National Academies Press: OpenBook

Tanker Spills: Prevention by Design (1991)

Chapter: 5 Design Alternatives

« Previous: 4 Engineering Considerations
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

5
Design Alternatives

This chapter will introduce and explore tank vessel designs and operational alternatives that are intended to mitigate pollution in an accident. The committee's technical assessment is based on how the various characteristics of these designs influence safety, operation, and cargo outflow in an accident. The physical phenomena and engineering issues underlying this analysis were discussed in the previous two chapters.

The designs considered in this chapter were chosen from several dozen alternatives gathered or solicited from various sources and, in a few cases, suggested by committee members. Proposals ranged from the conceptual to the tested and operational. Some originated in the United States, others in one of several foreign countries; some were suggested by individuals, others by major international corporations or industry associations. All of the suggestions were reviewed, but not all were included in the formal evaluation. The exclusion criteria are explained in the following paragraphs.

Proposals were screened to eliminate the clearly impractical esoteric visions, as well as those judged beyond the bounds of industry compatibility. The surviving proposals then were rendered into 17 broad examples—designs generalized enough to encompass the key concepts and details.

For this chapter, the alternatives were grouped according to common physical principles. This resulted in the following three categories: secondary ''barriers" to oil intermingling with water; the mitigation of pollution via "outflow management" techniques; and the reduction of pollution potential through "increased penetration resistance." Operational options for "accident response" were grouped separately; these options can be employed

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

with most hull design alternatives. An itemized matrix listing all of these alternatives and operational options can be found in the next section of this chapter.

The committee evaluated the 16 alternatives based on the following criteria: (1) previous studies or documented experience with the design; (2) concerns about engineering, safety, and practicality derived from the experience of committee members; and (3) theoretical effectiveness of the design in preventing/mitigating pollution in collisions and groundings. As part of its analysis, the committee judged each design according to its developmental status; those not ready for immediate use were evaluated in terms of future promise.

The first half of the chapter covers general technical aspects of each design. Based on these considerations, some of the alternatives were eliminated from further committee consideration. The second half of the chapter mathematically assesses the pollution-mitigation potential of the remaining alternatives, and several possible combinations of these designs, in regard to groundings and collisions. For this section, the committee made use of a study developed by Det norske Veritas (DnV), contained in Appendix F. (This study will be referred to as the DnV analysis.) The methodology is explained in relevant sections of that report.

The designs and combinations assessed for pollution-mitigation potential also are subjected to cost-effectiveness analysis, described in Chapter 6.

THE MATRIX

The matrix presented on the following pages was prepared by the committee to combine, in one document, all of the technical considerations discussed in Chapters 3, 4 and 5. The matrix is intended to apply to both tankers and barges. The following description is intended to assist in understanding of the document.

Column 1 (Alternative Description) lists the design alternative considered in that row. Column 2 (Effectiveness) indicates, in a general sense, the type of accident in which the proposed alternative will be effective, in terms of controlling cargo outflow. If an alternative is effective in the accident type noted, then a dot (•) appears in the proper column. The terms "high" and "low" damage severity reflect the relative speed of the vessel prior to the accident. For collisions, speed refers to the ramming vessel; effectiveness applies to the vessel that is struck. The matrix does not grade the relative effectiveness of the various designs; that subject is taken up in the latter half of this chapter.

In Column 3 (Implementation), the committee has indicated technology status and technical constraints. Regulatory and financial constraints are

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

not considered in the first two sections. The last three sections of Column 3 indicate whether the alternative, in the committee's view: (1) could be applied to new tankers at the present time (within the next two years); (2) should be considered at some time in the future; or (3) could be applied (retrofitted) to existing vessels. A dot (•) in the column indicates applicability, and a question mark (?) indicates technical applicability but economic difficulties. A blank space (no dot) indicates no applicability, from a technical standpoint. Retrofitting involves some unique complications, which will be discussed in the text as applicable.

In Column 4 (Concerns), the headings follow the format of Chapter 4 through the subject of explosions. A dot (•) in the appropriate section indicates that the design alternative, to some extent, creates that concern. A general explanation of these concerns can be found in Chapter 4.

The last four sections refer to the following concerns:

  • Safety Downgrade—A dot in this section means that, to some extent, existing safety practices or requirements are diminished. (An example is an alternative that prevents the use of inert gas systems.) In all cases, safety downgrade probably could be overcome, but the committee felt it important to note the concern so that alternatives would not be viewed as panaceas.

  • Operations Complexity—Because crew size has been reduced and crew fatigue has played a role in some pollution incidents, the committee felt that alternatives requiring extra vigilance in operation should be identified.

  • Design Integration—A dot indicates alternatives involving either new technology or new design practices. This concern is highlighted to ensure that the advantages afforded by an alternative are not offset by some additional problem.

  • Rules and Regulation—A dot indicates alternatives requiring interpretation or revision of existing rules.

The last section of Column 4, entitled "Comments," is intended to summarize the major technical or operational concerns to give the reader a sense of the priorities.

Column 5 (Pivotal Argument) is the committee's summary of all preceding columns; the comments represent the major arguments for and against each design alternative. These arguments are restricted to technical and operational matters.

Finally, Column 6 (Warrant Committee's Economic Evaluation) indicates which alternatives will be pursued in the benefit/cost assessment (Chapter 6). Those alternatives requiring significant time and/or research and development to implement, and those lacking sufficient information for a benefit/cost assessment, are indicated by dots in the "No" section.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

TANK VESSEL DESIGN ALTERNATIVES

1. Alternative Description

2. Effectiveness

3. Implementation

CONTROL METHOD:

Grounding

Collision

TECHNOLOGY STATUS (YEARS TO DEVELOP)

Constraints

Applicability

Barrier

Damage Severity

Damage Severity

CONCEPT

RESEARCH

NEW CONSTRUCTION

RE-FITS

HIGH

LOW

HIGH

LOW

DEVELOPMENT

EXISTING

NOW

FUTURE

1. Protectively Located Segregated Ballast; (Marpol Tanker); Ballast tanks isolated from cargo tanks. Located to restrict possible outflow. This is current regulation.

 

 

Existing

No constraints - present standard

2. Double bottom; A non-cargo space between the cargo tank bottom plating and the ship's hull bottom plating.

 

Existing

Structural and weight- complications w/refit

?

3. Double Sides; A non-cargo space between the cargo tank side plating and the plating of the ship's hull.

 

Existing

Structural and weight complications w/refit

?

4. Double Hull; A non—cargo space between the cargo tank and the hull.

Existing

Structural and weight-complications w/refit

?

5. Resilient Membrane; A tough, pliable, nonstructural barrier separating the cargo from the ship's structure and acting to maintain separation of cargo and water in the event of being breached.

 

 

Concept - minimum 10 years to develop

Technology not sufficiently developed to support even a 'proof of concept' case.

 

 

 

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

4. Concerns

5. Pivotal Argument

6. Warrants Committee Economic Assessment

STRENGTH

MAINTENANCE

STABILITY

SALVAGEABILITY

EXPLOSION

SAFETY DOWNGRADE

OPERATIONS COMPLEX

DESIGN INTEGRATION

RULES & REGULATIONS

Comments

For

Against

YES

NO

 

 

 

 

 

 

Damaged stability concerns if ballast tank is ruptured.

Has eliminated a major cause of world's oil pollution (ballasting of oil tanks). Inexpensive and in existence.

Does not effectively handle damage conditions. Will not reduce oil outflow unless only ballast tank is ruptured.

 

 

 

 

 

 

Explosion concerns due to gas in voids.

Will prevent pollution in groundings where inner hull is not breached. It can be effective in limiting or preventing pollution in case of low to moderate damage.

Does not assist in collision damage. Possible increased pollution potential due to collision because salt water ballast (SWB) in D.B. will reduce amount of protectively located segregated ballast. Explosion concerns due to gas in voids.

 

 

 

 

 

 

Explosion concerns due to gas in voids.

Will prevent pollution outflow in collisions where inner hull is not breached. It can be effective in limiting or preventing damage in low to moderate damage cases.

Does not assist in grounding damage. Loss of buoyancy and resultant heel may cause increased ground reaction. Increased structural maintenance Explosion concerns due to gas in voids.

 

 

 

 

 

 

Design depth of double hull could provide structural and capacity restraints.

Will prevent pollution or mitigate extent of immediate pollution in the event of all but the most severe accidents.

Increased structural maintenance Explosion concerns due to gas in voids.

 

 

 

 

No technical support for the concept Required materials, operations, design demands not investigated.

Simplicity of concept. If problem of integrating membrane into ship's structure can be overcome, this solution could be quite beneficial.

The practical limitations of material properties. Impact of operational functions and the behavior of resilient membranes in contact with complex shapes and arrangements are virtually unexplored.

 

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

1. Alternative Description

2. Effectiveness

3. Implementation

CONTROL METHOD: Outflow Management

Grounding

Collision

TECHNOLOGY STATUS (YEARS TO DEVELOP)

Constraints

Applicability

Damage Severity

Damage Severity

CONCEPT

RESEARCH

NEW CONSTRUCTION

RE-FITS

HIGH

LOW

HIGH

LOW

DEVELOPMENT

EXISTING

NOW

FUTURE

6. Passive Control Hydrostatically Balanced Loading Concept; The establishment of the potential oil/water hydrostatic equilibrium at a height above the highest designed, or incurred for, or incurred point of damage.;

 

 

 

 

 

 

 

 

 

 

Would limit (via loading criteria) the cargo's head pressure at tank bottom to equal to or less than draft's water head at tank bottom such that bottom damage results in less oil outflow.

 

 

Existing.

No modifications required.

7. Intermediate Oil Tight Deck; A structural deck running the full length of the cargo area about 1/4 to 1/2 the depth above the bottom.

 

 

 

 

 

 

 

 

 

 

7a. Independent Tanks; The top and bottom tanks would be independent of each other. Would require upper and lower cargo piping.

 

Development — minimum 2 years to incorporate.

Design of piping system and strength of int. deck. Outflow performance uncertain—testing needed.

 

?

7b. Convertible Tanks; The top and bottom tank would be integrated through sluice valves.

 

Development — minimum 2 years to implement.

Need to develop and test system.

 

?

8a. Mechanically Driven Vacuum; The ullage space is subject to a mechanically induced vacuum such that the combined vacuum plus cargo head favors water inflow in the event of bottom damage.

 

 

 

Development — minimum 4-5 years to implement in full-size ship tests.

Need to prove practicality of providing tight deck and fail safe valving.

 

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

4. Concerns

5. Pivotal Argument

6. Warrants Committee Economic Assessment

STRENGTH

MAINTENANCE

STABILITY

SALVAGEABILITY

EXPLOSION

SAFETY DOWNGRADE

OPERATIONS COMPLEX

DESIGN INTEGRATION

RULES & REGULATIONS

Comments

For

Against

YES

NO

 

 

 

 

Operating with all slack tanks could create a free surface problem.

Mitigates pollution in the first few hours after the accident; instantaneously on-line. Low cost option to retrofit.

Reduces deadweight of vessel which increases cost to transport and requires more vessels. Given sufficient time after accident and no other response, the pollution effect would be the same as a vessel carrying the same amount of cargo. Vessel may be subject to high sloshing loads, which is an especially critical problem in retrofit application. May be difficult to administer operationally•

 

 

 

 

 

Magnitude of piping produces operational concerns.

Would mitigate pollution in high-energy accidents except for the most catastrophic ones.

Complex. Cracks in intermediate deck, or open cargo valves would void the function of the intermediate deck.

 

 

 

 

 

 

Magnitude of valving produces operational concerns.

Would mitigate pollution in all but the most catastrophic accidents.

Operationally burdensome and complex. Hydrostatic isolation of upper and lower tanks could be in jeopardy.

 

 

Major operations hazard. Existing deck structure insufficient. Requires constantly maintained tight ship.

Mitigates pollution outflow from groundings.

Adversely affects vessel and personnel safety. Generates serious operational problems (i.e. vapor disposal, maintenance & reliability).

 

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

1. Alternative Description

2. Effectiveness

3. Implementation

CONTROL METHOD: Outflow Management (cont.)

