SED04

Structural Engineering Documents ry Ole Damgaard LARSEN ate) WITH BRIDGES The Interaction between Vessel Traffic and Bridge Structures fi \nternational Association for Bridge and Structural Engineering IABSE Association Internationale des Ponts et Charpentes AIPC Internationale Vereinigung fir Briickenbau und Hochbau ——« [VBHBor 1942, Ole Dampaard LARSEN graduated from the Technical University of Den- mark in 1965. Most of his professional career has been spent with the Danish ‘consulting engingering ‘company COWiconsult, now as department managarin, ‘the transportation division. ‘He hes been responsible for ‘the planning ond design of ‘alarge number of major bridge, wnnel anc offshore projects. He has 15 years of ‘expertise {n ship collision ‘studies, mainly for the initial ‘Coriceptual phases af bridge ‘Schemes, where fundemen- Denmark ond the proposed Gibraltar Strait Crossing between Moraceo and ‘Spain, Among other publi- ‘cations, he js co-author of tary Design of Highway Bridges.Structural Engineering Documents 4 Ole Damgaard LARSEN SHIP COLLISION WITH BRIDGES The Interaction between Vessel Traffic and Bridge Structures International Association for Bridge and Structural Engineering IABSE Association Internationale des Ponts et Charpentes AIPC Internationale Vereinigung fur Brickenbau und Hochbau IVBHCopyright © 1993 by International Association for Bridge and Structural Engianeering All rights reserved. No part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system, without permission in writing from the publisher. ISBN 3-85748-079-3 Printed in Switzerland Publisher: TABSE - AIPC - IVBH ETH - Honggerberg CH-8093 Zurich, Switzerland Phone: Int, +411-3772647 from 27 September 1993: Int, +411-6332647 Fax: Int. +411-3712131PREFACE In June 1983 a colloquium on “Ship Collision with Bridges and Offshore Structures” was held in Copenhagen under the auspices of the International Association for Bridge and Structural Engineering (IABSE). The colloquium was initiated by the Danish Group of IABSE with the background of comprehensive investigations in this field in connection with the fixed crossing of the Great Belt in Denmark. The colloquium brought together bridge and offshore engineers, naval architects, navigational experts, risk assessment specialists, etc. with the view to exchange information on this subject. After the colloquium, the [ABSE Working Commission 1 “Structural Performance, Safety and Analysis”, established a Working Group for preparing a guideline or a state-of-the-art report. Members of the Working Group, appointed originally or during the work, were the following: ~ Dr. Yahei Fujii, Electronic Navigation Research Institute, Tokyo, Japan Mr. Comelis Q. Klap, Rijkswaterstaat, Voorburg, The Netherlands Mr: Michael A. Knott, Greiner Inc., Richmond, USA Mr. Thomas R. Kuesel, Parsons Brinckerhoff, New York, USA. Mr, Ole Damgaard Larsen, COWIconsult, Copenhagen, Denmark ~ Dr. Henrik 0. Madsen, Danish Engineering Academy, Copenhagen, Denmark — Mr, Holger S. Svensson, Leonhardt, Andrii u. Partner, Stuttgart, Germany. 1 The Working Group was chaired by Ole Damgaard Larsen, who also prepared the present IABSE publication. Yahei Fujii, Thomas R. Kuesel and Michael A. Knott made valuable contributions to the publication. The development of this publication took its basis from the proceedings of the international colloquium in Copenhagen in 1983. Since the colloquium, important research has been carried out in connection with development of national codes and standards for bridge design and in connection with planning and design of major bridge projects. It was decided in 1989 to leave out offshore structures from the publication. The reason was that it was found inexpedient to treat the two types of structures in the same publication. The final editing of the publication was carried out in the course of 1991, when conclusive data could be made available from the following major projects: — development of vessel collision guide specifications for the Federal Highway Administration in the USA — development of vessel collision design criteria for the Great Belt Crossing in Denmark,‘This document represents the conclusions of the work as accepted by the IABSE Working Com- mission | at the IABSE annual meeting in St. Petersburg, September 1991. The Working Group was dissolved at the same occasion and inquiries regarding this publication should therefore be addressed to the IABSE Working Commission 1 ‘The Working Commission 1 is greatly indebted to the author and much appreciates his efforts. Johan Blaauwendraad, Chairman of Working Commission 1TABLE OF CONTENTS 0. Preface 1. Introduction Ti Background 12 Validity 13 Application 2. Initial Planning 21 Siting of Bridge Structure 22 Navigation Channel Layout 23 Overall Bridge Layout 24 Vertical Clearance 25 Horizontal Clearance Seieeee eae ene 3.1 ‘Traffic Routes 32 Passage Statistics 33 Vessel Characteristics 3.4 Traffic Forecast 4, Risk Acceptance 4.1 Consequences of Collision 42 Disruption Risk Acceptance Criteria 43 Fatality Acceptance Criteria 44 Optimum Cost Acceptance Criteria 5, Collision Risk Su Collision Risk Model 52 Causation Probability 5.3 Geometrical Probability 34 Failure Probability 6. Vessel Impact Forces 6.1 Head-on Bow Impact Forces 62 Sideways Impact Forces 63 Deckhouse Impact Forces 64 Local Collision Forces 65 Barge Vessel Impact Forces Page aare EF eoneorns 19 20 21 24 27 27 28 32 34 37 38 39 44 49 BE) 54 70 71 aan 737. Bridge Design ez) 7A Energy Considerations 1 72 Structural Analysis 80 73 Design of Substructure 92 14 Design of Superstructure 83 8. Prevention Measures 85 8.1 Aids to Navigation 85 8.2 Vessel Traffic Regulations 87 83 Vessel Traffic Management Systems 87 9. Protection Measures 1 9.1 Fender Systems 91 92 Pile Supported Systems 93 93 Dolphin Protection 95 94 Artificial Island or Reef Protection 95 9.5 Floating Protection Systems 7 10. Protection of the Public 101 10.1 Collision Hazard Detection 101 10.2 Bridge Traffic Control 102 Appendices 103 A. Vessel Size and Geometry Data 103 B. _ Records of Collision Accidents 113 C. Selected Literature 1191, INTRODUCTION 4.1 Background Any structure established in navigable waters constitutes a hazard to shipping and is itself vulnerable to damage or destruction in the event of vessel collision. Among the most significant structures exposed to this hazard are bridges crossing coastal or inland waterways. A list of serious accidents recorded during the period 1960-1991 has been included in Appen- dix B, The list represents an updated version of the records published by Frandsen [1-1] in 1983 ‘The records indicate an average of one serious vessel/bridge collision accident per year worldwide, see Figure 1.1, More than 100 persons died in these accidents and large economic losses were incurred directly in repair or replacement costs as well as indirectly in the form of lost transportation service. ‘The photos in Figures 1.2, 1.3 and 1.4 from actual collision accidents illustrate the seriousness. of the subject. The accidents shown are all briefly described in Appendix B. Number of Serious Accidents A Year 1960 1965 1970 1975 1980 1985 1990 Figure 1.1 The number of serious vesselibridge collision accidents per year in the period 1960-1991 (32 years)2 Introduction Many factors are involved in the problem of vessel/bridge collision confronting most countries around the world: ~ The vessel traffic intensity has increased considerably in number as well as in size of vessels. Since 1960 the worldwide water borne tonnage has increased at an average yearly rate of 4%, — Bridges to cross navigable waterways are being planned and constructed at an increasing rate. — Vessel impact loads have not been considered in the design of most bridges more than 25 years old, — Vessels built today are often constructed with bulbous bows which increase the impact load. — Bridges designed today, poorly sited and with inadequate navigation clearance, indicate that the problem is still not fully appreciated. This situation has led to increased concern over the safety of bridges crossing navigable waterways and research into the vessel collision problem has been initiated in several countries of the world in connection with: — evaluation of vulnerability of existing bridges ~ establishment of design criteria for new bridges — development of national codes and specifications regarding vessel/bridge collisions. References are made to [1-2] through [1-17]. The main purpose of the present publication is to promote the understanding internationally that the interaction between bridge structures and vessel traffic should be considered a decisive design aspect because of concern for: ~ the risk of disruption of the bridge — the safety of the users of the bridge the safety of the vessels ~ the risk of environmental pollution. A list of important publications on the vessel/bridge collision topic, available today, is included in Appendix C.Background 3 vet! Figure 1.2 Vessel/bridge collision accidents; at the top, the Maracaibo Bridge accident in Venezuela in 1964 (from Ostenfeld [1-13]) and below, the Hopewell Bridge accident in Virginia, USA in 1977 (from US National Research Council {1-17])4 Introduction Figure 1.3 Vessellbridge collision accidents: at the top, the Tjérn Bridge accident in Sweden in 1980 (photo: Aksel G. Frandsen) and below, the Sunshine Skyway Bridge accident in Florida, USA also in 1980 (from Knott et al. [1-5]).Background 5 Figure 14 Vesselibridge collision accidents: at the top, the Strangnds Bridge accident in Sweden in 1990 (photo: Mats Johnsson) and below, the Carnafuli Bridge accident in Burma in 1991 (courtesy of Dr. ¥. Fujii)6 Introduction 1.2 Vali y It has been found that the state-of-the-art of the vessel/bridge collision problem is not so well established that specific regulations for worldwide use can be prescribed at this time. Such a task would also be impeded by national differences in design philosophies, code formats, etc. Thus, this publication is intended to provide guidelines to enable engineers—faced with the question of preventing or protecting vessels and structures from collision—to consider the subject in a ratio nal way, and to take advantage of up-to-date knowledge. ‘A basic concern during the preparation of this publication has been not to restrict ingenuity of design. The information and advice presented should not be interpreted as covering all aspects of vessel collision with bridges. Engineering judgment must always be used when the problems related to a particular bridge site are to be evaluated. It is expected that state-of-the-art philosophies, methodologies and key assumptions will change with time as engineers gain more knowledge about the vessel collision problem. There- fore, the lifetime of this document may be limited. Prescriptive codes and standards on collision design criteria exist on a national basis and within particular technical fields. This publication has benefitted from such codes and future codes may benefit from the present publication. Further development of specific design criteria on a national and international basis is encour- aged in order to counteract inadequate practices by individuals and bodies not influenced by IABSE publi 1.3 Application ‘The present publication contains information and advice applicable for: ~ planning and design of new bridges — planning and design of navigation channels ~ planning and design of prevention measures — planning and design of protection measures — upgrading or retrofitting of existing bridges — upgrading or retrofitting of navigation channels — evaluation of the safety of existing bridges — evaluation of the safety of vessels evaluation of the safety of persons — evaluation of the safety of environment.2. INITIAL PLANNING As considerations for the interaction between bridge structures and vessel traffic will often influence fundamental decisions such as location and type of bridge crossing, it is essential that the vessel collision aspects are properly investigated as early as possible in the planning process, A number of important aspects of the bridge design, siting, and aids to navigation can be evaluated by relatively simple means on the basis of the initial knowledge of the waterway and the navigation, applying experience and common sense, This chapter contains such general considerations concerning the ini the bridge structure and the waterway. al planning and layout of -1 Siting of Bridge Structure ‘The purpose of the bridge often determines a specific location. However, minor modifications can normally be introduced and if major problems are identified it might still be possible to consider alternative locations. For the siting of a bridge crossing, the following aspects, among others, should be considered, ref, [2-1], [2-2], [2-3], [2-13], [2-14], [2-15], and [2-18]: — Locations with congested navigation should be avoided. ~ Locations with difficult navigation conditions (shoals, cross currents, etc.) should be avoided. — A straight and unencumbered navigation channel approach of adequate length before the bridge-passage should be achieved. — Adequate distance to locations where berthing manoeuvres take place should be provided. — The bridge's alignment should preferably be perpendicular to the navigation channel — The centre of the navigation span should coincide with the centre line of the navigation channel. Locations where bridge piers can be placed in shallow water so that vessels out of control may not reach the bridge structures without first running aground should be preferred. With regard to the adequate length of unencumbered channel approach, the following empirical estimate has been reported by Shoji et al. [2-18], [2-19], and [2-20] based on analyses of collisions with bridges world-wide: The minimum distance from a bridge line to the position of the nearest turn in the navigation route should be at least 8 L and preferably 20 L, L being the length of the vessel. If the distance is smaller, the turn will influence the navigation at the bridge crossing.8 Initial Planning 2.2. Navigation Channel Layout The aim of the bridge designer to attain optimal safety against vessel collision will as a rule also lead to requirements for the layout and operation of the navigation channel. The bridge designer is, however, often limited in this respect since the responsibility and authority for implementing such navigation improvements often belongs to navigation authorities. Regardless of the question of design responsibility, the following indicates various aspects of navigation channel layout to be considered by the designer: — The waterway should have clearly defined and properly declared navigation channel(s). — If the traffic density is high, i.e. involving frequent meeting or overtaking close to the bridge crossing, two-way traffic should be established by introduction of a traffic separation scheme. — If conditions allow, the traffic from minor and leisure vessels should be separated from that of large (merchant) vessels. — The water depth in the navigation channel should be at least 1.2 times the maximum vessel draught. The direction of the navigation channel and the location of bends may be adjusted to improve the navigation conditions in general. 2.3. Overall Bridge Layout The primary area of vessel collision risk to the bridge is the region near the navigation span. ‘Therefore the layout of the bridge in this region should be developed to maximize the horizontal and vertical clearance for navigation, In section 2.4 and 2.5 empirical rules are outlined with the aim of ensuring that the clearance of the navigation span is sufficient to provide safe passage under normal conditions. ‘The philosophy is that collision with the bridge structures should only occur as a result of navigation error or technical failure on board, possibly in combination with low visibility and adverse weather conditions, and not because of particular navigation difficulties created by the presence of the bridge. It should be realized that all parts of a bridge crossing situated where the water depth allows vessels, including vessels in ballast, to navigate are exposed to collision risks. ‘The lengths and heights of approach spans should be planned considering these risks. During the planning phase, thought should also be given to protection alternatives in cases where a cost-effective design of the bridge structure cannot be achieved directly. The alternatives include: — pier fender systems to reduce impact force and energy — independent protection structures, i.e. artificial islands or reefs, dolphins, floating arrestors, etc. to withstand or redirect the colliding vessel. The development of a suitable overall bridge layout is an iterative process comparing different alignments and differing degrees of protection for alternative bridge solutions,Vertical Clearance 9 2.4 Vertical Clearance According {o intemational practice, the vertical clearance in the navigation span of a bridge crossing should be planned to allow the passage of the highest vessel (incl. equipment such as mast, antennae etc.) using the waterway at that point in time when the construction of the bridge is planned and made public to the users of the shipping lane. The clearance should permit the passage of the highest vessel in a ballast condition at high water level with due allowance for vertical vessel movements. Data on air draughts of vessels are not available in public ship registers and therefore have to be obtained from owners and/or shipyards. For preliminary planning purposes the typical vertical clearance dimensions of different types of vessels, shown in Appendix A, may be used, The scatter in the vertical dimensions is very considerable as illustrated in Figure 2.1. The Figure shows chimney heights, which are less than the required air draught but represent the highest part of the vessels able to damage the bridge in case of collision Vessel Chimney Heights (m) A 60 * Oll Tankers + Bulk Carriers — © Container Vessels / Naval Vessels 10 0 > 0 100,000 200,000 300,000 400,000 500,000 DWT (Tonnes ) Figure 2.1 Typical heights of vessel chimneys above waterline of different types and sizes of vessels in ballast. To arrive at the air draught, the height of auxiliary equipment such as mast and antennae has to be added.10 Initial Planning For the existing world fleets of merchant and naval vessels a vertical clearance of 65 m above high water level will suffice. The vertical clearances of long span bridges world-wide are listed in Table 2.1 for comparison. Bridge Name Country Completion | Navigation | Vertical Year Span | Clearance (m) (mn) George Washington | USA, New York 1931 1067 65 West Bay USA, California 1936 704 x2 65 Golden Gate USA, California 1937 1280 67 Bronx-Whitestone USA, New York 1939 701 46 Tacoma Narrows USA, Washington 1950 853 56 Mackinac USA, Michigan 1957 1158 45 Forth Great Britain, Scotland 1964 1006 52 Verrazano Narrows | USA, New York 1964 1298 69 Severn Great Britain, England 1966 988 37 Tagus Portugal 1966 1013 70 Angustura Venezuela 1967 712 64 Kanmon Japan, Honshu-Shikoku 1973 712 61 Bosphorus (1*) Turkey 1973 1074 64 Humber Great Britain, England 1981 1410 30 Innoshima Japan, Honshu-Shikoku 1983 770 50 Ohnaruto Japan, Honshu-Shikoku 1985 876 41 Minami Bisan-Seto | Japan, Honshu-Shikoku 1988 1100 65 Kita Bisan-Seto Japan, Honshu-Shikoku 1988 990 65 Shimotsui-Seto Japan, Honshu-Shikoku 1988 940 31 Bosphorus (2) Turkey 1989 1090 64 Table 2.1 Vertical clearances of long span bridges world-wide. It is to be noted, however, that certain special-purpose vessels, such as crane vessels, offshore drilling rigs during transport and others, may reach more than 100 m above sea level. In connection with the construction of the Danish Great Belt East Bridge across the international navigation route between the Baltic Sea and the North Sea, the question of free passage of very high offshore drilling rigs was referred to the International Court of Justice in the Hague in 1991 by Finland. In 1992 the case was withdrawn from the court after an agreement had been reached between the Finnish and Danish Governments implying that the vertical clearance of 65 m could be kept.Horizontal Clearance 11 2.5 Horizontal Clearance As the risk of vessel collision depends decisively on the horizontal clearance of the navigation span, this should be as large as can be technically and economically justified. An estimate of the horizontal clearance necessary to obtain a high level of safety under normal conditions of passage can be achieved by empirical methods. The following methods are available: ~ consultation of literature or codes (may not be reliable and up-to-date) — analysis based on observations of navigation behaviour, e.g. “ship domain” — manoeuvring simulations (should preferably be applied after the overall layout of the bridge has been decided). As two-way traffic is much more demanding than one-way traffic, it has to be established if the vessel traffic density is so high that a two-way situation is required. For this purpose a method has been developed based on the “ship domain” theory, i.e.: “close encounter” analysis (described at the end of this section). Literature or Codes Based on the ship domain theory, the Japanese Government passed a Maritime Safety Law [2-11] in 1973, requiring that the minimum width of a fairway for international vessel traffic is 700 m for one-way traffic and 1400 m for two-way traffic. Shoji et al. {2-18}, [2-19] and [2-20] have investigated the relation between vessel size and horizontal span clearance for actual cases of vessel/bridge collision accidents. It was found that in all investigated cases the span clearance was less than 3 times the ship length. Assuming one-way iraffic for the investigated cases, this is in good agreement with the ship domain theory. Based on the papers by Shoji et al., the US Guide Specification [2-14] contains the following guideline for the planning of navigation span clearance: “Bridges with main span less than 2 or 3 times the design vessel length are particularly vulnerable to vessel collision.” Ship Domain Analysis Observations have shown that in order to navigete safely, the master of a vessel tries to keep a certain distance from other vessels, fixed objects, shallow water, etc. The distance varies with the vessel speed, visibility, type of encounter, and a number of other navigational aspects. This method of explaining navigational behaviour is called the “Ship Domain” theory. The domain concept was introduced by Fujii et al. [2-8] based on comprehensive studies of navigation patterns in Japanese waterways and Goodwin [2-10] for use in open sea situations. In this theory, the “Bumper Area” is the area a vessel actually occupies in a waterway including a zone around the vessel within which other vessels’ bumper areas should not overlap. Observa- tion shows that vessel encounters with overlapping bumper areas often result in unwanted12 Initial Planning Navigation Opening Ir] 80L2 ¢ Lr ay Q 3 . : y 8 {4 8.0L; Bridge Figure 2.2 Vessels and respective bumper areas in a waterway where navigation is impeded by a bridge with a narrow navigation opening. evasive manoeuvres involving increased accident risk. Figure 2.2 shows two vessels approaching a bridge with a narrow navigation opening. The indicated sizes of the bumper areas show that the encounter is critical, : Fujii and several other Japanese researchers have estimated the size of the bumper area on the basis of data obtained through radar observations. Reference is made to [2-5], [2-6], [2-8], (2-16], and [2-21] ~ For waterways with sufficient width to provide free navigation at service speed (5-8 m/sec.) and with no obstructions (islands, shallow water, etc.) in the channel, the following average values for the size of the bumper area (approximated by an ellipse) has been found: 8.0 L in the course direction and 3.2 L in the side direction, L being the length of the vessel. ~ In narrow channels and harbours, where the conditions require vessels to travel at a reduced speed (3-4 m/sec.) and where no head-on encounters, overtaking or crossing encounters take place, the following average bumper area size has been found: 6.0 L in the course direction and 1.6 L in the side direction, L. being the length of the vessel. The above results have been derived from waters with a high traffic density and with a large fraction of small vessels. Open sea situations with lower traffic density have been studied by Goodwin [2-10], resulting in larger bumper area sizes than discussed above. For confined, protected