ML110030901

Exhibit Pwa 00014, Kamiar Jamali, Use of Risk Measures in Design and Licensing of Future Reactors

ML110030901
Person / Time
Site:	Pilgrim
Issue date:	01/03/2011
From:	Jamali K Elsevier, US Dept of Energy, Office of Nuclear Energy
To:	Atomic Safety and Licensing Board Panel
SECY RAS
Shared Package
ML110030897	List: ML110030898 ML110030901 ML110030903 ML110030907 ML110030908 ML110030909 ML110030910 ML110030911 ML110600894 ML110600896 ML110600897 ML110600899 ML110600900
References
RAS 19369, 50-293-LR, ASLBP 06-848-02-LR, PWA 00014
Download: ML110030901 (7)

Text

PWA00014 Pilgrim LR Proceeding SO-293-LR, 06-848-02-LR Kamiar Jamaili -PRA's Subject to Uncertainties Co n te n t!l-I i'St-g, .av a i I BIble .at Sde.n ce DI ReHabi l ity Enginee,.-ng and System Safety j.o u r n a l h o m.e p.a g.e: 'W'Ww.e l:S<e vl e r.oo m/l oca t e/r es.s. Use of r i sk measures in des i gn and li cens i ng of fu ure r eactors K amiar J amali Unit.a:::l Srars nq,aRm rnt Of for of ll ucmr .f.nagy , J ooo W ammg;DD. DC 2lJ.s.as. IN F O ABS T R AC T ... .rDd!!-h imlry:

5 April 2I[J 1 U m in.c 1 10 A p ril :2il l () "'.,....,.,.., iN 1U d-t!D.T!fIe.KJtm" S.J.: R!tl' P'.rdI:ab iti stic: risk CP U) J't.cc.cpt;mCc a.iltcria tNext :gener.artia 11 nJdea:r pla: n t rm odu w C1I:n.'II!! lise o 1'I rr t:o. m.ad o n aBd j n:sigtn: s fr o.m , p nna h l li:5l:k:: li s..l , ase:5f , rnems. (P I i:A5.) 1 111 f'eacro r :§.'l f er y .i!l p p l l cad o n:s. h.as. been I r.cre.a: s:.ing by tt.e nlide.a r lOOliS'tfy aDd! '!:he boI: h do:m.esdc.a U y aDd! If llte r-n.a t T hi s. ls: .i!I dies JI'.i!l b J e ueBd. a s. flaw demon:s: t r.are d. ca , p.ab lli cy t o. I ID p mve s.a1e'ly aOO o , pera t l O I\aJ t1ex;l b il Jty beYDnd '[ ha t p rovi ded 'I:hmt,tgtlt derennin l sa: k:

aJ o...e.. B u t [I.e re c.a n be pt)'tef1 i[.La I , pi rf a li s.. l 111 e I i m l tatloilS o t' fi s.k ass.es.s.me m tee be m..t t hi rough l ap , pr oadhes tha't A:!, l y 0., 0 q ua r llt l ta[Ne P I tA , re:5U l ts (r efe r red! to as r.t s.k , measu r es , I n [h is .I laj)ef'). bec.a1Hoe o.'f t , Be L IIR.i!I nb l gllous b t U , por.e , l u l aJ ty rnis.l e.a d ing , mess.ag:e

'C l u t caR be diet l w , re d by n s;k;.ba:s.ed r mmbe rs.... T h is Is , p.ar d Cli I.i!J Cl y u-uefo'r f llrut'e 1<R As..are IiSedI d l1 d ngt:t.e deS Jg;R.a , OO I ice R S..l B8 P roces.ses..

FO' tt.e:se .a , p.P-=.a'[J O BS".It I:s. lmpo r taB't 'to ensure t hr.a t 'tile de fa cto.. (If' even peroet\red lISe o:f rt S..K Il'1le.1sure s.l n '[I.e o.'t'etciler .acce p't.anoe crt t en.a I s..a 'de d.. W h il e '[Be iss-lN!S d iscusse d Ie ca n Il ave a s Jgn i lica l u iIl'lillle nee on design ce rt i1ica t l o.n o.r co mb i ned! li cense .a pp.li ca d o.BS. f, flft1i r e re.act o , ts.. 'they ca n al so , BaW :SeCO.Bd!.a

, ry impa ctS 0., 1} cu r re n l1 y o ,J)ef'.i!l t Ing reactot"S...