Grounding

Collision

TECHNOLOGY STATUS (YEARS TO DEVELOP)

Constraints

Applicability

Damage Severity

Damage Severity

CONCEPT

RESEARCH

NEW CONSTRUCTION

RE-FITS

HIGH

LOW

HIGH

LOW

DEVELOPMENT

EXISTING

NOW

FUTURE

8b. Hydrostatically Driven Vacuum (Passive); A vacuum which occurs due to the run out of oil cargo.

 

 

 

Development—minimum 2 years to implement.

Need to prove effectiveness and environmental safety of chemicals.

 

8c. Imaginary Double Bottom; A passive vacuum system coupled with a water layer below the oil cargo.

 

 

 

Development—minimum 2 years to implement.

Need to prove practicality of providing tight deck and fail safe valving.

 

9. Smaller Tanks; Increase compartmentalization to reduce oil outflow exposure.

 

 

Existing

 

 

9a. Service Tank Location; Position all oil service and oil waste tanks clear of the vessel's hull in a defensive location relative to hull damage.

Existing.

 

 

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

TANK VESSEL DESIGN ALTERNATIVES

4. Concerns

5. Pivotal Argument

6. Warrants Committee Economic Assessment

STRENGTH

MAINTENANCE

STABILITY

SALVAGEABILITY

EXPLOSION

SAFETY DOWNGRADE

OPERATIONS COMPLEX

DESIGN INTEGRATION

RULES & REGULATIONS

Comments

For

Against

YES

NO

 

Major operations hazard. Existing deck structure insufficient. Requires constantly maintained tight ship.

Mitigate pollution outflow for groundings.

Adversely affects vessel and personnel safety. Generates serious operational problems (i.e. vapor disposal, maintenance & reliability).

 

 

Major operations hazard. Existing deck structure insufficient. Requires constantly maintained tight ship.

Mitigate pollution outflow for groundings.

Adversely affects vessel and personnel safety. Generates serious operational problems (i.e. vapor disposal, maintenance & reliability).

 

 

 

 

 

 

 

Increased maintenance and piping.

Smaller tanks reduce pollution for a specified amount of hull damage. Effective for smaller vessels and barges.

Major capital and operating expense. Will allow slightly less pollution than existing PL/SBT in event of major hull penetration.

 

 

 

 

 

 

 

 

No negative impact on existing protocols.

Positive pollution preventing requirement easily achievable on new buildings.

Inefficient use of space. Only addresses minor pollution source of ship's outflow.

 

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

1. Alternative Description

2. Effectiveness

3. Implementation

CONTROL METHOD: Increased Penetration Resistance

Grounding

Collision

TECHNOLOGY STATUS (YEARS TO DEVELOP)

Constraints

Applicability

Damage Severity

Damage Severity

CONCEPT

RESEARCH

NEW CONSTRUCTION

RE-FITS

HIGH

LOW

HIGH

LOW

DEVELOPMENT

EXISTING

NOW

FUTURE

10. Internal Deflecting Hull; An internal, forward, inner lower and bottom tank structure of exceptional strength similar to existing icebreaker hulls; Conventional outer hull would combine to form partial double hull. Upon grounding the vessel rides up and the inner hull deflects the vessel away from the obstacle. Existing SBT is retained.

 

Development—minimum 2 years to develop.

Need to develop structure and perform model tests.

 

 

11. Grinding Bow; Forebody bottom structure to be double bottom with internal transverse structure designed to act in a grounding as a "rock rasp". Thereby grinding down underwater obstacles and allowing safe override by remaining structure.

 

Concept—minimum 10 years to develop.

Need to develop & test both materials and structure.

 

 

12. Unidirectionally Stiffened Bottom Structure; Bottom structure designed to have "crushable" transverses in combination with longitudinal structure such that bottom structure moves (in the event of grounding) as a unified pliable member.

 

Research—minimum 3 years to develop.

Existing crushable structures would need to be reviewed and expanded. Need to model test structure.

 

 

13. Honeycomb Hull Structure; Utilize high energy absorbing deep honeycomb steel structure, sandwiched between steel plates.

 

Research—minimum 5 years to develop.

Need to develop structure and perform model tests

 

 

14. High Yield Steel Bottom Structure; Construct bottom plating and structure using high yield type steel in conjunction with design stress limitations associated with mild steel.

 

 

Existing.

 

 

15. Concrete; Construct hull using reinforced concrete internally molded to outer steel cargo tank structure.

 

 

Concept—minimum 10 years to develop.

Need to examine methods of utilizing concrete for this application.

 

 

16. Ceramics; Place ceramic coating on the exterior of the hull.

 

 

Concept—minimum 10 years to develop.

 

 

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

4. Concerns

5. Pivotal Argument

6. Warrants Committee Economic Assessment

STRENGTH

MAINTENANCE

STABILITY

SALVAGEABILITY

EXPLOSION

SAFETY DOWNGRADE

OPERATIONS COMPLEX

DESIGN INTEGRATION

RULES & REGULATIONS

Comments

For

Against

YES

NO

 

 

 

 

 

 

 

 

Provides both a forward secondary barrier to grounding induced damage as well as acting to divert vessel from continued damage. Provides superior protective support in the area most prone to grounding.

Does not address side damage

Will cause considerable damage if the vessel rams another vessel.

 

 

 

 

 

 

 

Success largely dependent upon strength of rock.

Provides secondary bottom structure which must be breached before cargo is exposed to the sea. In groundings, could provide better protection of cargo areas.

Practical materials and applications seem unlikely.

 

 

 

 

 

 

 

Could compromise bottom strength in a seaway.

Provides structural support for grounding without losing cargo capacity.

Does not address collision. Research needed to insure structure would protect against plate tearing.

 

 

 

 

Increases safety concerns without improving upon conventional double hull.

Provides secondary structural barrier via high impact absorbing structure. Could be less weight and cost than conventional double hull.

All of the negatives of double hull structure plus a quantum increase in risks associated with uninspected voids adjacent to cargo tanks.

 

 

 

 

 

 

 

 

 

 

Raises the severity of bottom impact that can occur without hull rupture. Should be considered in combination with other options.

Single barrier defense which, once breached, offers no improvement over conventional single hull design. Repair at yards unfamiliar with the material is a concern.

 

 

 

 

 

 

 

Raises the severity of bottom impact that can occur without incurring oil outflow.

Unique torsion and tension properties of concrete do not match those required to manage ship structural dynamics.

 

 

 

 

 

 

Effects of adhering ceramics to the hull.

Raises the severity of bottom impact that can occur without incurring oil outflow. Could conceivably provide a smoother cleaner hull surface.

Torsional and impact properties of ceramics are not understood for this application. Could be cost prohibitive.

 

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

1. Alternative Description

2. Effectiveness

3. Implementation

CONTROL METHOD: Accident Response

Grounding

Collision

TECHNOLOGY STATUS (YEARS TO DEVELOP)

Constraints

Applicability

Damage Severity

Damage Severity

CONCEPT

RESEARCH

NEW CONSTRUCTION

RE-FITS

HIGH

LOW

HIGH

LOW

DEVELOPMENT

EXISTING

NOW

FUTURE

17. Enhanced Information Processing; Incorporation of pipe and structure sensing devices connected to a central processing facility that would advise crew immediately of any off-limit structure and cargo system integrity, and would recommend pollution mitigating action consistent with overall vessel limitations.

 

Concept — minimum 5 years to develop.

Considerable wiring, sensing and control system design and testing.

 

 

18. Towing Fittings; Installation on board vessel of easily access fittings to facilitate emergency hookup to service and tow/tug boats.

 

 

Existing.

 

19. Distressed-Ship Cargo Transfer System; The cargo tanks are fitted with a high suction dedicated auxiliary pipe/valve system that could transfer oil out of the upper part of any damaged tank to undamaged ballast tanks. Transfer would be used to equalize oil/water hydrostatic head above the damaged area.

 

 

 

Development — minimum 2 years to implement.

Need to develop sluice control system.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

4. Concerns

5. Pivotal Argument

6. Warrants Committee Economic Assessment

STRENGTH

MAINTENANCE

STABILITY

SALVAGEABILITY

EXPLOSION

SAFETY DOWNGRADE

OPERATIONS COMPLEX

DESIGN INTEGRATION

RULES & REGULATIONS

Comments

For

Against

YES

NO

 

 

 

 

Improved response time and salvage decision process. Conceptually attractive.

Malfunction/

erroneous signal could worsen situation. For data display to be positive factor would require integrity and reliability improvements. Not currently available for use in marine/oil environment.

 

 

 

 

 

 

 

 

 

 

Hardware addition.

Improves stabilizing and movement of damaged vessel. Easy to implement.

Fittings are not a major issue. Vessel currently fitted at both ends and sides with mooring fittings.

 

 

Safety issues raised by introducing hydrocarbons into noninerted spaces.

Potentially mitigating pollution system offering simplicity, ease of installation, modest capital cost and no deadweight impact with respect to existing tonnage.

Potential in leakage of explosive gases into voids; impacts I.G.S. and tank venting.

High reliability required for rare usage.

 

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

BARRIERS

All of the design concepts in this category act initially to prevent the loss of cargo containment integrity during the sequence of an accident that ruptures the vessel's primary, or outer, hull. Barriers are, in effect, secondary physical obstacles to the loss of cargo containment integrity. They are inevitably passive (requiring no action by personnel or machinery to initiate) and, once brought into play, their performance is both irreversible and uncontrolled other than by accidental events.

A barrier can take the form of an inner hull or a resilient non-structural membrane. The concept, application, and operational characteristics of various types of inner-hull secondary structures are well known. Those protecting against groundings generally are referred to as double bottoms, those guarding against collisions as double sides, and those protecting against both as double hulls (or double skins). For tankers, the voids between the outer hull and inner structure almost always are utilized as ballast tanks, in order to immerse the hull to meet regulatory requirements tied to propulsion and maneuvering. Ballast is less likely to be carried in the voids of barges, as their propulsion source is independent of the vessel.

A double structure may not extend throughout the entire vessel. For example, while a double bottom generally is combined with some double-side tankage, a double-sided vessel seldom features a double bottom because sufficient ballast can be carried in the side tanks to meet regulatory requirements.

Double bottoms, double sides, and double hulls are in use today, albeit to a moderate degree. (Approximately 6001 tankers over 10,000 DWT—or 20 percent of the world fleet of approximately 3,0002 —have some form of double structure.) Their protective advantages must be weighed against increased risk of explosion, higher susceptibility to risks associated with the vagaries of workmanship, and increased construction and operating costs. Engineering issues related to double structures, including strength, stability, and safety, were discussed in Chapter 4.

An alternative to structural barriers is a tough, resilient membrane (or tank liner), a concept akin to the inner tube within a bicycle tire. This idea, in numerous forms, has the advantage of simplicity. Whether protecting against bottom damage or side damage, the principle is the same: The pliable membrane activates and adjusts to fit the structural dislocation caused by the accident. In so doing, it becomes a physical barrier between the oil cargo and the water. The membrane is not now in use, due to a host of practical obstacles that have confounded its inventors and proponents over the years.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

1. Protectively Located Segregated Ballast (PL/SBT or MARPOL) Tanker

This design (see Figure 5-1) is considered the baseline alternative, as it is the current international standard and represents about 15 percent of the current world fleet (10,000 DWT and larger); about 35 percent of the fleet has segregated ballast, either with or without protective location.3 MARPOL design provisions and additional requirements for tankers trading in U.S. waters were described in Chapter 2.

2. Double Bottom

Double bottoms (see Figure 5-2) are intended to provide the maximum protection for all but the highest energy groundings. Even for grievous damage, double bottoms could protect tanks on the periphery of the damage area.4 A double bottom of the MARPOL-required depth (2 meters or B/15, whichever is less) will not hold sufficient ballast to meet regulatory requirements for the ballast voyage, so some side wing tanks are designated for ballast. The arrangement reduces the side hull area protected by ballast tanks, compared to the MARPOL tanker.

FIGURE 5-1 Alternative 1: Protectively located segregated ballast— the PLBST tanker is separated by longitudinal bulkheads into wing and center tanks. Transverse bulkheads separate one set of tanks from another. (MARPOL Tanker.)

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

FIGURE 5-2 Alternative 2: Double bottom—the void space between the cargo tank bottom plating and the bottom hull plating.

