all plants licensed to operate, NRC has found that routine operations will have no measurable radiological impact on any member of the public. Therefore, the object of the experimental policy is to develop safety goals that limit to an acceptable level the additional potential radiological risk that might be imposed on the public as a result of accidents at nuclear powerplants.

In establishing the numerical guidelines to support these safety goals, NRC noted that progress in developing probabilistic risk assessment (PRA) techniques and in accumulating relevant data since the 1974 Reactor Safety Study (24) has led to recognition that it is feasible to begin to use quantitative assessments for limited purposes. However, because of the sizable uncertainties still present in the methods and the gaps in the data base-essential elements in gaging whether the guidelines have been met-NRC indicated that the quantitative guidelines should be viewed as goals or numerical benchmarks that are subject to revision as further improvements are made. Many of the participants in the Safety Goal Workshops held by NRC agreed that quantitative goals were not feasible at this time, but numerical guidelines could be used to support qualitative goals. Finally, in setting the numerical guidelines, NRC specified that no death attributable to a reacor accident ever will be "acceptable" in the ense that the Commission would regard it as a outine or permissible event. NRC intends that o such accidents occur but recognizes that the possibility cannot be eliminated entirely.

With these caveats, NRC established four ex›erimental numerical guidelines: two for individial and societal mortality risks for prompt and lelayed deaths; a benefit-cost guideline for use n decisions on safety improvements that would educe those risks below the levels specified in ccordance with the longstanding regulatory priniple that risks from nuclear power should be "as ow as reasonably achievable"; and a plant perɔrmance guideline that proposes a limitation on n the probability of a core melt as a provisional uideline for NRC staff use in reviewing and evalating PRAs of nuclear powerplants. These guide

nes are:

The risk to an individual or to the population in the vicinity of a nuclear powerplant site of

prompt fatalities that might result from reactor accidents should not exceed 0.1 percent of the sum of prompt fatality risks resulting from other accidents to which members of the U.S population are generally exposed.

The risk to an individual or to the population in the area near a nuclear powerplant site of cancer fatalities that might result from reactor accidents should not exceed 0.1 percent of the sum of cancer fatality risks resulting from all other causes.

The benefit of an incremental reduction of risk below the numerical guidelines for societal mortality risks should be compared with the associated costs on the basis of $1,000/man-rem averted.

The likelihood of a nuclear reactor accident that results in a large-scale core melt should normally be less than 1 in 10,000 per year of reactor operation (25).

In its experimental safety goal proposal, NRC left open a number of questions for future consideration. These include: whether the benefit side of the tradeoffs should include the economic benefit of reducing the risk of economic loss due to plant damage and contamination outside the plant; whether a numerical guideline on availability of containment systems to mitigate the effects of a large-scale core melt should be added; and whether there should be a specific provision for risk aversion and, if so, what it should be. In addition, the proposal sought further guidance on developing a detailed approach to implementing the safety policy, including decisionmaking under uncertainty; resolving possible conflicts among quantitative aspects of issues; the approach to be used for accident initiators that are difficult to quantify (e.g., seismic events, sabotage, human and design errors); the terms for definition of the numerical guidelines (e.g., median, mean, 90-percent confidence); and identifying the individuals to whom the numerical guidelines should be applied (e.g., the individual at greatest risk, the average risk).

Shifting from prescriptive regulation to a safety goal approach could have far-reaching consequences. Such a change might contribute to a more favorable regulatory environment for the nuclear utilities since the number and unpredictability of regulatory actions probably would be reduced. Furthermore, utilities would be allowed

to select the least costly route to compliance, with resultant gains in efficiency. Another result of the safety goal approach might be to encourage diversity and innovation in developing alternatives for improving safety. Such activities, however, may not be consistent with the standardization of nuclear powerplants (4).

