6
Design, Construction, Operation, and Maintenance of Double-Hull Vessels

The performance of double-hull tank vessels with respect to such matters as structural integrity, safety, and prevention of oil spills in the event of accidents has been a subject of investigation for more that 20 years. This chapter begins with a discussion of the results of the committee-commissioned study of hull designs (see Appendix K),1 notably oil outflow and ship stability characteristics. Next, the chapter reviews industry experience in design, construction, operation, and maintenance of double-hull tank vessels. The chapter concludes with a discussion of current design issues, including the need for revised design standards for double-hull vessels and the need for research on improved design tools. Additional information about research on double-hull vessel technology since 1990 is provided in Appendix L.

Comparative Analysis of Double-Hull and Single-Hull Designs

Because few large double-hull tankers had been built before 1990, the promulgation of Oil Pollution Act of 1990 (P.L. 101-380) (OPA 90) and MARPOL Regulations 13F and 13G (MARPOL 13F and 13G) confronted naval architects with new design issues; existing national and international design regulations had been developed with single-hull tankers in mind. This new challenge stimulated creativity in the design process, as illustrated by the varied hull arrangements of

1  

The comparative study of double-hull and single-hull designs was performed under subcontract to the committee by Herbert Engineering Corporation, whose president is committee member Keith Michel.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 115
--> 6 Design, Construction, Operation, and Maintenance of Double-Hull Vessels The performance of double-hull tank vessels with respect to such matters as structural integrity, safety, and prevention of oil spills in the event of accidents has been a subject of investigation for more that 20 years. This chapter begins with a discussion of the results of the committee-commissioned study of hull designs (see Appendix K),1 notably oil outflow and ship stability characteristics. Next, the chapter reviews industry experience in design, construction, operation, and maintenance of double-hull tank vessels. The chapter concludes with a discussion of current design issues, including the need for revised design standards for double-hull vessels and the need for research on improved design tools. Additional information about research on double-hull vessel technology since 1990 is provided in Appendix L. Comparative Analysis of Double-Hull and Single-Hull Designs Because few large double-hull tankers had been built before 1990, the promulgation of Oil Pollution Act of 1990 (P.L. 101-380) (OPA 90) and MARPOL Regulations 13F and 13G (MARPOL 13F and 13G) confronted naval architects with new design issues; existing national and international design regulations had been developed with single-hull tankers in mind. This new challenge stimulated creativity in the design process, as illustrated by the varied hull arrangements of 1   The comparative study of double-hull and single-hull designs was performed under subcontract to the committee by Herbert Engineering Corporation, whose president is committee member Keith Michel.

OCR for page 115
--> the double-hull tankers constructed since the passage of OPA 90. Some of these designs, however, do not provide the high levels of environmental protection that can be achieved with double-hull vessels. The comparative study of single-hull and double-hull vessels commissioned by the committee investigated the following: the effectiveness of hull design in reducing potential oil outflow following collisions and groundings ship stability as indicated by survivability characteristics after experiencing a collision and intact stability during load and discharge operations ship structural integrity as reflected by ballast condition, hull girder strength, and draft considerations when in ballast Oil outflow, survivability, intact stability, ballast draft, and strength were evaluated for 27 tankers that either have been delivered or are under contract. Oil outflow and survivability calculations were also made for nine barges. Double-Hull Tank Arrangements The arrangement of tank vessel cargo tanks and ballast tanks has a major influence on a vessel's effectiveness in reducing oil outflow after an accident as well as its damage and intact stability. In particular, the subdivision of cargo and ballast tanks by centerline bulkheads can have important implications for oil outflow in the event of a collision or a grounding. Box 6-1 shows the three most common cargo tank arrangements. Nearly all double-hull tankers exceeding 200,000 deadweight tons (DWT)—very large cargo carriers (VLCCs)—built to date have cargo tanks arranged three across. The cargo tanks on double-hull tankers of less than 160,000 DWT are usually arranged in ''single-tank-across" or "two-tank-across" configurations. Approximately 60 percent of these vessels have single-tank-across cargo tanks in all or part of the cargo block.2 All tankers exceeding 120,000 DWT delivered in the last three years have oil-tight longitudinal bulkheads subdividing the cargo tanks. This is partly because of concerns regarding the outflow and stability characteristics of single-tank-across tankers and partly because of economic considerations. Suezmax tankers (about 150,000 DWT) that do not have oil-tight centerline bulkheads require a large number of transverse bulkheads to satisfy MARPOL regulations for tank size and hypothetical outflow. As a result, construction costs for single-tank-across and two-tank-across double-hull tankers of approximately 2   Data on cargo and ballast tank arrangements are from a compilation by Exxon Company International of configurations for 327 double-hull tankers comprising more than 95 percent of the world double-hull tanker fleet greater than 5,000 GT (gross tons). The compilation was derived from the Oil Companies International Marine Forum ship information questionnaires provided to Exxon by shipowners and from Exxon's internal inspection records.

