Probabilistic Analysis of Safe Return to Port (Public Version)

Probabilistic Analysis of Systems Availability

Designing Ships for Post-Damage Return to Port

Safe Return To Port• Ship abandonment is a very dangerous

undertaking. There is a broad category of accidents for which it is safer to stay within the ship in the wake of the accident than abandoning her in lifeboats.

• In such cases ships must return to a nearby port to offload her occupants. This return journey must, despite the damages, satisfy some basic minimum conditions in order to be safely viable.

Functionality Critical to Safe Return to Port

• Habitability : The ability of the ship to ensure health and safety of her occupants.

• Propulsion and Steering : The ability of the ship to travel to one of the nearest ports for offloading.

Conditions for Viable Return to Port

• A large ship has to serve as a (temporarily) self-sufficient microcosm of urban settlements in terms of facilities such as fresh-water supply, food supply, illumination, sewage, air conditioning, waste treatment, communication, fire safety services, emergency medical provisions etc.

• Most of these provisions are critically important for people to live in the ship an extended period of time.

A Natural Next Step from Index-A

• Index-A : Probability of post-damage sustenance of stability.

In addition we could target higher degrees of robustness through a similar approach to other functions besides ‘stability’.

• Index-X : Probability of post-damage sustenance of X.

X could be functions such as habitability, propulsion, steering.

Dependency Tree Based AnalysisThe dependency tree for a system represents a Boolean

expression describing whether the system can function properly based on the condition (``failure'' or ``working'') of its components and dependencies.

Here the systems don’t necessarily refer to tangible machinery or equipment – they can be functions or facilities.

The high level functionality is expressed in terms of their dependency on their sub-functions or sub-systems.

Dependency Model for Safe Return to Port

“Safe return to port” would be the top level function, and hence the root of the tree.

The leaves of the tree would be localized subsystems or machinery components.

Increasing Level of Detail

High Level Functionality

Localized Subsystems

Dependency Notation (the negated form of usual fault trees)

Z works if all of A,B and C work.

A

B

C

A

B

C

Z

Z works if of A or B or C works.

B

C

Z

A

Z works if A works and, B or C works.

Z

Locality of SystemsBelow a certain level of detail, systems and sub-systems

are localized, i.e. they are located in a confined region onboard the ship.

Several high level systems and functions are not localized – they have their components spread out widely across the ship.

In order to associate systems with damage cases, it is necessary to reason about their location in the shipboard environment. The damages are expressed in terms of subdivisions of the shipboard space. This is addressed by the spatial data model of Helios.

Damage CasesThe damage cases defined for probabilistic damage stability are the

damage cases considered for probabilistic analysis of safe return to port. Maybe the non-survivable cases could be excluded because they render return-to-port irrelevant.

If p(i) is the probability of the i-th damage case, and g(F,i) is a function that takes the value 1 if the functionality F fails in the i-th damage case and takes the value zero otherwise.

Then we can define a probabilistic failure index for that functionality as follows:

idx(F)=

N

i

N

i

ip

iFgip

1

1

)(

),()( [as Proposed for simplicity, but the expression could be slightly different if Monte-Carlo trials are used to accommodate probabilistic failure rates for flooded systems]

Towards Computing the Probabilistic Index

As to what will be the exact mathematical formulation for the index may be decided at a later stage, but to start with we must devise the technology for evaluation of system availability vis-à-vis damage situations.

We would need a computing tool that allows (i) modelling of the ship systems and their inter-dependence, (ii) modelling of damage cases and their probabilities and (iii) calculate availability of systems and functions relevant to “return to port” for each damage case.

An extension of our simulation tool viz. Helios has been developed to address the above. It allows both the modelling and analysis tasks involving ship systems and damage cases.

The Modelling Task

Produce a labelled drawing showing the subdivisions along with their names as labels.

Hull surface (in IGES or any other surface format is nice to have.

Loading the 3d Subdivision in Helios

The drawing is automatically converted into a 3D spatial database faithfully representing the subdivisions and located objects.

Interactive 3d interface is available for grouping of spaces, if necessary.