1 In t n xl: uct:ion Pro ba.b ili s t i c ri.s k .a.sse.s.s me:n t (PM) res ults .an d in.s i g h ts h ave h e l ped ID imp ro ve nu d e.a.r po.......e r pl a n t s a f e t y a n d o pe: r a ti o n a J n e:x i b iJ i ty f or m o r e t ha.n .3 0 T hi.s s u a:e:s.s h as k!d , to i_n c m.a.sed u se of P RAs by t h e n'uc: l e.a: r i nd u st:Jy a n d reg ul at o ry a u t ho ri t i e.s 1NO rl dv.r i d e.. W hj I e t.h is t:r 1!n d is l arg e ly pcs i ti ve.. t h e re ca n be: po t e n t i a l n e g.a: t i ve co n seq ue n ces t h a't h a ve n ot b ee n \<'Vi d e l y d i.scus.sed in. r e l ated Ii tera t: u re:. "W i th :so m , e e::x£I!P'ti o ns (e..g... (1 1)" I t "W as beca u s e o f t hi.s po s iti ve (D J]t ribu tio n 1D s af e ty th at th e U S N: u c Je.a r R egu l a'l n ty Com m is s i o n (NR C) g r ad u a ll y r e: fi ned the:: i r o r ig in a J det e rmin_istic-ba.ser::)

J]u d e B , r .sa..fety b y i n c o r po r at: i n g 'the u se o f ri.s k in fo rma.'t i o n a n d in.s i g h t s "Wi't hi n a ri.s k-in fo rrn ed fr.a.m ev."CJ rk. R is k-inf o rm ed r egulat i o n.s fo r th e c ur re n t f.l.eet of o p e-r a tin g l igh t-"Wa'lE r r eact o r s (L W Rs) are d e fi ned t hr o u g h a co m b in atio n 0 f rul e-m a kin g a nd pub Li ca t i o n o f 1 0VW!r-t i e r doc u me n ts, s u c h as r egu l a'lD ry g ui d es o r NR C's e n d o r se:: m , e n t o f ce rta in n uc l ea r in d u s try docu m ents. Thu s. in a ri.s k-i n Jo rrn ed fr a m ewo r k., ri.s k inf o rm.a d o n a n d in s i g b ts s u p m , e nt t h e tr ad i ti o n a l det e rm in i s t i c a p pro ach e.s a n d f o rm a pa rt o f t h e 0 ve: r.a.11 .sa..fe:ty ca.se ("W b.i c h i.s s o m eti m es re f e rr ed ID as th e safety ba.s is) f o r a n'u d e.a r pl a nt. Th e Co m mi.ss i o n h as aJ.so ca l f o r i.nc::re.a.seCI u se o f P RA tec h no l ogy in a ll r egu l aID ry in a m a n_n e r t ha.t co mpl e me n ts NR C's p red o min a n t l y det e rmini stic: ap p roach , e.s "W i th.i n the (D nfin.e.s of a ri.s k-i n_f o rm ed as 0 pposed ID a £.mail .adclress.:

bnxia T () l..8:!20 tt .. fro m e lT.l.l itlE!' P:u h1 iish.ed by lA:d.. d eli: 1 U.l () 1 6/j.Jrcss.2lO l (u)4.00 1 l:oub ll s;t.edi by Else vier LId.. rl.s k-ba.sec::'I reg ul a'lD ry co n.st ru ct. So m e 0 f 't b e d is ti n g ui.s h_in.g feat u res bet:wee: n 'the: 'tv.'O are a l.s o d i..scu s.s ed in 't bi.s pa pe;L T he: n uc: l e.a r in dustry aJ.s o has u.seCI PRA tech ni q u e.s extE: n s i ve: l y "W i t b b e n e: fi , d a l res ul ts.. i nc l ud in g i o n 'th e des i g n o f advanc ed o r evo l u ti o n a ry nu d ea r Thes e be n e fi ts a r e:.. in part.. l1!.I.a ted ID t h.e f.ar:t 't h at 'these s.arn.e u.se: r s ca n a l.so co n tro l a n d lim.i t'the: infl u e nc e o'f th e in (D m p l ete sa f e t y inf o rma t i o n t h a: t i.s p rov i ded th o ro u g h th e r es u I ts o f the: P RA a l o n e.. F acto r s 't h a't are u s u a ll y no't full y ac co un ted fo r in a P RA m , ode: 1 but are g e rma.n e to 'the: (D ns i d e r a ti o n o f adeq u acy 0 f s af e ty fe.a.t u res fo r a s p e c ifi c i.ss\E o r aa:: i.d e n t scena ri o ma y i n d:u de:: m.agn itu de:s of r e eva n t .sa..fety ma.rg i n.s. i n CD r poa r at i o n 0 f d e f e n.se i.n doq:: r t h. pote n t i a l f o r c o rr ecti ve 0 r co m pe;ns a'lDry att i o n.s. degree o f c o n.se: rva ti.s m in a n a l ys i.s. a nd m.a n y o'th e r s.. Th e very sa m , e P RA in fo rm atio n.. h owever. w h , e n us ed t o (D mpl y "Wi t h -we.11-in't e: n t i o n ed l'1!g ul at o ry JX>l i cie:s a n d a ppr oa c hes ca n l e.ad to so m e un des ir ab l e (D ns eq u e nc es.. So m e o'f th e u n des i_r ab I e (D ns eq u e n CE.5 in a pp U catio n.s i_n vo l v i.n g fu tur e re.actD r'S a r e a l so d i.scussed 'be l ow PRA s prov iCI e bo t h q u a li ta ti ve a n d q u a n'titative:

inf o rm a ti o n_ Recent: 'tr e n d s in t he: de:v e kJ pm e n t: o f n ,eN ri s k-re l a t ed a p pr o ac h es.. "W he:t h.e r 't h ey are pe r f o rm ed b y 'the: r egu l aID ry sta ff. nu clea r in d 0 r o'th e r do m esti c o r i_n te: rn a ti o n a l 'bo d i es.. are tl:JWards heav i e r e mph.a.s i.s in u se o f q u a n t it a: t i ve P RA r es'ul ts (in't e n:.h.a nga bl y re.f erred 'to as'" r is k m eas u res'" i.n t.hi.s pape r).. I t i.s ""'e.U-kn ow n th a t q u a n'ti tat:ive r es ults o f P RAs, in a r e: s u'bj ec t 'to va r io us types o f un ce-: r t a in ti es.. Exa mp l e.s of 'th ese 'u nce:rta..i nd e.s in cl ud e pr o b a b ili st i c q u a n't ifi.c.a: ti o n o f s in g l e a lX1 (Dc a: u.se h.a r d"Wa r e o r softwa r e fa i I u r es.. o a: u rr e n ce o f I] ph y s i c.a.1 ph e n o m.e: n a.. hu m.a n e r ro r s o f 0 m_iss i o n a n d c o m_m.i ss i o n" magn itu des 0 f SO U r<l! t e rms. r.ul i o nu d i de r elease a n d tr a n spert. atmos p he ri c dispers i o n, b i olo gi ca l elfucts o f r ad i a ti o n , dose ca l cu l a ti o n s. an d m any o th e r s. U nli k e d<!l:e rmini stic un a.rta in t i es re l ated to ph ys i ca l ph e n o m e n a (e.g.. ne u tro n ics:, t h e rmal-h yd r a uPRII un a.rta in t i es aR! n ot R!.ad il y r ed u cib l e in m ost in stances. Uncertai nti es associatm w ith ph ys i ca l p he n o m e n a ca n 0 FtP.n b e red u md by tests:, expe ri m e n ts:, 0 ""ra ti n g expe ri e nc e o n act u a l o r p rototype des i g n s. o r imp rove m ents: in a n a l ytica l m odels o r ID mputa t i o n a l c a pa b i l iti es. Des pi te th.i s we ll-kn ow n limi tatio n. if q uan t i tat i ve PRA r esu l ts: are u sed in th e ID n text o f r is k .aa:epta n<l! cr i te ri a (c e., w he n t h.ey a r e co m pa red aga i n st a set o f thres h o l d va l u es estab I is h ed by e ith e r the i nd u stry o r th e r egu l aID r). it wo u I d be d i Ffi c ul t to ID un t e r the un a m b igu ous b u t pot e n tia l ly m.is l ead i ng or i n ID rr ocl: m essage th at i s by such a num ber-b ased pro<l!55; i..e.., impl y ing th at a des i g n i s un aa:eptab l e o r unsa f e because i t d i d n ot m ee t a pa rticu l a r r iskbased num e ri ca l t h res h o l d (l ab eled as a ri s k .aa:epta n<l! cr i te ri o n). An i m pe rtan t i ss u e th at i s o u ts id e of th e SID pe o f th i s paper. b u t i s wo rth y o f deta il ed d i scuss i o n s o f its ow n , is t h a t th e in trod u ct i o n a n d i m(>act 0 f P RIIs: i nth e des i g n a n d ng stilg es fo r a fu ture reacto r i s by a n d l arge d iff ere n t fro m th e way t ha t r is k-in fo rm ed r egulatio n s h ave b ee n a (>pli m to exis tin g reacto r *. C u rre n tly oper.lting R!iLClD r s h.ad a dem rmini stica ll y li a m s in g basis (w hi c h in d u ded t h e plan t's safety bas i s) b e fo r e pl a n t-s p eci fi co r g e n e ri c ri sk inf o rm atio n and insigh ts: we r e m ade ava i l ab l e th ro u g h P RIIs. T he P RIIs: gene r a ll y co n f irm ed t h at th e o ri gi n a l deter m.i ni stic a p pr oac h to des ign a n d Ii a.n s i n g was co n serva ti ve (e.g.. plan ts co uJ d respe n d ID so m e a ccide n t sce n a ri os in m a nn ers t h at we r e n ot cred i ted in the deter m.in i£ti c a n a l yses) a n d furt he r i de n tified cha n ges th at co uJ d impr ove pl a n t des i g n o r oper.lt i o n a l safely. M ee ti n g th e dete rmini st i c r eq uirm e n ts m ea n t t ha t impl e m e n tatio n o f th e ir atb!n