Risk-Based Ship Design, Notas de estudo de Engenharia Naval

Baixe Risk-Based Ship Design e outras Notas de estudo em PDF para Engenharia Naval, somente na Docsity!

Risk-Based Ship Design

Editor

Prof. Apostolos Papanikolaou National Technical University Athens School of Naval Architecture & Marine Engineering Ship Design Laboratory Heroon Polytechniou St. 9 157 73 Athens Zografou Campus Greece papa@deslab.ntua.gr

ISBN: 978-3-540-89041-6 e-ISBN: 978-3-540-89042-

DOI 10.1007/978-3-540-89042-

Library of Congress Control Number: 2008939068

©c (^) Springer-Verlag Berlin Heidelberg 2009

This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable to prosecution under the German Copyright Law.

The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

Cover design: eStudio Calamar S.L., F. Steinen-Broo, Girona, Spain

Printed on acid-free paper

9 8 7 6 5 4 3 2 1

springer.com

Preface

Risk-based ship design is a new scientific and engineering field of growing inter- est to researchers, engineers and professionals from various disciplines related to ship design, construction, operation and regulation. Applications of risk-based ap- proaches in the maritime industry started in the early 1960s with the introduction of the concept of probabilistic ship’s damage stability. In the following, they were widely applied within the offshore sector and are now being adapted and utilized within the ship technology and shipping sector. The main motivation to use risk-based approaches is twofold: implement a novel ship design which is considered safe but – for some formal reason – cannot be approved today and/or rationally optimise an existing design with respect to safety, without compromising on efficiency and performance. The present book derives from the knowledge gained in the course of the project SAFEDOR (Design, Operation and Regulation for Safety), an Integrated Project under the 6th framework programme of the European Commission (IP 516278). The topic of SAFEDOR is risk-based ship design, operation and regulation. The project started in February 2005 and will be completed in April 2009. Under the coordination of Germanischer Lloyd, 52 European organizations – representing all stakeholders of the maritime industry – took part in this important R&D project. The present book does not aim to be a textbook for postgraduate studies, as con- tributions to the subject topic are still evolving and some time will be necessary until maturity. However, as the topic of risk-based design, operation and regulation is almost absent from today’s universities’ curricula, the book aims to contribute to the necessary enhancement of academic curricula to address this important subject to the maritime industry. Therefore, the aim of the book is to provide the readers with an understanding of the fundamentals and details of the integration of risk- based approaches into the ship design process. The book facilitates the transfer of knowledge from the research conducted within the SAFEDOR project to the wider maritime community and nurtures inculcation upon scientific approaches dealing with risk-based design and ship safety. The book is introduced by an overview of risk-based approaches to the maritime industry in Chap. 1 by Dr. Pierre C. Sames (Germanischer Lloyd). The risk-based

v

Contents

1 Introduction to Risk-Based Approaches in the Maritime Industry... 1 Pierre C. Sames

2 Risk-Based Ship Design...................................... 17 Dracos Vassalos

3 Regulatory Framework...................................... 97 Rolf Skjong

4 Risk-Based Approval........................................ 153 Jeppe Skovbakke Juhl

5 Methods and Tools.......................................... 195 Jørgen Jensen, Carlos Guedes Soares and Apostolos Papanikolaou

6 Applications................................................ 303 Dag McGeorge, Bjørn Høyning, Henrik Nordhammar, Apostolos Papanikolaou, Andrzej Jasionowski and Esa P¨oyli¨o

Acronyms and Glossary.......................................... 359

Authors Biography................................................. 365

Index............................................................. 369

vii

Contributors

Bjørn Høyning FiReCo AS, Fredrikstad, Norway

Andrzej Jasionowski Safety At Sea Ltd, Glasgow, UK, a.jasionowski@na-me.ac.uk

Jørgen Jensen Technical University of Denmark, Kgs. Lyngby, Denmark, jjj@mek.dtu.dk