The hull or superstructure does not play a role in the computation involved in this analysis but may be used for virtual reality look-and-feel.

Placement of SystemsSystems can be placed on the spatial subdivision model by mouse clicks.

They may be moved around, renamed or deleted interactively.

All such interactive operations are recorded in a journal file for inclusion in the model file.

One can also assign the probability of failure of a system in the event of its space being flooded. This is 1.0 by default but may be 0.0 or any other value for sealed systems.

Specification of System Functionality and Interdependence

Interdependence of systems can be entered by mouse clicks.

Simple and/or dependencies can be specified only using mouse clicks ( e.g. Select n systems, and click the and-dependency hyperlink, this specifies the first selected system to be dependent on all of the rest of the selected system through an and node.

More complicated dependencies may be typed in easily, e.g.

{& A {| B C}} represents the following tree branch :

B

C

A

The dependencies may also be scripted as follows:

declareDependency Z {& A {| B C}}

Systems may be declared as non-local, by which they are not directly damaged by the location of their graphic avatar in a damaged compartment.

Description of Damage Cases

Damage cases as used for index-A calculation may be used directly.

Since “return to port” is relevant to survivable damages, severe damage cases may be excluded.

Probability Case ID Rooms Flooded

The Required Information Model

R7002

3D Subdivision/GA

Systems Dependency and Location

Room Names

Damage Cases

B

C

A

Graphic User Interface : Just another Helios Input Page

Display of Model Structure and Results

•The dependency graph and its expanded form is visualized and interactively manipulated.

•Results are displayed on the 3d viewer and dependency graph viewer.

•Numerical results are also saved in an Excel file for post-processing.

The Algorithm used for Availability Calculation

The dependency expressions are syntactically expanded as much as possible. That is :

If A refers to {| B {& P Q}}, B refers to {& R Q} and Q refers to {| M N} then A would be expanded into

{| {& R {| M N}} {& P {| M N}}, assuming M,N,P,R are leaf level components. The failures are also translated into such expanded expressions. For example

the failure of Q becomes the expression {~ {| M N}}.

Binary decision diagrams (BDDs) are made out of such expanded expressions, where the leaf level systems represent independent Boolean variables. System expressions are composed by conjunction (i.e. &-node) with failure expressions to give BDDs representing the state of the systems.

Satisfiability of such a BDD for a system is equivalent to the propositional value that it can work. Satisfiability problem for a BDD is a simple vertex reachability problem which has been implemented within our system.

Binary Decision DiagramBased on the idea of Shannon expansion i.e. the following expansion:

f(X1,X2,...,Xn) = X1f(1,X2,...,Xn) + X1'f(0,X2,...,Xn) followed by the expansion for two variables, and noting that expansion can be continued for any number of variables.

Shanon expansion leads to a tree structure which can be folded into a DAG by eliminating duplicate sub-expressions. Whether folded or not, “SATISFIABILITY” of the original expression is equivalent to “REACHABILITY” on this graph. The general problem itself is NP-hard (i.e. there is no guarantee of efficient solution yet), but BDDs are “usually” efficient for most expressions.

f(x1, x2, x3) = -x1 * -x2 * -x3 + x1 * x2 + x2 * x3

http://upload.wikimedia.org/wikipedia/commons/1/14/BDD_simple.svg

References• Fault Tree Handbook For Aerospace Applications: Dr. Michael

Stamatelatos, NASA HQ, OSMA, Dr. William Vesely, SAIC, Report prepared for NASA Office of Safety and Mission Assurance, NASA Headquarters, Washington, DC 20546, August, 2002

• Graph Based Algorithms for Boolean Function Manipulation, Randal Bryant, IEEE Transactions on Computers, C-35-8, pp. 677-691, August, 1986

• Fault Tree Analysis and Binary Decision Diagrams, R. Sinnamon and J. Andreas, Proceedings of the Reliability and Maintainability Symposium, January 1996, pp 215-222.

• New Algorithms for Fault Tree Analysis, A. Rauzy, Reliability Engineering and System Safety, Vol. 40, 1993, pp 203-211.