da n t pr ovis i o n s e m bod i ed w i t hin the IDnceptso f defense in de p t h. safety m a r gi ns. co n serva ti ve ass um pti o ns a n d an a l yses. q u a l i Iy ass ur a n ce, a n d nu mero us ot h e r facto r s (man y of w hi ch are n ot R!.ad il y m eas urable w i t hin a P RII m ode l) created a safety cus hi o n o r m arg in t ha t pr oto::ted t h ese pl a n ts f ro m u ncerta inti es. i nclud in g th ose f ro m " un k n ow n unk noWIlS" (fo r w hi ch a e u ph e mi s m ca n be "e m e r gi n g safety as d iscuss:ed in Sec ti o n 1 O n th e o th e r h a nd , P RA m ode l s h.ave to R!l y o n r ea li stic inpu ts: to e llS ure t h at ri sk s i g nifi ca n t in s i g h ts a r e n ot obscu r ed by a rti f i c i a U y b i ased res ul ts: de r ill'ed fro m the a p p n 0 f un eve n ID llS e rva ti sm5. Th emfo r e, great c aR! mu st be exe r cised in b rin gi n g P RAs in to t h e des i gn p racess to e n su re t ha t t h e fu nda m enta l pi U a r s o f deter m.i n i£ti c sa f ety as:su r anc e process m e n t i o n ed above a r e not u nd u I y co m pro mi sed. Thus , f o r futu r e r eacto r s. 'U se 0 f ri sk in fo rroati o n c a n ha v e a fa r m ore si gn i fica n t im (>iLct o n th e safety bas i s o f th e p l a n t, in d u di n g the pete nti a l ID d ri ve so m e key des i g n decis i o ns. Th e in t e n t of ri s k-in fo rm ed R!gu l ations i s to ens u re t h ei r influ e nc e i s pas it i ve in sa f ety t r adeo f f do:: i s i ons. N: R C pu b l i s h ed t h e Safely Goa l s P o l i cy Staterne n to n A ug ust B , 1 986 W hil e t h e tex t o f t h is I'll I icy stat l!!Tl e n t does u se th e p h laSe " aca.p t.m I e ri sk," the ti t l e and th e r est o f th e d i scuss io n s were c a r e ful to avo i d th e u se o f th.e Q,Ja nti ta ti , ,,, Hea lth Objectives (Q: HOs) o f pr o mp t fata liti es (p r s) and l ate n t cance r fa t a liti es {[L r s) as R!gu l a t ory ri sk--.aa::e p ta n<l! c rit e ri a. In o th.e r wo ms. the selectio n o f th e t l'rm.i no l ogy o f " safely goa l s-was very de l ib e ra1E. A n im perta nt attr i bu te o f t h e ca l cu l atio n of pl a ntspec ifi c P Fs a n d las fo r ID m pa r oo n w ith th e d u a l Q H Os: is t ha t bot h are by n e<l!55 ity " in teg r a r" q u a nti t i es t ha t a r e de ri ,"ed fro m the ID n t ri bu ti o ns of a ll a a: i de n t scena ri os t ha t a r e co n s i dered in the pl a n t-spec ifi c P RA model. Th e Com m i£s i o n '5 1 005 P RA I'Il I i cy State men t 0 n u se 0 f P RII m et bads in nu dea r r egu l aIDry act i v iti es [3 1. w hi ch was i ss u ed in t h e aFtP.rm a th o f th e ID mpl e ti o n o f P RIIs fo r a ll o"" r at ing nu dea r pl a n ts in aa:orda n<J! w i t h th e I nd i v i d u a l Pl a n t E xa mi na t i o n s Generic lett e r l 41 states. in (>a rt: Th e u se o fP RII tec h no l ogy.ba ul d be in creased in a U R!gu l atory m att e r. ID th e exb!n t s up perl:l!d by t h e state-af -the-a rt in P RII m e th ods a n d data and in a mann e r th.at co mpl e me n ts: th e N: R C's deter mi n i stic ap proach a n d s u!