Among its attributes, a double bottom provides a very smooth inner cargo-tank surface compared to the baffled and egg-crate-type tank bottom in a conventional tanker. Cargo flow (or any liquid drainage) to the discharge suction is optimized, facilitating tank cleaning. The cargo transfer system could be placed in the double-bottom void to enhance pumpout efficiency; however, such a system, besides being a source of potentially explosive hydrocarbon gas in those empty tanks, most likely would be damaged in a grounding. Therefore, from a damage and outflow standpoint, cargo transfer piping should be placed within the cargo tank. This arrangement also is better suited to the transfer of cargo from a damaged tank.

Drawbacks to a double bottom include increased risks associated with poor workmanship, corrosion, and obstacles to personnel access. Proper construction, inspection, and maintenance are required. Conceptually, any extra void space increases the risk of fire or explosion, although, as noted in Chapter 4, there is no hard evidence of greater incidence. The risk is reduced by the periodic clearing of tanker void spaces with ballast (this may not be the case on barges). This process, however, increases the risk of ballast tank corrosion.

Double bottoms do not increase oil outflow in grounding, although some critics have made this claim. The committee's reasoning was explained in Chapter 4, in the discussion of perceptions about salvage of double-hull tankers.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

Should a double-bottom vessel suffer damage to the outer hull only, seawater would fill the void and the vessel would increase draft, trim, and heel accordingly. However, as noted in Chapter 4, tankers are exceptionally stable compared to other large vessels and, loaded or not, are tolerant of most hull damage.

A double bottom offers one definite advantage in groundings where the obstacle reaches high enough to pierce both the outer and inner hulls: The extra structure may diminish the longitudinal extent of damage by absorbing energy and perhaps will reduce the number of tanks affected.

In the committee's view, the double bottom can prevent outflow in some accidents and deserves further consideration.

Retrofitting Tankers and Ocean-Going Barges

Retrofitting an existing single-skin vessel with a new double bottom would cost an estimated 25 percent of full replacement value.5 Thus, retrofitting a 30-year-old U.S.-built tanker of 40,000 DWT would cost about $15 million, or about 25 percent of its replacement cost. For tankers, additional regulatory costs would be involved as a double bottom would be considered a ''major" conversion; the vessel would have to be upgraded in safety and other areas. These costs, plus renewal of wasted hull steel and machinery upgrades and rebuilds, could bring the total retrofit cost to 40 to 50 percent of replacement cost. Such an investment would have to be analyzed carefully in light of the longevity of the tanker. As tankers normally are depreciated over 20 to 25 years, the owner of a 15-year-old tanker faces the dilemma of the retrofit cost plus the non-depreciated value of his tanker, balanced against the remaining life of the tanker.

Due to their flat profile, barges could be retrofitted with double bottoms at a reasonable cost (15 to 20 percent of replacement cost).

3. Double Sides

Double sides (see Figure 5-3) can be found in some crude oil and product tankers. As with double bottoms, the minimum MARPOL-required width is 2 meters or B/15, whichever is less; this requirement may not provide sufficient ballast capacity. Therefore, because no other void space exists, the double-side tanker employs a greater tank width (B/7 to B/9) so as to meet ballast requirements.

Because the cargo is separated from the outer side hull by a void, the design offers good protection from collisions. Double sides also offer some of the advantages of double bottoms, because the wide side tanks protect as much as 15 to 20 percent of the outboard region of the bottom. This design could be improved, in terms of limiting outflow following a grounding that

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

FIGURE 5-3 Alternative 3: Double sides—the void space formed between the cargo tank side plating and the side hull plating. Note: In Figures 5-3 and 5-4, longitudinal cargo bulkheads have been omitted for clarity.

breached the hull, by the use of hydrostatically balanced loading. This operational alternative is discussed later in this chapter (alternative #6).

The double-side configuration improves cargo operations somewhat, due to the smooth inner sides of the cargo spaces (the complex "structure" is on the ballast side of the bulkheads). The smooth surface reduces cargo clingage, enhances tank coating life, and simplifies tank cleaning and inspection. However, if the cargo bottom is damaged, the cargo transfer system could be rendered inoperative (as its suction opening, located close to the bottom hull, would be under water). In this respect, the double-side tanker is no better than the single-skin vessel.

After incurring side damage, a double-side vessel is highly susceptible to asymmetric flooding. However, as with double bottoms, the extent of settling and heeling depends on several factors, principally ballast tank arrangement. The question of heel is somewhat different in a double-side vessel, in comparison to a double-bottom design, due to the outboard placement and capacity of the ballast tanks. However, as discussed in Chapter 4, double-side tankers can be designed to be stable.

Retrofitting Tankers and Ocean-Going Barges

Retrofitting double sides to an existing vessel, combined with other additional U.S. regulatory requirements, could cost more than 50 percent of the replacement value of the vessel. With the relatively low incidence of colli

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

FIGURE 5-4 Alternative 4: Double hull—the double hull is a void formed between the vessel's outer side and bottom hull plating and the cargo tank plating both at the sides and bottom.

sions in U.S. waters, double sides would appear to be an unattractive retrofit option from both economic and pollution-abatement standpoints.

Retrofit for barges may be more desirable if a detailed assessment of accidents reveals that pollution was caused as a result of side damage. Retrofitting a barge generally is simpler and less costly than for tankers, due to less complex hull and structural formation and absence of requirements for concurrent upgrading of crew and accommodation-related standards.

4. Double Hull

For preventing oil outflow in low-energy collisions or groundings, the double hull is, logically, the most effective design (see Figure 5-4). Some product tankers, combination carriers, and a few crude carriers already have a double hull, which essentially is a combination of a double bottom and double sides. The width of the side tanks may be less than in the doubleside design however, because the required ballast can be divided among the side and bottom spaces. Consequently, the double-hull vessel may have a lower threshold of sustainable side damage than a double-side vessel in a major collision.

From the cargo operations point of view, the double hull is the best of the barrier designs. With inner tank sides smooth, the flow of cargo to the suction inlets is unhindered, side clingage is minimized, tank cleaning is

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

optimized, and cargo tank coatings are easier to maintain. With the cargo transfer system in the cargo tanks, the double-hull design retains cargo transfer capability after an accident, as long as the inner hull is not damaged and the cargo transfer system still works.

At the same time, however, double hulls magnify all the risks noted for double bottoms and double sides related to construction, maintenance, inspection, and safety. The large structural area subject to cracks, for example, puts high demand on access for inspection. All of this implies high standards on the part of designers and builders, and added vigilance on the part of operators, builders and classification societies.

The pollution-control performance of a double hull in some high-energy accident scenarios could be improved by the use of hydrostatically balanced loading. But this compound design alternative would involve additional concerns, as described later in this chapter (in the discussion of alternative #6) and in Chapter 6.

In the committee's view, the double hull would prevent outflow in many accidents, and the related concerns are manageable if the following conditions are met:

  • inspection access;

  • minimum inter-hull spacing requirements;

  • minimum outer-hull plate thickness;

  • exclusion of cargo piping from ballast spaces;

  • constraints on the longitudinal extent of each double hull void;

  • maintained corrosion-prevention systems; and

  • damage-stability requirements that assure residual (post-accident) damage stability consistent with that of single-hull vessels subjected to the same accident scenario.

Retrofitting Tankers and Ocean-Going Barges

Retrofitting double hulls on existing vessels would be possible but very expensive; the previous discussions of double-bottom and double-side retrofit give some indication of the cost and other concerns. In addition, potential difficulties could result from combining new and old structural materials. Replacement of the entire ship forward of the machinery space would entail fewer technical difficulties and would require less time in a shipyard; however, it might be more expensive than retrofitting a full double hull.

5. Resilient Membrane

In this concept, not yet in use, cargo would be separated from the outer hull by a resilient membrane sheeting (see Figure 5-5). In theory, a breach of the outer hull would activate the membrane; during normal passage, the

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

FIGURE 5-5 Alternative 5: Resilient membrane— a pliable, non-structural tank liner.

membrane would be inactive, integrated into the ship's design to allow unhindered operations.

The membrane, in some form or other, has become a familiar concept over the years. Large, inflatable, compliant bags employing various materials have been tested and used on a limited basis for storage and marine transport of petroleum products. But the committee could find no evidence that this concept has been utilized successfully in a cargo tank. Its total absence in the tank vessel industry likely is due to practical obstacles, which have been insurmountable so far.

Internal tank structure and equipment are not physically conducive to the fitting of liners. Piping, pumps, heating coils, washing machinery, ladders, operating rods, and gauging equipment all thwart the freedom of motion required by a liner. Furthermore, operating practices are not compatible with liner characteristics. Cargo pumping requires a free cargo flow into the suction openings; cleaning requirements introduce concentrated hydraulic forces that would tear the membrane; maintenance and repair requires unhindered access to structure and systems; and normal inspections must cover the very areas hidden by an inactive membrane.

Other drawbacks noted in previous evaluations are equally difficult to overcome. Liner strength factors, folding characteristics, fatigue, and abrasion remain major concerns. Furthermore, all these factors suggest that

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

liners are unlikely to perform in the event of fire, explosion, or significant bottom damage. The membrane might be torn easily by the sharp edges of a damaged hull.

Although membranes are not in use today in cargo tanks, they might be suitable in smaller tanks having fewer structural, systems, or operating constraints. Fuel oil or waste oil tanks may be environments that, with modest research and development, would benefit from a pollution-prevention strategy incorporating membrane liners. However, in the committee's view, this concept would not become viable for cargo tanks in the foreseeable future and will not be considered further in this study.

OUTFLOW MANAGEMENT

While barriers would prevent the onset of cargo outflow, "outflow management" restricts the amount of cargo subject to outflow. Outflow management can be passive (smaller tanks) but generally is active, requiring either personnel or machinery to activate. Barriers and outflow management techniques are not mutually exclusive; in fact, an environmentally optimized tanker could draw upon a mixture of features, as noted previously.

Outflow management amounts to either a manipulation of the hydrostatic equilibrium acting on the oil/water interface (explained in Chapter 3), or downsizing or special placement of the cargo volume subject to outflow. In all cases, some initial oil outflow is expected, however slight.

Among the outflow-management concepts promoted by several sources are vacuum systems to retain cargo in a breached tank. These systems reduce atmospheric pressures on cargo surfaces, manipulating oil/water interface hydrostatics. There are two basic schemes. The first is active, employing mechanical systems to maintain low pressure in the ullage space so that, theoretically, little or no oil can escape. The second is semi-active, relying on reduced pressures induced by the falling oil level in a damaged tank. This scheme limits rather than prevents outflow.

6. Hydrostatically Balanced Loading Concept (passive)

This concept involves less-than-full loading of cargo tanks. An accident then would result in the damaged tank's oil/water interface hydrostatics favoring water inflow, or, at least, limiting cargo oil outflow. The physical basis of this concept was explained in Chapter 3. (See Figure 3-7, Case 2.) This condition is achieved by the carriage of ballast simultaneously with cargo and/or the light loading of all cargo tanks.

This concept is unique among the alternatives considered because it could be achieved almost immediately on most ships. Hydrostatically balanced

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

loading (or hydrostatic control) does not require new structural arrangements or major physical changes to most existing ships.

Hydrostatic control involves six major drawbacks. First, it is an operational approach requiring continued compliance in order to be effective. Thus, it is not equivalent to a structural solution in terms of actual pollution reduction. Monitoring would be difficult, although a precedent for monitoring compliance through regulation was set with the LOT system (described in Chapter 2).

Second, cargo would be lost after a hull breach; the amount would depend on tidal variation, sea state, and damage location. Over time, there also is likely to be some intensive mixing of oil and water inside the cargo tank due to hydrodynamic effects, including turbulence in the oil and/or water, surface tension dynamics, and associated stratified flow near the hull rupture opening. Loss of cargo may be aggravated when a moving vessel is penetrated. Little is known about these phenomena under large-scale conditions, but the effect could be to encourage oil outflow. However, there is some evidence from accidents that the amount of oil actually lost closely approximates the loss indicated by calculations based on hydrostatic balance, without accounting for violent ship motion.

Third, there are technical problems related to cargo sloshing and the loss of stability due to the "free surface" effect (discussed in Chapter 4). Baffles and special design of transverse structures can ameliorate sloshing to avoid pitch resonance with ship motions; ballast and cargo distribution for specific classes of ships also may obviate much of the sloshing potential. Similarly, dividing cargo spaces into smaller tanks would reduce free surface effects. The forces resulting from sloshing during light loading may be a significant concern for some existing vessels, particularly if they are weakened by corrosion. Vessels would have to be checked individually; those unable to operate with reduced cargo loads either would have to be restricted from U.S. waters, or would require strengthened or additional bulkheads. Preliminary data (Lloyd's Register of Shipping, 1989) indicate that about 15 percent of the current tanker population would be so affected. New tank vessels and combination carriers, especially ore/bulk/oil (OBO) carriers, also would need to be evaluated, as they may have been built to standards preventing partial loading.

