Review of NAS Report
In Response to H. Res. 602, Coast Guard Authorization Act of 1998 Section 423.
Double Hull Alternative Design Study

November 14, 2001

This review is submitted by MH Systems Inc, a marine systems design group that is in the process of testing under contract with Office of Naval Research (ONR) a cargo spill control system for single hull and double hull tankers. MH Systems believes that the future of its own proposed system, together with that of several alternative systems, is directly linked to the NAS report and to its interpretation and implementation.

We believe that the logical outcome of the NAS study would have been to set standards that have to be met, and then encourage research and development leading to equivalent or improved ways of meeting them. If these standards or criteria were met, then the "rulemaking" would have been set in motion for acceptance.

However, the outcome of the NAS study is different than what was expected, and what, in the opinion of MH Systems, might have hoped. Instead of setting standards, the NAS only identified criteria in which the standards might usefully be set, leaving the standard setting to the US Coast Guard. In essence the NAS produced a "road map" of detail design specifics, leaving the approval process of each detail design (not concepts) to the US Coast Guard.

We believe that the NAS report should be further expanded to define, in light of the identified guidelines, certain simplified, objective, measures by which the performance/non- performance of an spill containment system be objectively assessed. We also believe that the NAS report should be further developed (i) in which simplified criteria are developed that define system capabilities sufficiently to justify rulemaking and (ii) to provide a quantitative basis for formal regulations. This simpler approach would replace current practice in which these decisions are made subjectively.

Review of the National Academy of Sciences Report,
Environmental Performance of Tanker Designs
in Collision and Grounding

This is a brief review of a preprint copy of the Double Hull Alternative Design Study prepared by the National Academy of Sciences (NAS) at the direction of Congress (H Res 602, Coast Guard Authorization Act of 1998, Section 423). The review is submitted by MH Systems Inc, a marine systems design group that is in the process of testing, under contract with ONR, a cargo spill control system for single and double hull tankers. MH Systems believes that the future of its own proposed system, together with that of several alternative systems, is directly linked to the NAS report and to its proper interpretation and implementation.

Summary

The Oil Pollution Act of 1990 (OPA 90) was enacted in order to require protective measures against tanker cargo loss caused by collision, grounding and other incidents. The primary practical effect of this legislation is to mandate the use of the double hull in new construction. This is what the NAS refers to as a prescriptive approach. But the legislation also encourages the development of alternative systems that are equivalent to the double hull in spill prevention effectiveness.

Both the conventional and double hull tankers are currently regarded as acceptable concepts. It is only necessary that each individual design meet statutory technical requirements. New or innovative approaches, however, are not yet acceptable concepts, were not anticipated in formulating current requirements, and can require new rulemaking. And there are no rules under which new rulemaking can be initiated.

Since the enactment of OPA 90, a number of alternative systems have been proposed as alternatives to the double hull. Up to now, these proposals have been evaluated using arbitrary, essentially non-numeric criteria, or without any specific criteria at all. All proposals have been rejected, essentially by withholding the rulemaking process.

It is this situation that led the Congress to request guidance from the engineering and scientific communities in its review of the alternative systems contemplated by OPA 90. What was wanted was a set of criteria, which, if met, would lead regulatory authorities to set the rules under which a given concept could be implemented. It was recognized that the formulation of those rules, or the evaluation of individual designs, might require further detailed engineering or environmental study.

MH Systems believes that the primary focus of the NAS report is on the detailed evaluation of individual, specific designs. It is a roadmap for the final, conclusive approval of a specific design before it is placed in service, and is aimed at the development of rules such as would eventually appear as statutory requirements imposed before a specific design is approved for construction. This is an entirely reasonable objective, and the report is an excellent guideline for the development of quantitative standards.

The objective of the Congress is to advance promising concepts to the next stage of consideration. The double hull, after all, has its own vulnerabilities and, as the report shows, is sometimes less effective in preventing cargo loss than the conventional tanker. And it is costly. The development of relatively simple selection criteria would allow attention to be focused on any promising alternative concepts, after which detailed rulemaking would follow. Effectiveness and safety, ultimately, would be assured by statutory rules, generated eventually by a process evolved from the recommendations of the NAS study.

We believe that the process advocated by the report should be developed further to provide a quantitative basis for formal regulations. At the same time, however, we also believe that a parallel path should be followed in which simplified criteria are developed that define system capabilities sufficiently to justify rulemaking. This simpler approach would replace current practice in which these decisions are made subjectively.

Background

The Oil Pollution Act of 1990 (OPA 90) was enacted in order to require protective measures against tanker cargo loss caused by collision, grounding and other incidents. OPA 90 specifies that tank vessels trading into US waters must meet strict design and operational standards, and a schedule is defined for the phase out of nonconforming vessels. The primary practical effect of this legislation is to mandate the eventual use of the double hull or its equivalent in spill prevention effectiveness. But it explicitly states that alternative systems are to be sought and considered.

