the menu bar. 
  • Select the "SOLAS RoRo" criterion. 
  • Enter the criterion parameters. 
  • Select RoRo compartment's identification layer. 
  • Enter the average height of swell in service area. 
  • Define a reasonable heel range, for example from -20° to 80° by 2.5° steps. 
  • Tick the optional results to output. 
  • Click [OK] to start the RoRo stability calculation. 

Exceptionally, depending on damage and free surface effect's importance, ship's buoyancy / residuary stability may be insufficient and calculation may fail.


 How to calculate a Deterministic Damaged Stability ?

 As the deterministic damaged stability analysis is generally based on the calculation of a large number of damage / loading combinations, MAAT Hydro provides various functions for automating this process:

- The 'Transverse Stability Results' dialog box provides a [Retain All Cases] button, which allows repeating automatically the current calculation, with its current options, criterion and data, for all the current damage / loading combinations. This means that, after having created the damage and loading conditions to combine in the current workspace, you just have to start the 'Tools / Transverse Stability' calculation, select the criterion in the results dialog box and tick the report options before clicking on the [Retain All Cases] button and waiting for repeating the current calculation on all the damage / loading combinations (reports are directly created in the [2D] section).

- The 'Tools / Deterministic Stability' function provides a complementary and more synthetic approach, allowing to locate the KG of every loading condition according to the maximum KGs calculated for all the damages in a given drafts / trims range. When the criterion is selected in the dialog box and the heel / draft / trim ranges are specified (the [Calculation Range] button allows to preset automatically these ranges by calculating the extreme drafts and trims corresponding to the current loading conditions in order to insure that their KG will appear on the calculated diagrams). A special attention must be paid to the increments before validating the calculation data, as the response time totally depends on the number of intermediary curves to calculate. Moreover, as this analysis is focused on max KG calcuilations, there is usually no reason to include liquid effect in this calculation, although it is supported if necessary. See function's help for further details.

At last, using the 'ReportDamageData' command at the end of criterion's script is highly recommended to allow controlling the listed data and their layout in the generated reports (see STC Scripting Guide for more details).


 What is the difference between 'Deterministic' and 'Probabilistic' criteria ?

As the damaged stability calculations are usually repetitive because of the large number of damages to calculate, their automation quickly appears as crucial, a fortiori for the probabilistic stability in which the number of damages may be very high.

The '/Tools/Transverse Stability' function, intended primarily to allow analyzing specific situations, already provides a first level of automation of the repetitive calculations thanks to its [Retain All Cases] button, which allows chaining the current calculations for all the current loading and damage conditions. 

As a complement,to this automated intact or damaged stability calculation, MAAT Hydro also provides a '/Tools/Deterministic Stability' function (see above), focused on a synthetic Max KG analysis .Nevertheless, as this function requires certain complementary data, only the STC criteria who provide them to MAAT Hydro, i.e. whose name start with 'Deterministic', must be used.

At last, as the probabilistic stability calculations (see below) also require specific complementary data, only the STC criteria who provide them to MAAT Hydro, i.e. whose name start with 'Probabilistic', must be used.

In a nutshell, then, only the criteria whose name start with 'Deterministic' must be used with the 'Deterministic Stability' function and only the criteria whose name start with 'Probabilistic' must be used with the '/Probabilistic Stability/Calculate Damage System' function.

In this regard, although the 'Transverse Stability' function accepts any STC criterion (the specific data provided to MAAT Hydro are not used) but it is recalled that the  'Dredge stability' and 'WoD Stability' functions can only operate on compatible STC criteria, i.e. calculating and providing the specific data needed by MAAT Hydro (see items above).

How to calculate a Probabilistic DamagedStability ?            

1 Abstract:

Probabilistic stability calculations being exhaustive by nature, are based on analyzing a large number of damages, supposed to represent all the possible cases. Nevertheless, for obvious reasons, only the most probable connex damages are taken into account, in order to restrain this huge number as much as possible.

The analysis of a large number of real damages has provided statistical models  allowing to associate a probability to any of these damages and a rule (similar to the deterministic SOLAS rule) also allows quantifying damage's survivability Si between 0 and 1.

More accurately speaking, damage's probability is considered to be the product of 2 independent probabilities, the first one (Pi) only depending on damage's transverse / longitudinal extent and the second one (Nuj) on its vertical extent.