The fourth major drawback is economic: Cargo is traded for ballast. A ship would carry 15 to 20 percent less cargo, depending on its design and on the level of cargo required for hydrostatic control. As a result, up to 15 percent more ships (and/or larger ships) would be required to carry the equivalent cargo, and, as a result, the frequency of collisions and groundings likely would increase.

Fifth, if a cargo tank is breached above the hydrostatic balance level, oil will escape relatively quickly, especially if there were a large hole. In fact,

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

for large holes, the limitation may be the rate at which the ship's cargo venting system will allow air to enter the expanding ullage space. With a hole equivalent to 1 percent of tank area, a tank may lose 20 percent of its cargo in only 3 to 5 minutes. Obviously, that is insufficient time to take physical measures to curb outflow. Setting a potential oil/water interface equilibrium at a height above a probable point of damage can be accomplished by passive or active means, as described in the following design options.

The sixth concern relates to practicalities of tanker operations. Ships frequently have multiple port discharges, and sometimes loadings, particularly where ships are being lightered. If a ship were to ground while partly discharged, the hydrostatic balance might have been disturbed, and, therefore, the beneficial effect of the concept would be lost. The protection against such a circumstance would be to lower the cargo level in all tanks while the ship is being discharged; from the standpoint of cargo handling and logistics, this may be impossible. Also, dropping all tanks to a hydrostatically balanced level in light-loaded ships could induce instability or aggravate sloshing, especially in OBOs.

In the unusual situation where cargo oil density is greater than that of water, hydrostatic control would not be applicable, as explained in Chapter 3.

Hydrostatic control, in the committee's view, offers a means to quickly reduce pollution potential of many existing vessels, and, despite operational limitations, it merits further consideration.

7a. Intermediate Oil-Tight Deck (independent tanks)

Hydrostatics favoring water inflow can be achieved by dividing cargo tanks into separate upper and lower chambers (see Figure 5-6). This greatly reduces the cargo head acting on the lower chamber and the hull bottom plating (the lower chamber's cargo head equates to its height, rather than the full depth of the vessel). In the event of bottom damage, hydrostatics would strongly favor water inflow, at least initially; in addition, the reduced volume of the lower chamber (compared to a full-depth tank) would limit the amount of oil subject to outflow.6 Coupled with a double-side arrangement of ballast tanks, the design also could provide good protection against outflow resulting from collisions.

An intermediate oil-tight deck (IOTD) would reduce slightly the available cargo cubic capacity of a standard MARPOL tanker. Besides the steel weight increase, roughly equivalent to that for double hulls, the need to vent lower tanks and to arrange for personnel access would reduce the volume of cargo spaces. The amount depends on design details. However, the intermediate oil-tight deck results in less loss of cargo capacity than simple hydrostatic control, with potentially much greater protection against oil outflow.

The IOTD, which is not a new concept, has a number of attractive at-

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

FIGURE 5-6 Alternatives 7a and 7b: Intermediate oil tight deck.

tributes, particularly when coupled with double sides (IOTD w/DS). (See Figures 4-3 and 5-6.) The following discussion compares the theoretical performance of the IOTD w/DS with that of other design options, particularly the double hull.

To comply with MARPOL space requirements for segregated ballast tanks, the double side-voids in an IOTD w/DS vessel would be about twice the width of those in a double-hull vessel; therefore, the IOTD w/DS would offer greater protection to cargo tanks in some collision scenarios.

The IOTD w/DS provides significant protection against oil outflow in the event of bottom rupture; the hydrostatic pressure favoring water inflow would be greater than in simple hydrostatic control. Furthermore, the wide side tanks would protect against grounding damage extending to the side of the ship. Overall, in theory, the IOTD w/DS likely would provide less protection than the double hull in low-energy groundings,7 but greater protection in high-energy groundings (when damage penetration would pierce the inner hull).

When the bottom of an IOTD w/DS is breached and seawater enters the lower chamber, the oil, providing it does not have access to the double-side voids, is lifted into the ullage space (as little as 2 percent) and forced up into the vent/access spaces. Hydrostatic balance is quickly established with the seawater only slightly above the tank bottom. In one Japanese 280,000 DWT design (Ministry of Transport-Japan, 1990), the cargo level could be as little as 0.5 meters above the bottom of the ship, making the design subject to possible outflow from vessel list or trim and the dynamics of oil/water mixing. If the ship remained afloat, vessel rolling could cause some leakage. Small-scale tests, however, indicate that the outside pressure of the sea is more than enough to hold all oil in the vessel even with signifi-

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

cant heeling or rolling. A smaller IOTD w/DS design would be subject to oil outflow from a receding tide following a grounding. In all cases, leakage would be significantly less than for simple hydrostatic loading. Nonetheless, an IOTD w/DS design is likely to lose some cargo following a low-energy grounding whereas a double-hull design would not (provided the inner hull remained intact).

An IOTD is more complex operationally than a double-hull tanker or any other design involving permanent structural pollution controls. Operator error when loading could negate the hydrostatic advantages of the IOTD design; cargo transfer systems must be properly controlled and separation of cargo in upper and lower tanks must be ensured.

Structurally, the IOTD design has both advantages and disadvantages. Ballast tanks in an IOTD w/DS have less surface area than in a double-hull, implying less risk of corrosion or other structural problems that could lead to oil leakage into the void/ballast spaces (posing an explosion hazard). The wider side tanks also mean that an IOTD w/DS design should be easier to construct, inspect, and maintain than a double hull. The IOTD w/DS has more cargo tank structure to inspect and maintain than either a single-hull MARPOL tanker or a double-hull tanker, but risks associated with structural deterioration or failure stemming from corrosion may be diminished. This follows from location of the intermediate deck at or about the neutral axis. Hence, the deck structure and its components will be subject to less primary hull bending stress, and will be less prone to failure than vital structure and components of some other designs.

The effect of an undetected leak or structural failure in the intermediate cargo deck would be less hazardous than a leak from a cargo tank to a ballast tank. (The cargo areas separated by the intermediate deck each would be subject to inert gas control and monitored for pressure changes.)

Overall, comparisons of the IOTD w/DS and the double hull based on structural and operational concerns are imprecise.

The principles of the IOTD w/DS design are simple and straightforward, and they do not entail any extreme departure from current shipbuilding practice. However, the concept has many permutations, with more undoubtedly to follow. Much work remains to be done to determine the most desirable proportions and arrangements.

Retrofitting. Conceptually, retrofit of an intermediate deck to an existing MARPOL vessel would require less steel than retrofit of a double bottom. However, a major modification for the cargo handling, ventilation, and cleaning systems would be required. The deck retrofit also could be carried out afloat, whereas adding a double bottom would mean extended drydock time. The deck installation in the wing tanks would require considerable fitting around existing structure, and/or slotting of that structure; this would increase the cost, as would the extra piping, vent and/or truck systems. Thus,

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

the deck retrofit would cost roughly the same (25 percent of replacement value) as would the double-bottom retrofit. Likewise, depending on the age and condition of the base vessel, and interpretations of regulations, the total retrofit cost could approach 40 to 50 percent of the replacement cost of the tanker. An exception would be the retrofit of an intermediate deck within the existing double-side vessel (the deck would only be fitted in the adjacent side, center cargo tanks). This could cost as little as 15 percent of total tanker replacement cost.

7b. Convertible Tanks

Convertible tanks are a variation on the intermediate oil-tight deck. The cargo tanks have an intermediate deck, which contains remote-activated, securable openings connecting the upper and lower chambers. During conventional cargo handling operations, the valves would be open. Once the tanks were loaded, the valves would be closed, thus dividing the cargo.

A bottom accident would expose only the lower chamber to the sea, and, for the same reasons as noted for the intermediate oil-tight deck, hydrostatics would dictate initial water inflow rather than cargo outflow.

This concept poses the same difficulties as the intermediate oil-tight deck and has the additional drawback of extreme dependance on reliability of valve operation, which demands diligent maintenance and repair. For these reasons, in the committee's view, this concept is not worthy of further consideration in the present study.

Summary—Hydrostatic Control Alternatives

Hydrostatic control, while not yet in use, appears to be an effective concept for reducing pollution. Furthermore, hydrostatic control could be implemented immediately on many tankers; barges are more problematic because the system requires attention from the crew, and barges generally are unmanned.

The intermediate oil-tight deck, especially in combination with double sides, is attractive. Some members of the committee consider the IOTD w/DS ready for implementation on new vessels and view it as at least equivalent to the double hull in overall performance. Other committee members, however, consider such findings premature; they recommend deferring a decision pending a thorough evaluation by the IMO. The results of such a study could be used to determine whether the IOTD w/DS design should be accepted as a substitute for the double hull in reference to the mandate of OPA 90.

Retrofitting of the IOTD may be possible for existing double-side tankers. While hydrostatic control is not practical for barges, the IOTD concept

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

could be considered as a means of limiting outflow resulting from groundings; it would be, in essence, a high double bottom.

8a. Mechanically Driven Vacuum (active)

Vacuum systems, as shown in Figure 5-7, work by reducing atmospheric pressures on cargo surfaces, thus manipulating oil/water interface hydrostatics. The mechanically induced vacuum has the benefit of not requiring major structural changes to existing tank vessel designs, provided the deck and bulkhead structure is adequate to handle the stress imposed by the pressure differentials. The system may be triggered either automatically or by the crew. The vacuum is created by securing all openings to the cargo tank(s) while simultaneously withdrawing air from the ullage space. Depending on the cargo type, the vacuum system design characteristics, and the contribution of hydrostatic control (if any), the force restraining oil outflow can be made to vary in accordance with the perceived risks.

The committee is aware of only one instance in which a prototype of this type of equipment was fitted on a tank vessel and tested under controlled conditions. The idea is believed to be fraught with practical concerns, even though the concept and principles are straighforward. It is not a question of whether the system can be made to work; it can, with an appropriate

FIGURE 5-7 Alternative 8a: Mechanically driven vacuum (active).

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

expenditure of funds. The real issue is whether it can be made to function on the one occasion it may be needed, after years in service with an unpredictable quality of maintenance. This system also would require modification of inert gas systems (IGS is discussed in Chapter 3) through a change in international conventions, because this safety system depends on overpressure of cargo tanks. Cargo, vent, and inerting systems would need to be automated, and fitted with elaborate sensing and monitoring instruments. Backup safeguards also would be required to prevent inadvertent activation of the vacuum system (i.e., during cargo discharge), which could rupture tanks.

In one version of this concept, tankers would operate with the vacuum permanently active, so that no crew intervention or automatic activation would be required after an accident. In this design, theoretically, no outflow would result from a grounding. But such a system appears to involve two major problems. First, a system would have to be developed to collect hydrocarbon boil-off vapors, which would be created rapidly in a vacuum. Vapor recovery systems for tankers have been under discussion for approximately 20 years, but they have not yet been implemented successfully to any major degree except in specialized trades (liquified petroleum gas, for example). Second, once a system has been developed, there is the question of vapor disposal. Liquefaction is very expensive, and burning of the vapors requires special modifications of propulsion engines, or boilers.

The committee is not aware that the full range of potential operational problems related to a vacuum has been addressed. Moreover, even with all necessary modifications, the machinery required to effect the vacuum (unless utilizing a stand-alone power source) would require that the ship's main machinery space be functional. That is not assured in the case of significant bottom damage.

In the committee's view, the mechanically driven vacuum system poses significant problems, although research aimed at enhancing the concept should not be discouraged. This alternative will not be considered further in this study.

8b. Hydrostatically Driven Vacuum (passive)

This systems, as shown in Figure 5-8, capitalizes on the physical effects of initial cargo outflow following a grounding. By securing all openings to cargo tank(s) with applicable vacuum relief safeguards, the tight ullage space will incur a progressive drop in pressure consistent with outflow of cargo. Theoretically, outflow would cease when the combined ullage vacuum and cargo head no longer exceeded the hydrostatic head of water.

