tech nical failure Feasibility real-time simulation Operations feasibility Safety Assessment and ranking Tool (SMART)

Transcription

1 SAFETY ASSESSMENT IN PORTS C van der Tak J. H. de Jong MSCN/MARIN P.O.Box 28, 6700 AA Wageningen 1 INTRODUCTION Public safety awareness and the related distribution of responsibilities to local authorities has increased the need for tools to evaluate the total safety in the port environment. The maritime operations determine an essential part of this safety. Traditionally expert opinion completed with simulation studies, either full-mission simulator or fast-time simulations, helped to evaluate the different design or existing port lay-outs and operational measures within a given environment and a given traffic distribution. This is still a viable option for the basic assessment of the feasibility of a design on the operational level but fails to predict the total levels of risk and consequences of measures once the total traffic distribution needs to be evaluated. Over the last two decades additional quantitative safety management assessment tools have been developed who take into account the total vessel traffic image and its related risks in the whole physical port environment and who are capable of evaluating the consequences of measures on a strategic level. More recent developments in the Netherlands and the UK (Port Marine Safety Code) suggest the application of formal safety assessment to ports as a panacee to all strategic safety issues, a promise eventually true if due consideration is given to the devil in the details. This paper describes different kinds of safety studies that are performed to assess the safety level in ports. The type of study to be performed is related to the type of questions to be answered. This paper tries to clarify the merits of the different type of safety studies against their area of application. Table 1 shows a subdivision of safety studies and their mutual relationships by presenting their common and different parameters. Which type of study will be performed depends on the main question that has to be answered. A rough subdivision of safety studies is presented in table 1.1. Different types of Safety Studies Fast-time simulation & expert opinion daily opera tions study parameters Generic vessel Port Environment Organization Policy level human failure tech nical failure traffic lay-out wind current waves on shore on ship type classification Feasibility real-time simulation Operations feasibility Safety Assessment and ranking Tool (SMART) Regulatory Formal Safety Assessment (FSA) Regulatory Table 1.1 Classification of safety studies

2 All study parameters presented in the table are dealt with in the different safety studies. Only the attention to an aspect varies from little (blank in the table), a rough description (+ in the table) to a detailed description (++). In reality the classification can not be performed as straight forward as described in the table. MSCN/MARIN has performed many studies of the first three types. Examples of these studies will be given in the sections 2, 3 and 4. In section 5 the difference between SMART and FSA is explained. In fact the general approach in POLSSS (see section 3) is rather similar with the approach in FSA. The difference is formulated in section 5 and it is indicated which synergy can be achieved by combining SMART and FSA. 2. SAFETY ASSESSMENT THROUGH MANOEUVRING SIMULATIONS 2.1 Introduction Safety assessment is a key issue from the early stage of port lay-out design or in designing port extensions. Port lay-out design and port extensions, including the further supportive infrastructure like pilotage systems, VTS, buoyage system, tug assistance, etc. determine for a great deal the inherent safety of a port. It reduces the likeliness of contacts and groundings and facilitates the smooth regulation of traffic flows in a port. Obviously the basic infrastructure of a port e.g. the approach channels to be dredged, the length and configuration of the breakwaters and the lay-out design of the quays are cost driving items and need an evaluation in an early stage of the design or port extension project in order to execute the C/B analysis. Based on the above considerations three phases could be distinguished within a new design. Feasibility study Concept design Detailing or confirmation study In port extension or safety assessment projects only the latter two phases are executed. Safety assessments in existing ports can reveal important differences in safety level. In the following paragraphs separate attention will be given to the different phases, the available tools and the relation with the overall safety assessment. 2.2 Feasibility study The PIANC and IAPH publication Approach channels, A guide for design, addresses the first phase of the design. The guide, which is as a computer program for the channel width freely available at MSCN, gives together with expert opinion a first estimation of the space required in the port entrance and the harbour basins to allow a safe operation of the incoming largest vessels. Through a variation of the size of vessels and the environmental conditions, as representative conditions under which port operations should be possible, a large matrix of configurations of port and harbours can be quickly evaluated. The safety assessment in this stage is largely based on collective experience and is valid for the individual vessel operation only. Contributions to the total safety level through the large numbers of vessel movements in a port are not evaluated in this stage.