In the ordinary process of ship design and construction, regulatory authorities must use detailed criteria to determine the acceptability of a specific ship design. A formal design is represented by a final set of drawings and specifications. Criteria affecting the safety of life at sea are established, in the United States, by the U.S. Coast Guard. Many other requirements are imposed by insurance underwriting societies. For both conventional and double hull designs, it is taken for granted that the concept represented by the design will be found acceptable providing the applicable requirements are met.

No review process is provided in OPA 90 to support its declared intention to consider alternative approaches. In practice, the acceptance or rejection of concepts for further rulemaking has been essentially a subjective one. There have been no quantitative criteria. All of the alternatives proposed have been rejected. Rejection has taken the form of refusing to establish requirements at all.

It is this difference that led the Congress to request guidance from the engineering/scientific community in its review of the alternative systems contemplated by OPA 90. What was wanted was a set of criteria, which, if met, would lead the regulatory authorities to set the rules under which a given concept could be implemented. In other words, the objective was not to write the rules, but to define the conditions that would lead to the rules being written. This intermediate step seemed to be necessary because the responsible authorities have thus far declined to consider the rulemaking necessary for concepts other than the double hull.

The NAS report accurately describes the current interpretation of OPA 90 as prescriptive, in that it meets the primary objectives of the legislation by defining only a single conceptual approach that it finds acceptable. The alternative is to set standards that have to be met, and then encourage research and development leading to equivalent or improved ways of meeting them.

Study Objectives defined by Congress

Section 423 of the Coast Guard Authorization Act of 1998 defined the task to be accomplished by the Marine Board of the National Research Council. The objectives, briefly paraphrased, were the following:

  1. Develop a database which relates total cost of an oil spill to the amount spilled.
  2. Develop vessel damage estimators based on data and research
  3. Develop an environmental index to assess overall outflow performance
  4. Apply the proposed index to double hull vessels and alternative designs

The objective of these efforts was to develop a rational basis for evaluating spill-prevention alternative concepts as contemplated by OPA 90.

The NAS Report

The report has essentially three sections, evaluation of accident damage and outflow, determination of oil spill consequences, and the application of these two areas of analysis to a comparison between alternative designs. These three general areas are discussed below.

Damage and Outflow

The report recommends the use of the program, DAMAGE, developed by an MIT-Industry group, as the analysis mechanism for grounding incidents, and the program SIMCOL, developed by Virginia Tech, for the analysis of collision incidents. Each of these computational procedures was exercised 10,000 times, for a variety of conditions, to establish a damage/outflow spectrum for a pair of nominal vessel designs. The resulting outflow distributions were then converted to oil spill consequences.

These programs are detailed and comprehensive, addressing the structural environment and the expected range of operational conditions. They probably represent our best knowledge of these complex and difficult events, and we can offer no comment as to specific ways in which they might be improved. They have the additional advantage of being closely related to common naval architectural procedures.

At the detail design stage of a conventional tanker or a double hull tanker, these analysis programs could key directly into the many tradeoffs that designers have to make: structural features, tank arrangements and operational procedures. We note, however, that these analysis tools relate to the design process only at the detail development level. There have to be rather complete plans and layouts, and an established structural concept. In short, the analysis process adopted in the report is visualized as an analysis process conducted to validate final design closure and as an ingredient of regulatory approval for a specific design.

The problem with genuinely new or different concepts is that the required level of design detail is not going to be available, or may be available only for a limited set of generic designs. This does not mean that the recommended analysis cannot be performed. And it can be, once it is decided that the concept is sufficiently promising to justify detailed design. But some of the newer concepts are not seeking regulatory approval as such, but only a regulatory framework within which a detailed design can be developed. And our system, for one, cannot be operated under current detailed regulations. It is as if the Wright brothers had been confronted simultaneously with (1) a prohibition against powered flight without prior approval, and (2) an approval mechanism based on the assumption that only vehicles lighter than air could fly.

There is a solution that would be more supportive to innovation. We suggest that the conclusions drawn from the two recommended analysis programs, or any other promising analysis tools, be parameterized insofar as possible so as to facilitate their use on a conceptual analysis basis. For example, statistical distributions could be developed for the probability of hull rupture, the depth of rupture and its extent. The result would be a set of criteria that resemble those developed by the International Maritime Organization. The IMO standards are quite general, and can be applied without great difficulty to conceptual designs. Damage models of this kind could then be updated from time to time as further analytical or empirical data became available.

This generalization of the structural/outflow basis of comparison should be added to the other objectives of additional effort advanced by the report.

Consequences of Oil Spillage

This section of the NAS report attempts to define a set of measures that can be used to quantify the negative consequences of each oil spill incident. The basis parameter is the spill quantity derived from the damage and outflow determination described above. Consequences are developed as functions of spill quantity. The consequences studied are of widely disparate types: financial losses, spill response costs, denial of resources such as beaches, environmental damage, fisheries losses and others. To translate them into a composite measure, weighting factors must be used.