• The Synthesis of Two-Terminal Switching Circuits, Claude Shannon, Bell System Technical Journal, vol.28, pp.59-98

System Availability Analysis of [Name Witheld]

• We have received a large dataset of drawings and documents and are currently in the process of interpreting the data.

• There was an initial phase of difficulty with

interpreting the data, it has been addressed by collection of data from the site.

• Systems availability analysis has been carried out with the collected data.

Data Collection

[Name Witheld] went to [Name Witheld] and collected data on the systems locations and inter-dependence.

He came back with a 880 node dependency tree involving 100 sub-systems of steering and propulsion.

Progress on [Name Witheld]• The ship subdivision and the damage cases have been

modelled for the availability analysis tool.

• Some ambiguities in the subdivision labels have bee resolved in consultation with Tony.

• A new feature of propagating damage along “compartment connections” was necessary for modelling the damage cases of [Name Witheld].

• Availability analysis has been performed for the propulsion+steering systems onboard [Name Witheld].

• Systems spanning multiple rooms is supported.

Damage Cases for [Name Witheld]

560 damage cases have been modelled

657 three-dimensional sub-divisions have been modelled.

Probabilistic Analysis of Propulsion and Steering

100 systems and functions were defined along with their dependencies.

Dependency Tree

Dependency Graph

and Systems Placement

Constructed from Data

Collected Onsite

Individual and Aggregated Damage Cases

Effect of individual damage cases can be studied.

Aggregated distribution of damage cases can be used as a design aid.

Bayesian analysis for probabilistic quantification criticality of components is being investigated.

Systems Placement MethodSystem placement can be made according to damage probability distributions (joint distributions over space pairs).

Results (1/14)

% of damage cases that lead to failure to return to port

for each damage category defined by the number of damage compartments

% Cases of Failure to Return to Port

0.0010.0020.0030.0040.0050.0060.0070.0080.00

1 co

mpartm

ent d

amag

e

1 to

2 co

mpa

rtmen

t dam

age

2 co

mpartm

ent d

amag

e

1 to

3 co

mpa

rtmen

t dam

age

2 to

3 co

mpa

rtmen

t dam

age

3 co

mpartm

ent d

amag

e

1 to

4 co

mpa

rtmen

t dam

age

2 to

4 co

mpa

rtmen

t dam

age

3 to

4 co

mpa

rtmen

t dam

age

4 co

mpartm

ent d

amag

e

Damage Case Category

Fa

ilure

in %

-ag

e o

f c

as

es

Results (2/14)

Probability of Failure to Return to Port

0.00000.05000.10000.15000.20000.25000.30000.35000.40000.45000.5000

1 compartmentdamage

2 compartmentdamage

All SOLAS 2009damage


Pro

ba

bili

ty o

f F

ailu

re

Probability of Steering System Failure

0.0000

0.0200

0.0400

0.0600

0.0800

0.1000

0.1200

0.1400

0.1600

1 compartment damage 2 compartment damage All SOLAS 2009damage


Pro

ba

bili

ty o

f F

ailu

re

Probability of failure to return to port for three significant damage categories, and probability of failure of the two main sub-functions thereof, viz. Propulsion and Steering.

Probability of Propulsion Failure

0.0000

0.0500

0.1000

0.1500

0.2000

0.2500

0.3000

0.3500

0.4000

1 compartmentdamage

2 compartmentdamage

All SOLAS 2009damage


Pro

ba

bili

ty o

f F

ailu

re

Results (3/14) : RTP Failure Probability for All Solas 2009 cases Failure Probability : All Systems

