OMAE DOWNTIME ANALYSIS TECHNIQUES FOR COMPLEX OFFSHORE AND DREDGING OPERATIONS

Transcription

1 Proceedings of OMAE04 23rd International Conference on Offshore Mechanics and Arctic Engineering June 20-25, 2004, Vancouver, British Columbia, Canada OMAE DOWNTIME ANALYSIS TECHNIQUES FOR COMPLEX OFFSHORE AND DREDGING OPERATIONS Remmelt J. van der Wal MARIN Wageningen, the Netherlands Gerrit de Boer MARIN Wageningen, the Netherlands ABSTRACT Offshore operations in open seas may be seriously affected by the weather. This can lead to a downtime during these operations. The question whether an offshore structure or dredger is able to operate in wind, waves and current is defined as workability. The two different methods are compared and discussed for a simplified offloading operation from a Catenary Anchor Leg Mooring (CALM) buoy. The differences between the methods will be presented and recommendations for further applications are given. In recent decades improvements have been made in the hydrodynamic modelling of offshore structures and dredgers. However, the coupling of these hydrodynamic models with methods to analyse the actual workability for a given offshore operation is less developed. The present paper focuses on techniques to determine the workability (or downtime) in an accurate manner. Two different methods of determining the downtime are described in the paper. The first method is widely used in the industry: prediction of downtime on basis of wave scatter diagrams. The second method is less common but results in a much more reliable downtime estimate: determination of the 'job duration' on basis of scenario simulations. The analysis using wave scatter diagrams is simple: the downtime is expressed as a percentage of the time (occurrences) that a certain operation can not be carried out. This method can also be used for a combination of operations however using this approach does not take into account critical events. This can lead to a significant underprediction of the downtime. For the determination of the downtime on basis of scenario simulations long term seastate time records are used. By checking for each subsequent time step which operational mode is applicable and if this mode can be carried out the workability is determined. Past events and weather forecast are taken into account. INTRODUCTION By nature operations offshore are exposed to the marine environment. This means that operations may be hampered by wind, waves and current or even may be suspended. The prediction of this 'Wait -on-weather Downtime', or its more optimistic conjugate 'Workability', can be important with respect to the total costs and duration of the operation. The total duration of an offshore operation, the gross duration, can be determined by correcting the ideal duration (i.e. duration of the operation in case no downtime is present) with a 'workability percentage'. Delay factors other than the environmental conditions may also increase the project duration, but these are no subject of this paper and left out of consideration. Which environmental factors have the most impact on the workability depends on their respective magnitudes, the 1 Copyright 2004by ASME

2 response of a floater or structure and the ability to deal with this response. Section 2 will mention some typical effects that wind, waves and current may have on typical offshore operations. In order to quantify the workability for a certain project execution period, the workability prediction should address the following is sues: and F(P)SO, is governed by phenomena such as fishtailing instabilities, resonant low frequency motions in wave groups and interactions with respect to wave, wind and current loading on the different vessels. In extreme conditions this can result in exceedance of the maximum allowable hawser loads, too large relative angles between shuttle tanker and F(P)SO or difficulties in mooring the shuttle tanker at the location. i) Assessment of the environmental conditions at the project location. This is still a challenging issue and subject to considerable research. It will however not be addressed in this paper. ii) iii) Determination of the operational limits of offshore equipment, in view of project requirements and operating methods. In last decades major improvements have been made in the (hydro-)dynamic analysis of marine structures, among them dredgers. Recent improvements for this purpose have been made for dredging equipme nt: 'Dredsim2000'. Combination of the environmental conditions and operational limits into a workability, with due account of the limitations and opportunities within the project. In this paper two essentially different methods to predict workability can be defined: - Directly based on scatter diagrams, further referred to as 'Scatter diagram approach'. - Project scenario analysis based on long term seastate time records, referred to as 'Scenario approach'. A simple example of an offshore operation will be used to illustrate the differences between the above two prediction methods and their practical consequences. In practice the operation will be more complex, however for the present example the difference between both approached are clearly