This concept has several inherent disadvantages. It poses the same difficulties as the mechanical vacuum in terms of interface with the ship's struc

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

FIGURE 5-8 Alternative 8b: Hydrostatically driven vacuum (passive).

ture and existing safety systems. Secondly, without instantaneous identification of the damaged tank(s), and closure of all vent lines and piping to the tank(s), the system would not react quickly enough to attain anywhere near the vacuum required.

If a passive vacuum system depends on automatic closure of all tank vents from a remote location, or on pressure and/or liquid-level sensors in tanks, malfunction of this system (shut-off of all vents during normal cargo discharge) could cause a catastrophe. With a ship discharging at full rate, the vents absolutely must remain open, or the IGS must be operating. Otherwise, a major collapse of the deck and/or other structure would ensue, with a strong likelihood of explosion and fire—all occurring in a port. Finally, the allowance for initial cargo outflow negates the limited advantages of this concept over the mechanically driven vacuum system. The concept will not be considered further here, but it deserves further research and development.

8c. "Imaginary" Double Bottom

A variation on the hydrostatically driven vacuum principle or hydrostatic control alternatives is the "imaginary" double bottom, where chemically treated water is placed in the bottom of the cargo tank to a depth of 1 to 2.5 meters (see Figure 5-9). Upon grounding, only this layer of treated water would escape from the hull as the vacuum is established; thus, conceptually, no oil would flow out.

The imaginary double bottom concept has limited application. This con-

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

FIGURE 5-9 Alternative 8c: Imaginary double bottom concept.

cept could not be used with any products that would be contaminated by water. Crude that must be heated could not be carried with a water layer underneath; otherwise, it would be impossible to heat the cargo.

The committee will not consider this concept further, but it deserves further consideration for possible research.

Summary—Vacuum System Alternatives

A vacuum has natural advantages, and development of a practical, fail-safe, vacuum system would be desirable. The committee encourages research along these lines. Proposals for manually closing a ship's vents when the ship is in shallow waters, or for connecting IGS branches to cargo tanks, may have merit. However, there is still a concern with additional cargo vapor pressure build-up.

In the committee's view, the various vacuum concepts lack sufficient cost data to be considered further in the present study.

9. Smaller Tanks

This concept represents a return to past practice. Dividing the cargo section into smaller tanks, with additional bulkheads, would limit the vol

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

ume of cargo exposed to a given point of damage. This concept would significantly reduce pollution in collisions, and it also would be effective in groundings.

The committee was able to obtain cost data for this option for new construction, but retrofit costs are highly dependent on base vessel configuration. Only the new construction alternative will be pursued further in later chapters.

9a. Service Tank Location

While pollution-control efforts have focused on restricting outflow from cargo tanks, most tankers carry 2 to 5 percent of their cargo deadweight in heavy fuel oil. In fact, all ships carry fuel oil in tanks subject to collision or grounding damage.

In new tanker designs, fuel tanks could be placed in defensive locations so as to minimize fuel oil pollution. In the committee's view, no further study is required.

INCREASED PENETRATION RESISTANCE

These proposals aim to limit pollution risk by making the ship's hull more resistant to damage. Where "barriers" would prevent any pollution, and "outflow management" would limit short-term pollution, "increased penetration resistance" would try to control the worst case. The proposals do not have a common principle, and they address diverse problems.

The pollution risk associated with tank vessel accidents is related to both material characteristics and the reaction characteristics of the design details. For example, tripling the scantlings of all lower hull structure would raise the threshold of grounding impact that could be tolerated before hull rupture. Alternatively, the same result could be achieved by selecting either a more damage-tolerant steel or a more grounding-tolerant bottom structural configuration.

Conventional tank vessel designs reflect an economic balance of widely available material and human resources. In all cases, designs tend toward the practical and are influenced strongly by historical factors, as well as by recognition that tank vessels may need structural repairs at remote locations. Unique or revolutionary designs, exotic or unusual materials, and unconventional work practices traditionally have been resisted, pending incentives that would support their research, development, and integration into the shipbuilding and ship repair industry.

10. Internal Deflecting Hull

In this design, shown in Figure 5-10, the forward 10 to 20 percent of the vessel would incorporate an inner double hull made of massive slab-type

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

FIGURE 5-10 Alternative 10: Internal deflecting hull.

structure of exceptional strength, configured much like the lower bow plating (forefoot) of an ice breaker. The outer hull, made of normal-grade steel plating, would be shaped (as now) to suit the most efficient hydrodynamic form. In addition to shielding cargo with the resultant voids, the vessel, upon grounding, would initially ride up upon (and/or have its course deflected by) the inner slab-shaped structure. This would deflect damage from the remaining 70 to 80 percent of the hull.

This design provides local hull protection for that portion of the hull (the forebody) most likely to be damaged in groundings. On the downside, the risks associated with double-bottom voids, including fire/explosion hazard from gas leaks and flooding in event of hull breach, also apply here. Another concern is increased potential damage to vessels struck by a ship with a deflecting hull.

While a deflecting hull might be suited to small tankers and barges, the committee believes that in large tankers, considering their mass and speed, the amount of energy the hull would have to withstand to deflect the vessel would be enormous. The questionable effectiveness of this diversion might compromise protection of aft structure from damage.

This concept does not exceed the bounds of existing technology nor analytical evaluation techniques, although it would require development. But it presupposes a specific accident scenario8 and in doing so, limits itself. Extensive cost and effectiveness evaluation, and small-scale testing, would be needed to provide even a preliminary assessment of various versions of this concept; the committee will not consider the concept further.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

11. Grinding Bow

In this design, shown in Figure 5-11, the forward 20 to 30 percent of the vessel would incorporate a double bottom with an internal transverse structure specifically engineered and constructed to suit rock-grinding criteria. In a grounding, the most vulnerable cargo tanks would be protected by a double bottom. In moving over the obstacle, the rasp-like transverse structure would grind down the material, thus creating a clear and unobstructed passage for the remaining 70 to 80 percent of the vessel (or a smooth bed for the ship to come to rest on).

The partial double bottom would involve the related increased risk of explosion. A greater practical concern is that the grinding-bow concept never has been studied, as far as the committee can ascertain. It is an entirely unproven application of technology, and would be applicable only to the pinnacle or submerged-reef types of grounding described in Chapter 4.

In accidents, this design conceivably could minimize the extent of hull penetration and the resulting, largely uncontrolled destructive forces that mangle ship structure. But the concept is not sufficiently developed to be considered further in the present study.

12. Unidirectionally Stiffened Hull Structure

The initiation of plate tearing requires greater stress than the continuation of plate tearing. Stress buildup leading to plate tearing occurs as a local function of the plate's resistance to the deflecting force or obstacle. In recent tank vessel designs, highly rigid transverse bottom stiffening attached to more flexible bottom plating creates points of peak stress as a moving vessel grounds (see Figure 5-12). If the bottom transverse structure were designed to fail early in the stress build-up, or if conventional transverse structure were eliminated altogether, the bottom would flex uniformly and thus could avert onset of plate tearing. Although the bottom structure would deform significantly, it would not necessarily fracture.

FIGURE 5-11 Alternative 11: Grinding bow—lower bow structured to sustain impact and grinding criteria.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

FIGURE 5-12 Alternative 12: Unidirectionally stiffened structure.

This concept has been applied to some ships. Its integration into tank vessels has required a new design approach; a Japanese shipyard has such tankers under construction for the product trade. Elimination of much transverse material would lower structural redundancy significantly and could increase susceptibility to catastrophic collapse in heavy weather or minor groundings.

Nevertheless, the proposal warrants strong research and development support, as its characteristics would enhance any existing or proposed design in terms of pollution prevention. The committee will not consider it further in the present study due to insufficient design and effectiveness data.

13. Honeycomb Hull Structure

This concept entails a "sandwich" double-plate construction of high-density material bonded to the inner and outer hull plating of a double-hull design. While this concept is intended to provide both a high-yield and an energy-absorbing hull structure, the committee found no evidence that the energy dissipation involved in the displacement and crushing of the honeycomb structure would provide significantly increased resistance to failure in comparison to conventional structures, such as the double hull.

This concept would require extensive research and testing. It involves novel material requiring unique production technology, and it introduces design and operational requirements never previously addressed by the industry. Maintenance and repair requirements would be unique, inspection plans would require special development, and the risk of cargo leaks into voids and the related explosion hazard would be considerable.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

Evaluation of this proposal requires considerably more support documentation than is available. The committee will not consider it further.

14. High-Yield Steel Bottom Structure

The behavioral characteristics of steel can be controlled by its composition and manufacturing process. Conventional shipbuilding steels generally are chosen to suit minimum strength criteria, balanced against the shipbuilder's best economic interest regarding production, labor skills, and steel cost. Some steels provide significantly better impact resistance than those generally used in tanker construction, but they require more costly fabrication. These steels (such as the HY80 steel used in submarine hull construction), if used as hull plating on otherwise conventional single-hull tankers, would increase resistance to grounding and collision damage.

However, the use of exotic materials would require a commensurate investment in production facilities and personnel training. At present, worldwide repair resources, in terms of both materials and skill, are not readily available.

This concept will not be considered further in the present study, but research and development should be encouraged. Research and development needs concerning materials for tank vessel hulls and structures are discussed in Chapters 4 and 7. The use of high-yield steel is one aspect of the overall research requirement for:

  • more robust and uniform-strength scantlings;

  • greater use of steels having high strain energy absorption and high strain-to-rupture characteristics; and

  • other materials or structural configurations that will enhance the resistance of the hull girder to rupture.

15. Concrete Hull Structure

Concrete has been used before in the marine industry. During World Wars I and II, a small number of moderate-size prototype concrete ships were constructed. Since then, concrete has been utilized periodically in offshore storage facilities, barges, and in some yachts. The benefits include good compressive qualities, relatively high tolerance to salt water, potentially seamless construction, and adaptability to fluid forms. In sufficient mass, concrete hulls may offer greater resistance to the onset of hull failure from groundings than would single-hull steel designs.

Concrete structure would seem to offer advantages in inland barge structure, where hull bending loads are limited. Such barges could be built economically with less-skilled labor than required for steel construction. The added mass provided by concrete would allow vessels to sustain minor

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

damage without rupture. As some barges are restricted in dimension and draft, the added mass would be detrimental to cargo deadweight capacity.

Considerable research is required to sort out the engineering issues. The greater concern, however, would be the willingness of the industry to adopt the necessary facilities and support infrastructure worldwide. Given these conditions, the committee will not consider this concept further.

16. Ceramic-Clad Outer Hull

A thick ceramic cladding might increase significantly hull surface hardness, so as to withstand greater impact without fracture. In addition, a ceramic surface could reduce resistance to vessel movement over minor obstacles, thereby increasing hull tolerance for minor groundings.

This technology is novel and unresearched. Other than noting that maintenance likely would require development of specialized facilities, the committee cannot evaluate this proposal other than by educated judgment. There are no data that encourage development of the concept at present. The committee will not consider it further.

ACCIDENT RESPONSE

The previous three categories dealt with structural hull designs that could prevent or mitigate pollution at the time of an accident. This category includes concepts that could mitigate pollution by aiding response to an accident. These options involve less drastic revisions than the structural designs discussed previously. They are, in effect, "add-ons" that could be combined with other designs to enhance overall effectiveness. Although the committee will not consider these proposals in outflow-reduction or economic assessments, some nonetheless warrant consideration for use.

17. Enhanced Information Processing

In this concept, the ship's hull, tank, and pipe structure would incorporate extensive sensing devices connected to a central processing facility. This monitoring system would alert the crew to any unacceptable condition: It could confirm structure and cargo system integrity, analyze response alternatives, and recommend action to mitigate pollution. In theory, quick and accurate knowledge of the status of the ship and its cargo following an accident could promote effective response. The following data would be useful, for example: Fluid levels in all cargo, ballast, and fuel tanks together with level trends; the draft of the tanker and trends in draft change; extent of flooding of cargo or ballast piping; the status of engine room function; and tidal and other environmental data.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

Some of the technology for this concept exists. Most modern tankers already have some type of computer(s). Most classification societies require stress computers to calculate loading, strength, trim, and stability. Many ships also have, for example, computerized cargo load and discharge sequence programs. In addition, programs exist that measure "damaged" tanker stress, trim and loading, and most modern tankers have installed liquid-level gauges that can be read in a central control room. However, these computers are stand-alone units and rely on prior calculations stored within. No sensors are used to update or provide input to the computer programs.