Systems/Functions

_Return_To_Port_

Port_Propeller

Prop_Shaft_P

Propeller_Pitch_Control_P

Prop_Shaft_S

Propeller_Pitch_Control_S

Propulsion_System

Starboard_Propeller

Cooling_Water_System_P

Engine_Driven_Pump_P

Gearbox_P

Hydraulic_Control_P

Lube_Oil_System_P

Main_Engine_3_P

Main_Engine_4_P

Prelube_Pump_P

Pumps_Engine_Heat_Exchangers_P

SW_Cooling_System_P

SW_Heat_Exchangers_P

Booster_Pumps_S

Separator_Settling_Tank_Service_Tank_Lub_Oil_Separation_S

AE_3_FW_Cooling_Lub_Oil_Storage_P


Main_Switchboard_P

AE_1_FW_Cooling_Lub_Oil_Storage_S

Cooling_Water_System_S

Engine_Driven_Pump_S

Gearbox_S

Hydraulic_Control_S

LO_Transfer_Pump_P

LO_Transfer_Pump_S

Lube_Oil_System_S

Main_Engine_1_S

Main_Engine_2_S

Main_Switchboard

Main_Switchboard_S

Prelube_Pump_S

Pumps_Engine_Heat_Exchangers_S

SW_Cooling_System_S

SW_Heat_Exchangers_S


HFO_Heating_Boiler_S

HFO_Heating_Boiler_P

AFS_Separator_Settling_Tank_Service_Tank_Lub_Oil_Separation_S

Port_Rudder_P

Rudder_Stock_P

Rudder_Stock_S

Starboard_Rudder_S

Steering_System

Service_Tank_S

AFS_Separator_Settling_Tank_Service_Tank_Lub_Oil_Separation_P

Fuel_System

Separator_Settling_Tank_Service_Tank_Lub_Oil_Separation_P

Service_Tank_P

FW_Cooling_S

Pneumatic_Fuel_Pump_Blackout_Pump_S

LO_Tank

Air_System_Control_P

Engine_Casing_P

Exhaust_Systems_P

Engine_Casing_S

Exhaust_Systems_S

Gearbox_Lubrication_P

Gearbox_Lubrication_S

Heat_Exchangers_S

ME_Room_Reservoir_S

Self_Contained_Lubricating_Oil_System_P

Self_Contained_Lubricating_Oil_System_S

AE_Room_Reservoir_P

Emergency_Closing_Control

Pro

babi

lity

0

0.2

0.4

Results (4/14) : Propulsion Failure Probability for All Solas 2009 cases Failure Probability : Propulsion Systems

Systems/Functions

Port_Propeller

Prop_Shaft_P


Prop_Shaft_S


Propulsion_System

Starboard_Propeller


Engine_Driven_Pum

p_P

Gearbox_P

Hydraulic_Control_P

Lube_Oil_System

_PMain_Engine_3_P

Main_Engine_4_P

Prelube_Pump_P


SW_Cooling_System

_PSW

_Heat_Exchangers_P

Booster_Pumps_S




Main_Sw

itchboard_P



Engine_Driven_Pum

p_S

Gearbox_S

Hydraulic_Control_S

LO_Transfer_Pump_P

LO_Transfer_Pump_S

Lube_Oil_System

_SMain_Engine_1_S

Main_Engine_2_S

Main_Sw

itchboard

Main_Sw

itchboard_S

Prelube_Pump_S


SW_Cooling_System

_SSW

_Heat_Exchangers_S


HFO

_Heating_Boiler_S

HFO

_Heating_Boiler_P


Service_Tank_S


Fuel_System


Service_Tank_P

FW_Cooling_S

Pneumatic_Fuel_Pump_Blackout_Pum

p_S

LO_Tank

Air_System

_Control_P

Engine_Casing_P

Exhaust_System

s_P

Engine_Casing_S

Exhaust_System

s_S



Heat_Exchangers_S

ME_Room_Reservoir_S

Pro

babi

lity

0

0.2

0.4

Results (5/14) : Steering Failure Probability for all Solas 2009 cases Failure Probability : Steering Systems