>p<lrts t h e N: R C. tr.ld i ti o nal defe n se-i n-<l e pth p hiJ oso p by. Th e Ca m mi ss i o n's safety goa l s f o r n uc l ea r power pl a n ts and s u b s i d i ary num er i ca l ob j ectives a r e to be used w ith a p pro pri ate co n s i de rati o n o f u nc e rtain*ti es in m aki n g r egu l atory ju d gm e n ts: a n th e n em f o r p ro p os in g and ba c kfi t ti.n g n ew ge n e r ic req uj 0 n n u dea r pav.", r pl a n t li a.n sees. Th.e Co mm iss i o n a (>p rm"ed th.e staff's W hi t e P ape r o n Ri skrn fo rm ed a n d P e r fo rm ance-Based Regu Ia tio n in March 1 900 [5 1 , w hj c h prov i ded defi ni t i o n s of ri sk-in.fo rm ed a n d r isk-based r egu L a ti o ns. I t R!i t e r ates t h at the Co mmi ss i o n does not e n do r se a n a (>p roac h t h at i s r isk-basm , w h erei n decis i o n-m a kin g is so l e l y based a n the num erica l r es ul ts o f a r is k assess m e n t. Re gul atory G u ide 1.174 1 6 1 est;Ib li s h ed the fr a m ewo r k fo r ri s k-in fo rm ed r egu l a ti o n s in a ppli cat i ons r<gam in g m ak in g pl a n tspeci fi c c h a ng es to t h e I i<J!osi n g basis. I ts ap pr oac h e n s ur es t ha t num e ri ca l P RA r es ul ts wo ul d n ot f o rm th e so l e bas i s f o r m aki n g nu dea r .afety do::i£i o n s by Ii sti n g fi ve key p ri n c i p l es (L e., m ee ti n g cu r R!l1t regu la tions l w hi ch a r e prim a ril y dete rmini st i q. m.ee ting defe n se-i n"4le pth p ri nc i pi es, m a in ta i ni n g .uffi de n t safely m a rgi n , kee pi n g i nc r eases in ri sk. mal J, a n d pe r fo rm a n<l! m.o ni to r ed) th at h ave ID b e m e t f o r a ri sk-i n fo rm ed a p proac h_ Clea rl y , c urr e n t r egu l a ti o n s are by a n d l arg e based 0 n dete nn inj stic req u ir e m e n ts. A key pert i o n of th e sect i o n a n SID pe (Sectio n 1.4) states: Th e N:RC h as chosen a m ar e rest ri c ti ve po l i cy th at wo u Id pe rmj t o nl y s mall i ncreases in ri sk, a n d the n o nl y whe n i t i s r easa n ab Iy ass u r ed. a m o n g ot h e r thj n gs, t h ats uffi d e n tdere: n se in dept h and s uffi c i e nt m a r gi n s are m a i n ta i n ed. Thj s pe l.i q i s ado p ted because 0 f un ce rta i nti es a n d to acID un t f o r th e fact t ha t sa f ety i ss u es co ntinu e ID e m e rg e r egardi.n g design , co n s tru ctio n , a n d 0 perat i o n a l m atte r. n otw i thsta n d i n g th e m a turi ly o f t h e nu dea r power in d u stry. Th ...... fac1D r s s u ggest t ha t nu clea r pcwer R!.aclD r s s h o u I d operate ro u ti n ely o n I y at a pr ud e n t m arg i n abo ve adeq u a t e pro t ectio n. Th e safety goa l s u bs i d i ary objectives are u sed as a n exa mp le o f s u ch a pru de n t m a r gi n. Th e cla u se abo u t ID n ti nu a l e m e rg e n<l! o f sa f e ly i ss u es fo r pL a n ts w ith m a n y yea r s o f ope r at in g exper i e n<J! i s a n a l ter n.ative way to state the ID nce rn rega r d i n g u n a.rt a in t i es abo u t th e " unk now n th at a r e a m are s i g nifi ca n t co n<J!rn fo r f uture R!.a clD r designs. O n e R!iL5o n th.at RI!gu l at o ry G u ide 1.174 h.as wo r ked we U in i s th at it was in te n ded fo r operati ng plan ts w i t h a prim ar il y dete rmjn is ti c bas i s a lr eady in p lace, w hj ch m ea n s t ha t th e pl a n ts ",,,, r e a l ready dete n:nin ed to b e safe: before a pp ly ing th e res ul ts o f pl a n t-speci fi c P RAs. ri.nall y , N'otel o f C hap t e r 19 o f the Sta n dam Rev i ew Pl a n (S RP) 1 7 1 states th at t h e Q:H:Qws u r regates 0 f OJ r e Da m ag e F R!Q u ency (CDf) a n d larg e Release Fr eq u e n cy (lRF) a r e goals a n d n ot r egu l aID ry r eq u i r e m ents. Th.e key co n d usio n fro m th e: above i s t h at th e N: R C Co mmi ss i oners ha ve n ot e n do r sed a " ri sk-b ased-a pp rnach to regu l a ti o n because of the u ncerta inti es in q u a nti tative R!S ul ts: o f P RAs.. Th.ese u n a>rta in