waterways with very high traffic density, such as narrow rivers or harbour passages, smaller bumper area sizes than mentioned above may be relevant.Horizontal Clearance 13, For a bridge crossing a navigation channel, it is thus suggested that the requirement for the navigation span clearance should be related to the width of the bumper area of a typical large vessel passing the bridge. The typical large vessel may be selected as the largest vessel which is able to safely pass the bridge without assistance, provided that larger vessels have pilots on board or their passages are regulated by a Vessel Traffic Service system. Taking the typical large vessel as the design vessel, with the length Lp, the following requirements for the horizontal clearance of the navigation span may then be derived from the bumper area widths indicated above: — Incase of one-way traffic the horizontal clearance C should be equal to the width of the bumper .2 Lp for waterways with vessels travelling at service speed area of the design vessel, i.e. C and C = 1.6 Lp for restricted waters — In case of two-way traffic the horizontal clearance should be equal to the width of two bumper areas of two meeting design vessels plus a separation zone between the two bumper areas. The necessary width of the separation zone has been addressed by Fujii et al. in [2-7], suggesting a separation zone width of 0.3 Ly - 1.8 Lp. Thus, the horizontal clearance C should be: C = 6.7 Lp - 8.2 Lp for waterways with vessels travelling at service speed and C = 3.5 Lp - 5.0 Lp for restricted waters. Figure 2.3 illustrates the above horizontal clearance indications. It is suggested that these clearance requirements be used in the planning stage for important bridges crossing open deep waters, where it is physically possible for vessels to depart substantially from the nominal sailing route. In addition to these general rules, the local navigation aspects, ic. risk-increasing effects stich as nearby bends and risk-reducing effects such as special navigational aids, should be accounted for. When complying with these requirements and suggestions, the presence of the bridge is expected not to give rise to particular difficulties to the navigation, and the risk of vessel collision may be expected to be acceptably low. This point of view is supported by the results of an analysis of vessel collision accidents involving North American bridges, ref. Frandsen et al. [2-3], where a marked lower collision frequency for bridges with spans fulfilling the above proposed requirements to the horizontal clearance compared to bridges with spans not fulfilling these requirements have been found, see Figure 2.4. Most existing bridges do not fulfil the above suggested requirements to the horizontal clearance and for many new bridges, economy does not allow full compliance either. In cases where the bridges span relatively narrow channels, for which the available width of water deep enough to float large loaded vessels limits the “domain” of these vessels, the suggested requirements are not fully relevant and the collision risk may still be acceptably low. In other cases where navigation clearances have been dictated by economy rather than collision risk considerations, more cautious navigation practices by vessel masters and pilots are required and attention should be given to collision preventive and protective measures in order to keep the collision risk at an acceptable low level.14° Initial Planning One - Way Traffic Free Navigation Restricted Waters 67-82L 3.5-5.0L Figure 2.3 Horizontal clearance requirements for important bridges according to the domain theory for one-way traffic and two-way traffic respectivelyHorizontal Clearance 15 Number of Collisions per 10,000 Passages A 25 20 15 1.0 05 o—* > 2 3 4 Existing Horizontal Clearance / Clearance required by Domain Theory Figure 24 Number of collision accidents per vessel passage for a series of North American bridges versus the ratio between the actual horizontal clearance and the (clearances) suggested by the domain theory (based on Frandsen et al. [2-3]).16 Initial Planning Manoeuvring Simulations with Pilots Computerized navigation simulators are able to model vessel manoeuvring capabilities very realistically in the sense that it is possible to take most important factors into account, .g.: — wind waves — current — visibility characteristics of vessels — vessel overtaking and meeting — geometry of waterway — geometry of bridge structures — location and type of navigational aids. This means that pilots are able to navigate almost as if they were sailing a real vessel in a real navigation channel, ref. Figure 2.5 and Figure 2.6. Manoeuvring simulations with professional navigators carried out in cooperation between bridge designers and maritime authorities have proven to be an efficient and reliable tool for the evaluation of requirements for horizontal bridge clearances, ref, Bay et al. [2-1], Frandsen et al. [2-4], Gardenier [2-9], and Meurs et al. [2-15]. Figure 2.5 A ship simulator set up for the investigation of two-way bridge passage (from Bay et al, [2-13]).Horizontal Clearance 17 \ \ \ \ x R / / / / / T / 4 | Figure 2.6 Typical plots showing the results of manoeuvring simulations (from Bay et al, [2-13]) Close Encounter Analysis A way to evaluate if the one-way traffic requirement to the navigation span clearance suggested by the domain theory is sufficient is to estimate how often a situation arises where two vessels meet which are so large that the clearance is not sufficient to provide safe passage. Such meetings of anti-directional vessels in the vicinity of the bridge are referred to as “Close Encounters”, Since several of the involved parameters depend on the vessel type and size, it is convenient to use a Monte Carlo simulation approach in order to calculate the yearly expected number of close vessel encounters as a function of the horizontal clearance of the bridge. The simulation of the occurrence of vessels of different sizes in the vicinity of the bridge requires a statistical description of the vessel traffic. The Poisson process is generally accepted as a relevant description of such events. The method is described in Ostenfeld-Rosenthal et al. [2-17] and Frandsen et al. [2-3]. Final advice about an acceptable level of annual number of close encounters for maintaining one-way conditions cannot generally be given, but should be defined in the particular case by the authority concerned. An example of the results of such analysis is illustrated in Figure 2.7.18 Initial Planning Number of Close Encounters per Year A Navigation Span Clearance (m ) Figure 2.7 Typical way of illustrating the results of a close encounter analysis.19 3. VESSEL TRAFFIC The characteristics of the vessel traffic passing under the bridge should be established in as much detail as possible for the design stage in question. Normally, the traffic characteristics in a specific situation can be established by utilizing available statistics and information from local authorities. On this basis, a forecast should then be made to cover the traffic development during the anticipated life-time of the bridge. Vessel traffic is conveniently defined in the following terms as described in the subsections below: ~ Traffic Routes Passage Statistics — Vessel Characteristics — Traffic Forecast. 3.1 Traffic Routes Vessel traffic routes in the vicinity of the bridge should be identified. All routes for which interaction between navigation and the bridge is possible should be considered. Ordinary routes as well as unofficial routes (short cuts etc.) should be identified. An example is shown in Figure 3.1 Figure 3.1 Vessel traffic routes in the vicinity of a bridge crossing (based on Olsen et al. [3-4])20. Vessel Traffic In general, the ordinary sailing pattern on the route is a Gaussian distribution across the route. Mean and standard deviations typically depend on width of waterway, size distribution of vessels and traffic density. This is further described in Section 5.3. 3.2. Passage Statistics Sources for obtaining data on vessel traffic in a particular waterway include the following: ‘The authority of the waterway in question often maintains logs of vessel trip data. Local port authorities and operators normally maintain logs of port call data. Local pilot stations often have fairly detailed records and knowledge of the navigation in the area, The Coast Guard authorities maintain records on navigational safety. Military or naval authorities may possess records on naval as well as commercial traffic. If no data or insufficient data are available, then recording of the vessel passages should be initiated. The following means are available: — visual observation (uncertain during night-time and fog periods) — surveillance by radar (uncertain with regard to vessel type and size) communication by VHF radio. Assuming that some general knowledge on the navigation pattern is available, a relatively short period of observation/surveillance will provide a fairly good insight into the number and types of vessels using the waterway. A longer period of observation/surveillance may be necessary for identifying unusual vessels or groups of vessels. Itmay be convenient to separate vessel traffic into selected vessel types which can be separately considered in the analysis. Vessel geometry, collision force, manoeuvering characteristics, types of cargo, ete. depend on vessel types. According to Couper [3-1], the world fleet of merchant ships (over 100 GRT) totals about 40,500 vessels and equals 400 million GRT. The distribution of the world fleet in main vessel types is given in Table 3.1. Typical profiles of different types and sizes of vessels are shown in Figure 3.2. If special-purpose vessels, such as chemical tankers, liquid natural gas carriers, vehicle carriers, offshore industry vessels, naval vessels, passenger vessels, leisure vessels, etc. form a significant proportion of the traffic, additional research and judgment by the designer will be required.Vessel Characteristics 21 Vessel type Percentage of world fleet tonnage Tankers 44% Bulk carriers 20% General cargo vessels 19% Combined bulk/tankers 6% Container vessels 3% Other types of vessels, 8% Total 100% Table 3.1 World fleet vessel type distribution (based on Couper [3-1]}. For all passing vessels the following data is of particular interest: — the loading condition as regards draught, air draught, collision force and manoeuvring capa- bility — the transit speed as regards collision energy and force and manoeuvring capabi the type of cargo as regards potential risk for environmental pollution — the use of pilot and/or other navigational aids as regards the level of safety during passage. ies. Such data is normally difficult to obtain and it may be necessary to make qualified estimates. Vessel draught (loaded and ballasted) in relation to the available water depths determines the upper limit sizes of vessels which are able to reach the different parts of the bridge structure. Con sideration should also be given to abnormally raised high water levels during storm conditions. Itshould be investigated if large vessels utilize the water level variations in the waterway and/or the possibility of transiting in partly loaded condition. Depending on the character of the shipping, a differentiation of loading conditions of vessels transiting in opposite directions may also be considered. 3.3 Vessel Characteri: ics ‘The characteristics of the particular types of vessels using the waterway should be established on the basis of specific information, However, if such data are not available, typical characteristics may be taken from the diagrams in Appendix A, which have been developed on the basis of [3-1], [3-2], [3-3], [3-5], [3-6], and [3-8]. Appendix A provides geometrical data for the following three types of vessels: tankers, bulk carriers and container vessels. Depending on the purpose of the analysis, other vessel types may be allocated to one of the three mentioned types.22 Vessel Trattic ‘Type: Crude Oil Carrier DWT: 270,000 T Speed: 6.2 m/s ‘Type: Ore Cartier DWT: 122,000 T Speed: 82 m/s ‘Type: Bulk/ Tanker DWT: 113,000 T Speed: 8.0 m/s ‘Type: Container Ship DWT: 32,000 T Speed: 11.8 m/s ‘Type: Bulk Carrier DWT: 27,000 T Speed: 8.0 m/s ‘Type: Product Tanker DWT: 25,600 T Speed: 8.0 m/s ‘Type: Ro-Ro Ferry Type: Trawler ay ely ‘Speer 5m/s 800 GRT. és i, i Type: Coaster a 500 GRT te 3 "a ‘Type: Cutter 100 GAT le t ia 0 50 100 150 200 250 300 350 400 Ship Length (LOA) (m) Figure 3.2 Examples of vessel profiles (partly based on Couper [3-1]Vessel Characteristics 23 Height of Deck House Figure 3.3 Vessel geometry definitions ‘Appendix A contains the following data for the three above-mentioned types of vessels: ~ air draught in ballast condition ~ air draught in loaded condition — height of deck house in ballast condition — height of deck house in loaded condition — draught in ballast condition — draught in loaded condition length — breadth — moulded depth ‘The definitions of geometry data are shown in Figure 3.3. ‘The data represents typical characteristics of present-day steel-hulled merchant vessels and the user should be aware that important variations exist. The vessel bow structure, partly determined by local vessel building practice, may be essential as regards magnitude of collision force. Different bow shapes are illustrated in Figure 3.4. Varying measurement traditions for size of vessel are used in different countries and for differ ent types of vessels. In this publication, Dead Weight Tonnage (DWT) and Displacement Tonnage (W) have been used to express vessel size and vessel weight respectively. To facilitate conversion from other units of measurement, conversion factors have been included in Appendix A.24 Vessel Trattic Tankers and Bulk Carriers Container Vessels Figure 3.4 Common vessel bow shapes. The bulbous bow shape is prevalent in the world fleet 3.4 Traffic Forecast The development of the vessel traffic should be forecast throughout the anticipated lifetime of the bridge. It has to be recognized, of course, that such a forecast can only be done with very limited certainty, because one of the main influencing factors is the political and economical climate, locally as well as internationally. ‘Alternatively, a certain year, say 10-20 years ahead, may be selected as a study year and the traffic in that particular year would then be forecast and used as a design basis. ‘Among the subjects to consider for the preparation of a forecast of the traffic would be, ref. Larsen et al. [3-3]: = Projection of international and national markets: + an internationally recognized rate of growth in volume of trade by sea may be considered. ~ Local circumstances having influence on the traffic: + deepening and widening of the navigation channel would allow transit by larger vessels ing of alternative or adjoining waterways might influence the traffic * upgrading or downgra volume in the channel in question + port development and establishment or closing down of local industries or vessel facilities might influence the volume of traffic considerably. Technological developments in transportation facilities: + fast growth of containerized traffic at the expense of other types of freight vessels is likely + new types of mixed purpose vessels might appear. Technological development in ship building: + the construction of 1,000,000 DWT tankers is technically feasible, ref. Spyrou [3-7] and Figure 3.5 + the construction of 100,000 DWT container vessels is technically feasible.Tratfic Forecast. 25 Vessel Size (DWT ) 1,000,000 500,000 Universe Apollo 100,000 }_______al Universe Leader e 50,000 4. 10,000 > 1950 1960 1970 1980 Year Figure 3.5 Development of the size of very large crude oil carriers (VLCC's). During the years 1950-1970, ref. Spyrou [3-7]. Although it is acknowledged that larger VLCC's are technically feasible, no new VLCC's have been constructed since the 1970's26 Vessel Traffic Recognizing that the development of navigation in reality might turn out to be very different from the forecast, future reviews should be instituted. Ideally, a spread in the forecast should be considered allowing for different possible trends in economic and technical development. An example of vessel traffic forecast covering a 20 year time span is illustrated in Figure 3.6. No of Vessel Passages per Year A 2000 — Ships of 40,000 DWT and above — ~~ Possible Trends 1000 Ships of 100,000 DWT and above Scenario 2 Figure 3.6 Example of forecast of the number of passages of vessels of 40,000 DWT and above in a Danish Strait. Two development scenarios were considered.27 4, RISK ACCEPTANCE Risk can be defined as the potential realization of unwanted consequences of an event (i.e. the product of the probability of an event and the consequences of the event). Both the probability of occurrence of an event and the magnitude of its consequences are thus involved. References are made to Philipson [4-15], Planeix [4-16], and Rowe [4-17]. Acceptance Criteria can be established either in the form of a predefined set of “Risk Accep- tance Criteria” or in the form of “Optimum Cost Criteria”: ~ Risk Acceptance Criteria are defined as acceptable limits to probabilities of certain unwanted consequences of collision and are expressed in terms of annual frequencies. Risk Acceptance Criteria are normally imposed by the Authorities to reflect the willingness of people and society to accept risk. In principle, such criteria do not consider the cost of observ- ing the criteria, ~ Optimum Cost Criteria are Acceptance Criteria based on cost-effectiveness analyses comparing the costs of bridge strengthening and protection measures against the benefit of risk reduction. Optimum Cost Criteria may be introduced in cases where it is not economical or technically feasible to design the bridge structures to comply with official Risk Acceptance Criteria. Furthermore, the Optimum Cost Criteria may be used for decisions on whether to reduce the risk to a level below what is required by the Risk Acceptance Criteria, because this can be justified economically. In order to make the use of the Acceptance Criteria unambigious and efficient they should be implemented together with guidelines on how complying with the criteria shall be documented. These guidelines should cover among others: — types of collision consequences to be considered ~ principles for estimating frequencies of the collision consequences considered ~ principles for addressing the uncertainties of the frequency estimates. 4.1 Consequences of Collision There are a number of different categories of coasequences that are associated with a vessel/bridge collision event. The main consequences may be classified into the following six categories (three categories of direct consequences and three categories of indirect consequences respectively):28 Risk Acceptance ~ damage to the bridge damage to the users of the bridge — damage to vessel and cargo — inconvenience cost to society and business ~ social losses ~ damage to the environment. For each category of consequences, a wide range of damage, from superficial damage to disas- ter, is possible. It should be noted, however, that if all imaginable categories and magnitudes of consequences are to be taken into account, then the number of cases to be considered would become very large and unworkable from a practical point of view. In most cases, the data and resources available will dictate that the formal risk acceptance procedure is limited to cover only the risk of bridge disruption and possibly also the risk of loss of lives. 4.2 Disruption Risk Acceptance Criteria Bridge disruption Risk Acceptance Criteria to cover vessel/bridge collision have been established in a few cases in connection with national codes and major bridge projects worldwide. In alll cases a risk comparison approach has been used in which the actual type of project has been compared with other engineering projects or natural events with comparable failure consequences. Typical data for risks associated with engineering projects and natural events, available in the literature, are shown in figures 4.1 and 4.2. Comparison of Risk Acceptance Criteria with the general safety level in codes and standards addressing normal loads is not meaningful because the failure criteria are for design of each element of the structure and for each load combination. The probability of failure in codes and standards is normally in the order of magnitude of 0.0001-0,001 in 100 years, ref, Nordic Com- mittee for Structural Design [4-14]. ‘The following examples of bridge disruption Risk Acceptance Criteria have been found in the literature: ~ For the Great Belt Bridge Project in Denmark (previous project discontinued in 1978) compris ing two approx. 8 km long road and rail bridges it was decided that the design of each of the bridges should be based on a mean time between catastrophic vessel collision accidents of 10,000 years. Reference is made to Frandsen et al. [4-1]. This corresponds to a probability of + 0.02 in 100 years for disruption of the total bridge link. — In 1985 the State of Louisiana, USA issued “Criteria for the Design of Bridge Piers with respect to Vessel Collision in Louisiana Waterways” [4-12]. The document distinguishes betweenDisruption Risk Acceptance Criteria 29 deep and shallow draught waterways. For deep draught waterways servicing vessels with up to 13 m draught the following acceptance criteria regarding vessel collision accidents was recommended: + 0,01 in 100 years for bridge disruption. — A working draft for an ISO Standard titled “Accidental Actions due to Human Activities” issued in 1987 [4-3], contains the following regulation on vessel impact on bridges from river and canal traffic: The design impact should be chosen in such a way that there is an assessed probability less than 0,01 that this or a higher impact energy will occur in a period of 50 years for one structure. This corresponds to: + 0.02 in 100 years for excessive impact energy. — The Guide Specification and Commentary for Vessel Collision Design of Highway Bridges, issued in 1990 by Federal Highway Administration in the USA [4-8] suggests the following acceptance criteria with regard to vessel collision accidents depending on the importance of the bridge. The acceptable frequency of collapse shalll be equal to or less than: + 0.01 in 100 years for collapse of “critical” bridges + 0.1 in 100 years for collapse of “regular” bridges. ~ For the revised Great Belt Project in Denmark, comprising three major structures, a combined 6.0 km long road and rail bridge, a 8.0 km long railway tunnel and a 6.7 km long road bridge, different Risk Acceptance Criteria were established for accidents causing disruption of both rail and road connection and accidents causing disruption of only road or only rail. Disruption was defined as closing of the link for traffic for more than one month. Reference is made to Vincentsen et al. [4-20] and Kampmann et al. [4-4]. The probabilities of accidents causing disruption of the link are not allowed to exceed: + 0.02 in 100 years for simultaneous disruption of rail and road links + 0.1 in 100 years for disruption of the road link + 0.1 in 100 years for disruption of the rail link. Tris to be noted that these Risk Acceptance Criteria are intended to cover the aggregate probability of accident due to several types of causes, including fire and explosion, ice impact, train accidents, etc. in addition to vessel collision. An order of magnitude of acceptable probability of bridge disruption of 0.1-0.01 in 100 years may be derived from the above examples. Due to the characteristics of individual bridge projects it is recommended that specific evaluation should be carried out in each case, Such evaluation should address, among others, the following aspects: — importance of the bridge connection to the public to society, to business and industry, to the national defence, etc. (ref, US Guide Specifications [4-8]) — fatality risk in case of high traffic volume (ref. Section 4.3), if specific Risk Acceptance Crite- ria for user fatality risk are not implemented — cost effectiveness of improving the safety of the bridge (ref. Section 4.4), Weighing the different considerations will often lead to compromises involving lower probabilities of disruption than reflected in the above examples.30 Risk Acceptance 10° MARGINALLY ACCEPTED” to" MERCHANT SHIPPING MOBILE DRILL RIGS 2 0 “ACCEPTED” GEYSERS SLOPE FF CANVEY LNG STABILITY STORAGE 103 to-4 10-5) ANNUAL PROBABILITY OF "FAILURE" “6 