Systems/Functions

Booster_Pumps_S



Main_Switchboard_P


Main_Switchboard

Main_Switchboard_S





Port_Rudder_P

Rudder_Stock_P

Rudder_Stock_S

Starboard_Rudder_S

Steering_System

Service_Tank_S


Service_Tank_P

FW_Cooling_S


Heat_Exchangers_S

ME_Room_Reservoir_S

AE_Room_Reservoir_P

Booster_Pumps_P

HFO_Storage_Tanks

Emergency_Steering_S

Hydraulic_Pumps_P

Hydraulic_Pumps_S

Steering_Gear_P

Steering_Gear_S

AE_Control_System

ECR

Working_Air_System_Service_Air

Bridge

Control_System

Em_Gen_Fuel_Tank

Em_Gen_PumpsControl_System

Emergency_Generator

Emergency_Switchboard

Pro

babi

lity

0

0.1

0.2

0.3

Results (6/14) : Common Systems for all Solas 2009 cases Failure Probability : Shared Systems

Systems/Functions

Booster_Pumps_S



Main_Sw

itchboard_P


Main_Sw

itchboard

Main_Sw

itchboard_S


HFO

_Heating_Boiler_S

HFO

_Heating_Boiler_P


Service_Tank_S


Service_Tank_P

FW_Cooling_S


p_S

Heat_Exchangers_S

ME_Room_Reservoir_S

AE_Room_Reservoir_P

Booster_Pumps_P

HFO

_Storage_Tanks

AE_Control_System

ECR

Working_Air_System

_Service_Air

Bridge

Em_Gen_Fuel_Tank

Em_Gen_Pum

psControl_System

Emergency_Generator

Emergency_Sw

itchboard

Pro

babi

lity

0

0.1

0.2

0.3

Results (7/14) : RTP Failure Probability for All 1 CMPT cases Failure Probability : All Systems

Systems/Functions

_Return_To_Port_

Port_Propeller

Prop_Shaft_P


Prop_Shaft_S


Propulsion_System

Starboard_Propeller


Engine_Driven_Pum

p_P

Gearbox_P

Hydraulic_Control_P

Lube_Oil_System

_PMain_Engine_3_P

Main_Engine_4_P

Prelube_Pump_P


SW_Cooling_System

_PSW

_Heat_Exchangers_P

Booster_Pumps_S




Main_Sw

itchboard_P

Port_Rudder_P

Rudder_Stock_P

Rudder_Stock_S

Starboard_Rudder_S

Steering_System



Engine_Driven_Pum

p_S

Gearbox_S

Hydraulic_Control_S

LO_Transfer_Pump_P

LO_Transfer_Pump_S

Lube_Oil_System

_SMain_Engine_1_S

Main_Engine_2_S

Main_Sw

itchboard

Main_Sw

itchboard_S

Prelube_Pump_S


SW_Cooling_System

_SSW

_Heat_Exchangers_S


HFO

_Heating_Boiler_S

HFO

_Heating_Boiler_P

FW_Cooling_S


Service_Tank_S


Fuel_System


Service_Tank_P

LO_Tank


Hydraulic_Pum

ps_P

Hydraulic_Pum

ps_S

Steering_Gear_P

Steering_Gear_S

Air_System

_Control_P

Engine_Casing_P

Engine_Casing_S

Exhaust_System

s_P

Exhaust_System

s_S



Heat_Exchangers_S

ME_Room_Reservoir_S

Pro

babi

lity

0

0.1

0.2

0.3

0.4

Results (8/14) : Propulsion Failure Probability for 1 CMPT cases Failure Probability : Propulsion Systems

Systems/Functions

Port_Propeller

Prop_Shaft_P


Prop_Shaft_S


Propulsion_System

Starboard_Propeller


Engine_Driven_Pump_P

Gearbox_P

Hydraulic_Control_P

Lube_Oil_System_P

Main_Engine_3_P

Main_Engine_4_P

Prelube_Pump_P


SW_Cooling_System_P

SW_Heat_Exchangers_P

Booster_Pumps_S




Main_Switchboard_P



Engine_Driven_Pump_S

Gearbox_S

Hydraulic_Control_S

LO_Transfer_Pump_P

LO_Transfer_Pump_S

Lube_Oil_System_S

Main_Engine_1_S

Main_Engine_2_S

Main_Switchboard

Main_Switchboard_S

Prelube_Pump_S


SW_Cooling_System_S

SW_Heat_Exchangers_S




FW_Cooling_S


Service_Tank_S


Fuel_System


Service_Tank_P

LO_Tank

Air_System_Control_P

Engine_Casing_P

Engine_Casing_S

Exhaust_Systems_P

Exhaust_Systems_S



Heat_Exchangers_S

ME_Room_Reservoir_S

Self_Contained_Lubricating_Oil_System_P

Pro

babi

lity

0

0.1

0.2

0.3

Results (9/14) : Steering Failure Probability for all 1 CMPT cases Failure Probability : Steering Systems