Figures 4-17 through 4-22 of the report show various fits to a postulated relationship between spill quantity and consequences. Because of the wide range of data variation, both in outflow quantity and in consequences, the logarithmic fit chosen for the report seems appropriate. The figures show straight-line fits of the form,

(Consequences) = (Spill Quantity)k

with the exponent, k, varying from about 0.30 to 0.45, and Figure 4-23 showing an exponent of 1.0, corresponding to a direct linear relationship between spill quantity and consequences.

To put this approach in perspective, note that a horizontal line, i.e., an exponent of zero, would be equivalent to the zero-outflow measure of effectiveness, that is, all outflow quantities impose the same penalty. So we have, except for a multiplier,

k = 0, Consequences are unrelated to spill quantity (the zero outflow criterion)

k = 0.3 to 0.45, the findings of the NAS study

k = 1, Consequences are proportional to spill quantity

The strength of this approach is that it addresses in some way essentially all of the significant negative effects associated with tanker incidents. Anyone viewing the analysis can see that his interest is represented in some way. The weakness is that the weighting factors are, in the final analysis, arbitrary. The report attempts to define the effect of this uncertainty by carrying out sensitivity analyses for some of the components. Figure 4-22 presents confidence limits within which variations in the calculated consequences range from ten to twenty to one. Probably the actual uncertainty would be much greater if full expression were given to a wide diversity of viewpoints.

At MH Systems, we feel that the work cited as a basis for the analysis is probably the best that can be had, that the form of presentation is appropriate and that comparisons can be drawn, at least in the sense that differences can be demonstrated. However, it is obvious that there is much uncertainty in this calculation. For the conduct of design level tradeoffs, e.g., comparisons among similar tank arrangements, this sort of criterion can be useful and appropriate. But for a major policy decision, such as a choice between two totally different approaches, the uncertainty in the consequences calculation is probably of at least of the same order as the differences in the systems themselves. Our calculations, admittedly much simpler and using only spill quantity as the selection criterion, show a relatively small variation in calculated system effectiveness over a wide range of design concepts.

We feel that to serve the intent of Congress, to create a basis for comparing alternative system concepts, the consequences calculation should be developed further if possible, especially in regard to the weighting factors. The model should then be simplified and standardized as a single function or perhaps as a very limited number of functions. The resulting index, or indices, would be used only to identify promising lines of analysis and design effort, subject always to the necessary rulemaking and design oversight.

References to Active Systems

The report refers to specific life-cycle risk analysis measures to be instituted for active systems. Risks are of two kinds. The first is the risk that the spill-prevention mechanism will not be operational when called upon. It is certainly true that a passive structural arrangement can be relied upon to remain in place, while an active system can sometimes fail to function. The criterion in this case is simply what is often termed availability, the probability that the system is "up", and is a consequence of design, design redundancy, component selection, maintenance policy, training and similar factors. A simple regulatory declaration that availability shall exceed some specified value, and shall be demonstrated as a condition of approval for each vessel proposed for construction, would be a sufficient guarantee of effectiveness. Such requirements are commonplace in military and other technical procurement. And formal methods exist for evaluating them. This solution is obviously much more logical than a refusal to set such a requirement on the grounds that it might not be met.

A second type of risk is that the proposed system might introduce a new risk mechanism, such as fire or some other collateral event. There is already an elaborate structure of requirements governing such risks under the Code of Federal Regulations, and additional requirements could be added as necessary to reflect any unique features of a new system. This solution, too, is much more logical than a refusal to set such requirements on the grounds that they might not be met.

In the specific case of the MH Systems spill prevention concept, it is indeed an active system and is indeed subject to component failure. But it uses only standard marine components, and primarily in standard ways. Establish a requirement that availability must be at least some minimum value, and let the requirement be comparable to those applied to other shipboard systems such as inert gas generation, collision avoidance and cargo handling.

Applicability to Retrofit Systems

In mandating the eventual use of the double hull, a system virtually impossible to retrofit, OPA 90 attempts to address the problem of the existing fleet by providing a phase out period. However, there is a significant period of time before the phase out can occur even if it is not extended. In the meantime, the world is exposed to the single hull disaster potential that the double hull ameliorates. There is also the possibility of retrofitting improvements that are effective in preventing or limiting cargo loss. Improvements gained in this way would represent an environmental dividend that would be separate and distinct from the eventual gain realized by the use of the double hull. Also, in the case of our system, we can show that it can further improve the effectiveness of the double hull.

Your further inquiries are invited. Write to: Corporate@mhsystemscorp.com

Copyright 2009 MH Systems, Inc. All rights reserved. | Terms of Use | Last Modified 11/17/02