According to this model, it becomes possible to calculate the surivival expectation dAi (first order statistical moment) associated to ith damage as a simple product: dAi = Pi.Nui.Si.

As soon as all the calculated damages form a complete damage system (in the probabilistic sense, i.e. no redundance between damages and sum of their probabilities equal to 1 ), a global survival expectation A can be calculated on it by summing all the elemetary expectations dAi associated to each damage.

By repeating this calculation for 'light', 'partial' and 'service' loading cases, the As, Ap and Al indexes can be calculated, allowing to calculate the global index A:

A = 0.4.As 0.4.Ap+0.2.Al

The SOLAS 2009 criterium therefore simply requires that ship's global survival expectation is greater than a threshold R, depending on ship's type, program and characteristics.

2 Modelling Ship's subdivision with MAAT Hydro:

The first step, probably the most important and tedious, consists in modelling the ship compartments onwhich the probabilistic calculations will be based.
First of all, it must be recalled that, as these calculations exclude tank's liquid effect (tank's incidence is supposed to be included in the loading cases as a constant effect). 
Moreover, as the probalistic stability calculations depend on the statistical concept of completeness, the complete damage system has to be generated first by MAAT Hydro, in order to ensure its completeness. This process automatically creates a damage system folder containing all the necessary damages in a hierarchized form, resulting from current ship's transverse, longitudinal and horizontal subdivision.
Nevertheless, as MAAT Hydro's subdivision analysis depends on watertight bulkhead's layers, it is important to pay special attention to respect the layer conventions onwhich the automatic damage system generation depends (check the subdivision layers on [Ship] tab's [Hydro] page and respect them during ship's subdivision): 
  • Transverse watertight bulkhead's outlines must be located in the 'Transverse Bulkhead' layer to be identified.
  • Longitudinal watertight bulkhead's outlines must be located in the 'Longitudinal Bulkhead' layer to be identified. 
  • Horizontal watertight bulkhead's outlines must be located in the 'Horizontal Bulkhead' layer to be identified.
In order to restrain as much as possible potential watertight bulkhead identification errors, any solid intersection by a canonic plane  ('/Solid/Split Ortho Plane' function) automatically selects the appropriate layer provided that the following sectionning mnemonics are used:
  • T for a Transverse watertight section.
  • L for a Longitudinal watertight section.
  • D (Deck) for a horizontal watertight section.
  • H for a non watertight section.
The other MAAT sectionning functions assume that the sectionned face layer is the layer currently displayed in the lower left selector; it is therefore important not to forget selecting the appropriate layer before non canonic solid intersections if they have to be considered as watertight.
As the automatic damage system generation only depends on this layer convention, it is important to pay as much attention as possible to the sectionning layers as to the compartment's geometry.
When ship's model is correctly subdivided, it is also possible to mark the compartments to flood progressively by setting their layer to the progressive flooding layer, controlled on [Ship] tab's [Hydro] page, with the number of intermediate flooding stages (take care to restrain this number as much as possible, as it may increase significantly the calculation time. Moreover, you must also remember that, as this layer will be responsible of compartment's liquid effect, its density must be equal to inflow's density, which is normally the default setting).
The wire descriptors  used for the regular stability  calculations (silhouette, sheer lines, openings, ...) must of course also be included in the model with their appropriate layer. 
At last, certain other layers displayed on [Ship] tab's [Hydro] page also allow controlling ship's behaviour: 
  • Unwettable Line Layer: The immersion of such a line (evacuation route etc...) at damaged ship's equilibrium will automatically force damage survivability Sito 0. 
  • Fatal Compartment Layer: Such compartment's damage (containing vital equipments, etc...) will automatically force damage survivability Si to 0. 

3: Complementary Modelling:

When the ship compartments are defined by this way, 3 loading conditions must be created, representing the standard 'light', 'partial' and 'service' loads. Two points must be carefully checked: 

  • The 'service' loading condition as well as the 'partial' one must correspond to a zero trim floatation on the intact ship. Conversely, the 'light' loading condition  has no contraint. 
  • As these loading conditions allow controlling the compartment permeabilities in addition to the masses and ship silhouette (and tank's contents, which are normally unused for such a calculation), it is important not to forget setting these data correctly if necessary, before creating each loading condition: Indeed, compartment permeabilities (like cargo holds) may vary depending on the loading condition, as well as ship's silhouette, and they must then be preliminarily configured.  
As soon as the 3 loading conditions needed by the probabilistic calculations are created in the current folder, in addition to the compartments described below, the damage system can, at last, be created by selecting '/Tools/Damage Stability/Make Damage System/SOLAS 2009 (PS)' in order to generate a Portside damage system (resp. /Tools/Damage Stability/Make Damage System/SOLAS 2009 (SB) for a Starboard system, when ship subdivision is not symmetric). 