3 2.3 Fast-time manoeuvring simulation safety assessment The use of fast-time manoeuvring simulation is a further iterative step in the process of balancing the spatial requirement and vessel operation, as the main parameters determining the basic safety issues and the most economic often smallest harbour configuration. Fast-time manoeuvring simulation still allows the performance of cheap, albeit less accurate and rapid trade-off studies. In this stage still trade-off analysis and rapid data collection are key-elements, whereas in the final stage the involvement of human (inter) action is paramount. An example of a track-plot of a fast-time simulation is shown in Fig This type of simulation gives the designer and port operator an insight into the inherent possibilities and restrictions of vessels in relation to the infrastructure and environmental conditions like wind, waves and current. It is particularly useful in comparative studies who has to evaluate several layouts. It is possible to judge the safety of individual vessel manoeuvres and operations in a harbour area on a fairly accurate level and address an important contribution of the total risk level. However the total risk level due to total traffic flows can not be evaluated nor the consequences of general safety measures like VTS and the application of pilotage. The consequences of tug assistance during emergencies can be evaluated as will be discussed below. The objectives that can be studied by means of fast-time simulation are: Determination of the factors having the strongest influence on the manoeuvrability Comparing alternative lay-out designs in terms of nautical safety and efficiency Proposing improvements on the layout or limits to admissibility Determination of requirements of tug assistance capacity Determination of the critical conditions for the manoeuvres under consideration It is clear from the above that some very important parameters (system influences) as identified within a formal safety assessment are addressed through the use of the fast-time simulation tool for the individual, most relevant ship or within a operational ship traffic context, which is likely to be the most important contributor to the probability of accidents. 2.4 Real time manoeuvring simulation safety assessment The most important aspect of the full-mission simulation is the incorporation of the complex human interference. The highly interactive complex port entrance manoeuvres during difficult environmental conditions, added up to ship to ship interactions in more complex traffic situations are typical scenarios for the safety evaluation of existing ports or ports under consideration. Human performance is a highly decisive parameter in the safety assessment of port entrance and harbour manoeuvres. To this end a study with real-time simulation is extremely useful. A mockup of a ship s bridge is used for real-time simulations. [Fig.2.2: A mock-up of a ship s bridge] The objectives, all aiming at safety assessment, that can be studied by real-time simulation are: 1. determining the admittance policy, e.g. tidal, wind or wave constraints 2. defining operational procedures and strategies, e.g. tug use 3. determining the (effectiveness of) aids to navigation (like buoys, beacons, leading lights, etc.)

4 These objectives can be reached in the simulator through a systematic evaluation of a preselected set of scenarios. Either within the modelled existing port or in the newly designed port. These scenarios are based upon an evaluation of the vessel traffic flows, and the dominant and determinative environmental conditions, often priorly tested through fast-time simulations and discussed against the background of tug assistance schemes in existence or required. The individual scenarios are straightforwardly judged and later on analysed by the evaluations of the trajectories, rudder and engine use, tug use, ship speed and space requirement. Per scenario the number of repeatedly executed runs is limited. Their number does not allow a thorough statistical evaluation but justify a heuristic judgement supported by the analysis per scenario combined sets of runs. The simulator often has an additional important role e.g. acting as a catalyst for the entire design or safety assessment team. The simulator allows many people who are to be involved and knowledgeable to contribute to the quality of the assessment. This team ideally consists of representatives of the following three categories: Designers: civil engineers, naval architects/nautical experts Administrator: local harbour authorities or government authorities Nautical users: (local) pilots, captains As representatives of each of the three categories can participate in the above system of simulation runs, this often leads to very productive and constructive discussions. 3. SAFETY MANAGEMENT ASSESSMENT RANKING TOOL (SMART) 3.1 INTRODUCTION What actions must be taken in order to improve the safety? Measures are not always assessed in a quantitative sense before they are introduced. Sometimes they do not have the expected outcomes. Firm and clear predictions of the effect of measures will help policy makers in finding the optimal mix of measures. A quantitative assessment requires a large number of databases and models. MSCN/MARIN, in co-operation with Marine Analytics, has built up during the last decade a toolbox within studies done for the Directorate-Transport Safety, the North Sea Directorate in the Netherlands and the European Commission. SMART is such a growing toolbox containing a lot of valuable data structures and descriptive models. Firstly a brief description of the methodology is given and next the two complete different examples are presented where SMART can be applied to. 3.2 METHODOLOGY The Marine Traffic System diagram is presented in Figure 1. SMART follows more or less the flows of the diagram. SMART can be applied in different type of areas, like within a port, for the approach to a port and within a full sea area. Depending on the area where SMART is applied some system elements are described in more detail. The most important elements are the traffic model and the casualty models that predict the number of casualties based on the traffic flows. The traffic model describes how the vessels are sailing within the area. The accident probability models describe what type of casualties can be expected within the area. In this approach the historical casualty data is used for the validation of the models. Local validation is very often difficult because relevant sets of accident data are most of the time not available. Hence it is important that the model itself inspires confidence through a