identified which is the main goal of the present paper. Under full appreciation of the effects that wind, waves and current have in reality, this paper will focus on workability determined by waves only. Both prediction methods can however also be used for workability limited by any combination of environmental conditions. FACTORS LIMITING THE OPERABILITY Each type / design of offshore equipment has its own particularities with regard to workability due to wind, waves and current. This section gives a few examples of typical offshore operations and activities causing downtime. Tandem/side-by-side offloading Offshore offloading operations take place in many locations around the world. Presently a great number of new deepwater developments are under consideration, not only for the offloading of oil but also offloading of LNG from LNG FPSOs. The behaviour of a combined system of shuttle tanker Trailing Suction Hopper Dredger (TSHD) Due to its hinging suction tube, a TSHD is relatively tolerable to wave induced motions, provided that the draghead lifting wire is heave compensated. Still a number of aspects deserve attention, like: - Deployment and recovery of the suction tube from a rolling vessel may involve the risk of the draghead damaging the hull. - Heave of the draghead gantry may exceed the heave compensator stroke, causing slack lifting wire and subsequent snap loads. - Connecting the floating discharge hose to the dredger's bow coupling. Some of the limiting factors can be made less harmful by changing working method or improvement of sub-systems. Furthermore not all factors illustrated above can be clearly measured and given clear operational criteria for the crew to decide upon. Therefore the specific skills and experience of the key crewmembers still remains a major factor in the final workability that can be achieved. Again, improvements in work method and subsystems combined with an experienced crew can improve the workability, but still the limitations are evident. DESCRIPTION OF DREDSIM2000 A great number of the workability phenomena addressed in the previous section, can be analysed by accepted (hydro) dynamic ship motion analysis programs. Dredsim2000 was a joint industry effort to integrate and customise numerical programs into a complete workability assessment tool for use by the dredging industry. The tool was developed specially for the dredging industry and typical dredging operations. The set-up of the tool is such that the hydrodynamic programs and downtime analysis programs are modelled separately. The tool can be used for other than dredging operations (e.g. offshore operations) by replacing the specific hydrodynamic response models for dredgers with 2 Copyright 2004by ASME

3 existing hydrodynamic response models for offshore structures like moored tankers or buoys. The present paper focuses on the differences between the scatter diagram approach and the scenario approach. The methods to determine the response of structures in wind, waves and current will not be discussed. This section briefly addresses the program and will be followed by descriptions of the Scatter diagram approach and Scenario approach which are applicable for offshore operations in general and the differences between both approaches. Joint Industry Project In 1998 the following dredging companies together with Maritime Research Institute Netherlands (Marin), launched a Joint Industry Project to investigate the feasibility of developing a computer program to predict the workability in a structured and integral manner, based on state-of-the-art analysis techniques. The participating companies were: - Ballast Nedam Baggeren BV* - Baggermaatschappij Boskalis BV - Hollandse Aanneming Maatschappij BV* - IHC Holland NV - Tideway Marine and Offshore BV - Van Oord ACZ BV* *) Now merged to Van Oord During the feasibility study and subsequent project execution phase the 'Ve reniging van waterbouwers in Bagger-, Kust- en Oeverwerken' (VBKO) acted as Marin s contract partner, on behalf of herself and above participants. These parties together jointly funded the project. Governmental support was obtained through Senter, the Dutch governmental agency responsible for the execution of grant schemes for developments in the field of technology, energy, environment, exports and international partnerships. Global set-up Dredsim2000 is basically an integrated suite of computer programs in the field of wave loading, dynamic response and operability prediction. The hydrodynamic programs were already existent within Marin, the workability programs were defined in close consultation with the participating dredging companies. Dredger Motion Characteristics Fig. 1: Global set-up Motions/loads Criteria and typical project information Downtime/workability Environmental conditions Waves, wind, current The tool is capable of calculating the workability of a dredger in wind, waves and current. Fig. 1 gives the global setup of the program. The dredger types which are incorporated in the program are: - Cutter Suction Dredger (CSD), - Trailing Suction Hopper Dredger (TSHD) - Plain Suction Dredger (PSD) - Stone Dumping Vessel (SDV) Dredsim2000 distinguishes the various operational