The proposed safety goals and numerical guidelines are not free of controversy. The proposed guidelines have been criticized as being "too remote from the nitty-gritty hardware decisions that have to be made every day by designers, builders, operators, and regulators to be of much use" (25). Most regulators and industry representatives agree that while, in principle, it would be nice to be able to use overall goals to supplant the myriad specific decisions NRC must make about the adequacy of hardware and procedures, they find the proposed goals too general and abstract to provide specific guidance for dealing with practical questions, and withhold judgment on whether they will prove useful. As one Commissioner noted, the only reliable guides to reactor safety remain time-tested engineering principles:

redundant and diverse means of protection against core damage, sound containment, sufficient distance from populated areas, effective emergency preparedness, and, of course, careful attention to quality assurance in construction and operation. To provide guidance to the NRC technical staff and the nuclear industry, and to inform the public, the Commission should distill its experience and state clearly and succinctly that each of these [engineering] principles must be satisfied separately, and how this is to be done. Unfortunately the Commission seems to be on an opposite course (25).

The nuclear critics object more strongly to the safety goal proposal, arguing that to adopt goal: with no viable means of confirming their achieve ment is a useless exercise. They do not believe there is any immediate prospect of PRA being de veloped sufficiently to provide a means of con firmation. Therefore, the critics argue that it is no feasible to use quantitative guidelines for limited purposes, and NRC only misleads the public in saying that PRA calculations will be used to sup port qualitative goals.

Nuclear powerplant licensing experience in the United States, for the most part, is based on the LWR design concept. The exceptions are the Fort St. Vrain high-temperature gas-cooled reactor (HTGR), which achieved full power in 1981, and the Clinch River breeder reactor. Yet variations on the LWR and other reactor design concepts are attracting attention for their possible safety and reliability advantages over the LWR, as discussed in chapter 4. Given the extent of the licensing and regulation experience with LWRs, it is reasonable to question whether a shift to a different design would entail substantial changes in the regulatory process, such that the same problems encountered in the regulation of LWRs would be repeated with alternative reactors, and whether the development of a licensing process for such reactors would delay their implementation.

Small LWRs contributed greatly to the origina development of commercial nuclear power in th United States. However, as operating experienc grew, apparent economies of scale motivated uti ities to purchase larger reactors. Today the norr is over 1,000 megawatts electric (MWe), but ir terest in smaller reactors is reemerging, primar ly for financial and system flexibility reasons.. shift to smaller reactors could not be accom plished by replicating existing small plant because the designs of those plants do not mee all current safety requirements. NRC has estał lished a systematic evaluation program specifica ly to review these older designs and improve the safety where possible. New small reactors woul require new designs based on current NRC re ulations, although such designs would no necessarily differ substantially from large LWF except in the size of the core and other pla

Components. Thus the regulatory process probbly would be similar to that for current large WR designs, including the potential for backfits, inless small LWRs were standardized within the ontext of preapproval of designs.

The high temperature gas reactor has little perating experience in the United States. The rimary safety concerns are quite different from he LWR and have not been studied as intensivey. As a result, the potential for the emergence of significant unforeseen safety concerns probbly is higher than for the LWR. On the other and, inherent characteristics of HTGRs make hem less susceptible to certain types of accidents hat can progress more quickly or have more serius consequences in a LWR. This eventually may implify the licensing process after any initial roblems are resolved.

During the early 1970's, several utilities made P applications for HTGRs. As a result, NRC nade a significant effort to formalize design reuirements and establish review plans for the ITGR. Nevertheless, several years would be reuired to make the regulatory process for this esign as mature as that for LWRs. Backfitting reuirements for the HTGR are uncertain but hould be reduced through the operating expe

rience of the Fort St. Vrain plant and the 900 MWe prototype that DOE is sponsoring.

The heavy water reactor, as represented by the well-proven Canadian CANDU design, has attractive safety and reliability features, but licensability is a major constraint on the adoption of this design by U.S. utilities. The NRC requirements for seismic protection and thicker pressure tube walls would require design changes in the CANDU that might reduce its efficiency and could lead to backfits until these changes were proven. NRC would have to establish new design criteria and standard review plans before a heavy water reactor could be licensed.