The committee believes that all of the hardware and software necessary to implement this concept exists. One classification society has even developed a prototype "black box" recorder, similar to those used on aircraft to record accident data. However, there are a number of problems inherent in an extensive computerized system. After a major hull rupture, cargo outflow can be very rapid, and the computer system would have to respond nearly as quickly. Furthermore, the most important unknown is often the extent of damage to the ship. This may not be detectable by computer. Even if all the necessary information is immediately available, the ship must retain its capability to transfer cargo from damaged tanks and must have intact, empty tank space available to receive it. Finally, performance of the system would depend heavily on maintenance and repair; malperformance would risk ship safety. Exposure to the marine environment and possible corrosive effects of cargo and ballast present maintenance problems for sensors and wiring.

The marine industry is becoming accustomed to the use of computers, but much of the industry still relies on experience, judgment, and past practice. With the increasingly broad application of computer assistance in work places, and the availability of most elements needed to enhance shipboard information processing, the committee considers this option worthy of further development. However, the committee was unable to evaluate the concept further within the scope of the present study.

18. Towing Fittings

Standardized towing fittings already are required by a number of tanker owners, although they are not required by any regulatory body. Towing fittings and their arrangements may enhance the ability of tugs to save disabled tankers before grounding or foundering. These fittings also assure that damaged tankers can be attached to positioning equipment, thus reducing the risk associated with uncontrolled movement of a stranded tanker.

Towing fittings can tolerate significantly stronger forces than standard mooring fittings; such forces are encountered when attempting to free and float a grounded and lodged vessel.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

All tankers regularly trading in Alaska carry emergency towing wires, as a result of a near-grounding in that region a decade ago. The concept has been evaluated and recommended by the IMO.

In the committee's view, towing fittings should be mandated for tank vessels, although insufficient data exist to allow further consideration in this study.

19. Distressed-Ship Cargo Transfer System

Tankers that run aground often remain stranded, requiring a major off-load prior to refloating. The cargo system often is rendered useless, and even if not, pumping cargo from a breached tank is impossible when the suction is at the bottom of the tank. At present, it is usually necessary to rely upon portable pumps with relatively low capacity (compared to operational pumps), which must be brought to the damaged ship and lowered into the tanks for off-loading. This can be time-consuming; for a large tanker, the process might take a week or more, leaving the ship exposed to greater damage or possible loss.

This risk might be reduced by two basic improvements. First, a means allowing for rapid transfer of cargo from a breached tank to intact tanks, such as segregated ballast tanks, could reduce oil outflow from a stranded ship as the tide falls, or due to wave action or ship motion. To be effective, such a system would have to be available to use within an hour or two after stranding. Second, if a ship were equipped with a system for transferring cargo from breached tanks at a level well above the bottom, where oil and water were mixing, this might facilitate much more rapid off-loading or lightering of cargo. This would enable more rapid refloating of a stranded ship than is common with the small, portable salvage pumps now used.

The committee is aware of three transfer-system concepts, as follows:

  • Each cargo tank could be fitted with permanent suction at about mid-tank height, capable of discharging oil from above the oil/water interface in a damaged tank. One proposal would have the mid-height suction line extended aft into the ship's pump room, where it could be installed so as to line up with the suction side of the segregated ballast system. A spool piece, not in place during normal operations, could be fitted following an accident, allowing the ship's ballast system to handle nearly immediate evacuation of oil from damaged cargo tanks into intact segregated ballast tanks.

  • Depending on a ship's arrangement of cargo and ballast tanks, it might be possible to fit sluice or transfer valves between cargo tanks and adjacent or nearby segregated ballast tanks. In the event of damage to a cargo tank, these sluice or transfer values would be opened, permitting the cargo to

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

flow by gravity (or by hydrostatic pressure in the IOTD designs) into the empty ballast tanks. Of course, before activating such a system, the crew would need to be sure that the receiving tank(s) were intact.

  • The third option is a portable emergency cargo transfer system employing high-capacity submersible pumps, which could be lowered into damaged cargo tanks while maintaining a vacuum in these tanks (to hold in the cargo). Such a system, which would be stored aboard on a moveable dolly, could be used to transfer cargo either to intact segregated ballast tanks or to a lightering vessel.

The committee found these concepts promising, as they employ, for the most part, existing hardware and sound operational principles. However, insufficient details were available for the committee to judge how rapidly these concepts could be activated following an accident. Because the initial outflow from a large hole takes place in a matter of minutes, the committee does not believe any of the systems described are likely to make significant reductions in immediate outflow. However, these systems could be useful in mitigating subsequent outflow, and in hastening the refloating of a ship.

Only rough figures were provided regarding costs, suggesting that such systems might add 1 to 2 percent or more to the cost of a new ship. Even without more precise application and cost data, the committee feels these concepts deserve further consideration by industry, the Coast Guard, and the IMO.

SUMMARY OF INITIAL TECHNICAL EVALUATION

Following are the significant findings drawn from the foregoing technical assessment:

  • Regardless of vessel design, the potential advantages associated with towing fittings argue for their serious consideration for all new and existing tank vessels.

  • Distressed-ship cargo transfer systems deserve further consideration by industry, the Coast Guard, and IMO.

  • The committee has eliminated from further consideration, regarding pollution control and cost, six design alternatives that, though promising in concept, lack the basic technical supporting data and involve major design and operational uncertainties. These include the resilient membrane, convertible tanks, deflecting hull, honeycomb hull, imaginary double bottoms, and ceramic-clad hull.

  • A number of other proposals should be considered for further research and/or development. These are the mechanically and hydrostatically driven vacuum systems, grinding bow, unidirectionally stiffened bottom structure, high-yield steel bottom, and enhanced information processing. A design combination worthy of study is the longitudinally reinforced bottom struc

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

ture, plus high-yield steel bottom plating and a grinding bow. In addition, the concrete hull structure for tank vessels requires research to resolve numerous application problems and to understand the lack of industry interest, reflected by the many decades of unsuccessful efforts to develop maritime markets.

  • Based on results of its technical assessment, the committee determined that the following design alternatives should be assessed on the basis of their pollution-control effectiveness:

    MARPOL tanker (as the reference vessel)

    double bottom

    double sides

    double hull

    hydrostatic control

    intermediate oil-tight deck

    smaller tanks

  • In conducting a technical assessment of the above designs, it was apparent that the following three compound design alternatives may offer improved outflow reduction potential for a range of accident scenarios. Moreover, these compound designs employ proven or well-understood technology:

    double sides with hydrostatic control

    double hull with hydrostatic control

    intermediate oil-tight deck with double sides

ASSESSMENT OF DESIGN ALTERNATIVES APPLICABLE TO BARGES

The committee's charge encompasses all ocean-going vessels 10,000 DWT and greater. Hence, ocean-going barges over approximately 5,000 gross tons are considered in this report.

The DnV study was prepared for tankers based on a tanker data base and could not be extended to cover barges in the time allowed for preparation of this report. However, as noted earlier, the technical assessment of design alternatives applies in some cases to barges as well as tankers.

The following discussion delineates how the various alternatives apply to barges and how application may be limited due to the unique nature of barges.

Barriers

The offshore barge can carry up to 50,000 tons deadweight and is configured much like a tanker of equivalent size (see Chapter 1, Figure 1-7).

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

Thus, all of the structural barriers would be applicable to offshore barges, with DnV pollution-effectiveness rankings similar to those of 40,000 DWT tankers.

Outflow Management

Barges are unmanned and are either towed or pushed by a power source (tugboat). Consequently, any outflow management alternative requiring manned intervention, or some automatic device electrically coupled to the tug, would be impossible to maintain in extreme weather conditions (when barge accidents are more likely to occur). Thus, the vacuum systems, and the oil transfer system using deck-operated valves, would not be applicable. The intermediate oil-tight deck and smaller tanks might be applicable.

Hydrostatic control appears to have limited, if any, application to towed barges. Barges carry a wide range of petroleum products of varying densities and properties. Thus, the cargo-loading levels required to achieve hydrostatic control would vary. Such precise control is at variance with barge crew manning levels and practices. Hydrostatic control may be somewhat safer and more feasible for some articulated barge/tug (pusher-tug and notched-barge) operations, subject to stringent cargo-loading controls.

Increased Penetration Resistance

The internal deflecting bow, grinding bow, high-yield bottom structure, honeycomb, and concrete structures would not be prime barge designs, because groundings usually take place either at low speeds (under power), or at very low speeds and unpredictable angles (when the barge is separated from the tug in heavy weather).

The longitudinally reinforced bottom structure, or any other design that manages bottom failure in high-energy accidents, would be poor choices, because barges are more prone to low-energy groundings.

Accident Response

Enhanced information processing would be difficult, as the system would have to be connected from the tug to the barge. Installation of towing fittings is routine on barges, and use of a distressed-barge cargo transfer concept poses no significant technical problem.

Smaller Offshore Barges and Inland Barges

While offshore barges of less than 10,000 DWT and inland barges are outside the scope of the present study, this brief discussion illustrates the dimensional constraints of small tank vessels.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

Due to these dimensional considerations, the Coast Guard, in information circular NVIC 2-90 (1990), recommends the following:

  • For vessels (tankers or barges) under 20,000 DWT, the width of a double hull can be reduced linearly from 2 meters for a 20,000 DWT vessel, to 1 meter for a 10,000 ton vessel.

  • For vessels under 10,000 DWT certified for inland routes (primarily barges), the void space in a double hull must maintain a clearance of 2 feet. This width is considered the minimum for construction, inspection, and maintenance.

  • Bow and stern compartments do not carry oil.

  • The protective spaces formed by the double hull should not contain oil. (The committee notes that, for survivability purposes, it would be preferable not to vent these spaces, as they are not ballasted.)

POLLUTION CONTROL ANALYSIS OF SELECTED DESIGNS AND DESIGN COMBINATIONS

The next assessment the committee applied to the 10 selected design alternatives and compound designs addressed the potential for reducing oil outflow in collisions and groundings. To aid in the analysis of the seven basic design alternatives, the committee sponsored an addition to a study conducted by Det norske Veritas (DnV) for the Royal Norwegian Council for Scientific and Industrial Research (Det norske Veritas, 1990). DnV is a Norwegian classification society that has developed a methodology for calculating oil outflow. The committee conducted its own evaluation of the three compound design approaches based on its own estimates and calculations derived from the results of the DnV analyses.

The calculations made in this analysis are based on a number of assumptions, and the committee does not necessarily accept all of these as the most reasonable of possible assumptions. Therefore, the committee regards the specific percentages for reduction of oil outflow derived from this analysis as indicative examples. These results represent only one attempt to quantify reduction in outflow; application of a different set of assumptions might produce different results.

Using a probabilistic approach to past accidents derived from Lloyd's Register, the U.S. Coast Guard, and DnV databases, the methodology is based on the damage severity experienced in collisions and groundings. Using this damage for predictive purposes, outflow is estimated for each alternative design, first for collisions and groundings separately, and then combined using a 40/60 weighting (based on the collision and grounding frequencies for tankers worldwide, per DnV statistics). The committee recognizes that the 40/60 (collisions/groundings) frequency weighting is approximately correct for the number of tanker spill incidents over 10,000

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

gallons in U.S. waters, while the ratio of oil volume lost in collisions versus groundings approaches 20/80 (in volume). Nevertheless, the committee and DnV agree that the relative outflow rankings of each design would be the same regardless of which of these two sets of ratios are used.

For purposes of the committee's study, the model was applied to 21 design arrangements. Outflow was calculated or estimated for two tanker sizes: VLCC (13 arrangements) and 40,000 DWT (eight arrangements). In addition, DnV assessed the implications of the design arrangements for outflow performance of an 80,000 DWT tanker; this assessment was based on extrapolations of the results for VLCCs and 40,000 DWT vessels, where appropriate. These three sizes represent a cross-section of the sizes of tankers generally employed near the U.S. coastline. The VLCC is involved in offshore oil trade and is lightered by tankers in the 80,000 DWT range, as described in Chapter 2. The 40,000 DWT tanker is typical of coastal carriers distributing refined products among U.S. ports.