OCR for page 115
--> 150,000 DWT are comparable. In contrast, many Aframax and Panamax tankers continue to be built with single-tank-across cargo tank arrangements. For tankers of less than 110,000 DWT, fewer transverse bulkheads are required within the cargo block, and the cost savings realized with the single-tank-across arrangement are more significant. Box 6-2 shows typical ballast tank arrangements. The L tank is by far the most common configuration; it is found in 88 percent of double-hull tankers. Ten percent of the tankers have a combination of U, L, and S types, and 2 percent have a U design only. Evaluating Oil Outflow The International Maritime Organization (IMO, 1995) guidelines for evaluating alternatives to double-hull tankers were used to assess the relative oil outflow of different designs. Although intended for evaluating the outflow performance of alternative arrangements to the double-hull concept, these guidelines are also well suited to comparing the outflow performance of single-hull and double-hull tankers. The guidelines take a probabilistic approach based on historical data from collisions and groundings. (Other sources of oil spillage, such as explosions and operational discharges, are not included in the analysis.) The IMO guidelines account for such factors as varying wing tank widths and double-bottom heights, internal tank subdivision, and the effects of tide.

OCR for page 115
--> Casualty statistics collected by classification societies were used to develop the expected distribution of side and bottom damage. The damage distribution functions were derived from about 60 tanker casualties involving primarily single-hull vessels. These distribution functions provide information on the expected penetration and the extent and location of damage from collisions and groundings. In the case of a single-hull tanker, if the outer hull is penetrated adjacent to a cargo tank, the cargo tank will be breached and oil will flow out. For a double-hull tanker, outflow will occur only if the extent of penetration is sufficient to extend beyond the protective double-bottom or wing tanks, thereby piercing the inner hull and penetrating the cargo tank. The size of the spill is directly related to the number of cargo tanks breached and their size. The likelihood that a double-hull tanker involved in a collision or grounding will spill oil is therefore largely influenced by the dimensions of the double-bottom and wing tanks. The amount of oil spillage is also impacted by the internal subdivision of the cargo tanks, which dictates tank sizes and the spacing of bulkheads forming tank boundaries. Naturally, larger cargo tanks will spill more oil. On the other hand, more closely spaced bulkheads increase the likelihood that more than one cargo tank will be damaged. The quantitative results of the outflow analysis should be used with care because of the limited size of the casualty database, the nature of the incidents included,3 and some simplifications in the calculation procedure. Nonetheless,

OCR for page 115
--> the IMO methodology provides a rational basis for comparing tanker designs and, in the view of the committee, is currently the best readily available analytical approach. Three outflow parameters were calculated: (1) the probability of zero outflow, (2) mean outflow, and (3) extreme outflow. (The mean and extreme outflow parameters measure volume rather than rate of outflow.) The probability of zero outflow is the likelihood that a collision or grounding will result in no oil spillage and indicates the effectiveness of a design in preventing oil spills. Mean (or expected) outflow is the weighted average of the cumulative oil outflow values for expected damage events and indicates the effectiveness of a design in mitigating the loss of oil due to collision or grounding. Extreme outflow is the weighted average of the cumulative outflow values for the most severe damage events and indicates the effectiveness of a design in reducing the number and size of large spills. The outflow parameters for tankers evaluated in the comparative study are shown in Figures 6-1 through 6-4. Data points are plotted for each of the single-hull, double-sided, and double-hull tankers evaluated. To facilitate comparison, curves representing the least-squares fit of the single-hull and double-hull data are shown. Zero Outflow Figure 6-1 shows the probability of zero outflow values for tankers evaluated in the study. The calculations indicated that the probability of zero outflow is four to six times higher for double-hull tankers than for single-hull tankers. In other words, the projected number of spills for double-hull tankers is one-fourth to one-sixth the number of spills projected for single-hull tankers. All cargo oil tanks on a double-hull tanker built to OPA 90 requirements are protected by ballast tanks or other non oil spaces. Thus, many scenarios that would culminate in oil spillage from single-hull tankers do not result in penetration of the cargo tanks of a double-hull tanker. The probability of zero outflow is a function of the double-bottom and wing tank dimensions and is not affected by internal subdivision within the cargo tanks. In other words, centerline or other longitudinal bulkheads within the cargo spaces or ballast tanks have no influence on the probability of zero outflow. 3   The committee recognizes that the probabilistic outflow methodology should ideally reflect the response of specific structural configurations. However, the same damage distributions are currently applied to both single-hull and double-hull vessels. This approach is likely to give conservative results (i.e., overestimates of outflow) when applied to double-hull designs, because recent studies have indicated that double-hull structures reduce the extent of damage from a collision or grounding. In certain cases the inner bottom or longitudinal bulkhead can withstand considerable deformation before being penetrated.