LIVES LOST | 10 100 1,000 10,000 COST in $ Im 10m 100m Ib 10b CONSEQUENCE OF FAILURE Figure 4.1 Probability of failure of selected engineering projects (from Whitman [4-21]}). In the above examples from the literature, excepting the Great Belt Project, the Risk Accep- tance Criteria cited are the only criteria to be met. Hence, these criteria should ensure also an adequate safety for the users of the bridge. For the Great Belt Project specific Risk Acceptance Criteria for the risk to users were established. Here the Risk Acceptance Criteria for bridge disruption and the Risk Acceptance Criteria for user fatality should all be met.Disruption Risk Acceptance Criteria 31 10, t rnadoes Total Natural < t 1/0) rricanes w > Oo \ g 1/100) Earthquakes 7 a SA. a N 5 1/1000} 2 Ww B 171,000 wd a we 1/100,000} 171,000,000 110,000,000 10 100 1000 10,000 100,000 1,000,000 FATALITIES Figure 4.2. Number of fatalities from natural events versus frequency of events (from Whitman [4221])32 Risk Acceptance 4.3 Fatality Acceptance Criteria The fatality risk of concern in most bridge projects is the risk to the users of the bridge, motor- ists or train passengers. The risk to be considered is one that is due to accidents that are unique to the bridge as opposed to normal road and railway risks on land. ‘Two risk categories are relevant: ~ individual fatality risk ~ societal risk. The individual fatality risk is the risk to each individual user passing the bridge. A reasonable Risk Acceptance Criterion for the individual fatality risk is that the additional risk due to accidents unique to the bridge is small compared to normal road- and rail-user risk, Normally, the added individual risk to users of the bridge due to accidents unique to the bridge, ship collisions, Frequency of Accidents ( Events per Year ) A Fatality Risk = Societal Risk © Zone of Acceptability a Number of Fatalities per Accident Figure 4.3 Principle diagramme illustrating the difference between Societal Risk Acceptance and Individual Fatality Risk Acceptance.Fatality Acceptance Criteria 33 Category of Risk Risk Acceptance Criteria Individual Fatality Risk Societal Risk + 1-19 fatalities + 20-200 fatalities 8-10" fatalities /year 3-10" accidents/year 5-10° accidents/year + more than 200 fatalities 1-10° accidents/year Table 4.1 Risk Acceptance Criteria for fatality risk in addition to the risk on land-based rail- ways for rail passengers using the Great Belt Railway Tunnel, ref. [4-4]. The criteria are based on a number of rail passengers per year of 14 millions. will be very low compared to normal road- and rail-user risks. Reference is made to Kampmann etal. [4-4], Larsen et al. [4-9], Sexsmith [4-18], and Vincentsen et al. [4-20]. The societal risk is expressed as a relationship between the number of fatalities in an accident and the frequency (events per year) of accidents with more than a certain number of fatalities (see Figure 4.2 and 4.3). This concept is introduced because it is observed experience that society reacts more vehemently to a few accidents with a large number of fatalities than to a large number of accidents with few fatalities. ‘The principle is illustrated in Figure 4,3. Asan example, ref. Kampmann et al. [4-4], the Fatality Acceptance Criteria for rail passengers using the Great Belt Railway Tunnel are shown in Table 4.1. ‘The users of the bridge benefit from using the bridge. A higher risk is acceptable in such cases than in cases where the persons subject to the risk have no benefit from the activity exposing them to risk. ‘The Risk Acceptance Criteria curve should therefore be established on the basis of experience from traffic and transportation rather than other types of societal risks. An acceptably low fatality risk may be obtained by: — ensuring a low probability of bridge disruption (see Section 4.2), and introducing bridge user warning systems on the bridge (see Section 10).84 Risk Acceptance 4.4 Optimum Cost Acceptance Criteria In many cases it is not economically or technically feasible to design the bridge structure to comply with strict pre-set Risk Acceptance Criteria, Illustrative examples from the literature are: — Tasman and Bowen Bridges, Australia. Ref. J. Leslie et al. [4-10] ~ Sunshine Skyway Bridge, USA. Ref. Knott et al. [4-2], [4-5], [4-6], and [4-7]. Situations in which optimum cost considerations may be relevant include: — bridges crossing very wide waterways with many piers exposed to vessel collision — bridges with piers located in very deep waters ~ existing bridges constructed without consideration of vessel collision, In these cases the objective is to choose the level of risk that provides the minimum aggregate sum of protection costs and expected present value of future consequences of collision accidents. Appropriate analyses methods have been reported by Knott et al. [4-8], Leslie et al. [4-10], and Sexsmith [4-18]. ‘An important clement in the analysis is the assessment of the costs involved in a vessel collision accident. With reference to the classification of consequences suggested in Section 4.1 the following details, among others, apply: Damage to the Bridge The direct consequences to the bridge owner are: — the costs of salvaging the wrecked parts of the bridge ~ the costs of repair/replacement of the bridge ~ loss of revenue in the repair/replacement period if the bridge is a toll bridge. ‘The owner may also consid — additional costs of repair/replacement due to more strict requirements imposed by the Author- ities after the accident. Damage to the Users of the Bridge The direct consequences to the bridge users are: injuries and loss of lives — damage to vehicles and goods.Optimum Cost Acceptance Criteria 35 If required, the loss of human lives may be converted into economic terms by assigning cost values for fatalities. In many countries such fatality cost values have been established for use in connection with assessment of the benefits of improving the safety of transportation facilities or evaluation of alternative transportation schemes. Damage to the Vessel and Cargo The direct consequences to the vessel owner are: ~ loss of lives — costs of salvaging the vessel — costs of repair/replacement of the vessel — lack of income during the repair period — damage to goods carried on the vessel claims for compensation from bridge owner and users increased insurance costs. Inconvenience Costs to Society and Business ‘The indirect consequences to industry, trade and society are: — road and railway inconvenience costs, which depend on the strategic importance of the bridge, i.e, the availability of alternative routes, the density and type of traffic on the bridge, etc. — port interruption costs if the bridge/vessel wreckage is blocking an important navigation channel — business and social losses due to disruption of traffic/transport over a substantial period of time. Social Losses The indirect societal consequences due to long term disruption of traffic ate very difficult to quantify in monetary terms. Damage to the Environment The indirect consequences to the environment due to spills and releases from vessel or from cargo carried on the bridge include: — cost of cleaning up — cost of physical restoration — long term ecological damage. The loss of environmental values in case of long term ecological damage ate very difficult to quantify in monetary terms.37 5. COLLISION RISK In general, it is not feasible to design all parts of a bridge structure to withstand worst case loads from ship impact. However, it is possible to estimate frequencies of overloading of bridge structures due to ship impact, With knowledge of the frequencies of overloading, a design can be selected which fulfils certain acceptance criteria as described in Section 4. Vessel collision accidents to bridge structures are relatively rare and conditions differ from bridge to bridge. Therefore, the estimation of the risk of collision can not be based on vessel/ bridge collision statistics alone. Collision risk models, simulating potential collision scenarios are necessary, In this section practical models for estimating frequencies of ship collisions with bridges are addressed. When establishing a vessel collision risk model the following approach is recommended: The navigation conditions in the vicinity of the planned bridge should be studied, and vessel accident data, i.e. groundings, vessel/vessel collisions, vessel/lighthouse collisions, etc., should be collected. — Vessel/bridge collision scenarios should be defined and modelled so as to incorporate the particular conditions of the bridge site. - Subsequently, data for estimation of the parameters to be used in the collision risk model should be collected. ‘Vessel/bridge collision risk models have been developed by a number of researchers, either for specific projects or for the establishment of general guidelines for bridge designs, State-of-the-art reports have been published in 1983 and 1992, ref. Larsen [5-19] and Olsen et al. [5-32]. Some examples of bridge collision risk models are as follows: — Establishment of vessel collision load specifications to comply with pre-set bridge disruption acceptance criteria for the proposed Great Belt Bridge Project in Denmark, 1978. The model was developed by CAP-Consult, ref, Frandsen et al. [5-5] — Assessment of the probability of disruption of the Tasman Bridge in Australia, reconstructed after a vessel collision accident, for the evaluation of benefits of constructing an alternative crossing, the Bowen Bridge, 1978. The model was developed by Maunsell & Partners, ref, [5-26] and Leslie [5-21]. — Assessment of the probability of disruption considering alternative protection measures, of the new Sunshine Skyway Bridge to replace the existing bridge which collapsed due to a vessel collision accident. Models were developed by COWIconsult, ref. Larsen [5-19] and Greiner, ref. Knott et al. [5-16].38 Collision Risk ~ Assessment of the probability of vessel collision with the bridges in the Tokyo Bay Crossing Scheme, ref. [5-37]. ~ Establishment of vessel collision design criteria for bridge piers in Louisiana Waterways in USA, 1985, The model was developed by Modjeski & Masters, ref. [5-30]. — Assessment of the probability of disruption and the probability of serious pollution for alterna tive bridge solutions and alternative protection arrangements for a bridge across the Straits of Gibraltar, 1986. The model was developed by COWIconsult, ref. Larsen et al. [5-20]. ~ Establishment of Guide Specification for Vessel Collision Design of Highway Bridges in the USA, including a model for the computation of annual frequency of collapse developed by Greiner assisted by COWIconsult, 1990, ref. Knott et al. [5-17] — Risk evaluation of crossing alternatives and establishment of vessel collision load specifica tions for the bridges in the Great Belt Link to ensure compliance with pre-defined bridge disruption and users’ fatality risk acceptance criteria, 1991. The model was developed by COWI- consult, ref. Olsen et al. [5-32], and Frandsen et al. [5-6]. 5.1 Collision Risk Model The general collision risk model is expressed in the following form, where the first summation refers to all ship classes considered and the second summation refers to all bridge piers and superstructure spans: Fo = EN Py? 2 Pow: in which F = Expected number of annual collisions to the bridge (bridge piers and/or superstructure spans) N, = Annual number of vessels belonging to a certain class (i) of the vessels passing the bridge. The distribution of transiting vessels into discrete groupings (classes) shall be developed considering the required accuracy of the analysis results and the categories of consequences requested. The distribution should normally account for variations in the following vessel characteristics: — Draught—in order to be able to sort out vessels that will ground prior to collision Air draught—in order to be able to assess the probability of collision to the bridge superstructure — Collision load characteristics—in order to be able to assess the degree of damage in case of collision — Amount of dangerous goods carried on board the vessel—in order to be able to assess the risk of environmental pollution in case of collisionCausation Probability 39 — Local navigation practices—in order to be able to account for variations in geometrical probability ~ Vessel standard—in order to be able to account for variations in causation probability. P., = The “causation probability” related to the actual class of vessel (i), i.e. the probability of not avoiding a collision, in case the vessel is sailing on a course that would lead to a collision and nothing is done to avoid collision, ref. Section 5.2. Poix= The “geometrical probability” or “rate of collision candidates” related to the actual class of vessel (i) and to the actual part (pier or span) of the bridge (k), i.e. the probability that a vessel is sailing on a collision course, ref. Section 5.3. Prix = The “failure probability” related to the actual class of vessels (i) and to the actual part of the bridge (k), i.e. the probability of failure, given that a vessel collision has taken place, ref. Section 5. 5.2. Causation Probability The causation probability, P., is the probability of failure to avoid an obstacle on the navigation route. Local values of P, can be established through analysis of ship accident observations in the bridge area, Estimates of P, on the basis of collisions with bridges are not available due to lack of sufficient data. More data is available on groundings and ship-ship collisions and most estimates of P, have therefore been based on these types of accidents. Estimates of P, have been made by many researchers since the early 70’s, Two basic estimates of the causation probability were published in 1974 by Fujii et al. [5-7] and by MacDuff [5-23]. Most estimations made by others are based on either Fujii’s or MacDuft’s approach. Some of the estimations are based on grounding or collision statistics in a specific waterway. Other estimations are based on P, values found for other waterways, adapted to the specific waterway using correcting factors for differences in environmental and navigational aspects. Selected estimations are mentioned in the following: ~ Analyses of platform collisions and groundings in Japanese Straits by Fujii [5-7] resulted in P, values ranging from 1.0 to 6.3 - 10, — In Matsui et al. [5-25], groundings, buoy collisions and different types of vessel/vessel collisions have been analyzed for 10 Japanese straits. For groundings the P, values range from 0.8 104.3 - 10°, ~ MacDuff [5-23] analyzed groundings in the Dover Strait and found a P, value of 1.4 - 10%. Vessel/vessel collisions were analyzed in two periods (before and after the introduction of traffic separation) resulting in P, values of 5.2 - 10 and 3.2 « 10“ respectively.40 Collision Risk — CAP-Consult estimated a P, value of 0.4 - 10° for bridge collisions in the Great Belt, ref. Lar- sen [5-19], on the basis of Fujii’s results for Japanese Straits, adjusted for traffic regulations, marking of route, detectability of collision object, etc. — Maunsell & Partners et al. [5-26] estimated the P, for collision with the Tasman Bridge to be within an interval of 0.7 to 1.0 - 10°, This value was found on the basis of MacDuff’s results for the Dover Strait, adjusted for visibility and other environmental conditions, human error, mechanical failure, and traffic density. = COWlconsult estimated the P, for collision with the Sunshine Skyway Bridge to be 0.5 + 104, ref, Larsen [5-19]. This value was found on the basis of Fujii’s results for Japanese Straits, adjusted for traffic density, use of pilots and traffic restrictions. ~ CBA-Buckland & Taylor calculated a P, value for the Fraser River, British Columbia, at the Annacis Island Bridge to be 3.6 +10°, ref. Sexsmith [5-35] — Greiner has carried out estimations of P, for 5 US bridges, ref. Knott et al. [5-17]: + Sunshine Skyway Bridge, Florida: 1.3 - 10“ for ships and 2.0 - 10% for barges + Francis Scott Key Bridge: 1.0 - 10° for ships and 2.0 - 10“ for barges ‘Wm Preston Lane, Jr. Mem, bridge, Maryland: 1.0 - 10" for ships and 2.0 - 10 for barges Chesapeake Bay Bridges and Tunnels, Virginia: 0.7 - 10“ + Dames Point Bridge, Florida: 1.3 - 10° for ships and 4.1 « 10 for barges. ‘The estimations are based on statistics of groundings and vessel/vessel collisions in the waterway near the bridge locations. When possible bridge collisions have also been included. — Modjeski & Masters [5-30] have calculated P, for bridge collisions in a number of waterways in Louisiana, For the Mississippi River bridges the results were: + Vicksburg Bridge: 5.4 + 10“ + Huey P. Long Bridge: 2.5 - 10 + Greater New Orleans Bridge: 1.3 - 10°. — For the planned fixed link across the Strait of Gibraltar, COWIconsult estimated P, on the basis of records on vessel traffic accidents in the Strait, ref. Larsen et al. [5-20]. Present values of 2.2 + 10* for strandings and 1.2 » 10“ for collision were found and for vessel/bridge collisions a future value of 0.6 - 10“ was assumed based on improved traffic safety. ‘The US Guide Specification [5-17] indicates that P, can be estimated using the basic values 0.6 10* for ships and 1.2 + 10* for barges and adjusted with correction factors for channel tum or bend, cross currents, longitudinal currents and vessel traffic density. — For the Great Belt Crossing East Bridge and West Bridge, COWIconsult estimated P, values of 1 + 10° for vessels with pilots on board and 3.2 - 10“ for vessels without pilots on board based on Fujii’s results for the Straits of Japan and adjusted for visibility, ref. Lund et al. [5-22]. The sizes of P, were verified by analysis of local accident data, ref. [5-28]. When assessing the P, value for a specific waterway, or when comparing the P, values for different waterways, a series of influential factors should be considered, Some of the most important factors influencing the causation probability or other parameters of the collision risk model are reviewed in the following:Causation Probability 41 Visibility Visibility is one of the most important factors affecting causation probability, especially when no radar facilities are available, The effect of visibility on the probability of vessel/vessel collision in the Dover Strait has been analyzed by Wheatley [5-38] and the probability of vessel/vessel collisions and groundings in Japanese Straits have been analyzed by Fujii et al. [5-8], [5-9] and [5-11]. These analyses show that the probabilities of collision and stranding increase drastically for reduced visibility conditions. The probability of collision is approximately inversely proportional to the visual range. As an example, the probability of collision in fog when the visual range is less than 200 m has been found to be about 100 times the risk in clear visibility conditions. Darkness The effect of darkness is closely related to that of the visibility. Based on analysis of collision and grounding accidents, Fujii et al. [5-7] proposed that the probability at night time should be taken to be 4 times the probability at day time. Wind Analyses of wind effects on a specific bridge passage by use of a manoeuvring simulator carried out by Meurs et al. [5-27] indicate that the probability of collision to the bridge piers was 3 times greater at wind speeds of 11-14 m/sec compared to 6-8 m/sec. Current ‘The US Guide Specification [5-17] proposes the following empirical factors, R, and R,., for the increase of probability of collision due to current: 1 + VJ/10 for parallel current 1+ V,. for cross current, in which V, is the current component parallel to the vessel path in knots and Vj. is the current component perpendicular to the vessel path in knots. An investigation made by Fujii et al. [5-8] did not show any significant influence of currents. Sea State The manoeuvrability of a ship and the visibility for the navigator is reduced in heavy sea conditions. In addition, water levels may be abnormally raised by storm surge tides, thus increasing the vulnerability of the bridge, However, during heavy sea conditions, the number of passing vessels will be reduced and one- way traffic may normally be assumed. No quantification of the influence on the collision probability has been found in the literature.42. Collision Risk Ice Ice conditions can cause severe difficulties for navigation as the manoeuvrability may be significantly reduced. However, no quantification of the influence on risk has been found in the literature. Vessel Type Fujii et al. [5-8] conclude for Japanese waters that passenger vessels are 6 times safer than cargo ships and tankers, and further state that cargo ships or tankers are 3 times safer than fishing boats, Part of this may be due to the better manoeuvring characteristics of ferries and part may be due to better knowledge of local conditions. Vessel Size and Speed According to Fujii et al. [5-7] the danger of vessel/vessel collision is almost proportional to the vessel length. According to Mizuki et al. [5-29] this tendency has, however, not been found in analyses of groundings and collisions to fixed objects. Vessel Loading Condition The manoeuvring capabilities of vessels in ballast are inferior to those of loaded vessels, in particular in the case of strong winds. However, no quantification of the effect on the collision probability has been found in the literature. Standard of Vessel, Crew and Equipment Cockcroft [5-2] and Fujii et al. [5-10] state that there exists a significant difference among vessel groups of different nationalities. Cockeroft [5-2] has found that the accident probability is remarkable higher for vessels more than 20 years old compared to vessels less than 10 years old. Det Norske Veritas has reported that 6.8% of all casualties are due to faults in the technical systems of the vessels, ref. [5-3] Pilot on Board The presence of a pilot on board reduces the probability of accident considerably. As an indica- tion, Fujii et al. [5-12] have found that the accident rate per trip is decreased by a factor of up to 6.Causetion Probability 43 ‘Traffic Density ‘The US Guide Specification [5-17] covers the increase of collision risk by introducing a factor R, defined as 1.0 when vessels rarely meet (low density), R, = 1.3 when vessels occasionally meet (average density) and R, = 1.6 when vessels routinely meet (high density). Based on analytical modelling of vessel/vessel collisions, Fujii et al. [5-8] and [5-10] suggest that the number of vessel/vessel collisions is proportional to the square of the traffic density. This proportionality should, however, not be applied directly in the case of vessel/bridge collision, Detectability of Collision Objects It seems that it should be easier for navigators to detect a bridge than a low island or an under- water shoal. However, no quantification of the effect of the detectability of collision objects has been found in the literature. Presence of Navigational Aids Matsui et al, [5-25] found the probability of collision to buoys in Japanese waters to be approximately the same probability as for groundings. It indicates that almost all navigators avoid hitting buoys. Therefore a suitable buoy arrangement may decrease the number of wandering vessels in a waterway. ‘Turns or Bends in a Route ‘The US Guide Specification [5-17] proposes the following empirical factor R, for the increase of the probability of collision due to turns or bends in the navigation route: Ry = 1+0/90° for a bridge located in a transition zone, i.e. between 3000 and 6000 feet from the turn point or between 0 and 3000 feet from the point where the bend staris. — Ry = 1+/45° for a bridge located in a turn or bend zone, i.e. within a distance of 3000 feet from the turn point or within the curve region of the bend. Modjeski & Masters [5-30] suggest a factor between 2.5 and 4 when a curvature in the waterway exists within 1 mile from the bridge. ‘Width of Navigation Channel The importance of the width of the navigation channel is obvious if the clearance of the bridge is significantly smaller than the width of the channel. In that case the bridge will constitute a “bottle- neck” and hence increase the number of “close encounters”, ref. Ostenfeld-Rosenthal et al. [5-33] Presence of Vessel Traffic Management System Studies of the effectiveness of Vessel Traffic Management Systems, often referred to as VTS systems, indicate risk reducing factors of approximately 2-3 for VTS systems comprising VHF communication, radar surveillance and automatic target tracking. Results from different studies of the effectiveness of introducing a VTS system are summarized in Table 5.1.44 Collision Risk Study Achievable risk reduction factor Great Belt VTS system, ref. Olsen et al. [5-31] 2-3 USS. VTS systems, ref. US Coast Guard [5-36] 2-3 Canadian VTS systems, ref. Canadian Coast Guard [5-1] 20-33 European VTS systems, ref. Fabre et al. [5-3] 22-25 Japanese traffic systems (maritime traffic, aviation and road traffic), ref. (5-29] eae Table 5.1 Results from different VIS effect studies. The studies may not be directly comparable due to differences in systems, estimation methods and data basis. 