5 sound and solid knowledge of the underlying physical processes and the mathematical modelling thereof. Additional models are used to model the consequences of casualties. Within the area of a port the individual risk and societal risk are important. The individual risk is the probability that a person dies by a casualty when he is the whole year at a certain location. The iso-contour of the individual risk connects points with equal individual risk. A norm (in the Netherlands) exists that no new building is allowed within the 10-6 contour. The societal risk is measured as the probability of more than 10, 100 etc people dying in a disaster. At sea the consequences in term of oil spills and danger to the passengers of vessels are important consequences that have to be minimised. Traffic Demand Cargo Passengers Fishing Recreational vessels Existing traffic management system Maritime Traffic System Traffic: Traffic intensity Traffic mix Traffic management measures: Routing (TSS) Waterway marking Piloting Vessel Traffic Services Characteristics of Port or sea area Ships: Technology on board Quality of ships Quality of crew Tactics (New traffic management measures) Traffic accidents Collisions Contacts Strandings Founderings Other accidents Fire and explosions Spontaneous hull accidents Cargo accidents Impacts Financial costs: investment cost operating costs Environmental consequences: oil spills amount of oil on coast chemical spills dead and affected organisms Economic consequences: loss of income repair costs cleaning costs delay costs caused by accidents extra sea miles caused by the use of a tactic Human safety: Individual risk (internal and external) Societal risk (internal and external) Search and Rescue Contingency planning Figure 3.1 System diagram for the Marine Traffic System

6 SMART is also used in combination with both fast-time simulation studies and real time simulation studies on our full-scale bridge simulator. SMART delivers additional insights that can not be achieved only with simulation studies. 3.3 EXAMPLES Two examples are presented for illustration where the SMART-model can be applied. The first example describes a very complex and extensive study concerning the safety at sea, approaches and in harbours. The second example describes a small study to the safety for the harbour of Kishon, THE POLSSS STUDY Background The project systematisering veiligheidsbeleid Noordzee (PSVN) has been describing the Dutch shipping safety policy. It is doing this by defining a framework for decision making with regard to the safety of shipping and the environment in the North Sea and areas around Dutch ports and harbours. This framework covers the evaluation of impacts of currently-used policy measures, as well as the prediction of the impacts of new measures that might be implemented sometime in the future. The final phase of PSVN called for: (1) an assessment of the costs and effects of individual policy instruments as well as combinations of these instruments; (2) a survey of the perceptions of stakeholders about the safety situation in the North Sea and areas around Dutch ports and harbours; and (3) a survey of the perceptions of stakeholders about the effectiveness of policies for maintaining safety in these areas. The policy for sea shipping safety (POLSSS) study constitutes the final phase of PSVN. The POLSSS project was commissioned by the Directorate for Transport Safety, which is part of the Directorate-General for Freight Transport (DGG) within the Ministry of Transport, Public Works, and Water Management. Begun in August 1997, the project was carried out by a team of researchers from RAND Europe and the Maritime Simulation Centre the Netherlands (MSCN), a department of MARIN. Work on POLSSS was divided into two parts: (1) a cost-effectiveness analysis of policy instruments, and (2) two surveys of stakeholders. The cost-effectiveness analysis considered most types of shipping accidents and maritime disasters, and aimed to: Define a set of new possible policy measures, Evaluate the costs and benefits of the new measures, Determine whether there are measures that can improve safety or reduce costs in the North Sea and surrounding areas. The second part of the POLSSS project consisted of two surveys that were conducted to obtain information on stakeholders perceptions about the safety situation in the North Sea and surrounding areas and about their perceptions of the current and possible new policy measures. The objectives of this part of the POLSSS project were to collect information from stakeholders about their: perceptions of the effectiveness of the existing policy for maintaining safety of sea shipping in North Sea and harbour areas, willingness to accept the risk of certain types of shipping accidents and maritime disasters, perceptions of the extent to which the policy objectives of current policies have been reached,