OCR for page 115
--> FIGURE 6-1 Probability of zero outflow for single-hull and double-hull tankers. Source: Herbert Engineering Corporation, 1996. Mean Outflow The mean outflow values for tankers evaluated in the comparative study are plotted in Figure 6-2. The mean outflow values for double-hull vessels are one-third to one-fourth the single-hull values, but mean outflow varies significantly even among double-hull tankers of the same size. Mean outflow is influenced by the double-hull dimensions as well as the extent of internal subdivision. Wider wing tanks and deeper double bottoms tend to reduce the likelihood of a spill, thereby increasing the number of collisions and groundings with no spillage. Hence, an increase in wing tank width and double-bottom depth reduces the mean spill value. Greater internal subdivision also tends to reduce the quantity of oil spilled. The variability in mean outflow values for double-hull tankers is primarily a result of differences in subdivision within the cargo block. Figure 6-3 is a plot of mean outflow, with tankers identified by the extent of longitudinal subdivision. Double-hull tankers without centerline bulkheads have approximately twice the expected outflow of designs with oil-tight centerline bulkheads in way of all cargo tanks. Single-tank-across designs and designs with oil-tight centerline bulkheads were found to have comparable outflow values when the vessel was subjected to bottom damage. However, single-tank-across designs performed less effectively when the vessel was subjected to side damage. The closer spacing of transverse

OCR for page 115
--> FIGURE 6-2 Mean outflow for single-hull and double-hull tankers. Source: Herbert Engineering Corporation, 1996. FIGURE 6-3 Variation in mean outflow with longitudinal subdivision for double-hull tankers. Source: Herbert Engineering Corporation, 1996.

OCR for page 115
--> bulkheads in these designs increases the probability that multiple cargo tanks will be breached. Once a single-tank-across cargo tank is breached, oil located all across the cargo compartment will flow out through the damaged side. Oil outflow is no longer limited to the oil being carried on one side of the vessel. Extreme Outflow Extreme outflow parameters are shown in Figure 6-4. There is considerable scatter in the data points, indicating that such characteristics as internal subdivision and draft-to-depth ratio have a significant impact on extreme outflow. Although the comparative analysis indicated that double hulls are very effective in reducing both the number of spills and the mean outflow values, their effectiveness in preventing large spills is less pronounced. Double-hull vessels with single-tank-across arrangements perform more poorly with regard to extreme outflow than both pre-MARPOL and MARPOL 78 vessels of comparable size. Data points representing two double-side tankers and three double-hull tankers lie above the single-hull tanker trend line in Figure 6-4. All five of these double-hull or double-side designs have single-tank-across arrangements. Despite their poor performance relative to double-hull tankers with one or more longitudinal bulkheads in the cargo tanks, the three double-hull designs with single-tank-across arrangements meet all current U.S. and international regulations. FIGURE 6-4 Extreme outflow for single-hull and double-hull tankers. Source: Herbert Engineering Corporation, 1996.

OCR for page 115
--> FIGURE 6-5 IMO pollution Index E for single-hull and double-hull tankers. Source: Herbert Engineering Corporation, 1996. Combined Outflow Performance The IMO Pollution Prevention Index E provides an overall picture of outflow performance. The three outflow parameters for a given design are combined, using weighting factors, and then compared to the outflow parameters for an IMO reference ship of similar size.4 An Index E greater than or equal to 1.0 indicates equivalency to IMO's reference designs. Figure 6-5 shows the Index E for single-hull and double-hull tankers evaluated in the comparative study. Single-hull tanker values generally fall between 0.3 and 0.4, whereas double-hull tanker values lie between 0.9 and 1.1. Sixty percent (9 of 15) of the double-hull designs have indices greater than 1.0, indicating equivalency to IMO reference ships. In general, ships with longitudinal oil-tight bulkheads in the cargo holds have the highest indices. Outflow Performance of Tank Barges The probability of zero outflow and the mean outflow values for tank barges evaluated in the comparative study are plotted in Figures 6-6 and 6-7, respectively. The results are similar to those for tankers. Double-hull tank barges exhibit substantial superiority in both probability of zero outflow and mean outflow when compared to single-hull tank barges.5 4   Sketches of the IMO reference ships are provided in Appendix K. These reference designs do not incorporate the minimum subdivision acceptable under current MARPOL regulations. They were selected because they exhibit a favorable oil outflow performance. 5   Although tankers in the 5,000 to 25,000 DWT range were not evaluated in this study, the committee believes that the outflow values for double-hull barges would be slightly better than for double-hull tankers. This observation is based on the relatively low freeboards (distances from sea surface to deck) of barge designs, which tend to reduce the outflow from bottom damage.

OCR for page 115
--> FIGURE 6-6 Probability of zero outflow for single-hull and double-hull tank barges. Source: Herbert Engineering Corporation, 1996. FIGURE 6-7 Mean outflow for single-hull and double-hull tank barges. Source: Herbert Engineering Corporation, 1996.