5.3 Geometrical Probability The probability that a vessel is sailing on a collision course, the geometrical probability P;, is a function of a series of circumstances and strongly dependent on the local conditions. The rate of collision candidates is calculated on basis of the following information: vessel traffic routes, number of passages and characteristics of the vessels as described in Section 3 distributions of the ship traffic across the routes — geometrical definition of bridge — natural protection by the seabed man-made protection works as described in Section 9. Based on experience from accident records, ref. Appendix B and Olsen et al. [5-32] and Lund et al. [5-22], the following accident scenarios are normally of interest: a) Vessels that follow normal navigation procedures during transit but are put at risk in the vicinity of the bridge due to human error or technical failure b) Vessels making evasive manoeuvres because of multiple encounter situations in the vicinity of the bridge. c) Vessels that fail to change course at a bend in the navigation channel close to the bridge. d) Vessels not following the ordinary navigation route (Wandering Vessels) and vessels drifting towards the bridge due to anchor or mooring failure, loss of propulsion, trouble with navigation equipment, caught in an ice drift, or the like (Drifting Vessels).Geomerrical Probability 45 The geometrical probability for scenario a) is defined as shown in Figure 5.1, as a function of the vessel track distribution before the bridge passage and the width of the collision zone. For a bridge pier, the width of the collision zone is equal to the breadth of the vessel plus the breadth of the pier. For a bridge superstructure, the width of the collision zone is the whole leagth of the bridge where the air draught to the top of the vessel deck house exceeds the high water clearance of the bridge girder. The relative distribution of the vessel passages in the 4 scenarios is specific to the bridge location in question. The aggregate vessel track distribution is arrived at by adding the track distribution resulting from each scenario. For many bridges it will be sufficient to consider scenario a) only. However, in the case of long bridges with multiple spans outside the navigation route it will often be reasonable to add a certain percentage of uniform vessel track distribution from scenario d) to ensure the robustness of the bridge structure at some distance from the navigation channel. ™ Bridge Pler Semen Centerlines of Navigation Routes y Vessel Track Distribution Figure 5.1 Definition of geometrical probability of collision for scenario a)46 Collision Risk Scenario a) The lateral distribution of vessel tracks in a straight, well-defined route is often approximated adequately with a normal distribution. The parameters in the normal distribution (mean value and standard deviation) may be obtained from direct radar observations of the actual waterway, from comparison with similar straits where radar observations have been made, or from empirical rules. Figure 5.2 shows an example of the results of the analyses of radar observations made for the Great Belt Eastern Channel. Relative Frequency - Centerline of Separation Zone Ua Histogram from 33 Vessel passages in East Channel, Great Belt Normal Distribution 048 474 0474948 1422 Distance from @ (m) East <——__»—___» West Figure 5.2. Lateral distribution of vessel tracks, calculated from radar observations. Based on a review of historical ship/bridge collision data, the US Guide Specification [5-17] indicates that a normal distribution should be utilized to model the errant vessel's sailing path near the bridge. The standard deviation, o,, is assumed to be equal to the length of the vessel. The location of the mean, p, is assumed to be the centre line of the navigation channel, ice. in case of two-way traffic the centre line of the inbound and the outbound half of the channel respectively.Geometrical Probability 47 Observations from the strait of Bisan Seto in Japan and the Great Belt in Denmark, ref. Figure 5.2, have shown that the standard deviation can be twice the length of a typical large ship. The typical large ship was defined as the 95% percentile in the ship size distribution, ref. Olsen et al. [5-32] Inoue [5-15] and Kuroda et al. [5-18] have presented an empirical formula for the distribution of two-way traffic in a navigational channel. They have concluded, divergent to the above indications, that the vessel track distribution can be approximated with a Gaussian distribution with the following mean value p, and standard deviation o: Bo =a-W = 0.105 - W (approximation) in which w width of navigation channel a = 0.2 if the channel is equipped with centre line markings and 0.1 if the channel is without centre line markings Figure 5.3 illustrates the principle of lateral distribution of vessel tracks in connection with a scenario a) bridge passage. Scenario b) If vessels often meet other vessels in the vicinity of the bridge line, the probability of collision will increase. The frequency of critical meetings may be analyzed by the “Close Encounter” analysis method, ref. section 2. Scenario c) Some vessels fail to make a proper change of course at a channel bend, Most will soon return to the correct route, but a small fraction will constitute a group of potential collision candidates. The geometrical probability may be calculated as for scenario a) using the vessel track distribution before the bend and reduced by a factor reflecting the probability of the vessel failing to correct the course before the bridge passage. With reference to Olsen et al. [5-32] this probability is a function of the distance from the bend to the bridge line and the probability of omission of position checks by the navigator. Some authors have included the effect of a channel bend in the causation probability, ref. Section 5.4. Figure 5.4 illustrates the principle of lateral distribution of vessel tracks in connection with a scenario c) bridge passage.48 Collision Risk Centerlines of Navigation Routes Distribution y Vessel Track Distribution Navigation Channel L Figure 5.3. Vessel track distribution for two-way bridge passage in a straight, well defined waterway: Scenario d) The distribution of the tracks of “Drifting Vessels” may be established as a function of the distance to locations where vessels might break away and the prevailing directions of wind, currents and ice drift. The distribution of the tracks of “Wandering Vessels” may be established on the basis of inter- views with local authorities to disclose the extent of non-compliance with local rules, such as short-cutting of the route, etc. In the absence of specific knowledge, a uniform distribution across the entire bridge length may be assumed. The intensity of the distribution may be assumed to be 1-5% of the general vessel traffic density in the waterway, ref. Pyman et al. [5-34]Failure Probability 49 Centerines of Navigation Routes Vessel Track Navigation Channel a Figure 5.4 Vessel track distribution for vessels at a critical bend in the navigation channel close to the bridge passage. 5.4 Failure Probability The probability of failure of the bridge structure, given that a vessel collision has taken place, is complex and is a function of the vessel size, type, speed, direction, mass, and the nature of the collision. It is also a function of the stiffness/strength characteristics of the bridge pier and span to resist the collision impact loads, ‘Two different approaches may be applied to assess the probability of failure: ~ the “Heinrich Ratio” approach ~ the probabilistic approach.50 Collision Risk The Heinrich Ratio Approach The “Heinrich Ratio” is defined as the ratio of severe accidents to all accidents, ref. Heinrich [5-14] and Olsen et al, [5-32]. In the present context the “Heinrich Ratio” is defined from the basis of vessel collision accidents, Collision statistics for vessel/vessel collisions have to be used because data on vessel/bridge collisions is sparse, ref. Fujii (5-13] By applying energy considerations regarding the differences between vessel/vessel and vessel/pier collisions, a Heinrich ratio for vessel/pier collision is obtained. Vessel/vessel collision data has been analyzed by Fujii [5-13]. Fujii proposed an empirical relation between the cumulative relative frequency of damage, F(x,y), the damage rate, x, and the vessel size ratio, y. The two parameters x and y are defined as follows: — The damage rate, x, is the ratio of the cost of repairing the vessel hit over the total value of the same vessel. ~ The vessel size ratio, y, is the ratio of the gross tonnage of the colliding vessel over the gross tonnage of the vessel being hit. Figure 5.5 shows the cumulative relative frequency of damage, F(x.y), reproduced from Fujii [5-13]. F(x,y) is equal to the Heinrich ratio. ‘The damage rate x where the bridge element is damaged more than acceptably, i.e. involving interruption of bridge traffic for an extended period of time, is not well defined, In different vessel collision studies it has been assumed that a value for the damage rate of about 0.10 results in serious damage to the bridge element. On the basis of Figure 5.5, energy considerations and the damage rate assumed, it is now possible to construct a Heinrich function giving the frequency of serious vessel/bridge collisions relative to all collisions. The Probal The probabilistic approach is based on a probabilistic model for the vessel impact force and a spatial stochastic model of the resistance properties of the bridge elements, ref. Madsen et al. [5-24]. Different failure modes of the bridge elements (e.g. crushing, rotation, sliding, etc.) shall be istic Approach considered. The probabilistic model for the vessel impact force is based on probability distributions for: the type, size, speed, and loading condition of the colliding vessel the collision force as a function of the vessel characteristics the collision energy as a function of the characteristics of the vessel and the water depth the angle of attack the position of the collision relative to the pier the friction coefficient between vessel and pier (in case of skew collisions).Failure Probability 51 Cumulative Relative Frequency of Damage Rate, F (xy) 1 04 y> 22 10 ~ 32 3.2 ~ 10 N1+-32 0.01 0.32 ~1 0.001 Ot ~ 0.92 lo y < 0.032 0.0001 0,001 0.01 0.1 1 Damage Rate, x Figure 5.5. Cumulative relative frequency of damage rate “Heinrich ratio” as a function of the damage rate (from Fujii [5-13]). For parameters where no probability distribution data are available, estimates may be used. Sensitivity studies should, however, be carried out in such cases.53 6. VESSEL IMPACT FORCES ‘The determination of the impact load on a bridge structure during a vessel collision accident is very complex as it depends on the vessel characteristics and the bridge structure, as well as the circumstances of the collision accident. Some important parameters are: ~ For the colliding vessel: type, size, shape, speed, loading condition, and strength and stiffness of bow, hull, and deckhouse. For the bridge elements in contact with the colliding vessel: size, shape, mass, and lateral resistance characteristics. ~ For the collision circumstances: eccentricity of impact and water depth. Most research has dealt with the ideal case of a vessel colliding with an infinitely rigid and immovable vertical plane wall structure. In this case the kinetic energy of the vessel together with the water surrounding and moving with the vessel is consumed totally through the deformation and crushing of the vessel. In a less ideal collision event, parts of the energy will be consumed in deformation and crushing of fenders, displacement of bridge pier and bridge superstructure and liberation of energy to the surrounding water. “Realistic” collision cases therefore involve lower impact forces, longer impact durations and less damage to the vessels than an “ideal” collision case. Collision impact load definitions are normally required in the following cases as described in the subsections below: — Head-on Bow Impact Forces ~ Sideways Impact Forces ~ Deckhouse Impact Forces — Local Impact Loads. An additional subsection: — Barge Vessel Impact Forces has been included to cover barge vessels which differ substantially from ship vessels and are of particular importance to bridges crossing inland waterways.54 Vessel Impact Forces 6.1 Head-on Bow Impact Forces Most literature on vessel collision deals with the forces involved in a head-on bow collision accident. An outline of the most important conclusions is given in the subparagraphs below: — Physical and Mathematical Models — Collision Force Definitions for Bridge Design — Forces Experienced in Vessel Collision Accidents, ~ Collision Force Definitions for Dynamic Analysis ~ Parameters Influencing the Impact Forces. Physical and Mathematical Models Various analysis models have been derived for estimation of the global loads involved in head-on bow collision events. The models have been based on: investigation of ship/ship collision cases ~ dynamic collision model tests = quasi-static bow indentation model tests = direct calculation of crushing resistance — combinations of the above. A detailed account of the development of the state-of-the-art has been given by Pedersen et al. [6-28], the most important contributions being the following: — A pioneer analysis model was published by V. U. Minorsky in 1959 [6-20]. Based on an investigation of 26 ship/ship collision cases, a linear relationship was found between the deformed steel volume and the absorbed impact energy. Minorsky’s well known diagram is reproduced in Figure 6.1. ~ In Japan, model tests have been carried out since 1958. The first of these were dynamic low energy tests in which the bow acted as a pendulum ramming into the ship side, ref. Kagami et al, [6-13]. — Dynamic model tests were carried out in Italy during the 1960's and the first half of the 1970’s as part of the design of the nuclear powered ship Enrico Fermi, ref. Euratom [6-3]. The tests were carried out by letting models of vessel bows run down a roller conveyor and ram into a model of the protective structure. ~ During the period 1967-1976, a number of similar high energy dynamic ship collision model tests were carried out in Germany by Woisin, ref. [6-38] and [6-39]. The purpose of these tests was also to give input to the design of strctures to protect nuclear powered ships. The set-up of the tests is illustrated in Figure 6.2. = Quasi-static indentation tests, where vessel models were deformed by means of hydraulic jacks, were carried out in connection with the design of the Bisan-Seto Bridges in Japan. Akita et al. [6-1] describe six collision tests and Hagiwara et al. [6-9] propose a method for prediction ofHead-on Bow Impact Forces 55 collision damage based on model tests. Nagasawa et al. [6-22] describe model tests simulating impact between small vessels and a rigid bridge pier. — Ohnishi et al. [6-25] describe similar model tests with two different bow structures, i.e. a 35,000 DWT container vessel and a 409,000 DWT tanker with bulbous bow against a rigid pier. Amdahl [6-2] has carried out a number of quasi-static load deformation tests on stiffened cylinders and models of ship bows. — Numerical procedures for the prediction of crushing loads of structures have been proposed by Reckling [6-30], Gerard [6-7], and Wierbicki [6-36]. On this basis more elaborate models with the specific purpose of evaluating vessel collision loads have been developed by Amdahl [6-2] and Yang and Caldwell {6-40}. The merits of the various methods have been summarized by Pedersen et al. [6-28]. 4000: 3000. x 8 3 8 LOW ENERGY POINTS b DEFORMED STEEL VOLUME (Ft? x In) 3 8 8 0 ‘500 1000 1500 ABSORBED ENERGY (1000 TONS - KNOTS)? Figure 6.1. The absorbed impact energy in shiplship collision accidents as a function of the deformed steel volume (reproduced from Minorsky [6-20]).56 Vessel Impact Forces Figure 6.2. Elevation view of the set-up of Woisin’s collision tests (from Frandsen et al. (6-4]) Most of the above research and testing has aimed at establishing the relationship between the impact force and the indentation (damage depth) when crushing of the vessel takes place during a collision accident. Two representative force-indentation relationship curves are shown in Figure 6.3 together with the corresponding force history curves. The curves A indicate dramatic fluctuations of the force during a very short (0.1-0.2 sec) initial phase of the collision event followed by a more or less constant force during the remaining time, ref, Woisin [6-38] and [6-39] The time duration of the maximum force, which has been estimated at twice the constant average force during the remaining collision time, is normally considered to be too brief to leave any influence on a relatively robust bridge structure. Therefore, bridge designers have as a rule utilized the average impact force as an equivalent static load for design against vessel collision. Reference is made to the examples on design force definition in the following. The curves B indicate a gradually increasing impact force during the whole course of the collision accident, ref. Pedersen et al. [6-28]. This may not involve higher average force but theHead-on Bow Impact Forces 57 Force ® a > Indentation ( Damage Length ) Time Figure 6.3. Schematic representation of impact force dynamics according to the traditional understanding, A, and according to recent research, B. longer duration of the relatively high force level during the last phase of the collision event imply that the maximum force rather than the average force should be used in design when based on static force analysis. Collision Force Definitions for Bridge Design A series of examples of head-on impact force estimates developed on the above basis for design purposes can be found in the literature (DWT indicates dead weight tonnage in metric tonnes): — Ina study, ref, [6-26], Von Olnhausen computed the impact load for a 40,000 DWT tanker and 45900 DWT dry cargo vessel using a collision energy formula developed by Minorsky [6-20]. ‘The computed load was 145 MN for the 40,000 DWT vessel. — The impact loads recommended by the Nordic Road Engineering Federation [6-23] were developed from the study by Olnhausen [6-26]. The recommended design loads as a function of vessel size and speed are reproduced in Figure 6.4. — Also based on the Minorsky relationship, Woisin et al, [6-37] developed an equation giving the average collision force directly: Py = W8- 171100 where:58 Vessel Impact Forces P, = the average collision impact load (MN) Vv the vessel velocity (m/sec) c the vessel’s length (m). This equation has been used in the design of the Bahrain Causeway and the Faroe Bridges, ref. Jensen et al. [6-12]. MN (Mp) = Navigation A _|_ Navigation Channel Channel 150, 75 (15.000) +~ (7.500) 16 Knot 10 Knot 100 | 60 (10.000) +~( 5.000) 5 Knot 50 [25 (5.000 ) ( 2.500 ) > 67 8 0 0) 12 Dreight (m) | be 2000 3300 5000 7500 11000 Ship Size ( BRT ) | > 3200 5000 7500 11000 20000 40000 80000 Ship Size ( DWT) Figure 6.4 Diagram showing vessel collision design loads as published by the Nordic Road Engineering Federation in 1975 [6-23]. 1 knot = 0.514 msecHead-on Bow Impact Forces 59 — Very detailed and thorough investigations on the subject of vessel collision with bridges were performed in connection with the proposed Great Belt Bridge project in Denmark in 1977-78, ref. Frandsen et al. [6-4] For large vessels (DWT > 40,000 tons) the collision forces were developed by Woisin [6-38] and [6-39], mainly based upon physical model tests. The tests indicated that the average impact load from a 250,000 DWT fully loaded tanker sailing at 16 knots was between 250 MN and 700 MN. The scatter was due to differences in bow construction, fore peak tank ballast, etc. Woisin suggested a triangular probability distribution of the average impact loads within the above-mentioned limits. It was chosen to use collision forces which were equal to the 70%- fractile of this distribution for the following reason: for a given design vessel the number of smaller vessels with a crushing strength greater than the 70%-fractile force of the design vessel would be approximately equal to the number of larger vessels with a crushing strength less than the 70%-fractile force of the design vessel. The collision forces for vessel sizes down to 40,000 DWT were established by assuming that the crushing forces for vessel sizes between 250,000 DWT and 160,000 DWT were proportional to (DWT), and for vessel sizes between 160,000 and 40,000 DWT the crushing forces were proportional to (DWT)', The collision force of a specific vessel was additionally assumed to be proportional to the vessel's velocity and to the square root of the displacement. IMPACT FORCE (10° TONS) 70% PROBABILITY DENSITY FUNCTION, 0% DWT(10? TONS) 50 100 150 200 250 Figure 6.5. Diagram showing collision design loads assumed for the Great Belt bridge project in 1978 (reproduced from Frandsen et al. [6-4]. The impact speed is 8misec 103 tons = 9.8 MN.60 Vessel Impact Forces The 70% fractile of average collision forces assumed in the design as a function of vessel size is reproduced in Figure 6.5. — Saul et al. [6-32], [6-33], and [6-34] have presented a modification of the Woisin approach. They have proposed the following relationship for bulk carriers: 0.88 (DWT)"" + 50% Praag where: P.yux = the effective maximum impact force of the vessel (MN) — Based on studies performed for the Sunshine Skyway replacement bridge in Florida, Knott [6-14] proposed that the Saul/Svensson equation be modified to reflect reductions in the maximum impact (P,,,) when the vessel is travelling at speeds lower than the maximum (estimated at 16 knots) and when the vessel is travelling in a partially loaded, ballasted, or a light (empty) condition. The equation utilized for design collision loads was as follows: Pao = 0.88 (DWT)! (V/8)"?* (Dyg/Dpas)"” where: V__ = the vessel velocity at the time of impact (m/sec) Da = the vessel displacement at the time of impact (tonnes) Dyaxc= the maximum (fully loade) displacement of the vessel (tonnes). — In 1985, the State of Louisiana, USA, issued “Criteria for the Design of Bridge Piers with Respect to Vessel Collision in Louisiana Waterways”, ref. Modjeski & Masters [6-21] The report contains design charts for deep draught waterways indicating collision forces for 14 knots and 7 knots impact speeds as a function of water depth and vessel size. The design chart for 14 knots impact speed is reproduced in Figure 6.6, — The Norwegian Public Roads Administration prescribes the following vessel collision loads for bridges and ferry ramps in the public road system, ref, [6-24]: P = 0.5 (DWT)? " where: P static equivalent collision force (MN) — For the Euroroute bridge/tunnel crossing proposal for the English Channel, vessel collision design basis was derived for a 270,000 DWT tanker, ref. Teyssandier [6-35]. A static force of 500 MN was assumed and 30% was addec for dynamic effects.Head-on Bow Impact Forces 61 Collision Force (1000 kips ) A 60 50 340 30 50 2 8 Deadweight Tonnage ( 1000 DWT ) mol TT 10 eg 0 i 20 30 40 50 60 Water Depth ( feet ) Figure 6.6 Collision load design charts for 14 knots impact speed as recommended by the State of Louisiana in 1985 (reproduced from Modjesky & Masters [6-21]) 1 knot = 0.514 misec. 1000 kips = 4.45 MN.62. Vessel Impact Forces Average Collision Force ( MN ) il A rt 1000 L i I 900 : I ! 800 t Deformable Vessels Loaded ! 