modes specifically for each type of dredger: dredging, stand-by, sailing, (dis-)connecting, discharging and survival, etc. For each mode the operational limits can be specified or calculated and their effect in the total workability can be quantified. Response analysis The response of the offshore equipment is calculated for a number of relevant single seastates. The single seastates are described by: - Wave direction relative to the vessel. - Seastate characteristics which are generally described by a wave spectrum. - Wind and current speeds and directions. Further the response analysis requires: - The (hydro-)dynamic properties of the vessel. - Loads or other effects resulting from the specific process (like draghead force or swing wire loads). Dredsim2000 implements two different response analysis methods: - Frequency domain analysis. A free floating vessel responds practically linear with the wave height. - Time domain simulations in case of strongly nonlinear dynamic systems. For both types of analysis the results are mainly expressed as statistically extreme responses for the subject irregular seastate. It is evident that these analyses are already a challenge on their own. They will however not be further detailed in this paper. Instead reference is made to previous publications [5] and []. OPERATIONAL LIMITS In general the operational limits of operational specific modes are expressed by 'downtime lines': the maximum allowable wave height, as function of primarily wave direction and period. Seastates higher than the downtime lines are considered not-workable for the subject operational mode, as they result in responses that exceed maximum allowable response values (the 'criteria'). Downtime lines are either defined by the user or computed from the extreme responses for a suitable grid of single seastate analyses. In the latter case the user has to specify the limiting criteria (e.g. the maximum allowable load in a hoisting wire). In principle there can be several downtime lines, each of them governed by its own criteria. Workability prediction based on the Scatter diagram approach Wave climates are normally given as wave scatter diagrams: the joint probability of occurrence of specific Hs Tz 3 Copyright 2004by ASME

4 combinations. Often these diagrams are given for different wave directions and seasons. Fig. 2 illustrates how only the area enclosed by all the downtime lines can be marked as workable. Significanr Wave Height Hs [m] Zero upcrossing period Tz [s] DTL_ DTL_1 0 DTL_ 'workable' Fig. 2: Determination of workability by scatter diagram approach. The sum of all workable entries (shaded area) is finally the workability for the subject operational mode. This workability may still depend on wave direction and operating season. The Scatter diagram approach is simple and gives a quick overview of the operational limits relative to the prevailing weather conditions. It should be noted however that each operational mode will have its own workability. The question is how the respective operational modes of a given operation contribute to the gross project duration. Here we encounter a major limitation of the Scatter diagram approach, as will be illustrated by the following example. Assume an operation offshore in wave conditions where only two operational modes are considered: - An offloading mode that can be sustained up to Hs = 4 m, in this example corresponding with workability of 70 %. For Hs > 4 m a floating discharge hose and hawser must be disconnected and the vessel will have to wait on better conditions. - Re-connecting the hawser and discharge hose is the other mode. This is a weather sensitive mode that can only be accomplished for Hs <2 m, now corresponding with 25 % workability. Connecting takes only 2 hours. The workab ility of the total project will be somewhere in the range %, but based on the Scatter diagram approach it is principally not possible to be more precise: - The number of periods with Hs > 4 m determines how often must be re-connected. - The time after a bad weather period before Hs < 2 m determines how long must be waited before re -connecting can commence. - The probability of a period with Hs < 2 m being longer than 2 hours determines how often re-connecting can actually be completed. Particularly in situations where there is a large difference in the operational limits of subsequent modes, the Scatter diagram approach is in principle unable to quantify the gross duration and one has to resort to a more advanced method: the Scenario approach. This will be illustrated as well in the project example further in this paper Workability prediction based on the Scenario approach A job scenario analysis simply follows a long term seastate time record and determines for each time step which operational mode is applicable and whether downtime occurs. Fig. 3 illustrates how the operational modes will follow a certain sequence, the so-called job scenario. Hs [m] offloading (connected) 1. disconnect waiting (to connect) 2. connect Fig. 3: Example of job scenario. Time [days] offloading (connected) criterion offloading -> disconnect criterion connect -> offloading Fig. 3 shows that the status of an offloading combination, in