OCR for page 115
--> TABLE 6-1 Survivability Indices for Single-Hull and Double-Hull Tankers   Survivability Index (%) Vessel Capacity (DWT) Single Hull Double Hull 35,000-50,000 95.0 95.9 80,000-100,000 99.6 99.7 135,000-160,000 100.0 99.7 265,000-300,000 99.9 100.0 It is important to remember that this study investigated the relative performance of different designs to mitigate outflow if they experienced a collision or grounding that breached the outer hull. The overall outflow performance must also take account of the likelihood that a given vessel will experience such an accident. Therefore, a comparison of barges and tankers cannot be made on the basis of outflow parameters alone. Ship Stability Survivability Survivability is a measure of a vessel's ability to survive (i.e., not capsize or sink) after sustaining damage to the hull. A probabilistic methodology was used to assess the survivability of single-hull and double-hull tankers. Probabilistic density distribution functions for side damage, as contained in the IMO guidelines,6 were used to determine all possible damage events and their probability. The average survivability indices for each tanker size are given in Table 6-1. There is no discernible difference between the survivability characteristics of single-hull and double-hull tankers. The survivability indices generally exceeded 99 percent for tankers of more than 80,000 DWT. However, two of the ships in the 35,000 to 50,000 DWT range had values of 87.2 percent and 92.5 percent, respectively. These values were more heavily influenced by the degree of compartmentation within the engine room and adjacent spaces than by differences between single-hull and double-hull arrangements. Intact Stability Single-hull tankers are generally stable under all loading conditions. Therefore, stability during operation when no damage has occurred (intact stability) 6   Assessment of survival or nonsurvival is based on damage survival requirements defined in Regulation 25 of Annex I of MARPOL. The index of survivability is developed by summing the probabilities for all damage cases that survive the damage criterion. A survivability index of 97 percent indicates that a vessel is expected to survive collisions breaching the outer hull 97 percent of the time.

OCR for page 115
--> There are no reliable data on the life expectancy of various types of coatings. Most double-hull tankers have been in operation for a relatively short time (less than five years), and no major corrosion problems have been reported on these vessels. Coating failures on relatively new vessels have been attributed to poor workmanship during construction. Some owners of 15- to 20-year-old double-hull tanker fleets have reported significant corrosion in double-hull spaces. Corrosion problems have led to major steel replacement in some double-hull vessels and have contributed to the scrapping of a number of double-hull vessels. Some tanker operators have detected "microbial-influenced corrosion" in the uncoated bottom plating of cargo tanks. Microbes present in the settled water and sludge in cargo tanks are believed to cause accelerated rates of corrosion.15 Although microbes have been present in tank sludge in the past, there is speculation that the higher temperature and other conditions found in some double-hull oil tankers foster explosive growth of the microbe population (Marine Log, 1996). This phenomenon is being studied, but its impact is not yet clear (Huang, 1996). Shipowners have made estimates of coating life. Many expect coatings to last throughout the life of the vessel; others expect coatings to last 10 to 15 years. All survey respondents emphasized the importance of surface preparation and adequate coating thickness in prolonging coating life. Coating replacement is expensive, and a replaced coating is generally less durable than the original. Some owners and operators believe that mandatory requirements for coating application are necessary to prevent owners from specifying low-quality coatings at the time of construction. The classification societies do not inspect coatings to verify application, and it is the owner's responsibility to see that a coating is applied in accordance with the manufacturer's recommendations. Construction Time Survey respondents indicated that the construction time for a double-hull tanker can be greater than that for an equivalent single-hull vessel, depending on ship size. The increase in construction time is related in part to the increased amount of steel needed for the more complex double-hull structure and the increased coating area. As a result of the longer construction time, shipyards, classification societies, and marine architects expect double-hull tankers to be 7 to 15 percent more expensive than equivalent single-hull designs. The committee's economic analysis indicated an increase in capital costs for double-hull tankers of between 9 and 17 percent. Problems encountered in double-hull construction include difficulties in painting narrow double-hull spaces, in providing adequate ventilation of double-hull spaces, and in providing adequate access to double-bottom ballast tanks when the side tanks and double bottom are divided to provide damage stability. It is 15   Corrosion rates of 0.55 to 1.00 mm per year have been reported (Marine Log, 1996).