700 600 500 400 Ridgid Vessels Loaded 300 Ridgid Vessels in Ballast 200 100 0 > 1.000 10.000 100.000 1.000.000 Tonnage ( DWT ) Figure 6.7 Average collision forces developed for the Gibraltar Strait crossing. Two types of vessels were considered, both types in ballast as well as loaded condition (reproduced from Larsen et al. [6-17])Head-on Bow Impact Forces 63 Ship Impact Force ( 1.000 kips ) A 80 70 40 0 2 4 6 8 10 2 14 16 Impact Speed ( knots ) Figure 6.8 Typical impact forces according to the US Guide Specification [6-15] J knot = 0.514 misec 1000 kips = 4.45 MN.64 Vessel Impact Forces ~ For the Gibraltar Strait Crossing a collision load design basis was developed in 1987-88, ref. Larsen et al. [6-17]. Based on the 1978-79 Great Belt studies, collision forces were extrapolated to cover possible future 1,000,000 DWT tankers. For container vessels and other rigid types of vessels an increase of the collision force compared to tanker and similar vessels was proposed. The collision force diagram developed is reproduced in Figure 6.7. — For the Dartford bridge across the River Thames in England, an equivalent static collision load of 350 MN has been assumed, corresponding to a 65,000 DWT vessel sailing at 5 m/sec, — For the Luling Bridge across the Mississ:ppi River near New Orleans, an equivalent static collision load of 270 MN was assumed, corresponding to a 40,000 DWT vessel. ~ The US Guide Specification [6-15] prescribes the following formula (translated into ST-units) for the computation of an equivalent static impact force for vessel collision: Bow Crushing Force, P hoy ( MN ) A 10.000 Pep eee ears > 500 1.000 5,000 10.00 60,000 100.000 500.000 Ship Size ( DWT) Figure 6.9 Diagram showing maximum vessel impact forces resulting from the investigations carried out in connection with the design of the Great Belt Bridge project (reproduced from Pedersen et al. [6-28])Head-on Bow Impact Forces 65 0.98 (DWT)'”- [V/8] ~ 0 impact force (MN) the vessel velocity (m/sec). < " A diagram from [6-16] showing typical ship impact forces is reproduced in Figure 6.8. Detailed and comprehensive vessel collision load regulations have been established for the Great Belt Bridge project in Denmark in 1991 based on existing knowledge as supplemented by extensive numerical calculations by Det Norske Veritas, ref. Frandsen et al. [6-5] and Pedersen et al. [6-28] The load-indentation relationships have been calculated, using the methods developed by Amdahl [6-2] and Yang & Caldwell [6-40]. Based on the dynamic characteristics found, it has been assessed that the maximum impact forces rather than the average forces should be used as the design force for design on the basis of equivalent static analysis. Maximum impact forces have been established for vessels between 500 DWT and 300,000 DWT. Different impact speeds and loading conditions were considered. Based on the results of the analysed collision cases, an empirical expression has been derived for the easy estimation of the maximum bow collision loads of ordinary merchant vessels: Pyoy = Py [E+ L? +(6.0-L) - L?]!” for E> L?* Prow = Pg: [5.0* E+ L]'? for E 1,000 5,000 10,000 50,000 100,000 500,000 1,000,000 Vessel DWT ( Tonnes ) Figure 6.10 Diagram showing vessel collision design forces as calculated from different sources. To facilitate comparison, all curves indicate design impact forces of fully loaded ordinary merchant vessels impacting at 7-8 m/sec collision speed.68 Vessel Impact Forces Figure 6.11 Vessel damage from vesselibridge collision accidents: At the top, the Newport Bridge collision accident in the USA in 1975 (from Kuesel [6-16]) in which the vessel was seriously damaged while the bridge suffered no permanent damage Below, the Tosteré Bridge collision accident in Sweden in 1991, ref. Appendix B (photo: Mats Fogeman) in which the vessel suffered only light damage while the bridge was partly destroyed.Head-on Bow Impact Forces 69 Crushing Load ( MN ) 1000 = 150,000 DWT > 0 5 10 15 20 Indentation of Bow (m) Crushing Load ( MN ) 100 50 L 3,000 DWT ie 2,000 DWT > 0 1 2 3 4 Indentation of Bow ( m ) Figure 6.12 Crushing loadlindentation relationships (based on Pedersen et al. [6-28}). At the top: ~ 150,000 DWT bulk carrier at an initial speed of 9.3 misec — 40,000 DWT container ship at an initial speed of 12.9 misec Below: - 3000 DWT general cargo carrier at an initial speed of 7.5 misec = 2000 DWT tanker at an initial speed of 7.0 misec70. Vessel Impact Forces Parameters Influencing the Impact Forces The above indications refer to “normal” vessels, colliding under “ideal” circumstances. Varia- tions from this basis should be considered in specific cases and judgment must be exercised in evaluating the influence on the collision load definition. The following typical variations should be considered: ~ The type of vessel: vessels constructed with particularly rigid hulls, i.e. container vessels, naval vessels including submarines, ice-strengthened vessels, etc., are likely to produce higher global and local collision forces than conventional merchant vessels, ref. Larsen et al, [6-17]. — The bow shape of the vessel: the maximum impact loads for bulbous bows are somewhat higher than the estimated values for conventional bows, ref. Pedersen et al. [6-28]. — The impact speed: due to the strain-hardening effect, the impact forces are dependent on the impact speed, ref. test results published by March et al. (6-18]. This is accounted for in some empirical expressions for collision loads, ref. Knott et al. [6-15] and Pedersen et al. [6-28] — The loading condition of the vessel: the maximum impact load for ballast condition is normally considered lower than the estimated impact load in the fully loaded condition. This is accounted for in some empirical expressions for collision loads, ref. Knott [6-14], Larsen et al. [6-17], and Pedersen et al. [6-28]. In the US Guide Specification [6-15] it is agreed that the reduction in the impact force due to reduced mass is offset by the increase in impact force caused by a stiffer bow due to the presence of water ballast in the vessel’s forepeak tanks. — The size of the bridge pier: in the case of collision against a pier with limited width or with a step or recess, the collision load may be adjusted by multiplying by the ratio of the deformed steel area in actual contact with the pier to the total steel area of the bow section, with the addition of shear forces equal to half the shear buckling loads of the undeformed parts of the bow outside the contact area, ref. Pedersen et al. [6-28] — The shape of bridge pier: in cases where the pier is shaped so that vessel collision accidents are likely to result in glancing blows rather than head-on impact, a large part of the collision energy will be consumed in friction and in the rotating of the vessel. The probability of development of full head-on collision forces is thus reduced, ref. Pedersen et al. [6-28]. = The eccentricity of the impact: in cases of eccentric impact, a significant part of the collision energy will be consumed in rotating the vessel. — The water depth: in cases of low underkeel clearance, the hydrodynamic mass, i.e. the mass of the water moving with the vessel, will inczease and thereby increase the collision energy. 6.2. Sideways Impact Forces For bridges crossing wide waterways, sideways collision to the bridge structure might be experienced in the case of a drifting vessel or in the case of an evasive manoeuvre to avoid a head-on collision.Deckhouse Impact Forces 71 For bridges crossing narrow channels, sideways collision might result from common navigation practice. In case of drift to one side of the channel the rotating of the vessel by using the rudder may swing the stern of the vessel sideways into piers or fenders lining the channel. Studies carried out in connection with the Great Belt Crossing, ref. Pedersen et al. [6-28], indicate that global sideways collision forces in the case of an immovable bridge structure are of the same magnitude as bow collision forces. Although the hydrodynamic mass involved with lateral vessel motion is increased by a factor of 5-10, compared to a head-on collision, the impact energy from a sideways collision is generally smaller due to the low speed. 6.3 Deckhouse Impact Forces For collision between the deckhouse of a vessel and a bridge superstructure, very little data and research have been reported in the literature In the US Guide Specification [6-15], the following relationship is prescribed: Pow = Ron’ Ps deckhouse impact force 0.1 for vessels greater than 100,000 DWT, 0.2 - 0.1 DWT/100,000 for vessels smaller than 100,000 DWT vessel collision force on pier. ‘This empirical relationship was based on Danish investigation results from the 1970°s. The US Guide Specification [6-15] further indicates that the impact force between ship mast and bridge superstructure may be figured at 10% of the deckhouse impact force. This is considered a conservative assumption in most cases. Most recent investigations have been made by Det Norske Veritas for the Great Belt project, ref, Pedersen et al. [6-28]. Various collision cases for typical deckhouses of vessels up to 5000 tons displacement have been considered. The impact loads have been calculated from the loads required to deform the structural elements of the deckhouse The calculations have resulted in recommended design loads depending on the size of the contact area, i.e. the product of the contact height, which is the vertical distance between the top of the ship’s deckhouse and the underside of the bridge girder, and the breadth of the deckhouse. Some results are summarized in Figure 6.13.72 Vessel Impact Forces Impact Force ( MN ) Contact height = 5.0 m A (Top deck and two next decks) 30 Contact height = 2.5 m (Top deck and next deck) 20 Contact height = 0.1 m (Top deck only) 10 9 > 0 10 20 30 Breadth of Deck House ( m ) Figure 6.13. Impact force versus deckhouse breadth and contact height. 6.4 Local Collision Forces In order to ensure that all bridge elements exposed to vessel collision are designed to a certain strength and robustness, local point loads and line loadings which will occur during a collision event should be considered. A review of the local impact forces found in the literature is presented below. The values indicated are difficult to compare since some represent peak forces and some are the average forces. In addition different types of vessels (tanker, container, etc.) are involved.Barge Vessel Impact Forces 73 ~ For the Faroe Bridges, ref. Jensen et al. [6-12], the design vessel for impact was a 2,250 DWT ship colliding with the bridge at 6.2 m/sec and the following design pressure applied to the piers: + From 5 m below to 5 m above the waterline: 1.0 MN/m? + From 5 to 10 m above the waterline: 0.5 MN/m* + From 10 to 15 m above the waterline: 0.05 MN/m? — According to Olnhausen [6-26], pier surface loadings of 2.0 MN/m? for the hull strength of “large” ships, and 0.5 MN/m? for “minor” impacts are valid. — Ohnishi et al. [6-25] report that for a 35,000 DWT container ship, an average impact force of 88 MN was estimated and an average impact pressure of 1.2 MN/m*; and for a 409,000 DWT tanker, an average impact force of 245 MN was estimated and an average impact pressure of 0.4 MN/m’. — Modjeski & Masters [6-21] recommend using a distributed surface load of 1.4 MN/m? combined with a line load of 2.9 MN/im for verifying the local strength of the pier and its structural members. The loads are to be applied at a height which corresponds to the elevation of the deck of the design vessel in ballast. ~ Using data reported by Kuesel [6-16] for the 1981 collision of a 31,800 DWT tanker with a main pier of the Newport Bridge, the average bow impact pressure was estimated to be approximately 0.6 MN/m’. — In connection with the Great Belt project, local design loadings were prescribed, together with detailed rules for their application, for all bridge elements exposed to vessel collision. For bridge elements in contact with the bow of the colliding vessel, two load cases are considered: + A system of 0.1 m wide line loads varying from 4.3 MN/m to 2.9 MN/m together with a uniform pressure level of 0.5 MN/m2 for exposed foundations. + Point loads varying from 17 MN to 21 MN acting over a 1.0 m square or 10 MN acting over a 0.5 m square, For bridge elements in contact with the superstructure of the colliding vessel three load cases are considered: + A uniform pressure level of 0.5 MN/m? acting over an area of 20 m’. + A system of 0.02 m wide line loads of 1.0 MN/m, + A point load of 2.5 MN acting over a 0.25 m square or 0.5 MN acting over 0.10 m square. The results of the investigations suggested that the magnitude of local loads does not vary very much with the vessel size. This allowed that the same loads could be specified for all bridge elements irrespective of the size of the design vessel governing the global loading. The results of the investigations summarized above for the Great Belt bridge project are considered to represent the state-of-the-art with respect to knowledge about local collision forces. 6.5 Barge Vessel Impact Forces Significant differences exist between the shapes and structures of ocean-going ship vessels and barge vessels using inland waterways. Regarding barges, very little research and data on behaviour during collision can be found in the literature.74 Vessel Impact Forces Impact Force ( MN ) 20 ment: 6,300 tonnes 15 10 5 oo 1.0 20 > Water Depth (m ) Figure 6.14 Design chart developed by Modjeski & Masters [6-21] for different classes of bargesiwaterways (transformed into SI units)Barge Vessel Impact Forces 75 Barge Impact Force ( MN ) A 25 15 10 > 0 1 2 3 4 5 6 Impact Speed (m/sec ) Figure 6.15 Design chart for different tow lengths of standard hopper barges (1900 tonnes displacement) included in the US Guide Specification [6-15] (transformed into Sl units). The following indications have been noted: ~ Modjeski & Masters [6-21] recommend the barge impact loadings shown in Figure 6.14 for barges in shallow draught waterways in Louisiana. ~ For bridges crossing the Rhine River, German Regulations [6-8] established design impact loads of 30 MN for head-on collision with piers near the navigation channel, and 6 MN for head-on collision with bridge piers located close to the river bank, — French regulations [6-6] specify that barge impact against bridge piers in rivers or canals of more than 12 m width shall be 8 MN in the direction of the current, For narrow waterways (those less than 12 m in width), the barge impact forces shall be 1 MN. — An ISO (International Organization for Standardization) Standard [6-10] on accident action is under preparation. This document combines the above-mentioned French and German regulations.76 Vessel Impact Forces — The most comprehensive studies reported was performed in 1983 by Meir-Domberg [6-19]. The studies included dynamic loadings with a pendulum hammer on barge models as well as numerical analysis. ~ The US Guide Specification [6-15] has adopted and elaborated the results of the above-mentioned German research. ‘The design chart shown in Figure 6.15 is taken from the US Guide Specification.7 7. BRIDGE DESIGN For bridge design purposes, design forces should be established for at least the following collision load cases: collision with the bridge piers and pier shafts, head-on by the vessel bow or sideways by the vessel hull — collision with the bridge superstructure by the vessel bow, hull, or deck house. Global forces for design for overall stability as well as local forces for design for local strength of bridge elements are required, Due to the low probability of collision, the vessel collision load case is traditionally considered to be an accidental action. This means that a reduced level of safety against failure is accepted. However, two or more levels of vessel collision load assumptions may be introduced in order also to cover lower vessel collision loads which might be experienced more often. As an example, the following three levels of safety, corresponding to collision loads with decreasing levels of probability, might be specified: ~ no damage accepted — minor damage accepted, provided continuous use of the bridge is possible and provided repair work can be effectuated without disturbing the traffic — major damage or local collapse is accepted, provided the remaining structure has sufficient redundancy to allow repair within a relatively short time (say, 3 months) to a degree that allows re-opening of the bridge for traffic. 7.1 Energy Considerations In order to determine the consequences of a vessel collision accident, the traditional method is to estimate the amount of kinetic energy available and thereafter determine how this energy is dissipated by displacement, deformation or crushing of the vessel and/or the bridge structure, including attached or free-standing fenders or protective works. The total kinetic energy KE is: KE My+ (M,+M,) + v? (7.1) where the velocity of the vessel is the mass of the vessel M, = the hydrodynamic added mass of the water surrounding the vessel and moving with it78 Bridge Design M, is the displacement tonnage of the vessel divided by the acceleration of gravity. In the absence of better information the displacement can be established approximately on the basis of the gross registered tonnage (GRT) or the dead weight tonnage (DWT), as indicated in Appendix A. ‘The hydrodynamic added mass, M,, which accounts for the hydrodynamic forces acting on the vessel during the course of the collision accident, depends on various factors such as: — the direction of the vessel travel — the shape of the vessel hull ~ the underkeel clearance (the distance between the bottom of the vessel and the waterway bed) the direction of the impact. The dependance of the different factors has been studied by several authors, ref. Petersen et al. [7-7] and Blok et al. [7-1]. In deep water, and for short impact durations, the following order of magnitude of the hydrodynamic added mass is normally recommended: = 0.05 M, - 0.10 M, for bow impact 0.40 M, - 0.50 M, for sideways impact. However, if the draught of the vessel approaches the water depth and/or the duration of the impact is longer than, say, 1 sec, the hydrodynamic resistance increases, particularly in the case of sideways impact. For the time being it is recommended to assume that the ratio M,/M, depends on the vessel draught/water depth ratio as indicated in Figure 7.1, The vessel travel direction may not correspond to the longitudinal axis of the vessel, owing to wind and current pressures and steering manoeuvres, In narrow channels it is common practice to keep the vessel’s bow headed along the channel, so the stern may oscillate laterally. The vessel velocity may be resolved into components parallel and normal to its axis for calcu- lating longitudinal and transverse kinetic energy. accounting for the appropriate hydrodynamic mass for each component. Owing to uncertainties in estimating lateral velocity components and lateral hydrodynamic ‘mass coefficients, including the effects of the yaw angle, it is recommended that the ratio of transverse kinetic energy to longitudinal kinetic energy be taken as not less than 0.5. For intermediate directions, it is recommended that an elliptical distribution be used, as shown in Figure 7.2. If the vessel comes to a complete stop, its total kinetic energy is dissipated in the collision. If the vessel is merely diverted, the collision energy may be determined from a vector analysis of the vessel's kinetic energy before and after the collision, Inan eccentric impact, part of the vessel’s kinetic energy will be dissipated through rotation and friction. Saul et al. [7-9] indicate that the portion of the kinetic energy to be absorbed by the vessel and the bridge structure is a function of the impact angle and the coefficient of friction between the colliding vessel and the bridge structure. However, the most common approach is to assume that the vessel is totally stopped by the bridge structure and all its kinetic energy is released in the collision.Energy Considerations 79 Hydrodynamic Mass ( M,,) /Mass of Vessel (My) Lateral Direction Longitudinal Direction 0.0 02 04 06 08 09 1.0 Draught of Vessel /Water Depth Figure 7.1 Hydrodynamic mass factors. ‘The kinetic energy involved in the collision is dissipated by displacement, deformation and crushing of: ~ the fender or buffer structure ~ the vessel — the bridge structure. The energy absorbed by elastic deformation is small and may be neglected.80 Bridge Design rect Ellipse Figure 7.2. Kinetic energy distribution (for well defined straight navigation channels, the angle 0. between the axis of the channel and the axis of the vessel may normally be taken as 0). It is recommended that all structural elements in the collision analysis be treated as rigid or plastic bodies. Each element has a different deformation threshold or yield point and a different postyield plastic deformation module. The general characteristics of the system are illustrated schematically in Figure 7.3. Figure 7.3 also illustrates an ideal design for a collision protection scheme. In this scheme, small collisions are absorbed by deformation of the fender. Under larger collisions, the fender is crushed, but before it collapses, crushing of the vessel structure takes over and absorbs the remainder of the collision energy. 7.2 Structural Analysis The response of the bridge structure (pier or superstructure together with fender or buffer systems) should be estimated using either static or dynamic calculation models depending on the circumstances, ref. Prucz [7-8] and Modjeski & Masters [7-5]. The calculation model should be based on a system of 3 elements: vessel, fender, and pier, as indicated in Figure 7.4.Structural Analysis @1 Force: Initial Fender Capacity A Time Fender Deformation / Figure 7.3 Forceldeformation characteristics. In cases where the energy absorbing capacity of the pier is less than required, the real problem is to design a protective structure that can make up for the energy absorbing deficiency of the vessel-pier system. Reference is made to Section 9. The following three approaches are available: static force analysis — equivalent static force analysis with dynamic amplification ~ full dynamic analysis. Static Force Analysis For simplicity and due to lack of data on impact load histories, this approach has been applied in most cases. Reference is made to Section 6.1. This approach assumes that the collision loading does not involve important dynamic amplification and therefore, caution should be exercised when selecting the level of the static design force.82 Bridge Design Equivalent Static Force with Dynamic Amplification This approach should be applied for slender structures with linear elastic response to loading. The structure is modelled by a simple mass-spring system which is exposed to a force-time relation, ref, Section 6.1. The maximum dynamic and static deflections are calculated and a dynamic amplification factor is estimated as the ratio between the dynamic and the static deflections, An equivalent static force is then obtained by multiplying the assumed maximum force by the dynamic amplification factor. Full Dynamic Analysis A full dynamic analysis should always be carried out for important structures, in particular if transient or permanent deflections or movements of the bridge structure and/or the fender or buffer system are introduced in the analysis. In this approach, a full dynamic model is conveniently established by applying Finite-Element- Model (FEM) techniques. The calculations are carried out in the time domain using a force-indentation relationship, ref. Section 6, as a force function and appropriate small time-steps with constant force. The step-wise calculations are repeated until all of the kinetic energy is dissipated. 