Jeppe Skovbakke Juhl Danish Maritime Authority, Copenhagen, Denmark, jj@dma.dk

Dag McGeorge Det Norske Veritas AS, Oslo, Norway, dag.mcgeorge@dnv.com

Henrik Nordhammar Stena Rederi AB G¨oteborg, Sweeden

Apostolos Papanikolaou National Technical University of Athens, Athens, Greece, papa@deslab.ntua.gr

Esa P¨oyli¨o Deltamarin, Helsinki, Finland

Pierre C. Sames Germanischer Lloyd, Hamburg, Germany, pierre.sames@gl-group.com

Rolf Skjong Det Norske Veritas AS, NO-1322 Hørik, Norway, rolf.skjong@dnv.com

Carlos Guedes Soares Instituto Superior Tecnico, Lisbon, Portugal, guedess@mar.ist.utl.pt

Dracos Vassalos Universities of Glasgow and Strathclyde, Glasgow, UK, d.vassalos@na-me.ac.uk

ix

2 P.C. Sames

design process leads to “risk-based design”. As applied to the design of ships, risk- based design and approval was introduced by Bainbridge et al. (2004) and is the focus of this book. Innovation in the transportation industry (aerospace, automotive and rail indus- try) has to a large extent been driven by safety. As an example of the automotive in- dustry, crash-performance tests of independent authorities have shown to customers that large vehicles with integrated crash energy dissipating elements, airbags for side or frontal impact protection etc. provide increased safety in accidents. On the other hand, ship safety is well regulated at United Nations’ level by the International Maritime Organization (IMO) instead of relying on individual manufacturers’ or na- tional administrations’ responsibility for safety. However, the development of mar- itime safety regulations has until recently been driven mainly by individual events instead of a pro-active and holistic approach. Every major catastrophic accident, in particular those in the industrialized world, has led to a new safety regulation and subsequent design measures imposed by the IMO and the classification societies. Today however, a clear tendency to move from prescriptive to goal-based regula- tions is emerging. Changes in scientific and technological developments at an ever increasing pace and an overall better technical capability at a much larger scale are fuelling in- novation in the shipping sector to meet the demand for larger, more complex and specialized ships. This is taking place in an environment that is still fragmented, un- dermanned and intensively competitive, while society is more demanding on issues related to human safety and the protection of the environment. Safety could easily be undermined and the consequences could be disastrous. Therefore, the way safety is being dealt with is changing and with the adoption of holistic and risk-based ap- proaches to maritime safety, balancing the elements affecting safety cost-effectively and throughout the life cycle of the vessel, safety will be dealt with as a key aspect with serious economic implications rather than a simplistic add-on in the design process seeking compliance with prescriptive regulations. Fuelled by expected continuous growth of maritime transport and the need to provide sustainable shipping, economic opportunities drive proposals for ever more innovative ships and shipping concepts. Recent examples include cruise ships with huge shopping malls inside the superstructure and compressed natural gas trans- porters. With risk-based approaches firmly established in the maritime industry, ship owners will be able to implement those innovative ships and maritime transport so- lutions which (partly) cannot be approved today because of the current rules and reg- ulations’ prescriptive limitations. Shipyards and equipment manufacturers will also benefit from the introduction of risk-based approaches through enabling novel and optimized ships and systems incorporating new functions and materials. The bene- fits arise from the fact that yards acquainted with risk-based approaches are among the first to respond to the increasing demand from ship owners for those novel ships. In addition, production costs may be reduced through application of risk-based ap- proaches when, e.g., novel systems allow for improved modularization. Although the recent focus of applying risk-based design was to passenger ships, examples for cargo ships also exist (for example, MSC 76/INF.15 and MSC 82/23/3).