OCR for page 115
--> anticipated that as shipyards build more double-hull vessels, there will be a reduction in labor hours as a result of optimization of the structural arrangements and details and more extensive use of computer-aided manufacturing processes. The slotting of longitudinals through webs and bulkheads, for example, occurs thousands of times during construction of a double-hull tanker. Developing more producible details that lend themselves to robotic welding is expected to yield substantial savings. Recent developments in this area have been reported by Odense Steel Shipyard in Denmark and Kawasaki Heavy Industries, Ltd., in Japan (Motoi et al., 1995; Tang-Jensen, 1995). Inspection and Maintenance Survey respondents said that frequent inspection and maintenance of coatings were essential for safe and economic operation of double-hull tankers and were also more critical than for single-hull vessels. Apart from this increased inspection requirement, inspection and maintenance practices do not differ significantly between single-hull and double-hull tankers. The structural and coating survey requirements for both single-hull and double-hull tankers have increased since the passage of OPA 90 and MARPOL 13G. In response to these regulations, the U.S. Coast Guard and International Association of Classification Societies (IACS) have established survey requirements. The main financial and operational impact of the hull portion of the Enhanced Survey Program stems from the requirement for a dry-dock survey to complete the special hull survey, provision of access for close-up surveys, periodic evaluation of coatings and increased monitoring of tanks with poor coatings, and thickness measurements during special and intermediate surveys.16 The opinion among those responding to the committee's survey was that more resources are required to maintain double-hull than single-hull vessels. Structural and coating inspections required by an owner are usually performed by the ship's crew. Inspection frequencies for each tank vary from every couple of months to once a year. In addition, some companies have inspections conducted by independent surveyors at two- to five-year intervals. Many of the companies monitor ballast tanks on double-hull tankers more closely and more frequently than those on single-hull tankers. If coating failure is detected, minor repairs are often carried out during inspections. 16   The IACS Enhanced Survey Program includes close-up surveys (visual inspection of the structure at close range), thickness measurements of areas subject to corrosion, and evaluation of the tank coating system. If the surveyor finds the tank coating to be in poor condition, the tank will be inspected annually.

OCR for page 115
--> Operational Safety Intact Stability Intact stability, which has not been a major concern for single-hull tank operators, has become a concern in the operation of some double-hull tankers, as noted in the committee's comparative study discussed earlier. Access Hazards and Explosion Risk in Ballast Spaces The structural arrangement of the double-side and double-bottom tanks of double-hull tankers is usually cellular. Safe access to these spaces is essential to monitor ballast tanks, conduct surveys required by classification societies, and maintain ballast piping. In addition, access may be needed to rescue an injured person from a double bottom in the event of an accident. Opinions varied regarding the accessibility of ballast tanks. Some operators considered access on double-hull vessels to be easier than on single-hull vessels and saw no need for regulations on access. Others reported that access to double-hull spaces was difficult, escape distances in an emergency were long, and design complexity required ship personnel to have good knowledge of tank configuration before entering. These respondents favored mandatory access standards.17 All respondents emphasized the need for adherence to stringent safety procedures by workers when entering tanks. Double-hull designs that have taken accessibility into account have horizontal stringers (or decks) and large longitudinal stiffeners or built-in walkways in side tanks to provide access. Large openings are used in intermediate decks for direct access to other levels; these openings can be a safety hazard unless railings are installed to prevent falls. Some designs have separate rescue hatches in every tank that allow direct access to the main deck in an emergency, and in double-bottom tanks the number of bays without direct access has been minimized. Questions have been raised about the ventilation of cellular double-hull spaces, and the opinions of operators on this issue varied greatly. Some believed that risks associated with a possible lack of adequate ventilation have been over emphasized. Others thought that even after forced ventilation, these spaces might contain pockets lacking oxygen or, in the case of oil leakage, pockets of flammable gases that could cause fires and explosions. The latter group urged installation of 17   IMO has some general requirements for accessibility, inerting, venting, and gas-freeing of cargo and double-hull spaces. These requirements include considerations governing safe access, the capability to supply fresh air to double-hull spaces, the capability to inert double-hull spaces, and the monitoring of other vapor concentrations. These requirements are included in the International Convention on Safety of Life at Sea (SOLAS), Chapter II-1, Regulation 12-2, and Chapter 11-2, Regulation 59.

OCR for page 115
--> fixed gas detection systems in the tanks, although some noted that instrumentation could lead to complacency. Many operators rely on portable fans and hatch openings for ventilation. According to TSCF guidelines, this is not effective for all ballast tanks. Some operators have had dedicated purge pipes installed in the tanks or use ballast pipes to provide air circulation in ballast spaces.18 Many operators consider it important to have the capability of making all ballast and void spaces inert (i.e., reducing oxygen content) by installing emergency connections from the inert gas system to ballast pipes. Others believe that fixed inert gas systems for ballast spaces should be a requirement. The use of ballast pipes to make a tank inert raises concern because inert gas can migrate to other ballast tanks through ballast pipes. Although some owners and operators expressed concern about the lack of mandatory requirements, the respondents generally considered existing guidelines for access, ventilation, and inerting of ballast spaces as defined in TSCF guidelines and the International Safety Guide for Oil Tankers and Terminals to be adequate. Structural Issues Although the structural design concept of a double-hull tanker is essentially the same as that for a single-hull tanker, the structural response of a double-hull tanker exhibits some distinctive features that may warrant special consideration. In particular, a double hull tends to be stiffer than its single-hull counterpart, and this can affect residual stresses induced during construction and local stresses induced by operational loads, both of which may initiate fatigue cracks. There are also concerns over high still-water bending moments, although their impact on ship structure is unclear. The strength design criteria typically employed for tanker hull structures are those set forth in classification society rules. Until very recently, the rules of most societies were semi-empirical, experience-based standards that reflected the experience of ships at sea but could not readily be extrapolated to new designs and new technology. These traditional rules were based on yielding as the primary structural failure mode, with other failure modes (such as buckling and fatigue) accounted for by safety margins in the criteria. It is generally recognized that a more rational and consistent method of establishing criteria is necessary to take account of the dominant structural failure modes of yielding, buckling, and fatigue. Since 1990, some classification societies have therefore taken steps to improve their structural standards by developing new hull structural strength criteria specifically for double-hull tankers and other ship types (see, for example, Chen et al., 1993). 18   Ballast pipes cannot be used for ventilation during ballast transfer operations.