7 future needs. This paper provides an overview of the cost effectiveness study. Analytical Framework The cost-effectiveness analysis was performed using a structured approach developed at RAND to assist policymakers in evaluating public policy options in situations involving complex systems with multiple measures of performance and involving competing interest groups with different, and frequently conflicting, goals. In the context of maritime navigation and safety in the North Sea, Dutch seaports and approaches, there are a wide variety of individual traffic management instruments and measures that could be implemented. A tactic is a change in the existing set of traffic management instruments and measures. Each tactic, alone or in combination with other tactics (a combination of tactics is called a strategy), will affect the occurrence or the severity of traffic accidents or mitigate their negative consequences (e.g., tactics related to search and rescue services). The accidents have many different types of effects, including effects on safety, the environment, and the economy. The tactics also have effects that are not related to accidents. For example, changes in routing may cause extra sea miles that introduce costs. The tactics also introduce financial costs. The term impacts is used to refer to the effects that result from the implementation of a tactic. Performance measures (in natural units, such as cubic meters of oil spilled) are defined for each of the impacts of interest. One of the major tasks in the costeffectiveness analysis was to estimate the effects of the tactics on the performance measures. The impacts of tactics are estimated for a variety of different types of accidents, including collisions, contacts, strandings, and some founderings by using the SMART-models. In addition to the type of accident, the consequences of an accident depend on the types of ships that are involved (e.g., ferries and oil tankers). The analysis is primarily focused on the annual cumulative effect of all accidents. However, some types of accidents happen rarely but have large consequences, the so-called disasters. The impacts of the tactics are likely to differ by type of disaster. Because of the magnitude of their consequences, a number of different types of disasters, called scenarios, are analysed separately. An example of a scenario studied in detail is a ferry colliding with an oil tanker. The results of the "impact assessment" (the impacts of tactics) are presented in the form of tables called scorecards. Each row of a scorecard represents an impact and each column represents a tactic. An entire column shows all of the impacts of a single tactic; an entire row shows each tactic's value for a single impact. Numbers or words appear in each cell of the scorecard to convey whatever is known about the size and direction of the impact. The POLSSS study analyses different regions separately. Because most of the tactics apply to some regions and not to others, and the impacts of the tactics differ by region, there are separate scorecards for each region. There are also scorecards showing the impacts of each tactic over the regions. Since the impacts of tactics and strategies are also assessed under future conditions, it is important to define what those conditions might be. A set of assumptions about the external conditions within which the tactics will be operating is called the context. Context variables (which remain fixed throughout the analysis) include such things as the traffic intensities, the types and sizes of ships, the quality of the vessels, and the level of training of the crews. The entire analysis was conducted separately for two different contexts. The first set of context variables describes the situation from the present up to the year The second set of assumptions is used to describe the situation within the time period Tactics A tactic is a single action taken to affect the maritime traffic system, such as widening the shipping lanes in the open sea by a certain percentage or removing all of the buoys used as reference marks. One way of characterising tactics is by their mechanisms, or the ways they operate. This gives rise to the following nine groups of tactics:

8 Routing tactics, which affect traffic separation schemes and deep water routes. They assign different shipping and offshore activities to different parts of the sea based on sailing direction, vessel size, and type or function. Position fixing tactics, which affect a variety of radio navigation systems that provide great accuracy in identifying ship positions. Waterway marking tactics, which refer to local (often visual) navigation aids that include fixed or floating lighted and unlighted objects (buoys and beacons), radar beacons of different types, and fixed land markings (e.g., lighthouses). Information supply tactics, which refer to the provision of traffic, hydrological, and other information to ships. This category includes the transmission of static information (such as nautical charts and publications) and dynamic information (such as weather forecasts). Piloting tactics are changes to the use of persons who help guide ships through the Dutch part of the North Sea and the approach areas to Dutch seaports, on board as well as from shore. Vessel traffic services (VTS) tactics, which affect the collection of information related to shipping traffic in port areas and the distribution of this information interactively to ships, pilots, and/or other officials. The main components of the VTS are manned traffic centres, radar tracking systems, communication systems, and data handling systems. Search and rescue service (SAR) tactics, which refer to actions that are taken after an accident has occurred, in order to limit the negative effects of the accident, including the preparation of such actions. Contingency planning tactics, which refer to plans that are developed in case a serious accident occurs. These plans deal with such things as the operational co-ordination of the handling of response equipment. Legislation and administration tactics, which affect to the rules that are intended to regulate traffic behaviour. Such rules are the International Regulations for preventing collisions at sea (COLREGS), the Shipping Traffic Act incorporated for the Dutch Waters, the Shipping Regulations Territorial Waters, and the traffic rules in inland waterways. Impacts As mentioned before, the tactics have a range of impacts. They affect the number or severity of accidents, either reducing or increasing the risk. The accidents themselves have further consequences, such as environmental and economic impacts. Some of the tactics affect routing -- resulting in changes in sailing miles. Six categories of impacts are identified to be included in the analysis. Within each impact category, several impacts were used to measure the performance of the tactics. For safety and some environmental consequences, the impacts from disasters were analysed separately. Table 3.1 lists the categories and specific impacts that were used. The models of SMART have been used successfully to estimate the various impacts.