Systems/Functions

Booster_Pumps_S



Main_Switchboard_P

Port_Rudder_P

Rudder_Stock_P

Rudder_Stock_S

Starboard_Rudder_S

Steering_System


Main_Switchboard

Main_Switchboard_S




FW_Cooling_S


Service_Tank_S


Service_Tank_P


Hydraulic_Pumps_P

Hydraulic_Pumps_S

Steering_Gear_P

Steering_Gear_S

Heat_Exchangers_S

ME_Room_Reservoir_S


HFO_Storage_Tanks

Booster_Pumps_P

AE_Room_Reservoir_P

AE_Control_System

ECR


Bridge

Control_System

Em_Gen_Fuel_Tank


Emergency_Generator


Pro

babi

lity

0

0.1

0.2

Results (10/14) : Common Systems for all 1 CMPT cases Failure Probability : Shared Systems

Systems/Functions

Booster_Pumps_S



Main_Switchboard_P


Main_Switchboard

Main_Switchboard_S




FW_Cooling_S


Service_Tank_S


Service_Tank_P

Heat_Exchangers_S

ME_Room_Reservoir_S


HFO_Storage_Tanks

Booster_Pumps_P

AE_Room_Reservoir_P

AE_Control_System

ECR


Bridge

Em_Gen_Fuel_Tank


Emergency_Generator


Pro

babi

lity

0

0.1

0.2

Results (11/14) : RTP Failure Probability for All 2 CMPT cases Failure Probability : All Systems

Systems/Functions

Port_Propeller

Prop_Shaft_P


_Return_To_Port_


Engine_Driven_Pum

p_P

Gearbox_P

Hydraulic_Control_P

Lube_Oil_System

_PMain_Engine_3_P

Main_Engine_4_P

Prelube_Pump_P


SW_Cooling_System

_PSW

_Heat_Exchangers_P

Prop_Shaft_S


Propulsion_System

Starboard_Propeller

Booster_Pumps_S




Main_Sw

itchboard_P



Engine_Driven_Pum

p_S

Gearbox_S

Hydraulic_Control_S

LO_Transfer_Pump_P

LO_Transfer_Pump_S

Lube_Oil_System

_SMain_Engine_1_S

Main_Engine_2_S

Main_Sw

itchboard

Main_Sw

itchboard_S

Prelube_Pump_S


SW_Cooling_System

_SSW

_Heat_Exchangers_S


HFO

_Heating_Boiler_S

HFO

_Heating_Boiler_P


FW_Cooling_S

Service_Tank_S


Fuel_System


Service_Tank_P

LO_Tank

Air_System

_Control_P

Engine_Casing_P

Engine_Casing_S

Exhaust_System

s_P

Exhaust_System

s_S



Heat_Exchangers_S

ME_Room_Reservoir_S

Self_Contained_Lubricating_Oil_System

_PSelf_Contained_Lubricating_Oil_System

_SPneumatic_Fuel_Pump_Blackout_Pum

p_S

Emergency_Closing_Control

AE_Room_Reservoir_P

HFO

_Storage_Tanks

Port_Rudder_P

Rudder_Stock_P

Rudder_Stock_S

Starboard_Rudder_S

Pro

babi

lity

0

0.2

0.4

Results (12/14) : Propulsion Failure Probability for 2 CMPT cases Failure Probability : Propulsion Systems