OCR for page 115
--> Salvage The committee was not able to find any new information on the performance of double-hull tankers in salvage situations. Thus, the issues raised in an earlier study on double-hull tankers remain unresolved (NRC, 1991). The salvors interviewed by the committee commented that the large number of ballast tanks in double-hull tankers provides flexibility in a salvage situation. Therefore, salvage of a double-hull tanker may be easier than salvage of a single-hull tanker. However, salvage procedures—and the associated benefits and disadvantages of double-hull spaces—are highly dependent on actual circumstances. Successful salvage of a damaged tanker is critical in minimizing oil outflow. Although the primary emphasis of the tanker industry is on accident prevention, at least one of the tanker companies interviewed has accorded high priority to the salvage capabilities of its tankers. Vessels have been equipped to facilitate salvage operations and have crews trained for salvage situations. Design of Double-Hull Tank Vessels Design Standards The regulations governing tanker design were developed primarily with single-hull vessels in mind, although the stability and strength characteristics of double-hull vessels are quite different from those of the traditional single-hull tanker. Existing and proposed regulations pertaining to oil outflows, intact stability, and survivability of double-hull tankers are summarized in Table 6-4. As demonstrated by the comparative study described in Appendix K, present regulations do not ensure consistently high levels of environmental performance by double-hull tankers. Where practical, IMO is committed to replacing the current deterministic regulations19 with probabilistic-based regulations. Work is under way at IMO to develop a performance-based regulation for evaluating tanker outflow. IMO is also harmonizing its damage stability criteria for all types of ships based on a probabilistic methodology that will eventually include tank vessels and chemical carriers. Performance-based criteria establish a minimum level of performance but do not specify the means of attaining this minimum. Such criteria generally take a probabilistic approach, so that the influence of a given incident on overall design is proportional to its likelihood of occurrence and to its severity or repercussions. 19   An example of a deterministic criterion is the IMO raking bottom damage regulation. This is a damage stability criterion that assumes extensive damage to the bottom shell while the double bottom remains intact. For tankers greater than 75,000 DWT, damage is assumed to begin at the bow and extend aft over 60 percent of the vessel's length. This type of criterion encourages designers to place bulkheads immediately beyond the specified damage extent but does not necessarily lead to optimum designs.

OCR for page 115
--> TABLE 6-4 Existing and Proposed Regulations Relating to Oil Outflow, Intact Stability, and Survivability Performance of Double-Hull Tankers Regulation Requirements, Scope, Status Oil outflow from collisions and groundings MARPOL 13F Establishes minimum dimensions for wing and double-bottom tanks comprising outer hull   Consistent with USCG requirements established in response to OPA 90 Regulations 22-24, Annex I to MARPOL 73/78 Define hypothetical outflow and tank length requirements governing extent of cargo tank subdivision   Regulations 22-24 being revised in light of probabilistic methodology for oil outflow analysis Intact stability of tankers None at present Intact stability to meet criteria recommended by IMO (Resolution A.749(18), 3.1.2.10)a normally exceeded by double-hull tankers through design   Two possible approaches: (I) through design only, and (2) through combination of design and operational procedures   Maritime Safety Committee of IMO addressing issue of intact stability for double-hull tankers. MARPOL Draft Regulation 1/25A calls for assurance of positive intact stability, both in port and at sea, through design onlyb Survivability of tankers Regulation 25, Annex I to MARPOL 73/78 Specifies extent of damage tanker must be able to survive MARPOL 13F Defines raking bottom damage criterion that supplements Regulation 25   Damage stability criteria for all types of ships being harmonized by IMO based on probabilistic methodology aIMO code on intact stability for all types of ships is covered by IMO instruments. bThe Marine Environment Protection Committee of IMO will circulate Draft Regulation 1/25A with a view toward adoption in September 1997.