9 Category Impacts Impacts on safety Number of accidents Internal individual risk over all traffic accidents (probability of a person dying per travelled kilometre) Impacts on the environment Oil spills from accidents (total cumulative amount per year) Chemical spills (probabilities for Ecological Risk Indicator) Amount of oil on coast (total cumulative amount per year) Total number of birds threatened by oil Operational CO 2 emission caused by the use of tactic (total cumulative amount per year) Operational oil spills caused by the use of tactic (total cumulative amount per year) Economic impacts Loss of income to shipping companies (the immobilisation costs during the period of repairing) Repair costs Cleaning costs (for all the accidents) Extra sea miles caused by the use of a tactic: sailing costs Delay costs caused by tactics Impacts from disasters Ferry Ship disaster, or foundering ferry: Individual risk (probability of a person dying per travelled kilometre) Societal risk (probabilities per year of more than 10, 30, and 100 deaths in one disaster) Ferry Oil/Chemical tanker disasters: Individual risk (probability of a person dying per travelled kilometre) Societal risk (probabilities per year of more than 10, 30, and 100 deaths in one disaster) Chemical/gas tanker Ship disaster: Individual risk (the area within which the probability of a person dying is more than one per million) Societal risk (probabilities per year of more than 10, 30, and 100 deaths in one disaster) Oil tanker Ship disaster: Cleaning costs (expected costs per year from a disaster) Large oil spill (probabilities per year of oil spills more than m 3 and m 3 ) Amount of oil on coast Number of birds threatened by oil Ship Platform disaster: Individual risk (probability per year that a person on platform dies) Societal risk (probabilities per year of more than 10 and 30 deaths in one disaster) Ship Harbour entrance: Economic consequences (delay costs) Other impacts Stakeholder acceptance (tactic is practical and effective) Financial costs Investment costs (except for piloting tactics) Operating costs (except for piloting tactics) The number of piloted trips that are affected by a tactic (piloting tactics) Change in man-hours (piloting tactics) Change in sea miles of pilot tenders (piloting tactics)

10 Table 3.1 Impacts Used in the Analysis Results from the Analysis of Tactics A key step in the policy analysis process is identifying possibilities for making changes in the system being studied. Considering the general maritime safety system, there are a wide variety of individual policy options (the so-called tactics) that could be implemented and that might contribute to meeting future goals or targets. The analysis started with a set of 78 tactics. This set was reduced through a series of steps to produce a set of 30 tactics that were analysed in detail. The following criteria were used to identify tactics that would not be analysed in detail: A. The tactic is not a traffic management measure. B. The tactic is part of another tactic. C. The tactic is not internationally feasible within 10 years. D. The tactic is not implementable. E. The effects of the tactic are not able to be measured. F. The tactic s description is vague or unclear. Detailed impacts of the remaining 30 tactics were assessed, and the results were presented in scorecards. The promising tactics were identified among the 30 tactics. A tactic was said to be promising if it was found to lead to an increase in safety (no matter how small) and/or a decrease in monetary costs (financial costs plus economic costs). No tradeoffs between financial costs and safety were made; such tradeoffs are in the policymakers domain. However, the change in the direct economic consequences from accidents was compared that a tactic would produce with its financial costs. A tactic was not considered to be promising if its financial savings were estimated to be less than the resulting increase in economic consequences from accidents. The results presented in this section should be considered screening results. The analysis started with a large number of tactics and produced a small set of promising tactics. The set of promising tactics should now be subjected to more detailed analysis including checks on feasibility, costs, and the underlying assumptions that were used in the analysis. After that, weights can be assigned to the different impacts in order to determine preferences among the tactics KISHON Background On behalf of the Ports and Railways Authorities of Israel (PRA) a Risk Assessment study has been performed to the extension of the Port of Haifa. A special harbour basin is planned for chemical, LPG and oil products in the future Haifa C port extension. However, at present there is an immediate need for two chemical/gas tanker berths. On practical grounds the only available location is along the Kishon harbour entrance channel.