1 Introduction to Risk-Based Approaches in the Maritime Industry 3

1.1.2 An Enhanced Design Process

Risk-based ship design introduces risk analysis into the traditional design process aiming to meet safety objectives cost effectively. This is facilitated by use of ad- vanced computational tools to quantify the risk level of a particular design and its variants. Risk is used to measure the safety performance. With safety becoming measurable, the design optimization can effectively be expanded and a new objec- tive – minimize risk – is addressed alongside traditional design objectives relating to earning potential, speed and cargo carrying capacity. It is expected that with the in- troduction of safety as an objective into the design optimization process rather than being treated as a constraint, new technical solutions will be explored: the design solution space becomes larger. Even though, deriving from the above, risk-based design is principally associated with introducing safety objectives explicitly in the design process; two clearly dis- tinct motivations for risk-based design could be identified. First, it is the realization of an idea for a new transport solution which challenges (possibly outdated) rules – meaning that the new solution cannot be approved. Risk-based design and approval are then used to identify the issues and prove that the new solution is at least as safe as required. A requirement can be either based on a reference vessel or defined by specified risk acceptance criteria. This approach is exemplified within regula- tion 17 of SOLAS-II.2 on fire safety. This first variant of risk-based ship design has become widely known as “Safety Equivalence”. Second, it is the optimization of a rule-compliant vessel aiming to increase the level of safety at the same costs or to increase earning potential at the same level of safety. An example for this variant of risk-based design is optimization within the new probabilistic damage stability regulations. For both variants of risk-based design and for risk-based design in general, the same technology and frameworks are needed, which derive from the introduction of safety as an objective in the design process. First, a design methodology needs to be developed, aligned with the traditional design process that includes safety as objective and integrate any associated computational tools to quantify pertinent risks. Second, the regulatory framework must be in place to facilitate risk-based design – core elements of this are risk evaluation criteria which preferably should be agreed at IMO.

1.2 How Did It Start?

1.2.1 Probabilistic Damage Stability

Risk-based approaches in the shipping industry started with the concept of proba- bilistic damage stability in the early sixties, but it took more than a decade for this concept to be introduced in the SOLAS regulations (SOLAS74) as an alternative to deterministic damage stability regulations. SOLAS II-1, regulation 25, indicates

1 Introduction to Risk-Based Approaches in the Maritime Industry 5

operators to the risks connected with their activities, see for example Skjong (1999). The legislation requires that the authorities be allowed having insight into the decision-making processes of the individual enterprise, including policies and tar- get safety levels, and that they have access to all safety relevant documentation. The Petroleum Safety Authority of Norway then acts – as regulator- on situa- tions that are considered not acceptable, but does not approve the documentation or the safety targets (as in the United Kingdom); this is the responsibility of the owner. The approach is called “self-regulatory”. The Norwegian offshore regula- tions are designed to reflect that the operators have full responsibility for their activities. For the approval of offshore activities in the United Kingdom, a safety case has to be produced since 1992 for submission to the Health & Safety Executive. The primary objective of a safety case is to ensure an adequate level of safety for a par- ticular installation, based upon the management and control of the risks associated with it. A central feature of a safety case is that the owner takes responsibility for assessing the risks associated with his installation, and for documenting how his safety management system limits those risks to an acceptable level. The safety case regime is mandatory, i.e. operations cannot legally be commenced or continued until a safety case has been compiled by the owner and submitted to the official regulator for scrutiny and approval (Peachey 1999). A safety case will include a comprehensive description of the installation itself, and of its operation and the environment within which it operates. Risks will be quantified to the extent it is appropriate to do so. Risk acceptance criteria will be set, relevant to the installation and its operational context, and usually in accordance with the ALARP (As Low As Reasonably Practical) principle. Typically, for a new installation, a design safety case would initially be compiled. This would subsequently be developed and expanded into an operational safety case as the installation enters service. Thereafter, the safety case would normally be subject to regular review, with updating as necessary, to take account of changing conditions, ownership, activities, modifications, etc. The effectiveness of the safety management system is usually monitored and verified by means of regular audits, and compliance with the requirements of the safety case is checked by means of inspections.

1.2.3 Structural Reliability Analysis

The development of structural reliability analysis started as a new discipline in en- gineering in the seventies, when it was shown that a probabilistic theory could be developed that linked reliability to rules. Structural reliability analysis represents a risk-based framework for developing and documenting rules for structures. The the- ory has now been continuously developed over a period of over 35 years and it is supported by standardized methods, textbooks and related software tools. The basis for the methods and terminology may be found in CEN (2002).