Systems/Functions

Port_Propeller

Prop_Shaft_P



Engine_Driven_Pum

p_P

Gearbox_P

Hydraulic_Control_P

Lube_Oil_System

_PMain_Engine_3_P

Main_Engine_4_P

Prelube_Pump_P


SW_Cooling_System

_PSW

_Heat_Exchangers_P

Prop_Shaft_S


Propulsion_System

Starboard_Propeller

Booster_Pumps_S




Main_Sw

itchboard_P



Engine_Driven_Pum

p_S

Gearbox_S

Hydraulic_Control_S

LO_Transfer_Pump_P

LO_Transfer_Pump_S

Lube_Oil_System

_SMain_Engine_1_S

Main_Engine_2_S

Main_Sw

itchboard

Main_Sw

itchboard_S

Prelube_Pump_S


SW_Cooling_System

_SSW

_Heat_Exchangers_S


HFO

_Heating_Boiler_S

HFO

_Heating_Boiler_P


FW_Cooling_S

Service_Tank_S


Fuel_System


Service_Tank_P

LO_Tank

Air_System

_Control_P

Engine_Casing_P

Engine_Casing_S

Exhaust_System

s_P

Exhaust_System

s_S



Heat_Exchangers_S

ME_Room_Reservoir_S

Self_Contained_Lubricating_Oil_System

_P

Pro

babi

lity

0

0.2

0.4

Results (13/14) : Steering Failure Probability for all 2 CMPT cases Failure Probability : Steering Systems

Systems/Functions

Booster_Pumps_S



Main_Switchboard_P


Main_Switchboard

Main_Switchboard_S





FW_Cooling_S

Service_Tank_S


Service_Tank_P

Heat_Exchangers_S

ME_Room_Reservoir_S


AE_Room_Reservoir_P

HFO_Storage_Tanks

Port_Rudder_P

Rudder_Stock_P

Rudder_Stock_S

Starboard_Rudder_S

Steering_System

Booster_Pumps_P

AE_Control_System

ECR



Hydraulic_Pumps_P

Hydraulic_Pumps_S

Steering_Gear_P

Steering_Gear_S

Bridge

Control_System

Em_Gen_Fuel_Tank


Emergency_Generator


Pro

babi

lity

0

0.1

0.2

0.3

0.4

Results (14/14) : Common Systems for all 2 CMPT cases Failure Probability : Shared Systems

Systems/Functions

Booster_Pumps_S



Main_Sw

itchboard_P


Main_Sw

itchboard

Main_Sw

itchboard_S


HFO

_Heating_Boiler_S

HFO

_Heating_Boiler_P


FW_Cooling_S

Service_Tank_S


Service_Tank_P

Heat_Exchangers_S

ME_Room_Reservoir_S


p_S

AE_Room_Reservoir_P

HFO

_Storage_Tanks

Booster_Pumps_P

AE_Control_System

ECR

Working_Air_System

_Service_Air

Bridge

Em_Gen_Fuel_Tank

Em_Gen_Pum

psControl_System

Emergency_Generator

Emergency_Sw

itchboard

Pro

babi

lity

0

0.1

0.2

0.3

0.4

Further Development• The development team has been extended to

develop a high quality visualisation and high-usability user interface for systems availability analysis. [Name Witheld] has been assigned to develop the new user interface for this tool.

• The following few slides give a few glimpses of this development.

• This tool will serve to make the systems availability analysis accessible to a wider audience – (e.g. so that it could be used by operational personnel to learn about criticality in the systems).

Dedicated Tool For Systems Availability Analysis

Systems and their Dependency are Marked by Line Diagrams

New Tool for Systems Availability Analysis

Systems can be placed and their connections routed in the 3D subdivision model of the ship.

Currently the data connection is set up using XML communication with helios.

Eventually it will be independent of helios.

Future Directions • This system would allow

block diagram method of modelling of system dependencies.

• This tool would also be used for coupling systems failure with flooding and fire models (similar to the way passenger evacuation is dynamically coupled with flooding and fire models).

• This line of development was initiated quite recently, we are expecting this to be a very efficacious design tool.

Thank You

Probabilistic Analysis of Safe Return to Port (Public Version)

Documents

binary decision diagrams

high level functionality

systems availability analysis

rtp failure probability

steering failure probability

propulsion failure probability

probabilistic analysis

damage case