The DnV report (except for its appendices) is presented in Appendix F. In addition, the committee interpretations of the assumptions used and some of the DnV conclusions can be found in the Appendix (where the committee has little or no comment, the committee accepts the DnV statements in the context of their report). Following is a summary of the designs analyzed and the rankings of these designs in terms of oil outflow.

General Features of the Tankers

All the designs analyzed have the following main particulars9:

 

VLCC

40,000 DWT

Length

315.0 m

190.1 m

Beam

57.2 m

27.4 m

Depth

30.4 m

17.8 m

Draft

20.8 m

12.9 m

Probabilistic Ranking of VLCC Tankers

Statements drawn from the DnV report are in italics (some are edited for brevity and style). Committee comments are in standard type.

An analysis of serious casualties carried out by DnV shows that, for tankers above 20,000 DWT worldwide, the probability of collisions is approximately 40 percent and for groundings approximately 60 percent. These percentages differ in other reports of worldwide accounts (some show nearly equal frequency) but approximate reports of major accidents in U.S. waters on a frequency basis; however, as noted previously, the relative rankings of various alternatives are not affected.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

Using the 40/60 ratio, the average oil outflow for different designs in collisions and groundings, respectively, have been combined in Figure 5-13 for 5-knot grounding speed, and in Figure 5-14 for 10-knot grounding speed. In these figures, the total index has been given as well as the portions derived from collisions and from groundings.

The different VLCC hull design arrangements and variations, and how they relate to the committee's design alternative categories, are listed in Table 5-1; sketches of specific arrangements and more detailed data regarding ship parameters are provided in Appendix F.

Results—VLCC Analysis

Statements drawn from the DnV report are in italics (some are edited for brevity and style). Committee comments are in standard type.

The double side/double bottom (i.e., double hull) VLCC designs (arrangements 1A, 1B) have the smallest potential oil outflow in collision and grounding, given the assumptions regarding damage location and extent. Compared with a modern conventional MARPOL VLCC, the double side/double bottom tanker is likely to spill, on the average, only about 33 percent of the standard amount of oil.

Comparing the two double side/double bottom designs, the alternative with two longitudinal bulkheads, and double sides (arrangement 1A) is likely

FIGURE 5-13 Combined ranking, 5 knots (low-energy). Reference DnV Report, Figure 3.25, Appendix F.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

FIGURE 5-14 Combined ranking, 10 knots (high energy). Reference DnV Report, Figure 3.26, Appendix F.

to leak less oil in about 75 percent of all collisions than the design with only one longitudinal bulkhead at the centerline (arrangement 1B), although both designs have the same overall combined performance.

The amount of oil likely to escape from a VLCC with a double bottom for ballast and segregated ballast side tanks (arrangements 4, 4A and 4B), is about 40 percent of the amount that would be spilled from a modern conventional MARPOL VLCC. The positive influence of locating the side ballast tanks forward, with regard to potential oil outflow in collisions, is evident comparing arrangements 4B and 4A.

Single-bottom designs with wide wing tanks for ballast (arrangement 3B) perform quite poorly overall (combined ranking) due to the large volume of oil likely to escape in grounding. In fact, this design performs only slightly better than modern conventional VLCCs.

Reducing the tank size to half of MARPOL requirements (arrangement 7) reduces the potential average oil outflow (combined ranking) to about 60 percent of the outflow from a conventional VLCC.

The partial double-bottom VLCC (arrangement 5) does not perform too well in groundings. The basic reason is that the double-bottom height is too low to be effective, given the study's assumptions. Increasing the height, and reducing the side tank width, would make this design similar to arrangement 1A and 1B.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

TABLE 5-1 VLCC Tanker—List of Alternatives/DnV Arrangements Analyzed

Committee Alternative

DnV Arrangement (Arrangement Code)

MARPOL Tanker (reference standard)

• Modern Conventional VLCC (OR)

Double Bottom

• Double bottom in whole cargo area (2)

 

• Double bottom with ballast in side tanks (options 4, 4A, and 4B)

 

• Double bottom in side tanks, single bottom in center tank (6)

Double Sides

• Double side, single bottom (3B)

 

• Double side in whole cargo area and partial double bottom (5)

Double Hull

• Double side and double bottom, with two longitudinal bulkheads (1A), with bulkhead at centerline (1B)

Intermediate Oil Tight Deck

• Intermediate Oil Tight Deck (8)

Hydrostatic Control

• Hydrostatically balanced loading (9)

Small Tanks

• Tank size half of MARPOL requirements (7)

The double-bottom designs analyzed would spill no oil at all in about 85 percent of all groundings, whereas some oil always escapes from single-bottom designs, irrespective of tank size and the presence of a vacuum tank system.

A vacuum system reduces significantly the amount of oil escaping in groundings for the single-bottom designs analyzed. However, the total amount of oil lost from the modern conventional VLCC with a vacuum system is still about twice the amount escaping from the double side/double bottom design.

Comparing VLCCs with double sides to VLCCs with single sides, the former provides an effective barrier against oil outflow in 20 percent of the all collisions for arrangements 1A and 1B, and in 42 percent of all collisions for arrangement 3B. The protection is particularly effective for low-energy collisions with limited damage penetration. The influence of increasing the double-bottom height as compared with increasing the double-side width is shown in Table 5-2. For example, increasing the VLCC double-bottom height from 2.0 to 3.9 m (B/15) reduces the probability of oil outflow in groundings from 58 to 14 percent, while increasing the width of side tanks from 2.0 to 3.0m only reduces the probability of outflow in collision from 88 to 80 percent.

The specific conclusions drawn by DnV can be found in Appendix F (page 275); the committee's commentary can be found on page 300.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

TABLE 5-2 Probability for NO Oil Leakage as Function of Double Bottom Height and Distance Between Double Sides*

GROUNDING (VLCC)

DISTANCE BETWEEN INNER/OUTER BOTTOM

PROBABILITY FOR NO LEAKAGE

 

METERS

B/—**

PERCENT

2.0

28.6

41.7

2.4

23.8

52.6

3.0

19.0

69.0

3.9

14.7

86.0

6.6

8.6

99.8

COLLISION (VLCC)

DISTANCE BETWEEN INNER/OUTER SKIN

PROBABILITY FOR NO LEAKAGE

 

METERS

B/—**

PERCENT

2.0

28.6

12.1

3.0

19.0

20.4

5.8

9.86

39.4

6.3

9.08

42.0

* Reference Table 3.3.1 in DnV Report, Appendix F.

** Ratio, beam to between hull width.

Probabilistic Ranking of 40,000 DWT Tankers

The 40,000 DWT designs chosen for the analysis are similar to the VLCCs. The objective in analyzing both very large (VLCC) and small (40,000 DWT) tankers was to detect trends, so the results could be applied, in a general sense, to other tanker sizes. The committee also selected the 40,000 DWT tanker for analysis by DnV because the size is typical in U.S. coastal activity.

Particular attention has been directed to investigating the influence of double-side width on oil outflow, as several 40,000 DWT product tankers have been built with narrow sides.

The arrangements listed in Table 5-3 were analyzed by DnV; Appendix F (page 278) provides a summary of the assumptions made regarding ballast capacity, side width, bottom height, and longitudinal bulkheads.

Sketches of the various arrangements are found in Appendix F (pages 281-285). Table 5-3 links the DnV arrangements to the general design alternatives considered by the committee.

The combined ratings for collision and grounding (using the 40/60 weighting) gives the combined ranking shown in Figure 5-15 for 5-knot grounding speed, and in Figure 5-16 for 10-knot ground speed.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

TABLE 5-3 Small Tanker (40,000 DWT Tanker)—List of Alternatives/DnV Arrangements Analyzed

Committee Alternative

DnV Arrangement (Arrangement Number)

MARPOL Tanker (reference standard)

• Modern 40,000 DWT SBT (1)

Double Bottom

• Double bottom and single sides (3)*

Double Sides

• Double sides and single bottom (2)

Double Hull

• Narrow (0.76 m apart) double sides and double bottom (4)*

 

• Double sides (1.2 m apart) and double bottom (5)*

 

• Wide (2.0 m apart) double sides and double bottom (6)*

 

• Wide (2.0 m apart) double sides and double bottom, short tanks (7), no centerline bulkhead

Intermediate Oil Tight Deck

• Intermediate Oil Tight Deck (8)

* With centerline bulkhead.

At 5 knots, design arrangement 6 with wide double side and low double bottom achieves a rating of 57, followed by arrangement 8 (the intermediate oil-tight deck) with an rating of -65. Only design arrangement 2, the double side without a double bottom, has an index above the reference vessel (MARPOL). The ranking does not change for the 10-knots scenario; design arrangement 6 (wide double sides with double bottom) remains the best with a rating of 38, followed by design arrangements 8 (intermediate oil-tight deck), 4 (narrow double sides and double bottom) and 5 (mid-width double sides and double bottom).

Results—40,000 DWT Tanker Analysis

This analysis demonstrates the value of double bottoms in preventing oil outflow.

Table 5-4 shows the effect of increasing double-bottom height and double-side width on the probability of spilling oil. The probability falls rapidly as double bottom height is increased. By contrast, increasing the width of double sides does not have the same dramatic effect.

Estimated Oil Outflow from a 80,000 DWT Tanker

Based on the studies of VLCCs and 40,000 DWT tankers, DnV appraised the potential oil outflow from a 80,000 DWT tanker designed with several hull configurations. No supporting calculations were carried out. The conclusions drawn by DnV can be found in Appendix F (page 296); the committee's commentary can be found on page 302.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

FIGURE 5-15 Combined ranking, 5 knots (low-energy), 40,000 DWT. Reference DnV Report, Figure 4.18, Appendix F.

FIGURE 5-16 Combined ranking, 10 knots (high energy), 40,000 DWT. Reference DnV Report, Figure 4.19, Appendix F.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

TABLE 5-4 Probability for NO Oil Outflow in Collision and Grounding*

GROUNDING (40,000 DWT)

DISTANCE BETWEEN

INNER/OUTER BOTTOM

 

PROBABILITY FOR

NO LEAKAGE

METERS

B/—**

PERCENT

1.83

15.0

85.1

2.0

13.7

90.0

2.6

10.5

98.4

3.9

7.0

99.9

COLLISION (40,000 DWT)

DISTANCE BETWEEN

INNER/OUTER SKIN

 

PROBABILITY FOR

NO LEAKAGE

METERS

B/—**

PERCENT

0.76

36

8.6

1.2

22.8

16.3

2.0

13.7

29.2

3.0

9.1

41.9

* Reference Table 4.3.1 in DnV Report, Appendix F.

** Ratio, beam to between hull width.

The Committee's Overall Conclusions from DnV Analysis

As noted earlier, DnV's conclusions, along with committee comments, are provided in Appendix F. The committee drew the following overall conclusions from the analysis of all tanker sizes:

  • Double hulls provide significant overall protection against oil outflow in low-energy collisions and groundings.

  • Wide tanks are likely to spill more oil than long, narrow tanks.

  • Double sides protect against oil outflow in collisions, particularly low-energy collisions.

  • Double bottoms protect against oil outflow in groundings, particularly low-energy groundings.

  • The intermediate oil-tight deck, when combined with double sides, should provide protection against both groundings and collisions.

  • Hydrostatic loading and/or smaller tanks should reduce oil outflow when used in conjunction with any design concept.

SUMMARY AND SIGNIFICANCE OF OUTFLOW ESTIMATES

The results of the DnV probabilistic study and the committee's own estimates for oil outflow relative to a standard of reference (100%), the

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

TABLE 5-5 Performance of Alternative Designs for Large Tankers (VLCC)—Oil Outflow

Design Alternative for VLCC (240,000 DWT) Tanker

Oil Outflow Relative to MARPOL* (100%) for Composite** of Collisions (40%) and Groundings (60%)

 

Low-Energy (5 kn)

High Energy (10 kn)

Double Bottom (B/15)

42

37

Double Sides

88

130

Double Hull

33

26

Hydrostatic Control (passive)

62

40

Smaller Tanks (1/2 MARPOL)

58

70

Intermediate Oil-Tight Deck with Double Sides

32***

23***

Double Sides with Hydrostatic Control (passive)

32***

21***

Double Hull with Hydrostatic Control (passive)

30***

22***

* MARPOL standard tankers have protectively located segregated ballast tanks.

** Composite based on frequency of collision and groundings.

*** Committee estimate (see Appendix K).

single-hull tanker with protectively located segregated ballast tanks (PL/SBT—MARPOL), are listed in Tables 5-5 and 5-6.