6 P.C. Sames

In the maritime area, the DNV offshore rules were the first international standards applying the new knowledge (see for example the review book on use of Structural Reliability Analysis, (Sundararajan 1995, Skjong 1995) and the review on risk and reliability in marine structures by Guedes Soares 1998). This was linked to the de- velopment of all-year offshore operations in the North Sea, which required a higher reliability level than required in the American Petroleum Institute’s offshore stan- dards for the Gulf of Mexico where offshore structures were abandoned in case of hurricanes. In shipping there was little published systematic use of structural reliability anal- ysis for rule development or ship design apart from the European funded research project SHIPREL which advocated the use of reliability theory in codes and pro- posed a reliability based format based on ultimate strength (Guedes Soares et al. 1996). Starting around 2000, new rules for the hull girder capacity of oil tankers were developed using structural reliability analysis within the so-called Joint Tanker Project of three major class societies which resulted in the Common Structural Rules for tankers (IACS 2006). The approach and selected results were also submitted to IMO as MSC 81/INF.6.

1.2.4 Alternative Design and Arrangement for Fire

Safety (SOLAS II.2/17)

The development that resulted in SOLAS II.2, Regulation17, started already back in the late eighties with the design of the cruise ship “Sovereign of the Seas”, which had an atrium, a public space extending to three or more decks, within one fire zone. The approval of this ship involved a reference to the standard for equivalent arrange- ments under SOLAS I/5. The atrium solutions were extended to three fire zones in the design of the cruise ship “Voyager of the Seas” delivered in 1999 and again involved equivalence considerations and reference to SOLAS I/5 (Bahamas 2001). The large RoPax/Cruise ferry “Color Fantasy” and the Ultra-Voyager-class of ves- sels have atria extending over four fire zones, and using the new SOLAS II-2/ for approval. The freedom in design introduced by these regulations facilitates op- timization of various design parameters. Various software tools, e.g., for analyzing evacuation performance of passenger ships, have been developed and can be used in design optimization. Guidelines are published to direct the fire engineering analysis (IMO 2001).

1.2.5 Alternative Design for Oil Tankers (MARPOL Annex I-4/19)

Regulation 19 addresses double hull and double bottom requirements for oil tankers. However, paragraph 5 of Reg.19 states that other methods of design and construction of oil tankers may also be accepted as alternatives to the requirements prescribed

8 P.C. Sames

introduces requirements related to operator management, similar to the International Safety Management code (ISM 2004) and operation limits (good weather, near place of refuge and rescue facilities available).

1.2.7 Formal Safety Assessment

Formal Safety Assessment (FSA) has been developed as tool to support decision making at IMO. Following a UK proposal in 1993, guidelines for FSA were even- tually adopted for use in the IMO rule making process (IMO 2002b), following a series of trial applications according to the interim guidelines. The guidelines have been updated recently (IMO 2007). With FSA, the maritime industry followed oth- ers sectors in adopting a risk-based approach to support rule-making. FSA deliv- ers in a transparent way the costs and benefits of proposed changes to the regula- tory framework and supports decision makers at IMO. FSA comprise five interre- lated steps:

1. Identification of hazards
2. Assessment of the risks arising from the hazards identified
3. Identification of options to control the risks
4. Cost/benefit assessment of the risk control options
5. Recommendations for decision making

To date, only a couple FSA studies performed within the maritime industry re- sulted in IMO decisions. One early application was related to the provision of he- licopter landing areas (HLA) on passenger ships and the FSA showed these to be not cost-effective for non RoPax passenger ships. The requirement was eventually dropped, though many ships, including non-Ro-Ro passenger ships, have in the meantime an HLA installed. More prominent is the bulk carrier safety “story” when a couple of FSA studies were prepared which concluded, among other issues, that double skins are cost-effective, see MSC 76/23. However, this recommendation was later also not adopted. A recent FSA study on cruise vessel navigation (NAV 51/10) focused on events leading to collisions and groundings. It concluded in documenting a number of risk control options related to navigation as being cost-effective, among them ECDIS (Electronic Chart Display and Information System). A dedicated FSA study on ECDIS addressing also other ship types was performed following the FSA on cruise vessel navigation. It confirms the cost-effectiveness of ECDIS for selected cargo vessels; see MSC 81/24/5. A series of so called high level FSA studies were performed recently for main ship types as follows (with the INF-papers containing the full studies):