The Process Inherent Ultimate Safety (PIUS) reactor is the least developed of the alternative design concepts. It has readily visible safety advantages, but they might not be accounted for in the regulatory process until significant operating and construction experience is established. If PIUS is forced to include all the engineered safety features of the LWR, it is not likely to be competitive. Successful development of PIUS, therefore, depends on NRC determining what level of safety is appropriate and crediting the inherent safety features of PIUS during the design approval and licensing process.

1. Cohen, S. C., Lobbin, F. B. and Thompon, S. R. "The Future of Conventional Nuclear Power: Nuclear Reactor Regulation," SC&A, Inc., Dec. 31, 1982.

2. Comptroller General of the United States, "Reducing Nuclear Powerplant Leadtimes: Many Obstacles Remain," EMD-77-15, Mar. 2, 1977. 3. Crowley, J. H., and Griffith,, J. "U.S. Construction Cost Rise Threatens Nuclear Option," Nuclear Engineering International, June 1982.

4. Fischoff, B., "Acceptable Risk: The Case of Nuclear Power," Journal of Policy Analysis and Management, vol. 2, No. 4, 1983.

5. Freeman, S. D., "The Future of the Nuclear Option," Environment, vol. 25, No. 7, 1983.

6. Hodel, D., statement before the Subcommittee on Nuclear Regulation of the Committee on Environment and Public Works, U.S. Senate, May 26, 1983.

7. Kemeny, J. G., "Report of the President's Commission on the Accident at Three Mile Island," October 1979, Washington, D.C.

8. Martel, L., Minnek, L. W., and Levy, S., "Summary of Discussions with Utilities and Resulting Conclusions," Electric Power Research Institute, June 1982.

9. "NRC Pulls the Plug on the Zimmer Nuclear Power Plant," The Energy Daily, vol. 10, No. 215, Nov. 16, 1982.

10. Office of Technology Assessment, U.S. Congress,

Nuclear Powerplant Standardization-Light Water
Reactors, OTA-E-134, June 1981.

11. Oregon Department of Energy and the U.S. Nuclear Regulatory Commission, Memorandum of Understanding, Jan. 4, 1980.

12. Patterson, D. R., "Revitalizing Nuclear Power Plant Design and Construction: Lessons Learned by TVA," October 1981.

13. Phung, D. L., "Light Water Reactor Safety Since the Three Mile Island Accident," Institute for Energy Analysis, July 1983.

14. Rogovin, M., et al., "Three Mile Island: A Report to the Commissioners and to the Public," U.S. Nuclear Regulatory Commission, NUREG/CR-1250, February 1980.

15. SC&A, Inc., "Nuclear Reactor Regulation and Nuclear Reactor Safety," Sept. 6, 1983.

16. Title 50 of the Code of Federal Regulations, Part 50.109(a).

17. Title 10 of the Code of Federal Regaulations, Part 50, Appendix E.

18. United Engineers & Constructors, Inc., "Nuclear Regulatory Reform," UE&C 810923, September 1981.

19. United Engineers & Constructors, Inc., "Regulatory Reform Case Studies," UE&C 820830, August 1982.

20. U.S. Department of Energy, "The Need for Powe and the Choice of Technologies: State Decisior on Electric Power Facilities," DOE/EP/10004-1 June 1981.

21. U.S. Department of Energy, "Nuclear Licensin and Regulatory Reform Act of 1983," draft bill sub mitted to Congress, March 1983.

22. U.S. Nuclear Regulatory Commission, "Nuclea Power Plant Licensing: Opportunities for Improve ment," NUREG-0292, June 1977.

23. U.S. Nuclear Regulatory Commission, "Nuclea Power Plant Licensing Reform Act of 1983," dra bill submitted to Congress, February 1983. 24. U.S. Nuclear Regulatory Commission, "Reactc Safety Study," WASH-1400, October 1975. 25. U.S. Nuclear Regulatory Commission, "Safet Goals for Nuclear Power Plants: A Discussio Paper," NUREG-0880, February 1982.

26. U.S. Nuclear Regulatory Commission, "A Surve by Senior NRC Management to Obtain Viewpoin on the Safety Impact of Regulatory Activities Fro Representative Utilities Operating and Construc ing Nuclear Power Plants," NUREG-0839, Augu 1981.