As noted earlier, the committee recognized the possibility of combining certain design alternatives to improve pollution control. The committee used DnV data to derive outflow estimates10 for three combinations: double sides with hydrostatic control; double hull with hydrostatic control; and intermediate oil-tight deck with double sides. The outflow ratings for these combinations were not evaluated directly by DnV. Therefore, these oil outflow percentages (as shown by ***) are to be considered indicative examples of performance but less rigorous than the other percentages.

Outflow Performance Relationships and the Underlying Reasons

To understand the significance of these performance values, it is important to identify the major differences between designs, to understand the reasons for these differences, and to understand to what extent this knowledge can be used for decision making.

1. The performance benefits from each design alternative are significantly better for the big ships (VLCCs) than for the little ships (40,000 DWT). This is true at both high and low speeds.

The reason is that, in big ships, the size and numbers of cargo tanks are not greatly changed among design alternatives. This, in turn, is the result of both IMO hypothetical oil outflow requirements and structural demands in

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

TABLE 5-6 Performance of Alternative Designs for Small Tankers (40,000 DWT)—Oil Outflow

Design Alternative for 40,000 DWT Tanker

Oil Outflow Relative to MARPOL* (100%) for Composite** of Collisions (40%) and Groundings (60%)

 

Low-Energy (5 kn)

High Energy (10 kn)

Double Bottom (B/15)

82

50

Double Sides

136

130

Double Hull

68

43

Hydrostatic Control (passive)

52

34***

Smaller Tanks

68

76***

Intermediate Oil-Tight Deck with Double Sides

57***

36***

Double Sides with Hydrostatic Control (passive)

70***

44***

Double Hull with Hydrostatic Control (passive)

61***

39***

* MARPOL standard tankers have protectively located segregated ballast tanks.

** Composite based on frequency of collision and groundings.

*** Committee estimate (see Appendix K).

big tankers. In smaller tankers, the presence of a protective skin (i.e., double bottom or double side) allowed the designer of the DnV series to select a smaller number of much larger tanks. Neither hypothetical oil outflow nor structural demands impose significant limits on cargo tank size in these small ships. The result is that in any accident piercing the inner skin of a smaller tanker, a significantly larger amount of cargo is exposed to release than in the base ship.

2. The performance improvement for all alternatives appears better at high speed (10 knots) than slow speed (5 knots) for both big and little tankers.

The reason is that, at higher speeds, the longitudinal extent of damage is greater in groundings (although collision, as assumed here, is independent of speed). The greater the longitudinal damage in the bottom, the more cargo potentially exposed to spillage, and accordingly protective alternatives appear to be, and are likely to be more useful at higher speeds. In this regard, it is important to recognize that the base ship performance value of 100 percent represents a very different amount of oil spillage in the two ships. For the base VLCC at 5 knots, 100 percent represents grounding outflow of about 11,000 tons while at 10 knots, 100 represents grounding outflow of about 21,000 tons—or an increase of almost twofold.

3. Four cases (double hull, double hull with hydrostatic control, double sides with the intermediate oil-tight deck, double sides with hydrostatic control) appear to provide nearly the same pollution prevention benefit.

This is due to a significant assumption used in making these estimates,

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

that is, that hydrostatic control works nearly perfectly with no loss due to tide, swell, or ship motion. This assumption may not be valid, and these estimates thereby may underestimate oil outflow.

These important apparent performance differences also can be explained, however, by recognizing that:

  • The double hull is definitely very effective in low-energy groundings, but it cannot prevent some collision outflow due to narrow side tank voids.

  • Because the double hull alone, in this analysis, prevents most of the outflow in groundings, adding hydrostatic control improves performance only marginally (outflow already is minimal); furthermore, this combination only improves performance in collision as a result of cargo tanks being less than full.

  • In this analysis, double sides with the intermediate oil-tight deck is better than a double hull in collision as a result of wider side ballast tanks, and is presumed to prevent nearly all outflow in low-energy groundings and some outflow in high-energy groundings (because the analysis takes no account of factors such as tides, current, or ship motion).

  • In this analysis, double sides with hydrostatic balance appears equivalent to double sides with the intermediate oil-tight deck for the above reasons, but, in fact, will not perform as well because tides, swells, and ship motion can be expected to ensure grounding outflow following the accidents.

Applicability of Results

From the preceding, the following observations can be made concerning the applicability.

These particular numbers, while providing a relative performance indicator for one particular set of designs and assumptions, should be regarded only as a sample of the type of analysis that can be made for these and other design possibilities (particularly with regard to number and size of cargo tanks in small ships).

Other assumptions about grounding and collision speeds, etc., also should be investigated.

Accidents other than grounding and collision (i.e., fire and explosion, structural failures) should be subjected to similar analysis.

Significance of Results

Because the committee made substantial use of data from the DnV study, as well as the committee estimates, it is important to understand the significance, limitations, and possible implications of these numbers before using them as the basis for the cost-benefit analysis discussed in Chapter 6.

These estimates are significant in that they show a substantial potential

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

reduction in oil outflow in collisions and groundings for several alternatives for one plausible set of designs, and grounding and collision circumstances. However, they show that a wide range of results are possible even for the same hull design concept, when these designs are applied to ships of different sizes with different cargo tank configurations and sizes.

Due to a variety of limitations, the estimates clearly should be regarded as only a sample of the type of work that could be done with more comprehensive analysis. The limitations include use of one particular statistical casualty profile, a number of simplifying assumptions regarding oil outflow immediately following collisions, and the disregard for tidal and wave effects in the grounding outflow analysis.

The DnV study points out the relationship between tank arrangements and oil outflow. Although the DnV study did not directly address damage stability, oil outflow in an accident and damage stability are linked directly to particular tank arrangements. The results of the DnV analysis have two important implications, which are consistent with the committee's findings regarding damage stability discussed in Chapter 4.

First, for MARPOL SBT tankers (having excess freeboard and a large amount of empty ballast tank volume during loaded passage), providing improved oil outflow reduction under all plausible accident scenarios would require careful consideration of both ballast and cargo tank size and arrangements. While there is no question that the use of ballast tanks as protective spaces can reduce oil outflow, it also can have adverse consequences on stability. Similarly, protecting cargo tanks with ballast spaces outboard, or beneath them, does not mean that internal cargo tanks can be made larger than is common in single-skin ships without possible increase in oil outflow. This suggests that, for smaller ships, both damage stability assumptions and hypothetical oil outflow should be reconsidered by IMO and the Coast Guard.

Second, because all the analyses submitted to the committee, by DnV and others, were based on investigations of conventional tanker designs, it is difficult to predict how combination carriers (OBOs and ore/oil, or O/Os) would be affected under similar accident scenarios and outflow limitations. While it is clear that combination carriers, when carrying oil, must comply with all tanker regulations, the committee recognizes the following difficulties that may be encountered, particularly in OBOs:

  • The very wide tanks in OBOs may lead to difficulty in hydrostatic loading, as OBOs cannot be sailed safely with slack tanks;

  • The lack of longitudinal subdivision of OBOs also may lead to complications if hypothetical oil outflow criteria for ships smaller than VLCCs are adjusted downward;

  • The narrow side tanks typical of OBOs also may lead to complications in complying with criteria applied to conventional tankers.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
  • The committee has not conducted studies of damage stability in OBOs but believes these ships must be able to comply with whatever criteria are deemed necessary for tankers;

  • The committee questions whether hatch covers on OBOs and O/Os would be able to withstand forces imposed by a vacuum system, were that option to be applied.

While it might be argued that combination carriers need not comply with each specific requirement applied to tankers, the committee feels it would be wrong to waive regulations for combination carriers when this would result in unequal protection against oil outflow from accidents of any type. Such a loophole could encourage increased construction of combination carriers, to circumvent requirements designed to protect the environment.

NOTES

1.  

Tanker Advisory Center, Guide for the Selection of Tankers, 1990, New York.

2.  

Clarkson Research Studies Ltd., 1990.

3.  

Clarkson Research Studies Ltd., 1990.

4.  

The Coast Guard estimates that an 11.5' (B/14.4) double bottom on the EXXON VALDEZ would have reduced oil outflow by 60 percent at most, and by a minimum of 25 percent— still a significant figure (U.S. Coast Guard, Marine Safety Center internal memorandum, May 25, 1989).

5.  

The committee estimated retrofit costs based on information from shipyard sources. These costs are highly dependent on the base vessel.

6.  

Assuming that cargo tank lengths and the ship's beam are the same as in the equivalent deadweight MARPOL standard.

7.  

The committee was shown videotapes of small-scale tests using a plexiglass model of one IOTD w/DS proposal, in a simulated low-energy grounding. While the results were impressive some committee members remain skeptical about actual performance under a wide range of operating conditions. Of special concern is the dynamic condition of a vessel running aground at service speeds (14 to 16 knots).

8.  

On the average, it would be applicable to about 40 percent of accidents and 60 percent of outflow volume in U.S. waters—those events where major grounding damage and spillage (more than 30 tons) have been incurred (see Figure 1-11).

9.  

Additional parameters are detailed in Appendix F (pages 252 and 278).

10.  

Committee estimates for design combinations were derived using DnV data in the combined ranking values. The value in the estimates for collision performance for one design is added to the grounding performance for another alternative, producing a composite estimate for a case not shown directly by DnV. For example, to generate an estimate for double sides with hydrostatic control at 5 knots, the committee combined the performance in collision of the double-side case 3B with the grounding performance of the hydrostatic control case 9, to generate a combined ranking of about 32. A more complete explanation can be found in Appendix K.

REFERENCES

Clarkson Research Studies, Ltd. FAX to D. Perkins, National Research Council, Washington, D.C., August 31, 1990.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×

Det norske Veritas. 1990. Potential Oil Spill from Tankers in Case of Collision and/or Grounding: A comparative study of different VLCC designs. Report conducted for the Royal Norwegian Council for Scientific and Industrial Research, Oslo. DnV 90-0074.


Lloyd's Register of Shipping. 1989. Maritime Overseas Corporation 64,000 DWT Product Oil Carrier Sloshing Investigation. Report prepared for MOC, New York, October 1989. CSD 89/33.


Ministry of Transport-Japan. 1990. Prevention of Oil Pollution. Report prepared for IMO Marine Environment Protection Committee, received by Committee on Tank Vessel Design, NRC, Washington, D.C., November, 1990. Toyko.


Tanker Advisory Center. 1990. Guide for the Selection of Tankers. New York: TAC.


U.S. Coast Guard. 1989. Marine Safety Center internal memorandum, May 25, 1989.

U.S. Coast Guard. 1990. Navigation and Inspection Circular 2-90. Published by the Coast Guard, Washington, D.C., September 21, 1990.

Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 101
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 102
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 103
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 104
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 105
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 106
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 107
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 108
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 109
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 110
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 111
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 112
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 113
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 114
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 115
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 116
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 117
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 118
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 119
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 120
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 121
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 122
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 123
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 124
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 125
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 126
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 127
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 128
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 129
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 130
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 131
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 132
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 133
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 134
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 135
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 136
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 137
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 138
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 139
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 140
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 141
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 142
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 143
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 144
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 145
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 146
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 147
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 148
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 149
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 150
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 151
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 152
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 153
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 154
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 155
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 156
Suggested Citation:"5 Design Alternatives." National Research Council. 1991. Tanker Spills: Prevention by Design. Washington, DC: The National Academies Press. doi: 10.17226/1621.
×
Page 157
Next: 6 Benefits and Costs of Design Changes »
Tanker Spills: Prevention by Design Get This Book
×
Buy Hardback | $39.95
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Can we design an oil tanker that meets our complex demands for environmental protection, economical operation, and crew safety? This volume evaluates and ranks a wide variety of tank ship hull designs proposed by experts around the world.

Based on extensive research and studies, the book explores the implications of our rising demand for petroleum and increase in tanker operations; U.S. government regulations and U.S. Coast Guard policies regarding designs for new tank vessel construction; how new ship design would affect crew safety, maintenance, inspection, and other technical issues; the prospects for retrofitting existing tankers to reduce the risk of oil spills; and more.

The conclusions and recommendations will be particularly important to maritime safety regulators in the United States and abroad; naval architects; ship operators and engineers; and officials in the petroleum, shipping, and marine insurance industries.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  7. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!