MAAT Hydro then scans all the present watertight bulkheads according to their layer in order to automatically create a folder containing all the damages corresponding to the complete damage system needed by the probabilistic calculations. 

A subdivision report is then also generated in the 2D browser, allowing to check ship's subdivision accurately before calculating the damage stability itself. 

It may also be interesting to expand and control accurately the structure of the damage system in the 3D browser (double clicking on a damage makes activates it and highlights the damaged compartments) before starting the calculation, in order to identify potential layer errors. 

By double clicking in the 3D browser, any of the created damages can therefore be immediately combined with any loading condition, allowing to calculate and control a few preliminary stability curves individually (/Tools/Transverse Stability)before calculating the complete damage system. The optimal heel range can also be determined ytough this preliminary step.

As long as they have not been calculated, all the created damages will appear in white in the 3D browser. Depending on the control statements present in the STC stability criterion, their color will be set from green to red after calculation, depending ongdamage's survivability, an overprinted exclamation mark identifying the fatal damages. 

It is important to mention that these damages can be renamed freely in the 3D browser but, as the damage system's structure is one of the key concepts of the probabilistic process, allowing a realistic survivabilty calculation, they must never be moved or duplicated in order to preserve system's completeness. 

System's data structure controls reports layout, the smallest (resp. greatest) calculated Si being automatically selected in tree's extremities.

IMPORTANT: Ship components (compartments, lines, etc...) names and pathes must no longer be modified after having created  loading and damage conditions, otherwise the renamed elements will no longer remain automatically controllable during the calculation.

At last, it is important to mention that, due to damage system's logic, certain damages may be duplicated in various tree branches, an optimization mechanism allowing to calculate all of them once.

 4 : Calculation and Results:

Starting the probabilistic damage stability calculation for all the damage and the 3 loading conditions present in the current directory can be done by selecting  '/Tools/Damaged Stability/Calculate Damage System' in the menu bar and by clicking on the damage system node in the 3D browser (the "SOLAS 2009 Damages (PS/SB)" folder which has been created below).

If the calculations are heavy, they can be processed branch by branch, provided that the final calculation will have to operate on the complete damage system in order to provide a complete calculation report (the previously calculated damages won't be re-calculated, so the global results are obtained very quickly as soon as all the branches have been already calculated).

When the damage system or sub-system has been selected in the 3D browser, a dialog box allows entering the calculation data:

  • Click on the [Criterion] button to select the 'SOLAS' script.
  • Read the correponding prompts and enter the necessary data in the right input grid, particularly the number of passengers and life raft's moment
  • Define an optimal heel range and step in the 'Calculation Range' frame and select the positive heel side, in relation with damage system's side (Port side or Starboard). The heel range must be wide enough in order to allow calculating the criterion for every damage / loading combination, otherwise corresponding survivabilities will be ignored in the final total. Don't forget that a too large heel range will make calculations longer for nothing. Moreover, the heel step must also be small enough to allow a sharp Si calculation, but not too small in order to avoid too long calculations . A [-10°, 80°] heel range with a 2.5° / 5° step is usually a good compromise.
  • It is advised to check the "List All Calculated Damages" checkbox in order to list the results for all the calculated damages in the reports instead of only the worst ones.
The damage calculation process will start as soon as the [OK] button will be clicked. As it may last hours (or even days in extreme cases...), depending on system's computation power, model's complexity and memory size, damage system's complexity and selected calculation options (integration resolution, heel range, number of intermediate flooding stages, etc...), its progression can be controlled in the command line.
After this potentially long calculation step,  different folders and documents are created in the 2D browser, containing all the calculated reports in order to make them viewable, printable and exportable. Moreover, the damages, initially white in the 3D browser, will then be become coloured, according to their Si survivability (see below): Green damages are the most survivable and the red ones are the less, an overprinted exclamation mark identifying the fatal damages. This damage colour is fully controllable by script and users can redefine it if necessary(see STC Scripting Manual for details) .
When the damage system has been totally or partially calculated (and its damages are coloured), it is advised to save the project immediately in order to avoid forgetting it and wasting all these results (the amount of data encapsulated in the damage nodes after calculation as well as the numerous reports created significantly increase the file size).
In addition to the general reports created in the 2D browser at the end of the calculation process, calculated (coloured) damages can also be controlled individually for a detailed damage analysis.  Just select the damage or damage system branch you want to control in the 3D browser and select '"Show Damage Analysis" in the right click popup menu. This function is also available in the menu bar (/Tools/Damaged Stability/Show Damage Analysis).
In all cases, attention must be paid not to select a too large amount of damages in order to avoid overflowing the system with a potentially huge amount of detailed reports (3 reports / damage).