a certain time step, is determined by the operational limits and the past weather conditions. When the offloading hose has to be disconnected because of the increasing seastates (point 1 in Fig 3), a re-connection has to be accomplished before actual offloading can be resumed (point 2). This requires lower wave heights and longer waiting time than when disconnection was not necessary (seastates smaller than the disconnection criterion). Depending on the weather forecast, the operator or captain could even decide to sail away to the harbour when he expects that the weather will get worse. Often the decisions based on these considerations are still made by arbitrary judgment. By repeating a job scenario a sufficient number of times, decisions can be made more rationally. The input required for a scenario analysis is: - The sequence of operational modes to facilitate the process of offshore activities. For the types of dredgers covered in Dredsim2000 these have already been implemented as logic decision trees. These can be easily replaced with decision trees which are more focussed on other offshore activities such as offloading. - The downtime lines for each operational mode of the total operation. These downtime lines can be directly specified by the user or can be calculated based on specified criteria. - Seastate time records that: apply for the project location and the appropriate seasons are sufficiently long for the gross duration of the dredging project 4 Copyright 2004by ASME

5 are available for a sufficient number of years to produce sufficient job scenario analysis, thus a statistically stable prediction of the gross project duration Discussion on the use of the numerical tools As mentioned before, the response prediction of dredging equipment or other offshore floaters like buoys and F(P)SO s/tankers is largely based on existing programs. It should be noted however, that an appropriate response assessment still requires expertise to select realistic input values and interpret the results in a correct manner. In case the vessel response which is used for the workability analysis is not correct the calculated workability will be incorrect as well. It is evident that the same is valid for the applied weather conditions: inaccuracies in the weather conditions will result in inaccuracies in the calculated workability. between the harbour (unloading) and the CALM buoy (offloading). For the present case a flow rate of 20,000 barrels per day is used. Which in this case results in an offloading time through the export tanker of 30 hours. The present case will focus on the project downtime in case of 300 hours of offloading at the CALM buoy. This means that the export tanker will make in total ten offloading cycles. This may not be realistic however for the present example it is used as a starting point and all results will be related to this. The operation starts with the export tanker sailing from the harbour to the CALM buoy. After offloading at the buoy, the tanker returns to the harbour and will start unloading. This cycle is repeated ten times. For the present case, the Base case, the offloading operation starts on June 1 st. Fig. 4 s hows the complete cycle. In essence the determination of downtime lines is a simple process and it is tempting to see it as a black box. In practice the results can, for instance, be dominated by a single phenomenon, which, when judged in a slightly different way, may lead to a considerably different evaluation. The user should always have a critical attitude to the results and / or the criteria used as input. Mode 4 Mode 3 Mode 1 Mode 2 CALM-buoy The traditional prediction of workability based on wave scatter diagrams gives a quick overview of the operability of a dredging spread in certain project conditions. In many cases they will provide a sufficiently accurate prediction of the downtime. Besides, these assessments take only little time. Mode 5,,7,8 waves However, in projects where downtime can be significant and a single, though short, operation may cause considerable waiting time, a more advanced scenario analysis will also give a more realistic prediction of the gross project duration. For a long time, seastate time records over a sufficient number of years, were scarce. These time records are a pre-requisite for job scenario analysis. Developments in wave hindcasting based on meteorological models have however continued, enabling the generation of these time records [8]. Also the emergence of satellite based wave measurements has given access to historical wave information. Together with improvements in the models that transform offshore waves also to nearshore waves it is now becoming more feasible to acquire realistic long term historical seastate information for any offshore / nearshore project location. PROJECT EXAMPLE CASE The following (fictive) project is meant as an example case to show in particular the effect of the used workability prediction method. It is evident that an actual project will be more comple x, however, principles are the same and both approaches will be fully comparable. Project description The project describes an offloading operation of an export tanker at a