• Container vessels, submitted as MSC 83/21/2 and MSC 83/INF.
• Liquefied natural gas tankers, submitted as MSC 83/21/1 and MSC 83/INF.
• Cruise vessels, submitted as MSC 85/17/1 and MSC 85/INF.
• RoPax ferries, submitted to MSC 85/17/2 and MSC 85/INF.
• Oil tankers, submitted to MEPC 58/17/2 and MEPC 58/INF.

1 Introduction to Risk-Based Approaches in the Maritime Industry 9

1.2.8 Selected Recent Research Activities

Following a number of tragic accidents with RoPax ferries in Europe, research was initiated to study possible means to improve the safety of those vessels. A thematic network was established in 1997 to coordinate and align related European research projects, mainly those funded by the EU-Commission. The theme was called “De- sign for Safety” which called for integrating safety as an objective into the design process; and it can be seen as first version of risk-based ship design (Vassalos et al. 2000, University of Strathclyde 2003). Coordinated projects focused on de- velopment of tools to predict the safety performance in accidental conditions like, e.g., collision and grounding (e.g., Otto et al. 2001,Vanem and Skjong 2004a), bow door and green water extreme hydrodynamic loads (e.g., Sames et al. 2001, Sames 2002), loss of structural integrity (e.g., Chan and Incecik 2000), fire (e.g. Vanem and Skjong 2004b), flooding (e.g., Papanikolaou et al. 2000, Vassalos 2004), mus- tering and evacuation (e.g., Vassalos et al. 2001, Dogliani et al. 2004). In addition, projects developed the basics for a new design framework which integrates safety and demonstrated the integration of tools for fast optimization of ship designs. Par- ticular attention was focused on developing a new probabilistic damage stability assessment concept for passenger and dry cargo ships that formed later the basis for the new harmonized damage stability regulations adopted by IMO. The most recent analysis, design and integration of risk-based approaches were performed for Aframax oil tankers (Papanikolaou et al. 2006). In the European research area, research into ship safety was later concentrated into the large project SAFEDOR which included also developments towards a modern regulatory framework and a large number of sample design applications for ships and ship systems (Breinholt et al. 2007b). A list of related research projects is provided in the references to this chapter. Research into risk-based approaches took also place outside Europe, in particu- lar in Japan and South-Korea. Kaneko (2002) presented a holistic methodology for risk evaluation of ships. He focused on prediction of collision probability and fire scenarios and showed a cabin fire as example application. An overview of current research activities in Asia is provided by Yoshida (2007). Kaneko (2007) presented an overview of approaches in risk modeling and pointed towards uncertainties in- volved. An ongoing development into a total risk management system was presented by Lee (2007) focusing on integrating available tools for design, regulation and op- eration. The system is supposed to run in real time delivering input for a simulator, too. Risks are computed using standard risk models, e.g., event and fault trees, for a number of scenarios. A database holding generic data aims to accelerate the com- putation.

1.2.9 Recent Regulatory Developments

Goal-based Standards (GBS) were put on the agenda of the Maritime Safety Com- mittee (MSC), by a decision of the IMO Council (89) in 2002. The first work-

1 Introduction to Risk-Based Approaches in the Maritime Industry 11

The risk of the design RDesign is typically the sum of partial risks coming from different accident categories like, e.g., collision, fire or grounding. Each partial risk can be computed with the help of risk models like, e.g., event trees or Bayesian networks. The choice of a risk model depends on the application. Fault trees are widely used for system analysis. Event trees and Bayesian networks have been used in FSA studies. Risk models expressed by mathematical formulae were developed for fast design optimization. The acceptable risk Racceptable is specified by the approval authority (flag state administration and/or classification society) in case of human life and environmental protection. The acceptable risk related to loss of property and business is usually defined by the owner or operator, and is not considered any further in the following. Two options exist to specify the acceptable risk: relative or absolute. In the first case, a reference design is selected which complies with current rules. In the second case, IMO risk acceptance criteria are used or referenced.