11 Objectives of the study The objectives of the study are to: 1. Provide a professional opinion on the suitability and effectiveness of the proposed layout. 2. Establish for the present and for the future traffic image whether the layout has an acceptable degree of safety with regard to the channel width and manoeuvring requirements. 3. Determine the influence of passing ships on the ships moored along the quays facing the channel (mooring safety). 4. Determine the risk of possible accidents which could occur in the entrance to the Kishon harbour, taking into account the ship sizes and type of cargo. 5. Formulate recommendations regarding operational guidance and restriction measures, if necessary, to avoid or contain such accidents (e.g. limiting size of passing ships, limiting environmental conditions, day/night manoeuvres, special aids to navigation, removal of moored vessels in case of a passage of a very large ship, etc.) Approach Fig.3.2 Kishon port lay-out In order to meet the objectives the following studies have been performed: a real time simulation study, a desk study; a safety assessment study The safety assessment study The safety assessment study results in the number of collisions that can be expected each year. Two types of collision risk are considered e.g. the drifting and the ramming risk. The drifting risk covers the cases where the moored ship, a quay or breakwater is collided by a ship that drifts after an engine/rudder failure. The ramming risk covers the cases where a moored ship is collided after a navigational error. In the Kishon study both the ramming and drifting model are applied. Figure 3.2 shows the layout of Kishon with the present and future berth locations (red). In the bar-diagram, fig. 3.3 the results are presented for different scenarios for the existing (e) and new (n) layout for drifting and ramming. The distribution in collision energy is assessed for each berth location. These figures are used to estimate the potential damage as consequence of the casualties.

12 probability on a collision / year Gadot_1 Gadot_2 Berth_3 Quays Breakwater 1997e_r 1997e_d 2010e_r 2010e_d 2005n_r 2005n_d 2010n_r 2010n_d scenarios (_r=ramming; _d=drifting) Fig. 3.3 Bar-diagram with collision probability per year Conclusions and recommendations The general conclusion was that the collision level in Kishon remains acceptable when a number of conditions are fulfilled. A number of recommendations are formulated that will further reduce the risk level in Kishon. 5. THE FORMAL SAFETY ASSESSMENT FRAMEWORK General The application of manoeuvring simulations is a well established mean throughout the port development industry. Port authorities, port engineers and dredging companies increasingly use the manoeuvring simulation tools to determine nautical safety aspects of port and fairways. To a lesser extent manoeuvring simulations are used to evaluate existing ports and their operational safety level. Often pilot training is logical consequence in a later stage of the earlier executed assessment and design studies. The training itself fits into the managerial safety assessment frame work. The safety management assessment models as described in the first part of the paper are applied on a completely voluntary base in the Netherlands. In the United Kingdom the Department of the Environment, Transport and the Regions is preparing a Port Marine Safety Code covering all marine operations in ports, including those which facilitate the safe use of a harbour by vessels. The code applies to ports of all sizes, irrespective of resources or levels of traffic. The Code aims to help those with these duties to understand and discharge them, and to achieve and maintain the highest standard for safe marine operations within their waters. It sets down a standard to which they will be required in future to account publicity For complying with the Port Marine Safety Code a full risk assessment and safety management system is required. The safety management system should be informed by and based upon the Formal Safety Assessment method, a documented, structured and systematic process comprising the identification and analysis of risks an assessment of these risks against an appropriate standard of acceptability a cost-benefit assessment of risk reducing measures where appropriate A trial application has been described in [3].