7.3. Design of Substructure Head-on Bow or Sideways Collision As it appears from Section 6.1, the forces involved in head-on bow collisions have been extensively investigated. Some main results together with recommendations as to the selection of design basis can be found in Section 6.1 ‘The forces from sideways collisions have normally been considered covered by the assumed bow collision forces. In narrow channels, however, where no head-on collision to piers is physically possible, it may still be necessary to consider sideways impact against structures lining the edges of the channel. Reference is made to Section 6.2. Local Loads Alll surfaces and structural members exposed to collision should be designed for local loads. ns of the order of magnitude of local forces: point loads, line loads, and area Some indi loads can be found in Section 6.4.Design of Superstructure 83 Location of Impact Forces The global collision loads should be in the direction of the vessel travel. It is normally specified that 50% of the global load shall be applied to the substructure in the direction perpendicular to the vessel travel, ref. Nordic Road Engineering Federation [7-6] and Figure 6.4 and Knott et al. (7-4. The purpose of this is to cover a series of possible collision scenarios which are very difficult to quantify, i.e.: — collision from vessel moving in other directions than the navigated channel direction ~ eccentric collision glancing collisions. Traditionally, the global impact force is applied at the mean high water level. However, the local water level variation as well as the bow shape of the design vessel should be taken into account, The local load components should be applied in accordance with the geometry of the colliding design vessel. In this connection it is very important to consider the shapes of the bulbous bows, the overhang or rake of the bows, etc. for all relevant ships and barges in the waterway, 7.4 Design of Superstructure Head-on Bow or Sideways Collision If the superstructure of the bridge is so low that it can be hit by the hull of a vessel, an upper limit of the design collision force would be the force valid for head-on bow collision. On this basis the US Guide Specification [7-4] specifies the following impact force relationship: Pan = Rau’ Ps where: Pen = impact force on exposed superstructure Ryq = ratio of exposed superstructure depth to total bow depth P, = impact force on pier. This magnitude of force can not be sustained by ordinary types of bridge superstructures, This is, in particular, the case for movable bridges which could be struck in the closed position in case of electrical or mechanical failure of the bridge machinery. Where possible, sufficient clearance of the bridge spans should be arranged to avoid bow collision,84 Bridge Design Deckhouse Collision The magnitude of impact forces involved in a collision between the deckhouse of a ship and an exposed superstructure span are of a more manageable order of magnitude when compared with the above-mentioned bow impact forces. However, very little research has been carried out on this, subject, Some indications can be found in Section 6.3. Local Loads All surfaces and structural members exposed to collision should be designed for local loads. Some indications of the order of magnitude of local forces (point loads, line loads, and area loads) can be found in Section 6.4. Location of Impact Forces The global collision loads should be applied in the direction of the vessel travel. Depending on the shape of the superstructure contact area, vertical load components may have to be considered in addition to the horizontal load. The local load component should be applied in accordance with the geometry of the colliding design vessel.85 8. PREVENTION MEASURES Planning and implementation of prevention measures to improve safe navigation in the waterway near a bridge crossing requires close cooperation with the relevant navigation authorities. Where the matter is of international concern, the national authorities will approach the international authorities, in most cases the International Maritime Organisation (IMO), ref. IMO [8-7]. Prevention or reduction of the frequency of collisions is achieved by providing assistance to navigation, The assistance may differ in extent and level of sophistication, depending on the waterway and the intensity of the navigation, In this connection, inspiration may be obtained from experience within the offshore sector, ref. Vendrell [8-12]. Three levels of assistance are discussed in the following: — installation of navigational aids on the bridge and in the waterway — introduction of navigation regulations — implementation of a vessel traffic management system. Comments on the effectiveness of the different types and levels of assistance are given in Section 5.2. 8.1 Aids to Navigation Visual, sonar as well as electronic aids to navigation should be designed to provide safe guidance in most weather conditions, For installation on the bridge structure, the following types of navigational aids can be considered to provide better detectability: colours (fluorescent) — signs — high intensity light beacons (flashing) — range lights — sound devices (fog horns) — racon installation at the centre line of the vessel track(s). An example of installations on a bridge to aid navigation is shown in Figure 8.1.86 Prevention Measures Central lantern Fog light Fog detector, Red /white central lantern Fog hom Navigation lantern Navigation lantern Red daymark Green daymark Figure 8.1 Navigation equipment mounted on the navigation span of the Great Belt western bridge for southbound vessels (from Olsen etal. [8-10)). For location in the waterway near the bridge crossing, the following types of navigational aids can be considered: lighthouse buoys beacons racon installations. ' Fewer but larger buoys with strong lights and fluorescent paint should be utilized rather than many small buoys. Racons (self-identifying radio beacons activated by the radar signals of passing vessels) should always be installed in connection with critical bridge passages. The most suitable arrangement of navigational aids may be obtained through manoeuvring simulations with pilots.Vessel Tratfic Management Systems 87 8.2 Vessel Traffic Regulations Traffic regulation measures should be studied in cooperation with local navigation authorities. The regulations may involve both mandatory and voluntary elements. Different possibilities exist: — separation of navigation routes for opposite travelling directions — limitation of maximum vessel speed (it should be remembered that the manoeuvring capabilities of a vessel are reduced with reduced speed) — requirement for pilotage (in national waters it is possible to introduce compulsory pilotage, whereas international waters require involvement of relevant organizations such as IMO) requirement for tug assistance (the comments related to pilotage also apply to tug assistance) — restriction on passage in bad weather or bad sea conditions — limitation of maximum vessel size — restriction on passages of large vessels to daylight transits only — demand on empty vessels to take on ballast for a minimum draught special demands on vessels with hazardous cargoes. 8.3. Vessel Traffic Management Systems Traffic management systems of varying sophistication can be considered, including facilities as follows: — radar surveillance — guard boat surveillance — transmission from data buoys of information on meteorological conditions, sea and tide, ete. — transmission of navigational information — navigational guidance by pilots directly or over radio — radio communication between vessel and bridge personnel, ‘The management system should cover the navigation channel in full width within a distance sailed in one hour, i.e. approximately 15 nautical miles on both sides of the navigation span. Information on Vessel ‘Traffic Management Systems is mostly published under the title Vessel ‘Traffic Services (VTS). Reference is made to Fujii et al. [8-2], Hara [8-4], and Mizuki et al. [8-9]. These systems include advanced shore-based radar systems with real-time surveillance capabilities, as proposed by Greneker et al. [8-3], and as implemented for the Lake Pontchartrain causeway in Louisiana, USA [8-13], and for the Great Belt crossing in Denmark [8-10] as well as small portable navigation units carried on board by the master or pilot of the vessel, as described by Knott [8-8] The Vessel Traffic Management System planned for the Great Belt crossing is shown in Figure 8.2.88 Prevention Measures / RADAR ale Hm, -/ Hov 9 CE 7,” The system includes. Radar installations (RADAR) VHF radio communication (VHF) VHF direction finder (DF) Closed circuit television (TV) Infrared television (IR) Vessel Traffic Service center Figure 8.2 Outline of VTS system covering the waters of the Great Belt in Denmark (from Olsen et al, [8-10]).Vessel Traffic Management Systems 89 A number of recent publications on VTS systems are briefly mentioned below: The “World VTS Guide” by IALA, IAPH, IMPA [8-5] contains information on existing VTS systems worldwide. The “Third Survey on VTS in the World” by Mizuki et al. [8-9] also contains information on existing VTS systems worldwide. The International Maritime Organization (IMO) has Services” ref. (8-6). The “Vessel Traffic Services Equipment and Technology Report (VTS Handbook)” [8-11] is intended to be used as a guide by the US Coast Guard in the selection of equipment and technology for VTS systems implementation and/or enhancements. In “COST 301. Shore-based marine navigation aid systems”, ref. Fabre et al. [8-1], functions and effectiveness of existing VTS systems in the EEC area are covered. issued “Guidelines for Vessel Traffic1 9, PROTECTION MEASURES In addition to bridge damage, vessel collisions may result in serious environmental damage such as the spilling of oil and other chemicals. ‘The consequences of a vessel collision may therefore reach far beyond the direct costs of repairing/replacing the vessel and possibly the bridge. ‘The bridge elements can be designed to withstand the impact loads, or a fender or protection system can be developed to prevent, redirect, or reduce the impact loads on bridge elements to non-destructive levels. If the force resistance of the protection system is higher than the vessel crushing force, the bow of the vessel will crush and the impact energy will be primarily absorbed by the vessel. If the vessel crushing force is higher than the resistance of the protection system, the impact energy will be primarily absorbed by the deflection and crushing of the protection system. ‘The protection system should be designed not only to protect the bridge structure, but also to protect the vessel and the environment against serious damage. This may be achieved by combin- ing different types of protective systems. Protection systems may be located directly on the bridge structure (such as a bridge pier fender), or independent of the bridge (such as a dolphin). The geometry of the protective structure should be developed to prevent the rake (overhang) of the design vessel’s bow from striking and causing damage to any exposed portion of the bridge. Generally, the analysis and design of bridge protection structures requires the use of engineering judgment to arrive at a reasonable solution, Surveys of protection measures can be found in Frandsen et al. [9-5], Knott et al, [9-10], Larsen et al. [9-12], Ostenfeld [9-20], and Saul et al. (9-21], [9-22], and [9-23]. The various types of protective structures commonly used for bridges will be briefly discussed in the subsections below: Fender Systems — Pile Supported Systems Dolphin Protection ~ Artificial Island or Reef Protection — Floating Protection Systems. 9.1 Fender Systems Timber fenders are composed of vertical and horizontal timber members in a grillage geometry attached to the face of the bridge pier, or erected as an independent structure adjacent to the92 Protection Measures pier. Energy is absorbed by elastic deformation and crushing of the timber members. Because of their relatively low cost, timber fenders have frequently been used on bridge projects for protecting piers from minor vessel impact forces. However, for relatively large collision impact loads the resulting timber fenders would have to be extremely large, and might be uneconomical in most circumstances. Rubber fenders are commercially available in a wide variety of extruded and built-up shapes. Impact energy is absorbed through the elastic deformation of the rubber elements either in compression, bending, shear deformations, or a combination of all three. Conerete fenders consist of hollow, thin-walled, concrete box structures attached to the bridge pier. Usually, a timber fender is also attached to the outer face of the concrete box fender. Impact energy is absorbed by the buckling and crushing of the concrete walls composing the fender system, An example of this type of fender is shown in Figure 9.1. —— Concrete Pier Timber Fender 6-0" Reinforced Concrete Box Figure 9.1 Crushable concrete box fender on the Francis Scott Key Bridge, Baltimore, USA (from Knott et al. [9-10]). Steel fenders consist of thin-walled membranes and bracing elements composed in a variety of box-like arrays and assemblies attached to the bridge pier. Impact energy is absorbed by compression, bending, and buckling of the steel elements in the fender, ref. Matsuzaki et al. (9-15] and Namita et al. [9-17]. Timber facing should be attached to the steel fender to prevent sparks resulting from direct contact with steel hulled vessels. An example of this type of fender is shown in Figure 9.2.Pile Supported Systems 93 Horizontal Plate Horizontal Plate Horizontal Plate Figure 9.2. Framed steel fender system, also referred to as a multi-cell type buffer, used for protection of the Bisan-Seto Bridges in Japan (from Matsuzaki et al. [9-15]). 9.2. Pile Supported Systems Pile groups connected together by rigid caps may be used for protection to resist vessel impact forces. Free standing piles and piles connected by relatively flexible caps are also used for bridge protection. The pile groups may consist of vertical piles, which primarily absorb energy by bending, or batter piles which absorb energy by compression and bending. Asa result of the high impact design loads associated with vessel collision, plastic deformation and crushing of the pile structure is permitted provided that the vessel is stopped before striking the pier, or the resulting impact is below the resistance strength of the pier and foundation. Fender systems may be attached to the pile structure to help resist a portion of the impact loads. ‘Timber, steel, or concrete piles may be utilized depending on site conditions, impact loads, and economics. An example of this type of protective structure is shown in Figure 9.3.94 Protection Measures Main Columns ‘Fender “Ring” (reinforced concrete) (reinforced concrete) Steel Pipe 900x20, St. 52-3 ELEVATION Figure 9.3 Pile supported protection system for the Tromso Bridge in Norway (from Tambs- Lycke [9-25]}.Artificial Island or Reef Protection 95 9.3 Dolphin Protection Large diameter dolphins may be used for protection of bridge piers, see Englot [9-3], Hahn et al, [9-6], Heins [9-8], Heins et al. [9-9], Knott [9-11], and Naoi et al. {9-18} Dolphins are typically circular cells constructed of driven steel sheet piling, filled with rock or concrete, and topped by a concrete cap. Dolphins may also be constructed of precast concrete sections, or precast entirely off-site and floated into final position, Driven pilings are sometimes incorporated in the cell design. Design procedures for dolphins are usually based on an estimate of the energy changes that take place during the design impact loading. Energy-displacement relationships are typically developed for the following energy dissipating mechanisms: — crushing of the vessel’s bow ~ lifting of the vessel’s bow — generation of water waves and turbulence — friction between the vessel and the dolphin ~ friction between the vessel and the river bottom — sliding of the dolphin ~ rotation of the dolphin — deformation of the dolphin. Deformation of the vessel/dolphin system is assumed to follow a path of least energy. For each potential displacement configuration of dolphin and vessel, a deformation path can be developed. Deformation stops when all the kinetic energy of the impact has been absorbed. For purposes of desiga, it is recommended that the maximum dolphin deformation be limited to less than one-half the diameter of the cell. Under design loading conditions, the cell is permitted to undergo large plastic deformation and partial collapse. Examples of this type of protective structure are shown in Figure 9.4. 9.4 Artificial Island or Reef Protection Artificial islands around bridge piers or artificial reefs in front of bridge piers provide highly effective vessel collision protection, ref. Brink-Kjaer [9-1], Denver 9-2}, Fletcher et al. [9-4], Havnoe et al. [9-7], Minorsky [9-14], and Sexsmith [9-24]. If the waterflow through the navigation channel is important to the surrounding environment, investigations about the net waterflow before and after installation of the protection islands should be carried out.96 Protection Measures Navigation Light Elevation 17.0 s0 Elevation 15.0 ff r Timber Fender System ere Concrete Wall and Ca M.S.L Elevation 0,00 —. a Elevation-5.00 ‘Steel Sheet Piles 40" Elevation-30 : | Crushed Stone Fill Existing Bay Scour Protection en “\— Hawthome Formation Bottom Figure 9.4 Examples of Dolphin protection. At the top, the Sunshine Skyway Bridge in USA (from Knott [9-11]). Below, the Rio-Niteroi Bridge in Brazil (photo: Niels Jorgen Gimsing).Floating Protection Systems 97 Islands typically consist of a sand or rock core which is protected by outer layers of heavy rock armour to provide protection against wave, current, and ice actions, The island geometry should be developed in accordance with the following criteria: — The vessel impact force transmitted through the island to the bridge pier must not exceed the lateral capacity of the pier and pier foundation. The island dimensions should be such that vessel penetration into the island during a collision will not result in physical contact between the vessel and any part of the bridge pier. The second requirement is particularly critical for empty or ballasted ships and barges which can slide up on the slopes of an island and travel relatively large distances before coming to a stop. The design of the surface armour protection of the islands for wave, current, and ice attack may be based on methodologies used for rubble mound breakwater design. ‘The following items have been identified as sources of energy absorption/dissipation during a vessel impact with an island: ~ crushing of the hull of the vessel lifting of the vessel generation of water waves and turbulence lifting of island material displacement, shear, and compaction of the island material friction between the vessel and the island generation of shock waves in the island crushing of particles of island material Inclusion of these items in a design analysis is difficult since their effects are only partially understood. Physical model studies, as well as mathematical simulations, are usually required when protective islands are designed. An example of the arrangement of artificial islands for vessel collision protection is shown in Figure 9.5. 9.5 Floating Protection Systems Various types of floating protective systems may be considered by the engineer. Several of these systems include: — Cable net systems. Vessels are stopped by a system of cables anchored to the waterway bottom and suspended by buoys located in front of the bridge piers, ref. Larsen et al. [9-12], Oda et al. {9-19}, and Vitalis [9-26].98 Protection Measures Figure 9.5 Arrangement of protection islands for the Annacis Island Bridge at Vancouver in Canada (from Sexsmith [9-24]). — Anchored Pontoons. Large floating pontoons anchored to the waterway bottom in front of the piers absorb vessel impact, ref, Mondorf [9-16]. — Floating Shear Booms. Floating structures anchored to the waterway bottom deflect vessels away from piers and absorb impact energy. Special consideration for corrosion protection must be made for all systems involving under- water steel cables and anchorages. Special consideration should be given to the function and vulnerability/durability of floating systems during winter time in waters subject :o icing or ice drift. Floating systems are vulnerable to overrun by vessels with sharply raked bows. An example of protection by cable net system is shown in Figure 9.6.Floating Protection Systems 99 Re Tro mm CONT im: bo or Figure 9.6 Arrangement of cable net system proposed for vessel collision protection of the Gibraltar Bridge (from Larsen et al. [9-12])101 10, PROTECTION OF THE PUBLIC ‘To minimize the loss of life which may occur in the event of a catastrophic collapse of a bridge during a vessel collision, bridge user warning systems may be introduced, ref. [10-1], [10-2], and [10-3]. 10.1 Collision Hazard Detection Devices to detect vessel/bridge collision hazards include the following: — Vessel Impact Vibration Detectors. Placed on bridge piers, these vibration sensors would be capable of distinguishing between normal structural vibrations and movements associated with substantial vessel impacts. Continuity Circuits. This electrical system would utilize pairs of conductors terminating with end-of-line devices attached to the bridge superstructure. Collapse of some portion of the bridge deck would interrupt the circuit continuity, VHF Radio Link. The use of this device would be in advance of imminent danger, as foreseen by the pilot or master of a vessel which had, for instance, lost steerage. If the mariner anticipated a possible vessel/bridge collision, he would radio the bridge personnel, or other appropriate agency, via VHF marine emergency channel in order to halt traffic on the bridge. Either of the first two of the above devices could activate traffic control/information systems automatically or through a machine-man-machine interface with the human intermediary verifying hazards before interrupting traffic. VHF radio units are readily available in the deckhouse of virtually every merchant vessel. The use of such a system would require the installation of a relatively inexpensive VHF set and continuous monitoring by the bridge personnel who could make appropriate traffic control decisions. Virtually any detection device can be electronically linked to traffic control/information equi ment in order to automatically warn or stop traffic. However, in actual practice, considerable difficulty can be experienced with false alarms and unnecessary interruptions of traffic. Included among possible verification methods to be carried out by bridge personnel before traffic control actions are taken, are the following: — Closed Circuit Television (CCTV). Cameras can be placed strategically to allow personnel at a monitor site to view the bridge main span, the navigation channel, the roadway, or any other feature desired. — Visual Delineation. The top of the bridge parapet or guardrail would be fitted with a series of reflectors or lights, immediately revealing the collapse of a portion of the bridge superstructure,102 Protection of the Public 10.2 Bridge Traffic Control Whether the hazard detector information is used automatically or is manually verified, the ultimate function of a traffic safety system is to appropriately control traffic or inform bridge users of hazards. The following devices can be used to accomplish this function: — Variable Message Signs. Virtually any message can be transmitted via this device, including warnings of catastrophic bridge failure, environmental hazards, traffic congestion, construction/maintenance activity, ete. — Flashing Beacons. Used in conjunction with standard format warning signs, these devices can be used to bring attention to a warning message. — Movable Gates, usually fitted with flashing red lights and an audio alarm (siren or bell), this device can be lowered across traffic lanes to halt traffic (as at railroad crossings).103 Appendix A. Vessel Size and Geometry Data AL Vessel Size Measurements Dead Weight Tonnage (DWT) The Dead Weight Tonnage expresses the carrying capacity of a vessel, ie. the maximum weight in metric tonnes of cargo, fuel, water, stores, etc. on board the vessel when fully loaded. DWT equals the difference between a vessel's displacement at fully loaded draught and light ballast draught. It is the normal unit for bulk carriers and tankers. Displacement Tonnage (W) The Displacement Tonnage expresses the total weight in metric tonnes of the vessel including cargo, fuel, water, stores, etc. The displacement thus depends on the loading condition. The weight of the ship when in a ‘light’ or unloaded state, but including the weight of water in boilers and any permanent ballast is called “light displacement”. The light displacement weight plus the weight of cargo, fuel, stores, fresh water, and water, ballast is called “load displacement”, W equals the weight of water displaced by a vessel when floating at the specific draught. W is the normal unit for warships. The Displacement Tonnage for a fully loaded vessel may be approximated to Dead Weight ‘Tonnage as follows: ~ Tankers: ~ Bulk Carriers: — Container Vessels: DWT = 0.6 x W Gross Register Tonnage (GRT) The Gross Register Tonnage expresses the internal volume of a vessel (less certain exempted spaces) measured in cubic feet divided by 100. GRT is thus not a measurement of weight. GRT is a normal unit for passenger ships and cargo liners and is used as a basis for safety requirements and manning. The Gross Register Tonnage may be approximated to Dead Weight Tonnage as follows: Tankers: DWT =1.6x GRT — Bulk Carriers: DWT = 1.9x GRT — Container Vessels: DWT = 0.9 x GRT.104 Appendix A Net Register Tonnage (NRT) The Net Register Tonnage expresses the internal volume of a vessel used for the carriage of cargo or passengers, measured in cubic feet divided by 100. NRT is thus not a measurement of weight. The NRT is the GRT less the volumes of certain spaces not used for the carriage of cargo (engine room, certain water tanks, etc.). NRT is frequently the basis on which harbour duties and pilotage fees are levied. It is a normal unit for passenger ships and cargo liners. A.2. Typical Ship Vessel Geometry Data A2d1 A222 A23 A24 A2S A26 A.27 A28 A29 Air Draught in Ballast Condition Air Draught in Loaded Condition Height of Deck House in Ballast Condition Height of Deck House in Loaded Condition Draught in Ballast Condi Draught in Loaded Condition Length (LOA) Breadth (Beam) Moulded DepthA2L Air Draught in Ballast Condition Metres 60 60 40 30 20 10 1,000 A22 —Container ~ ~Bulk Carrier - 5Tar inker Typical Ship Vessel Geometry Data 105 10,000 Air Draught in Loaded Condition Metres 60 50 40 30 10 1,000 — Container inker 100,000 1,000,000 Tonnes Bulk Carrier +--+ Tat Ty 10,000 100,000 4,000,000 DWT in Tonnes108 Appendix A A.2.3_ Height of Deck House in Ballast Condition Metres. 