1.3.2 How Risk-Based Design and Approval Work Together

Currently accepted and used risk-based design approaches are two-step approaches involving qualitative and quantitative steps (Breinholt et al. 2007a). The currently proposed risk-based approval process is also a two-step process. The qualitative step ends with a preliminary approval which documents the requirements for the full approval. The obvious question for future risk-based design is how much effort is needed upfront to explore the design solution space without preliminary approval from an approval authority. Additional activities within risk-based design and ap- proval processes have to be aligned with existing schedules for owners, yards and suppliers. Ideally, a yard seeks to build-up complete knowledge of the expected risk analysis and its results before the contract with the owner is signed. This means that a significant amount of analysis may need to be carried out prior to the application for preliminary approval and before the detailed approval requirements are issued. On the other hand, investing too much effort before an indication of feasibility is not advisable. Key milestones in the combined design and approval schedules are design con- cept, final design, letter of intent, contract, preliminary approval and final approval. It is emphasized that the alignment of these milestones will vary according to the ac- tual case. The alignment of schedules for a smaller risk-based design case indicates that after signature of the letter of intent, the yard starts to produce a full design con- cept which is then previewed with the approval authority to decide whether a risk- based approval is needed or not. If needed, the qualitative phase of the design and approval is entered which concludes with the preliminary approval by the approval authority. Once the conditions attached to the preliminary approval are known – and are acceptable – the yard approaches the owner to sign the contract. Following this key milestone, a quantitative analysis is started which – together with the traditional design activities in this stage – eventually results in an approved design.

12 P.C. Sames

For truly challenging and larger risk-based design projects, the quantitative part of the risk assessment is most likely carried out before the letter of intent and, therefore, well before the preliminary approval. The main reason is that yards do not want the process to be interrupted by the relatively late preliminary approval. Yards ideally seek to have all issues affecting the design and approval process solved be- fore applying for approval. It is noted in this context that one additional objective of risk-based design is to increase the knowledge about the ship design in the early design phase and, therefore, to facilitate an advance of the decision making. Thus, with advanced tools available, a risk analysis on key aspects can be performed cost- effectively before a letter of intent is signed.

1.4 What is Needed to Make Risk-Based Design

and Approval Work?

1.4.1 Regulatory Framework

The regulatory framework comprises IMO regulations, classification societies’ rules, regional and national regulations and industry standards. Details of a modern risk- based regulatory framework are described in Chap. 3. This includes a comprehen- sive review of Formal Safety Assessment developments in the shipping and other industries. The approval of risk-based design is detailed in Chap. 4. To facilitate risk-based design and approval, three main elements are needed and most of these are already in place:

• Provisions for risk-based designs SOLAS I/5 and MARPOL Annex I, I/5 have the necessary provision to allow al- ternative designs and arrangements. In addition, alternatives are possible related to fire safety and in the near future for electrical systems and lifeboats.
• Approval procedures A number of IMO documents exist to guide the approval process for alternative designs. In addition, SAFEDOR developed a high-level approval process and a system-level approval process for risk-based designs.
• Risk evaluation and acceptance criteria The FSA guidelines detail criteria related to human life safety, addressing indi- vidual and societal risks. Risk acceptance criteria related to the environment are not yet agreed at IMO but were proposed by Skjong et al. (2006). Furthermore, all FSA studies submitted and reviewed by IMO can be used as a reference.

1.4.2 Design Framework and Tools

The design framework couples traditional design with risk-based thinking. It de- scribes the integration of safety as an additional design objective. Risk-based design