13 Manoeuvring simulations The generic shipping model as used within the formal safety assessment gives a categorisation of system influences e.g. technical, managerial and environmental, which defines a framework for further (strategic) safety analysis. The environmental influences and their (operational) safety consequences for shipping into ports can be evaluated by the execution of full-mission manoeuvring simulations, albeit for the individual vessel or a few single traffic situations rather than for the aggregated total numbers of vessels calling at a port. In addition the real-time full mission simulations allow the incorporation of a relevant part of the human element into the safety assessment of the environmental influences, the incorporation of the full extent of regulatory and even some of the social environment influences. The latter two require the use of local pilots, tug masters and VTS operators to man the simulators. It should be beared in mind that in the first place the levels of skills and knowledge of rules of the human are put to a test during the assessment of the inherent safety of a port environment during full mission simulations and for a minor part the lack of persistence of these qualities during day to day operation. However back-up safety systems, as further addressed throughout the formal safety assessment, like for instance VTS-involvement, good bridge resource management training including tug masters and back up aids to navigation systems, should be in place to cover these short-comings and reduce the contribution to the total level of risk. In addition to the environmental influences a considerable number of managerial influences and technical influences can be assessed. Many of the fleet operations, port operations (admission policy), regulations concerning tug usage, emergency response through tug-escorting requirements and salvage measures, resource management for instance through requirements for pilot, tug masters, VTS-personnel training and education. It could be said that whereas Formal Safety Assessment applies general knowledge on safety per category of influences where often (port) specific knowledge is lacking, whereas full-mission simulation addresses directly the operationally relevant (high risk) and port specific risk levels. The full-mission simulation result is in the first place a qualitative safety assessment although with a high degree of fidelity, accuracy and relevance in the port under investigation and giving an safety assessment on the best practice circumstances. SMART All steps identified in the Formal Safety Assessment approach are also performed in the POLSSS-study, only the nomenclature is different. The main difference between the two approaches is the way in which the impact of a tactic (Called Risk Control Measure in FSA) is assessed: - The risk assessment models used in POLSSS are more based on physical processes, in which the historical data are used for the assessment of parameters. This approach resulted in casualty models that can be used for prediction of casualties under new circumstances or in different areas. The lay out, traffic intensities, traffic composition, environment is part of the input for the risk assessment models. It makes the models applicable in a wide range. - On the other hand in the Formal Safety Assessment the Regulatory Impact Diagram 1 is used for the assessment of the result of Risk Control Measures. This Regulatory Impact Diagram describes much more risk influencing aspects of the regulatory type than modelled in POLSSS. However the impact of a Risk Control Measure is assessed by the change of the index given to the total or a selected part of the Regulatory Impact Diagram and an arbitrary chosen relationship between the risk level and this index. In particular the latter is a mere 1 The Regulatory Impact Diagram defines all the dependencies (index), their weight and their influence, between the factors governing the probability of a type of accident happening in a certain context.

14 guess and not based on any evidence relating the actual measure with known effects on probability levels. Conclusions Manoeuvring simulations as a safety management tool 1. The application of manoeuvring simulations to assess safety will keep its own position in the safety assessment process. In the port (extension) design process it will be an extremely useful tool to address design parameters of paramount importance to the final safety and economy of the port. In the status quo safety assessment of a port it can be a mean to identify high risk areas related to the operational use of the port and a managerial measure if applied as a training tool for pilots, tug masters and VTS operators. 2. The application of manoeuvring simulation tools will be primarily qualitatively although the accuracy and fidelity, through the involvement of many highly committed people, and the high relevance make its application from an early stage on, worthwhile. SMART & FSA 3. The application of the FSA methodology in port safety assessment is increasing and recognised as a valuable tool to identify the risk determining factors. However the quantification of the actual risk level and the consequences of measures still requires considerable input from analysed accident databases, preferably locally based. 4. The SMART tool is based upon a lot of analysed accident data and a maximum of physical modelling, which allows the most direct quantitative connection between the occurrence of accidents and known physical processes within the vessel traffic flows. 5. The combination of both approaches, SMART and FSA, will result in improved risk assessment models, based on more physical relationships and in which also the impact of regulatory aspects are modelled. Because the outcome of the risk assessment model is directly used in the decision process whether a Risk Control measure is effectiveness or not, it is very important to improve this model where possible. References [1] W.E. Walker, M. Pöyhönen E (RAND Europe), C. van der Tak, J.H. de Jong (MARIN) POLSSS-Policy for Sea Shipping Safety: Executive Summary December 1998 [2] Port Marine Safety Code Draft version of the Ports Division, department of the Environment, transport and the Regions [3] Formal Safety Assessment Trial Application to high speed passenger catamaran vessels MSC 68/14/2