CALM buoy offshore. The tanker commutes Fig.4: Overview of operational modes (Base case) Operational limitations A summary of the operational modes and limits for the present example are shown in Fig. 4 and Table 1. The presented downtime lines (operational limits) are just assumed here, but normally they result from a preceding analysis or are based on experience records. Mode Duration Downtime line [max. Hs] 1 Sail harbour CALM buoy 45 hr. 2. Sail CALM buoy harb our 90 hr. 3. Unloading (in harbour) 30 hr. 4. Wait in harbour for sailing n.a. 5. Wait at CALM buoy for reconnection n.a. Connect hr. 2.0 m 7 Disconnect 1 hr. 4.0 m 8 Offloading (at CALM-buoy) 30 hr. 4.0 m Duration of ideal cycle 202 hr. Table 1: Overview of operational modes and limits (Base case) In case no downtime occurs, the ideal duration of 300 hours offloading (ten cycles) is equal to 2020 hours. 5 Copyright 2004by ASME

6 Environmental conditions The basic data apply for the offloading location (CALM-buoy) and reflects a 20 year time history trace, consisting of significant wave height Hs and zero -up crossing period Tz. For simplicity the effect of wave direction is neglected. The weather forecast is assumed 100 % reliable over each next 48 hrs. The following figures give an impression of the wave conditions. Hs [m] Hs mean Hs,mean - 2* StDev. Hs,mean + 2* StDev. Operational considerations An operator is typically confronted with questions like: - What is the expected (gross) project duration or what will be duration of a single cycle and what kind of variation can be expected? - What is the effect of the seasonal variation during the year? - What are the critical modes for the present operation and how can the gross project duration be reduced? For example, what is the effect on the duration of a cycle time in case the connection of the hawser / offloading line can be done up to 2.5 m instead of 2.0 m? The following section will discuss how these questions can be answered by respectively the scatter diagram and the scenario approach. Results and discussion Month Fig. 5: Seasonal variation (monthly average Hs and 95 % confidence interval). All years combined / All seasons combined / All directions combined Zero upcrossing Tz [s] Total Total [%] Significanr Wave Height Hs [m] Fig. : Year round Hs Tz scatter diagram based on the average of the 20 years. Hs [m] Mean Hs June, per year Average Hs in June (over 20 yr) Mean Hs October, per year Average Hs in October (over 20 yr) Year [-] Fig. 7: Variation of monthly average Hs over the subsequent years for months June and October. The downtime predictions using wave scatter diagrams are based on monthly scatter diagrams. This section will discuss the Base case (see Table 1) and a few variations: Case: Case description: 1 Base Case (see Table 1) 2 Base Case, start date is August 1 st (instead of June 1 st ) 3 Base Case, Connection possible up to Hs = 2.5 m 4 Base Case, Connection possible up to Hs = 1.5 m 5 Base Case, Offloading possible up to Hs = 5.0 m Base Case, Offloading possible up to Hs = 3.0 m Table 2: Overview of cases The following sections will discuss the various results obtained with the two approaches. Tables 3a and 3b already give an overview of the key results. Overview results Scenario approach Case Duration [hours] Ideal Mean Minimum Maximu m Table 3a: Overview of results from calculations using Scenario approach Overview results Scatter diagram approach Duration [hours] Case Mean duration month Ideal June August October Table 3b: Overview of results from calculations using Scenario approach Copyright 2004by ASME

7 Comparis on of scatter diagram and scenario approach Table 4 gives the expected gross durations of the various modes for the envisaged execution periods based on the Scatter diagram approach. Note that the scatter diagrams were derived from the same time records as used for the Scenario approach. Thus the subject wave climate is exactly the same. The results in Table 4 fit in the trends of Fig. 5. Mode Total Ideal duration June August October Table 4: Consumed (gross) duration of Base Case using Scatter diagram approach The 20 year of wave data allowed performing 20 job scenario analyses. Each job consists of 10 cycles; see also Fig. 4 and Table 1. Fig. 8 illustrates a sample of the scenario analysis for the present case. The waiting time before a re -connection can be made is clearly presented, as are other steps in the operational sequence. The scenario analysis of a part icular tear continues until the total offloading hours equals the required 300 hr (ten cycles). The next year the same scenario analysis is started. The following typical events can be distinguished (see also Fig. 8): 1: After sailing from harbour and the tanker arrives at the buoy the hawser/offloading hose are connected. Offloading starts. 2: The wave height exceeds the maximum allowable wave height for offloading and the vessel is disconnected. 3: At these points (indicated 3 ) the seastate is lower than the offloading downtime line so offloading could be resumed. However since the connection mode needs to be done first offloading is not possible (vessel will wait for lower seastates). 