25 —Container = Bulk Carrier -- Tanker T 20 6 10 1,000 10,000 100,000 A.2.4 Height of Deck House in Loaded Condition Metres 26 1,000,000 DWT in Tonnes —Container = ~Bulk Carrier «+ Tanker T 20 10 1,000 10,000 100,000 4,000,000 DWT in TonnesTypical Ship Vessel Geometry Data 107 A.2.5 Draught in Ballast Condition Metres 60 —=Container 40 30 20 10 1000 10,000 100,000 1,000,000 DWT in Tonnes A.2.6 Draught in Loaded Condition Motres 50 —Container = =Bulk Carrier -- Tanker i 40 30 20 10 1,000 10,000 100,000 1,000,000 DWT in Tonnes108 Appendix A A2.7 Length (LOA) Metres 500 - — Container - ~Bulk Carrier -- Tanker 400 200 fo | 100 f 1,000 || | 10,000 100,000 1,000,000 DWT in Tonnes A28 Breadth (Beam) Metres 80 = — Container = =Bulk Carrier ++ Tanker 70 60 50 40 30 20 10 1000 10,000 100,000 1,000,000 DWT in TonnesTypical Ship Vessel Geometry Data 109 A.2.9 Moulded Depth Metres 40 =——Container = =Bulk Carrier + * Tanker 30 +} 20 10 - 1,000 10,000 100,000 4,000,000 DWT in Tonnes110 Appendix A A.3. Typical Barge Vessel Geometry Data A3.1 Barges and Barge Tows in the USA ‘Typical dimensions and capacities of barge vessels and tows in use in US inland waterways are shown in the table below. Reference is made to: — Modjeski & Masters, Consulting Engineers: “Criteria for: The Design of Bridge Piers with Respect to Vessel Collision in Louisiana Water- ways”, Prepared for the Louisiana Department of Transportation and Development and the Federal Highway Administration, New Orleans, July 1985. Knott, M.A. and Larsen, O. Damgaard: “Guide Specification and Commentary for Vessel Collision Design of Highway Bridges”, US Department of Transportation, Federal Highway Administration, Publ. No. FHWA-RD-91- 006, Dec. 1990. Barge Type Jumbo Hopper Oversize Tank | Special Deck Length (m) 59.4 88.4 76.2 Width (m) 10.7 16.2 219 Depth of Vessel (m) 3.7 37 5.2 Empty Draught (m) os 05 08 Loaded Draught (m) 27 27 38 Empty Displacement (tonnes) 180 540 1200 Loaded Displacement (tonnes) 1700 3000 5700 Average number of barges along the length of a tow 2 1‘Typical Barge Vessel Geometry Data 111 A.3.2. Barges and Barge Tows in Europe Typical dimensions and capacities of barge vessels and tows in use in European inland waterways, classes I to VI, ref. The European Conference of Ministers of Transport (ECMT), are shown in the figure below, Eom Typical vessel Motor vse el 1 Spits (3001) 39x 5.1.x 24m cD aie tee Sy 0 Kempenaar (8001) 55 x 6.6 x 2.5m <=EEED Motor vessel m Dortmund - Eems Canal ship (10001) 67.8 x 8.2 x 2.6m Motor vessel: Rhine Herne canal ship Push tow unit: (17001) (13501) 85 x 9.5 x 2.8m 70.5 x 9.5 x 3.2m ee, —_——_— 5s ‘| e=F=—p Is Motor vessel Large Rhine vessel Push tow unit: (22500) (200%) 95 x 11.5 x 2.7m 76.5 x11.4x 3.2m) EEE Push tow uit (2 x 22508) ___ IED {Also navigable in IV and V) Push tow unit : i (2x 22501) (a 76.5 x 22.8 x 3.2m c 153 x 11.4.x 3.2m vi Push tow unit (4 x 22500) a 7 Push tow unit An (6 x 22601) YB 7113, Appendix B. Records of Collision Accidents Serious Vessel Collision Accidents recorded in the Period 1960-1991 1960 1964 1964 1964 1967 Severn River Railway Crossing, England Vessel: Tug pulling two barges each 450 tonnes displacement Accident : Broadside collision with pier Damage : Two spans and one supporting pier collapsed causing five fatalities Cause: Tug pilot’s negligence in dense fog Lit Peter Mason: “An investigation into the cause of damage to the Severn Railway Bridge”. The Structural Engineer No. 2, 1963 Maracaibo Lake, Venezuela Vessel 36,000 DWT loaded tanker Accident: Broadside collision with two piers more than 600 m from the navigational spans Damage : Three spans collapsed Cause: Failure in electrical system affecting steering gear Lit Engineering News Record, 1964-04-16 and 1964-12-24 Pontchartrain Lake, Louisiana, USA Vessel: Tug towing two loaded barges Accident : Three trestles were struck by the tug and the two barges Damage : Four spans collapsed, causing six fatalities Cause: Helmsman’s lack of attention Lit. : Engineering News Record, 1964-06-25 Pontchartrain Lake, Louisiana, USA Vessel: Tug towing two barges Accident: Tug hit a pile bent Damage : One pile bent was destroyed and two spans collapsed Cause Tug pilot’s inattention (possibly asleep) Lit : Engineering News Record, 1964-07-30 Chesapeake Bay, Virgina, USA Vessel: Drifting coal barge Accident : Vessel thrown repeatedly against the bridge deck Damage : Six spans were seriously damaged Cause: Barge torn loose from moorings in storm Lit. : Engineering News Record, 1967-12-14114 Appendix B 1970 1972 1972 1974 1974 Chesapeake Bay, Virgina, USA Vessel Accident Damage Cause Lit. 10,000 DWT US-navy cargo ship 1 hour battering against the bridge Five spans knocked down and 11 other spans damaged Vessel torn loose in stormy weather Engineering News Record, 1970-01-29 and 1970-03-12 Chesapeake Bay, Virginia, USA Vessel ‘Accident Damage Cause Lit. Drifting barge Barge thrown repeatedly against the bridge deck Two spans partially collapsed and five other spans were damaged Towline to tug broken in heavy wind Engineering News Record, 1972-09-28 Brunswick River, Georgia, USA Vessel Accident Damage Cause Lit. 13,000 DWT freighter ‘The superstructure hit by the bow of the ship ‘Three spans collapsed, causing ten fatalities ‘The helmsman misunderstood the pilot’s instructions US National Transport Safety Board and US Coast Guard: African Neptune collision with the Sidney Lanier Bridge”, Marine Accident Report, 1973 Pontchartrain Lake, Louisiana, USA Vessel Accident Damage Cause Lit. ‘Tug pulling four empty barges Tug hit pier some way from navigation span ‘Two pile bents were damaged and three spans collapsed, causing three fatalities Tug pilot asleep Engineering News Record, 1974-08-08 Welland Canal, Ontario, Canada Vessel Accident Damage Cause Lit. Ore carrier (length: 204 m) Ship rammed lifting span while opening Lift span fell in canal and lift towers were damagec Unknown Engineering News Record, 1974-08-29 and 1974-09-121975 1975 1976 1977 1977 Records of Collision Accidents 115 Derwent River, Hobart, Tasmania, Australia Vessel : 7,200 DWT bulk carrier Accident : Head-on and broadside collision with two. piers Damage : Three spans collapsed, causing 15 fatalities Cause Loss of steering ability due to engine failure Lit. Engineering News Record, 1975-01-09 and 1975-01-16 Fraser River, New Westminster, British Columbia, Canada ‘Vessel Barge in ballast (length: 183 m) Accident : Barge hit the bridge superstructure Damage : One 120m span collapsed Cause: Barge tore loose from mooring in heavy rainstorm Lit. Canadian newspaper “Vancouver Sun”, 1975-12-27 Pass Manchac Channel, Louisiana, USA Vessel Barge towed by a tug Accident : Barge hit a pile bent Damage : Pile bent destroyed and three spans collapsed, causing at least one fatality Cause: Barge off course due to careless navigation Lit. : Engineering News Record, 1976-09-23 James River, Hopewell, Virginia, USA Vessel: 25,000 DWT tanker in ballast Accident : The stem of the ship destroyed a pier bent about 120 m from the navigational span centerline Damage : Two spans collapsed Cause: Electrical fault in steering gear Lit. : US National Transport Safety Board: “US Tank Ship SS Marine Floridian, Collision with Benjamin Harrison Memorial Bridge”, Marine Accident Report, 1978 Passaic River, Union Avenue, New Jersey, USA Vessel: Empty oil barge Accident : Collision with a pier Damage Two spans collapsed Cause: Broken towline to ug Lit : Engineering News Record, 1977-08-05116 Appendix B ae 1978 1979 1980 1980 Gothenburg Harbour, Tingstad, Sweden Vessel Accident Damage Cause Lit 1600 DWT gas tanker in ballast Ship hit approach spans ‘Two approach spans destroyed Electrical fault in steering gear Danish newspaper POLITIKEN, 1977-09-11 Berwick Bay, Southern Pacific Railroad, Louisiana, USA Vessel Accident Damage Cause Lit. ‘Tug pushing four barges Lead barge hit the side span bridge superstructure One 70 m steel span fell into water and sank Tug skipper navigating with underpowered tow US National Transport Safety Board: “Collision of M/V Stud with the Southern Pacific Railroad Bridge. Marine Accident Report, 1980 Second Narrows Railway Crossing, Vancouver, Canada Vessel Accident Damage Cause Lit 22,000 DWT bulk carrier Stem of ship struck the superstructure in side span about 100 m from the navigational span centre line One span collapsed Captain's misjudgment of land marks due to dense fog New Civil Engineer International, June 1980 ‘Tjém, Alm6 Sound, Sweden Vessel Accident Damage Cause Lit. 27,000 DWT product carrier in ballast Deck house of ship struck the arch structure near the foundation on shore, about 100 m from the navigation channel centre line Total collapse of the main span causing eight fatalities Steering difficulties in rough weather and dense fog due to reduced engine power Statens Havarikommission: “Utredningsrapport betriffande... Star Clippers pasegling af Almébroen...”, Stockholm, April 1981 ‘Tampa Bay, Sunshine Skyway, Florida, USA Vessel Accident Damage Cause Lit. 35,000 DWT bulk carrier in ballast Stem of ship struck bridge column above pier top about 250 m from navigational channel Three spans collapsed, causing 35 fatalities Pilot’s careless navigation in rough weather with reduced visibility National Transport Safety Board: “Ramming of the Sunshine Skyway Bridge”, Marine Accident Report, 19811981 1982 1982 1983 Records of Collision Accidents 117 Narragansett Bay, Newport, Rhode Island, USA Vessel Accident Damage Cause Lit. 45,000 tonnes displacement tanker Ship struck main tower pier of suspension bridge head-on at 3 m/sec velocity Only superficial damage although the collision force substantially exceeded the design force Pilot's careless navigation in dense fog LR. Kuesel: “Newport Bridge Collision”, ABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983 Mosel River, Richemont Gas Pipeline, France Vessel Accident Damage Cause Lit. Tug pushing two barges Barge struck bridge pier One pier was destroyed resulting in collapse of the gas pipeline causing seven fatalities Tug pilot's careless navigation in dense fog French newspaper “Le Figaro”, 1982-01-18 Mississippi River, Hannibal, Missouri, USA Vessel Accident Damage Cause Lit. ‘Tug pushing 15 barges Barges struck abutment while passing swing span, lost control and tug swung into approach span One approach span collapsed Careless navigation Engineering News Record, 1982-05-13 Sentosa Aerial Tramway, Singapore Vessel Accident Damage Cause Lit. Petroleum drill ship with 69 m high mast Ship drifted into tramway and the mast severed the aerial cables of the tramway Two tramway cars fell into harbour causing seven fatalities Tow line failed when ship was warped out of berth by tug Local newspapers118 Appendix B 1983 1990 1990 1991 1991 Volga River Railway Crossing, Ulyanovsk, Russia Vessel Accident Damage Cause Lit. Passenger vessel “Alexander Suwarow” Vessel strayed off course and ran too close to the stpport column for the arch bridge and the upper deck of vessel collided with bridge superstructure The deckhouse of the vessel, including a film hall, was torn off causing approx. 170 fatalities Captain’s careless navigation European newspapers, June 1983 Herbert C. Bonner Bridge, North Carolina, USA Vessel Accident Damage Cause Lit Hopper dredge “Northerly Isle” (length: 60 m) Vessel drifted into the substructure of the bridge 4 pile bents were demolished and five spans collapsed The vessel dragged its anchors during a storm Engineering News Record, 1990-11-01 Tosterd, Striingniis, Sweden Vessel Accident Damage Cause Lit. Freighter “Arosandra” (length: 60 m) ‘Vessel struck the swing span support pier and side span superstructure The pier was displaced and the bridge superstructure in side span was partly tom down Captain’s careless navigation due to drunkenness Local newspapers Carnafuli River, Chittagong, Burma Vessel Accident Damage Cause Lit. Unknown, Vessel drifted into the superstructure of the bridge One superstructure span torn down Vessel lost control due to cyclone Personal Communication from Dr. Y. Fujii, Dec. 1591 Hamburg Harbour, Kattwyk, Germany Vessel Accident Damage Cause Lit 21,450 tonnes freighter “Stanislaw Kulezynski” Freighter under tow by three tugs in dense fog lost control and crashed into the side span of the lift bridge The side span fell into the river and one of the lift towers was seriously damaged The vessel lost control due to loss of one of the three tow lines Construction Today, Oct. 1991119 Appendix C. Selected Literature The literature list is organized in accordance with the sections of the publication: SECTION 1 - INTRODUCTION (-1] 1-2] [1-3] 0-4] {1-5} [1-6] [1-7] (1-8) [1-9] [1-10] U-] meee (1-13] Frandsen, A. G.: “Accidents Involving Bridges”, [ABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Frandsen, A.G. and Langsoe, H.: “Ship Collision Problems: I. Great Belt Bridge and IL. International Enquiry”, IABSE Periodica No. 2, 1980. Frandsen, A.G., Olsen, D., Fujii, Y. and Spangenberg, S.: “Ship Collision Studies for the Great Belt East Bridge, Denmark”, IABSE Symposium on Bridges—Interaction between Construction Technology and Design, Leningrad, 1991. Jensen, A.O. and Sorensen, E.A.: “Ship Collision and the Faroe Bridges”, [ABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen 1983. Knott, M.A. and Bonyun, D.: “Ship Collision Against the Sunshine Skyway Bridge”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copen- hagen, 1983, Knott, M. A.: “Ship Collision with Bridges”, Bulletin of Permanent International Asso- ciation of Navigation Congresses (PIANC), No. 57, 1987. Knott, M.A. and Larsen, 0. Damgaaré: “Guide Specification and Commentary for Vessel Collision Design of Highway Bridges”, US Department of Transportation, Federal Highway Administration, Publ. No. FHWA-RD-91-006, Dec. 1990. Larsen, O. Damgaard and Benmoussa, C.: “Ship Collision Aspects for Bridge across the Strait of Gibraltar”, Third Intemational Colloquium on the Fixed Link Europe-Africa Through the Strait of Gibraltar, Marrakesh, May 1990. Maunsell & Partners and Brady, P.J.E.: “Second Hobart Bridge—Risk of Ship Colli- sion and Methods of Protection”, Technical Report prepared for Department of Main Roads, Tasmania, Australia, 1978. Modjeski & Masters, Consulting Engineers: “Criteria for: The Design of Bridge Piers with Respect to Vessel Collision in Louisiana Waterways”, Prepared for the Louisiana Department of Transportation and Development and the Federal Highway Administra- tion, New Orleans, July 1985, Norwegian Public Roads Administration: “Report on Vulnerable Bridges Crossing Navigable Waters” (in Norwegian), Oslo, 1982. Otnhausen, W. Von: “Ship Collisions with Bridges in Sweden”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Ostenfeld, Chr.: “Ship Collisions Against Bridge Piers”, [ABSE Publications, Vol. 25, 1965.120 Appendix (1-14 [1-15] [1-16] 1-17] Rasmussen, B. Hojlund: “Design Assumptions and Influence on Design of Bridges”, JABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copen- hagen, 1983. Tambs-Lyche, P.: “Vulnerability of Norwegian Bridges across Channels”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983 ‘Transport Canada: “Vulnerability of Bridges in Canadian Weters”, Canadian Coast Guard, January 1982. US National Research Council: “Ship Collisions with Bridges. The Nature of the ‘Accidents, their Prevention and Mitigation”, National Academy Press, Washington D.C, 1983, SECTION 2 - INITIAL PLANNING (2-1) [2-2] [2:3] [2-4] [25] [2-6] (2-7) [2-8] [2-9] (2-10) (2-11) [2-12] Bay, J., Spangenberg, S., Olsen, N.H. and Pedersen, P:T.: “Ship Simulations as an Integrated Part of the Design Process for Bridges Crossing Waterways”, Permanent International Association of Navigation Congresses (PIANC), Bulletin 72, 1991. Frandsen, A. G. and Langsoe, H.: “Ship Collision Problems: I. Great Belt Bridge and Il. International Enquiry”, IABSE Periodica No. 2, 1980. Frandsen, A.G., Olsen, D., Lund H.T. and Bach, P.E.: “Evaluation of Minimum Bridge Span Openings Applying Ship Domain Theory”, Transportation Research Record No. 1313. Annual Meeting of Transportation Research Board of the National Research Council, Washington D.C., 1991. Frandsen, A.G., Olsen, D., Fujii, Y. and Spangenberg, S.: “Ship Collision Studies for the Great Belt East Bridge, Denmark”, IABSE Symposium on Bridges—Interaction between Construction Technology and Design, Leningrad, 1991. Fujii, Y: “Integrated Study on Marine Traffic Accidents”, [ABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Fujii, Y: “Effective Areas of Ships”, Journal of Navigation, Vol. 24, Oct. 1971. Fujii, Y., Yamanouchi, H., Tanaka, Yamada, Okuyama and Hirano: “The Behaviour of Ships in Limited Water”, 24th International Navigation Congress, Leningrad, 1977. i, Y. and Tanaka, K.: “Traffic Capacity”, Journal of Navigation, Vol. 24, 1971 Gardenier, J.S.: “Safety of Bridges and Offshore Structures—the Role of Ship Simulation”, [ABSE Colloquium on Ship Collision with Bridges and Offshore Struc- tures, Copenhagen, 1983. Goodwin, E.M.: “A Statistical Study of Ship Domains”, Journal of Navigation, Vol. 28, 1975. Japanese Association for Preventing Marine Accidents (JAPMA): “Illustrated on Marine Safety Law”, Tokyo, 1985 Jensen, A.O, and Sgrensen, E.A.: “Ship Collision and the Faroe Bridges”, IABSE, Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983.(2-13] [2-14] [2-15] (2-16) (2-17) [2-18] [2-19] {2-20} (2-21) Selected Literature 121 Klap, Q.C.: “Ship Collision Analysis for the Westerschelde Crossing”, IABSE Collo- guium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Knott, M.A. and Larsen, O. Damgaard: “Guide Specification and Commentary for Vessel Collision Design of Highway Bridges”, US Department of Transportation, Fed- eral Highway Administration, Publ. No. FHWA-RD-91-006, Dec. 1990. Meurs, K. and Oosterbaan, J.W.: “Simulation of Bridge Passage in High Winds”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copen- hagen, 1983. Okuyama, Y., Hayafuji, Y., Sasaki, Y. and Nakatsuji, T.: “A Study on The Marine Traf- fic Behaviours—Establishment of Methods of Observations and Analysis”, Journal of the Port and Harbour Research Institute (Japan), Vol. 15, No. 1, 1976 Ostenfeld-Rosenthal, P. et al.: “The Concept of Close Ship Encounters—A Decision Tool for Bridge Navigation Span Widths”, To be published in 1992, Shoji, K.: “On the Design of the Waterways Passing Through the Bridges in View of the Analysis on Ship Collision Accidents” (in Japanese), Journal of the Tokyo Mercantile Marine, No. 36, Tokyo, 1985. Shoji, K. and Iwai, A.: “Preservation of Marine Structures against Ship Collision”, International Symposium on Ocean Space Utilization, Tokyo, 1985. Shoji, K. and Wakao, 'T: “On the Ships Waterways Passing Through Bridges”, Water Forum ‘86, World Water Issues in Evolution, San Francisco, 1986. Yamaguchi, A.: “Study Report on the Navigation Safety of the Honshu-Shikohu Bridges”, Honshu-Shikohu Bridge Authority, Tokyo, 1968. SECTION 3 - VESSEL TRAFFIC (3-1) [3-2] (3-3] [3-4] [3-5] [3-6] [3-7] Couper, A. (Editor): “The Times Atlas and Encyclopaedia of the Sea”, Times Books, 1989, Knott, M.A. and Larsen, O. Damgaard: “Guide Specification and Commentary for Vessel Collision Design of Highway Bridges”, US Department of Transportation, Fed- eral Highway Administration, Publ. No. FHWA-RD-91-006, Dec. 1990. Larsen, O. Damgaard and Benmoussa, C.: “Ship Collision Aspects for Bridge across the Strait of Gibraltar”, Third International Colloquium on the Fixed Link Europe-Africa ‘Through the Strait of Gibraltar, Marrakesh, May 1990. Olsen, D., Gotfredsen, H.-H. and Fujii, Y.: “Risk Reducing Effect of the Great Belt VTS System”, Seventh International VTS Symposium, Vancouver, June 1992. Scott, R.: “Standard Ship Designs. Dry Cargo, Container and Ro-Ro Vessels”, Fairplay Publications Ltd., London, 1984. Scott, R.: “Standard Ship Design, Bulk Carriers and Tankers”, Fairplay Publications Ltd., London, 1985. Spyrou, A.G.: “The Million-Ton Tanker Fleet—Is it Really Coming?”, Oil and Gas Journal, March 4, 1974.422 Appendix [3-8] Williamson, G.A.: “The Typical Dimension and Operating Draughts of Bulk Carriers”, Marine Transport Center, University of Liverpool, Dec. 1975. SECTION 4 - RISK ACCEPTANCE [4-1] [4-2] [4-3] [4-4] [4-5] [4-6] [4-7] [4-8] [4-9] [4-10] (4-11) [4-12] [4-13] Frandsen, A.G. and Langsoe, H.: “Ship Collision Problems: I. Great Belt Bridge and IL. International Enquiry”, IABSE Periodica No. 2, 1980. Havnoe, K. and Knott, M.: “Risk Analysis and Protective Island Design for Ship Col- lision", IABSE Symposium on Safety and Quality Assurance of Civil Engineering Structures, Tokyo, Japan, 1986 International Standardization Organization: “Accidental Actions due to Human Activities”, ISO-Working Group TC 98/SC3/WG4, Working Draft, Nov. 1987. Kampmann, J., Kieler, K., Kohl, B. and Spangenberg, S.: “Risk Analysis of the Railway Tunnel under the Great Belt”, Ist International Conference on Safety in Road and Rail Tunnels, Basel, November 1992. Knott, M.A.: “Pier Protection System for the Sunshine Skyway Bridge Replacement”, ‘Third Annual International Bridge Conference, Pittsburgh, June, 1986. Knott, M.A. and Bonyun, D.: “Ship Collision Against the Sunshine Skyway Bridge”, TABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copen- hagen, 1983. Knott, M.A. and Flanagan, M.: “Pier Protection for the Sunshine Skyway Bridge”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copen- hagen, 1983, Knott, M.A. and Larsen, O. Damgaard: “Guide Specification and Commentary for ‘Vessel Collision Design of Highway Bridges”, US Department of Transportation, Fed- eral Highway Administration, Publ. No. FHWA-RD-91-006, Dec. 1990. Larsen, O. Damgaard and Benmoussa, C.: “Ship Collision Aspects for Bridge across the Strait of Gibraltar”. Third International Colloquium on the Fixed Link Europe-Africa ‘Through the Strait of Gibraltar, Marrakesh, May 1990. Leslie, J., Clark, N. and Segal, L.: “Ship and Bridge Collisions—the Economics of Risk”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983, Maunsell and Partners and Brady, P.J.E.: “Second Hobart Bridge—Risk of Ship Colli- sion and Methods of Protection”, Technical Report prepared for Department of Main Roads, Tasmania, Australia, 1978. Modjeski & Masters, Consulting Engineers: “Criteria for: The Design of Bridge Piers with Respect to Vessel Collision in Louisiana Waterways”, Prepared for the Louisiana ‘Department of Transportation and Development and the Federal Highway Administra- tion, New Orleans, July 1985, National Transportation Safety Board: “Ramming of the Sunshine Skyway Bridge by the Liberian Bulk Carrier Summit Venture, Tampa Bay, Florida, May 9, 1980”, Marine Accident Report NTSB-MAR-81-3, Washington D.C., 1981[4-14] [4-15] [4-16] [4-17] [4-18] [4-19] [4-20] [4-21] Selected Literature 123, Nordic Committee for Structural Design: “Recommendations for Loading and Safety Regulations for Structural Design”, NBK-Report No. 55, June 1987. Philipson, L.L.: “Numerical Risk Acceptability and Mitigation Evaluation Criteria”, TABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copen- hagen, 1983. Planeix, Jean-Michel: “Risk—A Subjective Notion Differently Perceived”, IABSE Col- loquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Rowe, W.D.: “Acceptable Levels of Risk for Technological Undertakings”, [ABSE Col- loquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Sexsmith, R.G.: “Bridge Risk Assessment and Protective Design for Ship Collision”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenha- gen, 1983. Swedish State Accident Research Board: “Investigation Report Conceming the Ramming of Almgsund Bridge by Star Clipper” (in Swedish), Stockholm, 1981. Vincentsen, L.J. and Spangenberg, S.: “Safety Management System for the Great Belt Link”, Second Symposium on Strait Crossings, Trondheim, June 1990. Whitman, R.: “Evaluating Calculated Risk in Geotechnical Engineering”, ASCE Journal of Geotechnical Engineering, No. 2, 1984. SECTION 5 - COLLISION RISK [5-1] [5-2] (5-8] [5-9] Canadian Coast Guard: “Vessel ‘Traffic Services”, Transport Canada, October 1984. Cockcroft, A.N.: “A Comparison of Safety Records”, Journal of Navigation, Vol. 34, 1981. Det Norske Veritas: “Cause Relationships of Collisions and Groundings”, Report No. 81-0097, 1981. Fabre, F,, Klose, A. and Salvarani, R.: “COST 301. Shore Based Marine Navigation Aid Systems”, Directorat General Transport, BEC, 1988, Frandsen, A.G. and Langsoe, H.: “Ship Collision Problems: I. Great Belt Bridge and IL International Enquiry”, [ABSE Periodica No. 2, 1980. Frandsen, A.G., Olsen, D., Fujii, Y. and Spangenberg, S.: “Ship Collision Studies for the Great Belt East Bridge, Denmark”, [ABSE Symposium on Bridges—Interaction between Construction Technology and Design, Leningrad, 1991. Fujii, Y., Yamanouchi, H. and Mizuki, N.: “Some Factors Affecting the Frequency of Accidents in Marine Traffic. Il - The Probability of Stranding and III - The Effect of Darkness on the Probability of Collision and Stranding”, Journal of Navigation, Vol. 27, 1974. Fujii, Y. et al.: “Survey on Vessel Traffic Management Systems and Brief Introduction to Marine Traffic Studies”, Electronic Navigation Research Institute Papers (Japan), No. 45, 1984, Fujii, Y., Yamanouchi, H., Tanaka, Yamada, Okuyama and Hirano: “The Behaviour of Ships in Limited Water”, 24th International Navigation Congress, Leningrad, 1977.124 Appendix C [5-10] (5-11) (5-12 [5-13] 5-14] [5-15] [5-16] ae [5-18] [5-19] [5-20] (5-21] [5-22] [5-23] (5-24) (5-25] [5-26] [5-27] [5-28] Fujii, Y. and Shiobara: “The Analysis of Traffic Accidents”, Journal of Navigation, Vol. 24, Oct. 1971. Fujii, ¥.: “Visual Range and the Degree of Risk”, Journal of Navigation, Vol. 27, April 1974, Fujii, Y. et al: “Survey Report on Compulsory Pilotage”, Japan Association for Preventing Marine Accidents (JAPMA), Tokyo, 1983. Fujii, Y.: “The Estimation of Losses Resulting from Marine Accidents”, Journal of Navigation, Vol. 31, May 1978. Heinrich, H. W.: “Industrial Accident Preservation. A Scientific Approach”. 