OCR for page 115
--> For example, current regulations specify minimum wing tank and double-bottom clearances. A performance-based criterion might establish a minimum value for ''probability of zero outflow." Rather than a uniform double hull, a more effective design might have a deeper double bottom located below the forward cargo tank and narrower wing tanks located outboard of the aft cargo tanks. Performance-based criteria are more difficult to develop than the traditional deterministic criteria and are generally more complicated in their application. An assessment is required of both the relative probability of each possible event and the associated risks to the vessel's safety and to the marine environment. Thus, the costs and benefits of ship safety and spill mitigation measures must be understood before effective performance-based criteria can be developed. Nonetheless, properly developed performance-based criteria have many advantages. They give the designer the freedom to optimize a design for minimum construction cost while ensuring that safety and environmental performance standards are met. They are also more adaptable to new concepts. For instance, a performance-based probabilistic outflow criterion would have predicted the poor outflow performance of many of the single-tank-across double-hull tankers. The methodology used to develop performance-based criteria is independent of the required index or performance level, thereby allowing the required level of vessel performance to be readily revised in the light of experience or in response to changes in cost-benefit scenarios. Progress in Design In the years after the promulgation of OPA 90, coordinated research on the performance of double-hull tankers has been pursued at several centers in the United States, Japan, the Netherlands, Denmark, and Norway (see Appendix L). Structural research is proving beneficial in providing improved design tools to incorporate fatigue and structural performance in accident scenarios into double-hull tanker designs. Advances in finite element stress analysis techniques have made it possible to determine accurately the detailed local stresses due to operational loads, which can result in initiation of fatigue cracks in double-hull vessels. For the most part, analyses of this type are now carried out routinely as an integral part of the design process, and the application of fracture mechanics has already paid dividends in improving the fatigue life of ship hulls. Nevertheless, there is potential for further progress. Since V.U. Minorsky's efforts in the late 1950s to correlate the interpenetration of colliding ships using accident data (Minorsky, 1959), there has been continuing research aimed at accounting more accurately for the structural details and approach characteristics of colliding ships. Collision analysis has been greatly aided by modern nonlinear finite element methods, which are now being used to optimize double-hull designs with respect to plate thickness, steel strength, and

OCR for page 115
--> the positioning of inner and outer hull plates, side stringers, and transverse webs. Verification of analytical procedures by means of scale-model tests and actual collision data, where available, is a necessary part of the approach. Some full-scale collision tests have been conducted (Vredeveldt and Wevers, 1992; Vredeveldt and Wevers, 1995; Wevers et al., 1994). Most of the characteristics of structural failure in tanker grounding incidents can be analyzed with the same methods used to analyze failure in collisions. However, hull-girder failure (i.e., "breaking the back" of a tanker) and hull tearing are specific to groundings and require special approaches. Hull-girder failures have been examined with the aid of increasingly powerful numerical models within the last five to six years. Issues studied include the contribution of dynamic effects to hull-girder collapse and the influence of friction between the hull and the seabed. Some scale-model and full-scale grounding tests have been conducted, and the results are in reasonable agreement with mathematical models. Integration of a reliable fracture analysis approach into collision and grounding analyses would constitute a major advance in the analytical tools available to evaluate and design double-hull tankers. The approaches used today ensure that tankers have sufficient strength to withstand the loads encountered in regular operation, but there are no provisions for the loads encountered in accidents.20 Similarly, outflow performance is based on tank subdivision only, and no consideration is given to the performance of the ship's structure in collisions and groundings. The development of tools that can be used to design tanker structures for good performance in accidents will be an important advance. Ongoing research has this objective, but much work is still required before research findings can be translated into practical design tools. Other than work being conducted by the U.S. Navy,21 the research of Wierzbicki and his coworkers at MIT is the main activity in the United States dedicated to the development of advanced analysis methods for ship structures (see Appendix L). Nearly all of Wierzbicki's effort is supported by industry, mainly from abroad. Most of the computer codes used to analyze nonlinear deformation of structures have been developed in this country, and the expertise needed to extend them to include the effects of collapse and fracture also resides in this country. The committee is concerned that important research opportunities may be missed because of the absence of any significant federal funding for this type of work. 20   In-process accident behavior—such as that associated with collisions and groundings—is referred to as "crashworthiness" in the automotive and aviation industries. 21   Commercial tanker designers and operators were not involved in either the planning or the execution of the U.S. Navy's $25 million project on the development of advanced double-hull tanker technology.

OCR for page 115
--> Findings Finding 1. The results of the comparative study commissioned by the committee indicate that—with the exception noted in Finding 2—double-hull tankers perform significantly better than single-hull tankers in preventing oil spills and mitigating oil outflow in the event of a collision or grounding. If a vessel experiences a collision or grounding that penetrates the outer hull, double-hull tankers are four to six times less likely than single-hull tankers to spill oil. Expected or average outflow is three to four times less with a double-hull compared to a single-hull tank vessel. The benefits of fitting double hulls to tank barges are at least as significant as they are for tankers. Finding 2. The committee's analysis indicated that outflow performance is heavily influenced by the extent of subdivision within the cargo tanks of a double-hull tanker. Vessels with single-tank-across cargo tank arrangements (i.e., without longitudinal bulkheads) exhibit inferior performance with regard to both outflow and intact stability compared to other double-hull designs. Approximately half of the existing double-hull tankers of less than 160,000 DWT have single-tank-across configurations. Double-hull tankers with single-tank-across cargo tank arrangements have approximately twice the average outflow of double-hull tankers with an oil-tight centerline bulkhead in the cargo tanks. Double-hull tankers with all single-tank-across cargo tanks perform more poorly than pre-MARPOL and MARPOL single-hull vessels with regard to extreme outflow. All of the single-hull tankers analyzed in the committee's comparative study are inherently stable and do not experience intact stability problems. Four of the nineteen double-hull designs studied, however, may become unstable during loading and discharging operations. All four of these designs have single-tank-across cargo tank configurations. Finding 3. There is no discernible difference between the survivability characteristics of single-hull and double-hull tankers, indicating that the MARPOL 73/78 subdivision requirements, supplemented by the raking bottom damage criterion included in the 1992 Amendments of MARPOL, ensure a high level of survivability. Finding 4. Existing regulations governing the design of double-hull tankers do not ensure that the environmental advantages of double-hull tankers are consistently achieved. Despite their potentially poor performance, the aforementioned single-tank-across designs comply with existing regulations.