4: The wave height is equal or lower than the connection downtime line so the hawser and offloading line are connected again. Offloading is resumed until the required 30 hours of offloading is completed. Hs [m] Sailing offloading (connected) waiting (to connect) offloading (connected) 2. disconnect criterion offloading -> disconnect 4. connect - Stochastic variation over the years is considerable. Three out of the twenty analyses result in durations that are actually less than the gross duration based on the Scatter diagram approach (2042 hr, see Table 3b). These are the 'good luck seasons' where the wave conditions tend to be more favourable than in average. - Unfortunately there are also 'bad luck seasons'. In one case the gross duration is even 300 hours more than the mean job duration. - In average however the scenario approach yields a more conservative (longer) project duration than the scatter diagram approach: for the base case 2132 hr vs hr. Duration Job durations Average job duration - Base case Ideal duration (no downtime) Job no. [-] Fig. 9: Results of 20 job scenario analyses (Base Case). Effect of seasonal variation In order to check the effect of the seasonal variation on the total job duration the start date of the jobs is changed. For the base case the start date is June 1 st which is the mildest season of the year (see also Fig. 5). When the offloading operation takes place later on during the year the weather will be less favourable and longer duration can be expected. For this example August 1st was chosen. Fig. 10 compares the results for start date August 1 st with the results of the Base Case (start June 1 st ): - The average gross project duration is now 2323 hr for the Scenario approach against 2053 hr for the Scatter diagram approach, so the difference between the two approaches is now increased considerably. - As now more re-connections are required there is now only one job where the scenario analysis gives an equal or shorter duration than the scatter diagram approach. - The stochastic variation, presented by the standard deviation of the job durations, has also increased considerably. The duration of the worst case increased:500 hours more than the mean duration. criterion connect -> offloading 1. connect Fig 8: Sample Scenario analysis Time [days] Fig 9 summarises the scenario results for the Base Case: - The ideal duration line (no downtime) added for reference. 7 Copyright 2004by ASME

8 Start operation on June 1st ; mean duration=2132 hrs Start operation August 1st ; mean duration=2323 hrs Ideal (no downtime) Average job duration (June 1st) Average job duration (August 1st) Base Case Hs = 4.0 m Hs offloading = 5.0 m Hs offloading = 3.0 m total 20 jobs Ideal duration is 2020 hr Duration Frequency [-] Job No. [-] Fig. 10: Effect of seasonal variation date using the scenario approach (basic scope only). Effect of system improvements Duration of job Fig. 12: Effect of offloading limit (basic scope only) Comments on the practical application of the Scenario approach Just to demonstrate the value of the Scenario approach, the following system improvements are investigated: The connecting limit is varied by +/- 0.5 m. The effect is plotted in Fig. 11, which shows a histogram of the gross durations of the twenty simulations, and summarised in Tables 3a and 3b (cases 3, 4). For the base case, a reduction of the connecting limit from 2.0 to 1.5 m increases the mean job duration by 112 hours and the variability even more: a clear trend to longer durations can be observed (the maximum duration is now 2551 hr). Note that with the Scatter diagram approach these effects are not found. A reduction of the offloading limit by 1.0 m (see Fig. 12 and cases 5, in Tables 3a and 3b) has a comparable effect in this case. Frequency [-] Improvement of connecting and offloading limits with 0.5 m leads to roughly 40 hours reduction in mean job duration. For start date August 1 st this would be more. The Scenario approach could be helpful for the identification of critical modes and can support in deciding what part of the design needs improvement in order to reduce downtime. Base Case Hs connect=2.0 m Hs Connect = 1.5 m Hs Connect = 2.5 m total 20 jobs Ideal duration is 2020 hr The examples of the previous sections, and as summarised in Tables 3a and 3b, are just illustrations of the value of the Scenario approach. Its value is most apparent in cases where the downtime is relatively large and / or some operational modes can hold up the entire dredging process. For an accurate prediction of the workability, realistic operational limits and reliable local wave data are crucial. This applies however also for the Scatter diagram approach. The major additional requirement that the Scenario approach imposes is the need for wave history time traces. Still, even in the case of perfectly executed scenario analyses with sufficient number of job scenario analyses, the actual realisation can be far off the prediction. This is an unavoidable 'fact of life', but at least we now have an impression of this variation and can use it in settling the contract. The scatter diagram approach gives no insight in these variations at all. This brings us to the convergence of the gross duration prediction based on the scatter diagram approach. In the cases addressed above as much as 20 years were available, this means 20 realisations of the stochastic seastate process. In practice this will often be less. The question is how many realisations are required for a reliable prediction based on the scenario approach. A theoretical treatment of this matter is beyond the scope of this paper, but some feeling can be obtained by plotting the average duration for an increasing number of job realisations. See Fig. 13. The order of the realisations will have some influence, but is maintained as per original data set Duration of job Fig. 11: Effect of variation connecting limit (basic scope only) 8 Copyright 2004by ASME