1959, Inoue, K.: “On the Separation of Traffic at Straight Waterway by Distribution Model of Ship Paths”, Journal of Nautical Society of Japan, No. 56, 1977. Knott, M.A. and Bonyun, D.: “Ship Collision Against the Sunshine Skyway Bridge”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copen- hagen, 1983. Knott, M.A. and Larsen, O. Damgaard: “Guide Specification and Commentary for Vessel Collision Design of Highway Bridges”, US Department of Transportation, Federal Highway Administration, Publ. No. FHWA-RD-91-006, Dec. 1990. Kuroda, K. and Kita, H.: “Probabilistic Modelling of Ship Collision with Bridge Piers”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copen- hagen, 1983. Larsen, O. Damgaard: “Ship Collision Risk Assessment for Bridges”, IABSE Collo- quium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Larsen, O. Damgaard and Benmoussa, C.: “Ship Collision Aspects for Bridge across the Strait of Gibraltar”, Third International Colloquium on the Fixed Link Europe-Aftica Through the Strait of Gibraltar, Marrakesh, May 1990. Leslie, J.A.: “Ships and Bridges, 3rd International Conference on Application of Statistics and probability in Soil and Structural Engineering”, Sydney, Australia, 1979. Lund, H.T. and Olsen, D: “Causation Probabilities and Ship Accidents in Great Belt” To be published in 1992. MacDuff, T.: “The Probability of Vessel Collision”, Ocean Industry, Sept. 1974, Madsen, H.O. et al.: “Methods of Structural Safety”, Prentice-Hall, 1986. Matsui, T., Fujii, Y. and Yamanouchi, H.: “Risk and Probability of Marine Traffic Accidents” (in Japanese), Electronic Navigation Research Institute Papers, Vol. 50, 1985. Maunsell and Partners and Brady, P.J.E.: “Second Hobart Bridge—Risk of Ship Colli- sion and Methods of Protection”, Technical Report prepared for Department of Main Roads, Tasmania, Australia, 1978. Meurs, K. and Oosterbaan, J.W.: “Simulation of Bridge Passage in High Winds”, TABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenha- gen, 1983, Mikkelsen, A.: “Ship Collision wit Danish Lighthouses”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983.[5-29] [5-30] [5-31] [5-32] [5-34] [5-35] [5-36] [5-37] 5-38] Selected Literature 125 Mizuki, N., Yamanouchi, H. and Fujii, Y.: “The Result of Third Survey on Vessel Traffic Services in the World (Appendix V, New Data on Accident Probabilities)”, Electronic Navigation Research Institute Papers, Vol. 59, 1989. Modjeski and Masters, Consulting Engineers: “Criteria for: The Design of Bridge Piers with Respect to Vessel Collision in Louisiana Waterways”, Prepared for the Louisiana Department of Transportation and Development and the Federal Highway Admini- stration, New Orleans, July 1985. Olsen, D.R, Gotfredsen, H.-H. and Fujii, Y “Risk Reducing Effect of the Great Belt VTS System”. 7th Intemational VTS Symposium, Vancouver, June 1992. Olsen, D.F, Drejfeldt, $.R. and Olsen, N.H.: “Theoretical Modelling and Application of a Ship Collision Risk Model to Bridges crossing Navigable Waters”. To be published in 1993. Ostenfeld-Rosenthal, P., Lund H. T. and Drejfeldt, S. R.: “The Concept of Close Ship Encounters —A Decision Tool for Navigation Span Widths of Bridges”. To be published in 1993 Pyman, M.A. Austin, J.S. and Lyon, P.R.: “Ship/Platform Collision Risk in the U.K. Sector”. [ABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Sexsmith, R.G.: “Bridge Risk Assessment and Protective Design for Ship Collision”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copen- hagen, 1983. United States Coast Guard: “Vessel Traffic Systems. Analysis of Port Needs”, Depart- ment of Transportation, Washington D.C., August 1973. Wasa, Y. and Oshitari, M.: “Ship Collision with the Tokyo Bay Crossing Bridge- Tunnel”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Wheatley, J.H.W.: “Research at NPL cn Marine Traffic Systems”, National Physical Laboratory (UK), North East Coast Institute of Engineers and Ship Builders, Transac- tions of 49th Session, Vol. 89, 1972-73. SECTION 6 - VESSEL IMPACT FORCES [6-1] [6-2] [6-3] [6-4] Akita, Y., Ando, N., Fujita, Y, and Kitamura, K.: “Studies on Collision—Protective Structures in Nuclear Powered Ships”, Nuclear Engineering and Design, Vol. 19, 1972. Amdahl, J.: “Energy Absorption in Ship-Platform Inpacts”, Dr. Ing. Thesis, Report No. UR-83-34, The Norwegian Institute of Technology, Trondheim, 1983. Euratom: “Collision Tests with Ship Models”, Euratom Report Eur 4560 B, Luxemburg, 1971 Frandsen, A.G. and Langsoe, H.: “Ship Collision Problems: I. Great Belt Bridge and II. International Enquiry”, IABSE Periodica No. 2, 1980.126 Appendix [6-5] [6-6] [6-7] [6-8] [6-9] [6-10] [6-11] 6-12] [6-13] [6-14] [6-15] (6-16) [6-17] (6-18] [6-19] (6-20] [6-21] Frandsen, A.G., Olsen, D., Fujii, Y. and Spangenberg, S.: “Ship Collision Studies for the Great Belt East Bridge, Denmark”, IABSE Symposium on Bridges—Interaction between Construction Technology and Design, Leningrad, 1991, French Ministry of Public Works: “Public Regulations” (in French). Fascicule No. 61, Paris, 28 Dec. 1971. Gerard, G.: “The Crippling Strength of Compression Elements”, Journal of Aero- nautical Sciences, January 1958. German Ministry of Traffic: “Safeguarding of the Piers of Rhine Bridges against Impact Forces from Barges”. Notification of 8 Aug. 1974. Hagiwara, K., Takanabe, H. and Kawano, H.: “A Proposed Method of Predicting Ship Collision Damage”, Ship Strength Laboratory, Nagasaki Techn, Institute, Japan, 1982. International Standard Organization: “Accidental Actions due to Human Activities”, Working Group TC 98/SC3/WG4, Working Draft, November 1987. Ito, H., Kondo, K., Yoshimura, N., Kawashima, M, and Yamamoto, S.: “A Simplified Method to Analyse the Strength of Double Hulled Structures in Collision, Report no.1, 2and 3.”, J. of The Society of Naval Architects of Japan, vol. 156, 1984, vol. 158, 1985, vol. 160, 1986. Jensen, A.O. and Sgrensen, E.A.: “Ship Collision and the Faroe Bridges”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Kagami et al.: “Research on Collision Resisting Construction of the Sides of Ship”, Report No. 2, Mitsubishi Nippon Industries, Technical Review 2, 1961 Knott, M.A. and Bonyun, D.: “Ship Collision Against the Sunshine Skyway Bridge”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copen- hagen, 1983, Knott, M.A. and Larsen, O. Damgaard: “Guide Specification and Commentary for Vessel Collision Design of Highway Bridges”, US Department of Transportation, Federal Highway Administration, Publ. No. FHWA-RD-91-006, Dec. 1990. Kuesel, T.R.: “Newport Bridge Collision”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Larsen, O. Damgaard and Benmoussa, C.: “Ship Collision Aspects for Bridge across the Strait of Gibraltar”, Third International Colloquium on the Fixed Link Europe-Africa ‘Through the Strait of Gibraltar, Marrakesh, May 1990. Marsh, K.J. and Campbell, J.D.: “The Effect of Strain Rate on the Post-Yield Flow of Mild Steel”, Journal of Mechanics and Physics of Solids, 11, 49, 1963. Meir-Domberg, K.E.: “Ship Collisions, Safety Zones and Loading Assumptions for Structures on Inland Waterways” (in German), VDI-Berichte No, 496, 1983. Minorsky, V.U.: “An Analysis of Ship Collisions with Reference to Protection of Nuclear Power Plants”, Journal of Ship Research, Oct. 1959, Modjeski & Masters, Consulting Engineers: “Criteria for: The Design of Bridge Piers with Respect to Vessel Collision in Louisiana Waterways”, Prepared for the Louisiana Department of Transportation and Development and the Federal Highway Administra- tion, New Orleans, July 1985[6-22] (6-23] [6-24] [6-25] [6-26] [6-27] [6-28] [6-29] [6-30] (6-31) [6-32] {6-33] [6-34] [6-35] [6-36] [6-37] [6-38] [6-39] Selected Literature 127 Nagasawa, H., Arita, K., Tani, M. and Oka, S.: “A Study on the Collapse of Ship Structure in Collision with Bridge Piers”, Naval Architecture and Ocean Engincering, Society of Naval Architects, Japan, Vol. 19, 1981. Nordic Road Engineering Federation: “Load Regulations for Road Bridges” (in Nor- wegian), NVF, Report No. 4, 1980. Norwegian Public Roads Administration: “Load Regulations for Bridges and Ferry Ramps in the Public Road System” (in Norwegian), Preliminary Edition, Olso, 1986. Ohnishi, T., Kawakami, H., Yasukawa, W. and Nagasawa, H.: “Ultimate Strength of Bow Construction”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Olnhausen, W. Von: “Ship Collision with Bridge Piers” (in Swedish), Teknisk Tidskrift, No. 17, Stockholm, 1966. Ostenfeld, Chr.: “Ship Collisions Against Bridge Piers”, [ABSE Publications, Vol. 25, 1965. Pedersen, P.T, Valsgard, S., Olsen, D. and Spangenberg, $.: “Ship Impacts—Bow Collisions”. 3rd International Symposium on Structural Crashworthiness and Failure, Liverpool, April 1993. Prucz, Z. and Conway, W.B.: “Ship Collision with Bridge Piers—Dynamic Effects”. ‘Transportation Research Board 69th Annual Meeting, Washington D.C., 1990. Reckling, K.A.: “Beitrag der Elasto- und Plastomechanik zur Untersuchung von Schiffskollisionen”, Jahrbuch der Schiffbautechnischen Gesellschaft, Band 7 0, 1976. Requena, L.F.: “Détermination dynamique de 1a charge transitoire, lors d’une collision”. IABSE Colloquium on Ship Collision with Bridges and Offshore Structures. Copenhagen, 1983. Saul, R. and Svensson, H.: “Means of Reducing Consequences of Ship Collisions with Bridges and Offshore Structures”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Saul, R. and Svensson, H.: “On the Theory of Ship Collision against Bridge Piers”, TABSE Proceedings, No. 2, 1982. Saul, R. and Svensson, H.: “On Protection of Bridge Piers against Ship Collision” (in German), Die Bautechnik, Vol. 58, 1981. Teyssandier, J.-P. et al.: “Impact Protection from Ships in the Euroroute Project”, IABSE Symposium on Concrete Structures for the Future, Paris- Versailles, 1987. Wierzbicki, T.: “Crushing Behaviour of Plate Intersections”, Structural Crashworthi- ness, edited by N. Jones and T. Wierzbicki, Chapter 3, Butterworth & Co. (Publishers) Ltd., 1983. Woisin, G. and Gerlach, W.: “On the Estimation of Forces Developed in Collisions between Ships and Offshore Lighthouses”, 8. International Conference On Lighthouses and Other Aids to Navigation, Stockholm, 1970, Woisin, G.: “Design Against Collision”, International Symposium on Advances in Marine Technology, Trondheim, 1979. Woisin, G.: “Die Kollisionsversuche der GKSS”, Schiff und Hafen, Heft 2, 1977128 Appendix C [6-40] (7-1) {7-2} {7-3} [7-4] [7-5] (7-6] (7-7) [7-8] Yang, P.D.C. and Caldwell, J.B.: “Collision Energy Absorption of Ships’ Bow Struc- tures”, International Journal of Impact Engineering, No. 2, 1988. ‘TION 7 - BRIDGE DESIGN Blok, J.J. and Dekker, J.N.: “On Hydrodynamic Aspects of Ships Colliding with Fixed Structures”, IABSE Colloquium on Ship Collision with Bridges and Offshore Strue- tures, Copenhagen, 1983. Fauchart, J.: “Choe de bateau sur obstacle deformable”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Fauchart, J.: “Consequences of a Ship Collision with the Verdon Bridge”, IABSE Col- loguium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Knott, M. A. and Larsen, O. Damgaard: “Guide Specification and Commentary for Ves- sel Collision Design of Highway Bridges”, US Department of Transportation, Federal Highway Administration, Publ. No. FHWA-RD-91-006, Dec. 1990. Modjeski & Masters, Consulting Engineers: “Criteria for: The Design of Bridge Piers with Respect to Vessel Collision in Louisiana Waterways”, Prepared for the Louisiana Department of Transportation and Development and the Federal Highway Administra- tion, New Orleans, July 1985. Nordic Road Engineering Federation: Norwegian), NVF, Report No. 4, 1980. Petersen, M. J. and Pedersen, P. Terndrup: “Collision between Ships and Offshore Plat- forms”, 13th Annual Offshore Technology Conference, Paper No. OTC 4134, 1981. Pruez, Z. and Conway, W.B.: “Ship Collision with Bridge Piers—Dynamic Effects Annual Meeting of Transportation Research Board of the National Research Council, Washington D.C., 1990, Saul, R. and Svensson, H.: “On the Theory of Ship Collision against Bridge Piers”, IABSE Proceedings, No. 2, 1982. “Load Regulations for Road Bridges” (in SECTION 8 - PREVENTION MEASURES (8-1] [8-2] (8-3] [8-4] Fabre, F., Klose, A. and Salvarani, R.: “COST 301. Shore Based Marine Navigation Aid Systems”, Directorate General Transport. EEC, 1988. Fujii, ¥. et al.: “Survey on Vessel Traffic Management Systems and Brief Introduction to Marine Traffic Studies”. Electronic Navigation Research Inst tute Papers (Japan) No. 45, 1984. Greneker, E.F., Eaves, J.L. and McGee, M.C.: “Bridge Ship Collision Electronic Detection and Early Warning”, [ABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Hara, Kiyoshi: “Progress of VTS and its Studies in Japan”, International Symposium on Vessel Traffic Service, Gothenburg, Sweden, May 1988.[8-5] [8-6] [8-7] [8-8] [8-9] [8-10] [8-11] [8-12] [8-13] Selected Literature 129 “The IALA, IAPH, IMPA World VTS Guide”, Pergamon Press, 1989. International Maritime Organization (IMO): “Guidelines for Vessel Traffic Services”, Resolution A.578 (14) adopted on 20 November 1985. IMO, 1985. International Maritime Organization (IMO): “International Conference on Revision of the International Regulations for Preventing Collision at Sea, 1972”. IMO, 1990. Knott, M.A. and Flanagan, M.: “Pier Protection for the Sunshine Skyway Bridge”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Mizuki, N., Yamanouchi, H. and Fujii, Y.: “The Result of Third Survey on Vessel ‘Traffic Services in the World”. Electronic Navigation Research Institute Papers, Vol. 59, 1989. Olsen, D.F.,, Gotfredsen, H.-H. and Fujii, Y.: “Risk Reducing Effect of the Great Belt VTS System”, Seventh International VTS Symposium, Vancouver, Canada, June 1992, US Coast Guard: “Vessel ‘Traffic Services, Equipment and Technology Report (VTS Handbook)”, Washington D.C., May 1987. Vendrell, J.: “Minimizing the Risk with Vessel Traffic Management Systems”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. “VTMS for World’s Longest Bridge”, World Port Construction & Ocean Technology, April/May 1991. SECTION 9 - PROTECTION MEASURES [9-1] [9-2] Brink-Kjzr, O., Brodersen, F.P. & Nielsen, A.H.: “Modelling of Ship Collisions Against Protected Structures”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Denver, H.: “Geotechnical Model Tests for the Design of Protective Islands”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Englot, J. P.: “Collision Protection of Arthur Kill Bridges”, New York ASCE Section, Structures Conference, ASCE, New York, May 1988. Fletcher, M.S., May, R. W.P. and Perkins, J. A.: “Pier Protection by Man-Made Islands for Orwell Bridges, U.K.”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Frandsen, A. G, and Langsoe, H.: “Ship Collision Problems: 1. Great Belt Bridge and II. International Enquiry”, IABSE Periodica No. 2, 1980. Hahn, D.M. and Rama, H.E.: “Cofferdams Protecting New York Bridges from Ship Collisions”, Civil Engineering-ASCE, February 1982. Havnoe, K. and Knott, M.: “Risk Analysis and Protective Island Design for Ship Collision”. IABSE Symposium on Safety and Quality Assurance of Civil Engineering Structures, Tokyo, 1986.130 Appendix [9-8] [9-9] [9-10] (9-11) [9-12] [9-13] [9-14] [9-15] [9-16] [9-17] [9-18] (9-19] [9-20] [9-21] [9-22] [9-23] [9-24] Heins, C.P.: “Bridge Dolphins Subjected to Impact", IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Heins, C.P. and Chiu, L.Y-B.: “Dynamic Analysis of Dolphins subjected to Ship Impact”, Computers and Structures, vol.15, no.1, 1982. Knott, M.A. and Larsen, O. Damgaard: “Guide Specification and Commentary for Vessel Collision Design of Highway Bridge”, US Department of Transportation, Federal Highway Administration, Publ. No. FHWA-RD-91-006, Dec. 1990. Knott, M.: “Pier Protection for the Sunshine Skyway Bridge Replacement”, Third Annual International Bridge Conference, Pittsburg, June 1986. Larsen, O. Damgaard and Benmoussa, C.: “Ship Collision Aspects for Bridge across the Strait of Gibraltar”, Third International Colloquium on the Fixed Link Europe-Africa Through the Strait of Gibraltar, Marrakesh, May 1990. Luong, M.P.: “Modeles réduits de Protection de pile de pont”. IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Minorsky, V.U.; “Evaluation of Ship Bridge Pier Impact and of Islands as Protection”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenha- gen, 1983, Matsuzaki, Y, and Jin, H.: “Design Specification of Buffer Structure”, IABSE Collo- quium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Mondorf, P.E.: “Floating Pier Protections Anchored by Prestressing Tendons”, [ABSE. Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Namita, Y. and Nakanishi, H.: “Analysis of Framed Buffer Structure around Bridge Pier”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Navi, Y. and Ishikawa, T.: “Case Stories of Dolphin Accidents and Remedies”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Oda, K, and Kubo, S.: “Collision Prevention Device of Floating Guideline Type”, TABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copen- hagen, 1983. Ostenfeld, Chi 1965. Saul, R. and Svensson, H.: “Means of Reducing Consequences of Ship Collisions with Bridges and Offshore Structures”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Saul, R. and Svensson, H.: “On the Theory of Ship Collision against Bridge Piers”, TABSE Proceedings, No. 2, 1982. Saul, R. and Svensson, H.: “On Protection of Bridge Piers against Ship Collision” (in German), Die Bautechnik, Vol. 58, 1981. Sexsmith, R.G.: “Bridge Risk Assessment and Protection Design for Ship Collision” IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copen- hagen, 1983. ‘Ship Collisions Against Bridge Piers”, IABSE Publications, Vol. 25,[9-25] [9-26] Selected Literature 131 Tambs-Lyche, P.: “Vulnerability of Norwegian Bridges across Channels”, IABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Vitalis, A.: “Ecran de protection des ouvrages en mer contre les collisions”. IABSE Col- Joquium on Ship Collision with Bridges and Offshore Structures, Copenhagen 1983, SECTION 10 - PROTECTION OF THE PUBLIC [10-1] [10-2] [10-3] Greneker, E.F., Eaves, J.L. and McGee, M.C.: “Bridge Ship Collison Electronic Detection and Early Warning”, [ABSE Colloquium on Ship Collision with Bridges and Offshore Structures, Copenhagen, 1983. Knott, M.A. and Larsen, O. Damgaaré: “Guide Specification and Commentary for Vessel Collision Design of Highway Bridges”, US Department of Transportation, Federal Highway Administration, Publ. No. FHWA-RD-91-006, Dec. 1990. Transportation Research Board: “Considerations in the Development of an Early Warning Vessel/Bridge Collision System”, Conference Proceedings, Washington D.C., 1978.Structural Engineering Documents ‘Objective: Praviding in-depth Informa- ton to the practising engi- naer by publishing reports of aigh sciemtific ond technical level on structural ‘enginearing themes. Fields covered: Construction materials! siruetural analysis/ dynamic: analysis/constructian mathods/research/design/ execution/maintenance/ history af construction science and technology/ interaction between structurstenginesring and other fields {e.g) architec ture). Readership: Engineers throughout the ‘world who ara interested in clvit engineering structures: researchers, professors, designers, entraprenaurs, ‘administration and owners’ representatives. Publisher, TABSE - created in 1829 - lathe International Asso- lation for Bridge and ‘Structural Engineering, comprising more than 3400 members in B0 countries. A ‘scientific, sue lene|, nan= povernmental and non-prof sonducting conferences and Issuing publications, Besides the series “Structural Engineering Documents”, the {ABSE algo publishes athar serioée “Structural Engineering international’’=a quarterly technical journal JABSE REPORTS \ABSE CONGHESS REPORTS For further information: \ABSE ETH-HOnggerberg CH-093 Zurich, Switzerland Fax: Int.+ 41493712131Ship Collision with Bridges The Interaction between Vessel Traffic and Bridge Structures Any structure in navigable waters constit itself vulnerable to damage or destruction in the event of vessel collision: Worldwide vessel traffic and the average size of vessels continue to increase. At the same time, ever more bridges crossing navigable waterways are being planned and constructed, sometimes with inadequate navigation clearance and/or inadequate protection. es a hazard to shipping and is The objective of this publication is to provide information and guidelines for engineers charged with the planning and design of new bridges, navigation channels, and prevention and protection measures, It offers advice on upgrading and retrofitting existing bridges and navigation channels. And it ‘Provides: the means to evaluate the safety of bridges, vessels, persons and ‘the environment. After reviewing some basics of navigation and vessel traffic, and considering risk acceptance and collision risk, the publication examines vessel impact forces on bridges and proposes appropriate bridge design criteria. ‘Prevention measures, such as regulations and management systems, and protection measures and systems are also described. Major international research projects have provided the analytical basis for the publication, including the development of vessel collision guide specifications for the Federal Highway Administration in the | USA and the vessel collision design criteria developed for the Great Belt Crossing in Denmark, Prepared by Ole Damgaard LARSEN, Chairman of! the |ABSE Working Group “Ship Collision with Bridges”, this 132 page publication is a must for any engineer dealing with structures j in navigable waters,

SED04

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SED04

Uploaded by

Copyright:

Available Formats

You might also like