Why can negative densities appear on a Weight Curve ?            

MAAT Hydro calculates the weight curve by stacking individual mass distributions in the calculated folder.

The [Data] page shows that each mass is defined, longitudinally, by the Gx position of its center of gravity and by the Xmin / Xmax positions of its extremities, the load density being supposed to vary linearly between these extremities, depending on Gx's relative location:
  • Constant load density when Gx is located on the [Xmin, Xmax] middle.
  • Null density at Xmin and maximum at Xmax when Gx = Xmin + (Xmax-Xmin)/3
  • Maximum density at Xmin and null density at Xmax when Gx = Xmax - (Xmax-Xmin)/3
When Gx is out of this interval, its nearest extremity's density becomes as negative as it moves away.
It is therefore a good habit to confine Gx between Xmin + (Xmax-Xmin)/3 and Xmax - (Xmax-Xmin)/3. if you wish to avoid such negative load densities.
When such a situation cannot be avoided, just replace the singular mass by a few equivalent sub-masses whose induividually linear distribution will allow defining a more realistic polygonal distribution without any negative density. 
Negative masses can also be used for representing forces like stay tension, but must be used with care


How to display certain important hydrostatic data (free-boards, metacenters, floatation, waterlines, longitudinal strength…) in real time?            

In addition to the hydrostatic parameters displayed in the real-time hydrostatic window, several high level options allow a visual complementary control. 

These display options are available in the contextual menu that appear by a right click on the “Hydro Data” window title. They are often very useful, but don’t forget that they may significantly slow down MAAT Hydro’s response time: 

  • Show Floatation / Hide Floatation: Allows displaying / hiding float’s intersection with current floatation (best when combined with "Show Center Curve" option). 
  • Show BMT / Hide BMT: Allows displaying / hiding the transverse Metacentric radius and the local CB arc (may be hidden when rendered in the “opaque” mode. 
  • Show BML / Hide BML: Allows displaying / hiding the longitudinal Metacentric radius and the local CB arc (may be hidden when rendered in the “opaque” mode. 
  • Show Center Curve / Hide Center Curve: Allows displaying / hiding the center curve (i.e. curve joining centers of immersed stations). This option is best when combined with “Show Floatation” or “Show Heeled Lines”. 
  • Show Freeboards / Hide Freeboards: Allows displaying / hiding the minimum freeboard of all the freeboard and opening lines according to current floatation. 
  • Show Downfloodings / Hide downfloodings: Allows highlighting the downflooded compartments (i.e. compartments whose opening is immersed). 
  • Show Free Surfaces / Hide Free surfaces: Allows displaying / hiding current tank’s free surface. 
  • Show Tanks / Hide Tanks: Allows to show / hide tank’s silhouettes in the hydrostatic window. 
  • Show Longitudinal Strength / Hide Longitudinal Strength: Allows displaying / hiding the current longitudinal strength in the ship beam. This option is meaningless when no realistic mass distribution is available and / or when ship’s floatation doesn’t correspond to a longitudinal balance. 
  • Show Heeled Lines / Hide Heeled Lines: Allows displaying / hiding the heeled and trimmed ship lines (best when combined with "Show Center Curve" option). 

Moreover, the “Copy” function allows transferring current hydrostatic display to other applications like “MS Word”, thanks to Windows clipboard. 

 At last, the "CSV Export" function allows ewporting the current data to EXCEL thanks to the CSV format.