9 Duration Base Case (start June 1st) mean Base Case (start June 1st) Mean Start August 1st mean Start August latter gives a more conservative prediction (longer project duration), though more realistic because of its due inclusion of critical modes such as connecting a hawser and the related waiting time and inclusion of weather forecast. Furthermore the scenario approach gives much more insight in operational issues like: - stochastic variation of project durations - 'bottlenecks' in the operation - effects of operational strategies and system improvements Number of jobs used for average duration [-] Fig 13: Convergence of mean project duration. For the present case -10 job realisations seem to be sufficient to have a reasonable estimate of the mean duration. SUMMARY AND CONCLUSIONS The numerical tool Dredsim2000 is the result of a joint industry development to make the workability (or downtime) prediction of common dredging projects more structured and integral. By replacing the specific hydrodynamic response models with other offshore type response models as well as the logic decision trees it can be used for more offshore related operations. It implements existing and new methods for the determination of operational limits and the prediction of downtime, using site specific environmental data as input. The workability prediction based on scatter diagrams is simple and gives a quick overview of the operational limits relative to the prevailing environmental conditions. Many offshore operations have operational modes with a short duration and have clearly lower operational limits than other modes. Although the duration of these critical modes is short, they may hold-up the entire process, causing a considerable increase in downtime. Such effects cannot be quantified by the Scatter diagram approach, so that job scenario analyses need to be performed. Pre-requisites for the Scenario approach are the definition of the sequential relation between the various modes and sufficiently long seastate time records so that a project can be analysed for a sufficient number of job realisations. The first requirement can easily be fulfilled: for typical offshore operation such as tandem offloading or for the present case offloading at a CALM buoy the sequential relation between the various modes can be identified. The availability of representative long term seastate time records becomes increasingly accessible for the industry, due to developments in metocean engineering. For the present example only waves are used. In reality this will not be the case since wind and current play a role as well. However, both workability prediction methods can be used for offshore operations in combined wind, waves and current since the principle of the downtime analysis will be the same. In case wind and current will be involved as well the response of the vessel or buoy and the corresponding downtime lines will have to be determined for these conditions and, depending on the situation, resulting in a higher or lower workability. The more accurate prediction of downtime and the increased insight in above kind of effects may in many cases be well worth the effort that the Scenario approach requires. ACKNOWLEDGEMENTS The authors wish to thank the organisations that participated in Dredsim2000. REFERENCES [1] van Doorn, J.T.M. and Buchner, B. (2001) Design and operational evaluation of offloading operations for deep water FPSOs, DOT, Rio de Janeiro. [2] Feikema, G.J. (1995) Assessment of the Downtime of Dredges in Waves, WODCON, Amsterdam. [3] Grundlehner, G.J., van der Wal, R.J. and de Boer, G. Downtime Analysis Methods for Offshore Dredging Operations (2003), CEDA, Amsterdam. [4] Van den Bos, L. (1998) Bepaling van de werkbaarheid van materieel offshore, Master Thesis, Delft University of Technology, Delft (in Dutch). [5] Wichers, J.E.W. (1980) On the Forces on a Cutter Suction Dredger in Waves, WODCON IX, Vancouver [] Wichers, J.E.W. (1981) Hydrodynamic Behavior Of Cutter Suction Dredges in Waves, World Dredging and Marine Construction, November 1981 [7] Wichers, J.E.W. (1987) Handbook of Coastal and Ocean Engineering, Volume 3, Chapter 7. The behavior of dredging equipment operating in waves, ISBN [8] Wichers, J.E.W. and E.J.Claessens (2000) Prediction Downtime of Dredges Operating in the open sea. WEDA, Rhode Island A simple example describing a CALM buoy offloading case is evaluated by the Scatter diagram and Scenario approaches. The 9 Copyright 2004by ASME