OCR for page 115
--> Finding 5. Representatives of the tanker industry generally believe that double-hull tankers can be operated safely, albeit with more resources and attention than needed for single-hull tankers, including increased attention to rigorous inspection and proper maintenance of interior coatings. Operational guidelines for double-hull tankers issued by the Tanker Structure Cooperative Forum in conjunction with the International Association of Classification Societies and the International Safety Guide for Oil Tankers and Terminals are considered adequate by most operators. These guidelines address issues such as access, inspection, maintenance, ventilation, and making ballast spaces inert. Finding 6. Progress has been made in standardizing classification society regulations and in incorporating fatigue assessment and other analytical techniques into structural review of double-hull tanker design. However, some ship operators have expressed concern that minimum rule requirements do not result in a double-hull design that can be operated safely and economically throughout the expected life of a vessel. Finding 7. Progress has been made in applying computational tools to simulate structural performance in accident scenarios. New data from grounding and collision tests have become available and can be used to calibrate analytical methods. Apart from research conducted by the U.S. Navy with limited commercial application, the U.S. government and U.S. industry have funded little research on ship structural responses to collisions and groundings. References Chen, H.H., H.Y. Jan, J.F. Conlon, and D. Liu. 1993. New approach for the design and evaluation of double-hull tanker structures. SNAME Transactions 101:215-245. Goodwin, M.J., J.C. Card, and J.S. Spencer. 1996. Study of double-hull tanker lolling and its prevention. Marine Technology 33(3):183-202. Herbert Engineering Corporation. 1996. Comparative Study of Double-Hull vs. Single-Hull Tankers. Background paper prepared for the Marine Board Committee on Oil Pollution Act of 1990 (Section 4115) Implementation Review. San Francisco, Calif.: Herbert Engineering Corporation. Huang, R.T. 1996. Microbial Influenced Corrosion in Cargo Tanks. Presented at NACE Corrosion 96 T-14B Marine Vessel Corrosion, Denver, Colorado, March 27-28. International Maritime Organization (IMO). 1995. Interim Guidelines for the Approval of Alternative Methods of Design and Construction of Oil Tankers Under Regulation 13F(5) of Annex I of MARPOL 73/78. London: International Maritime Organization. Marine Log. 1996. Are bugs eating some double-hulled tankers? p. 31, October. Minorsky, V.U. 1959. An analysis of ship collisions with reference to protection of nuclear power plants. Journal of Ship Research 3(1): 1-4. Moore, C., J. Neumann, and D. Pippenger. 1996. Intact stability of double-hull tankers. Marine Technology 33(3):167-182.

OCR for page 115
--> Motoi, T., A. Murakami, A. Kohsaka, A. Kada, M. Unno, and T. Taniguchi. 1995. A new structural design of double-hull VLCC. Proceedings of MARIENV 95 1:191-197. National Research Council (NRC). 1991. Tanker Spills: Prevention by Design. Marine Board Washington, D.C.: National Academy Press. Tang-Jensen, P. 1995. Innovative structural design for high quality double-hull VLCCs. Proceedings of MARIENV 95, 1:144-151. Vredeveldt, A.W., and L.J. Wevers. 1992. Full-scale ship collision tests. Pp. 743-769 in Proceedings, First Conference on Marine Safety and Environment Ship Production, Delft, June 1-5, 1992. Delft: Delft University Press. Vredeveldt, A.W., and L.J. Wevers. 1995. Full-scale grounding experiments. Pp. 11-112 in Proceedings of Conference on Predictions Methodology of Tanker Structural Failure and Consequential Oil Spill, Tokyo, April 1995. Tokyo: Association for Structural Improvements of the Shipbuilding Industry in Japan. Wevers, L.J., J. van Vugt, and A.W. Vredeveldt. 1994. Full-scale six degrees of freedom motion measurements of two colliding 80 m long inland waterway tankers. Pp. 923-930 in Proceedings of the 10th International Conference on Experimental Mechanics, Lisbon. June 18-22, Rotterdam: A.A. Baldema.