ties a r e l a rg e f o r cu rr ent l y 0 pera ti ng nu dea r pl a n ts , pa: rticu larl y in the l.e!J'eI 2 and leve l 3 P Rlls. Th e fact that t h e l a rg e un a>rta in t i es in th e es tim al25 o f p mbab iliti es f o r a n d human e rr o[S. a n d un dersta n d ing and p robab i I i st i c q u a nti f i c<rtio n o f o cc u r m n ce o f so m e phy s i ca l p h e:no m e na in P RAs o f cu r re ntl y 0 pe r a tin g reacto r s s ee m l ess so beca u se o f repeated reuse s h o u I d n o t b e O!Ier l oo k e (L T r ea lm ent o f un a>rta in t i es in severe a c dd e n t a n d d e lin eatio n h as a l ways 'bee n 1 i m i ted in r is k ass es s m e nts perfo rm ed m d<lte, eve n in t h e stoo i es th<1l: we n t th e fu rthest in s u c h. a nal ys es, s uc h as N'URE G-115 0 A n ot h e r irn porta nt c o n s i derat i o n , a l so r e l ated to t h e genera l c<rll>gory o f u n c erta i nti es, i s t h e iss u e o f stab>-O in PRII m et hods a n d data. Th i s i s a n i ss u e fo r ri s k m od e l i ng o f all r eacto r d es ign s as a ll u Cled t o ab ove, a n d i t is es p e d aU y so f o r d es ign s t h at pri m a ri I y r e l y o n pa:ss i ve saf e ty fu n ct i o n s pe r fo rm ed by s af e l¥-related S ystems. Structu r es, a n d Co m po ne n ts (SSCs) and d igi ta l sy stems (e.g.. in i nst ru me n ta ti o n a n d co ntr o l I&<C). Th e s tab>-O f-t h e;Lrt d oes not pe rmit a high q u a li ty m odel ing f o r i ab il i t y eva luati o n s f o r thes e s ystems. rn pa rti c ul a r. the r e i s c o n s i d e r ab I e u n re rta i n ty w ith respect to t he m ntri butio n o f s o ftware f a ilur es (CtJ')1D d igi ta l sy s t e m r e U.ab i I i ty. F o r th e pot e n tia l ly safe r and m o pas s i ve adva n ced react o r d es ign s.. i t i s poss ib I e th at d igi ta l systems a n d hum a n e rror s o f co mmi ss i o n (d u e in partt o l o ng e r ti m eco n sta n see. e.g..I.1J1) migh t h ave a high e r r e lati ve ri s k m n t ri bu tio n, a m n tr i bub o n t h at m ay b e d i ffi cu I t to asses s w ith any s ig ni Ii ca n t l eve l 0 f co n Ei d e nce. T hes e i ss u es 0 ffer add i ti o nal reas o n s t o a p p I y q uan tit<rtiv e P RII r esu l ts j u d i d o u s l y fo r futu r e nu clear pl a n ts .. Th eCo mm iss i o n al so offered anothe r goa l o f l E-6J yr w ithin the Safl!!y Goals Pc U qr S1>II: em ent f o r fr lX! uenqr o f lar g e m1 mses to the e OYi 10 ome nt fo r f u rther sta ff exam in.atio n.. A CIeEin i tion f o r l a'll e reIe.ase was not o ffered in that doc um e n t fn 1 9 l the sta ff sever a l o p tio ns aoo Iinally r o mm ended th at a l a'll e r e lea.E 'be de li ned as a th<1l: h.as the po te n tia l f o r caus i ng an o ff s i te ea r ly fata Ut y. o the r S KY paper s (de notes pape-s s u b m ittul t o the Co mmi ssio n ers 'by the NR C staff}. S ta ff Memo r anda an d Advisory Co mmi ttee o n R ea<:to r Sai>guarCIs (ACRS) letters to the Co mmi ss io n (e.g.l lOl) deKItul m thi s s ub j ect Th e Co m mi ss io n d ir ected th e sta lf m that th e r ev a lu atio n o f la'll e mlease m a gn it OOe 'be m n siste n t v.ith. AC RS pr oposed gui de l in es I in kirlg the hi e ra Jt: hi ca l l weI!; o f th e saf e ty g oa l d bjertives, whe r e the Ia'll e gu iIlel.in e was m nsi Clerul the t hi rd leve l objerti ve (th e q uaUtati ve an d q uanti ta ti ve h eal th ob j ectives the level o n e aoo two Amn r d i ng t o these gui de U nes. each. subord ina te I_I o f the safl!!y !Jl al dbjocti\eS s h o ul d:

• b e m nsistent w i th the l eve l ab ove,

• n o t 'be SCI co n se rvativ e as t o C milt e a d e facID new po l.i cy.

• a s impUEi ca ti o n o f the p rev i o u s l evel

• pr O!l i d e a bas i s fo r ass uri n g th<rt the Saf e ty Goa l Pc U cy Ob j ecti ves 'be: in g m e t,

• b e deli n ed m ha v e broad g en e ri c a ppU cab il.i ty.

• b e st<1I: ed in ter m s t hat a r e und erstandab l e t o t h e publ ic. a n d

• g e n e r a l ly m mpl y w i t h. cu rr e nt PRII usag e a n d pr act i a.. fn 'the e n d. the s ta ff rea ch ed th e overa ll co n d u s i o n t h at d eve l o p m e nt o f a l a r ge r e leas e d e Ei n:iti o n a n d magn itud e, 'beyo nd a s i m pi e q ual i ta ti ve stata n e n t r e la ted t o th e freq u e ncy 0 f 1 E -6 Jy r i s n e i th e r p rac ti ca l n o r r eq uir ed fo r des ign o r r egu l am ry pu rpos es. fn aCJ(J iti o n , basul u po n th.e wo r k d o n e ela lu a bng l a'll e re l eases in N'U RE G*-l1 50 1 8 1 a n d o th e r re l ated ac ti v i ties. the s ta ff n otul that th e g e ne r a l *pe r f o rm a n Cl! gu i de l i n e o f 1 E -6Jy T a n d th e fi F stibs i d i ary dbjecti ve 0 f 1 E -4Jy r a re n ot m ns is te n t w i t h. th e o rigin a l Q R O s 1.11 1 (i.e.. th ey m ore co n se rvativ e. a n d th e degr ee o f m nservat i s m d e*pe n ds o n the s pec ili c pl.a n t). [n add i ti o n , t h e Co mm is s i o n reject ed th e u se o f lE-5 Jy r o f macID r o pe r a ti o n as a CD P goa l fo r adva n cul d es ig ru in. S E CY-1 6 1.12 a n d i ts SlUv t T hi s rej ect io n s h o u I d b e exa m i n ed m g e th.e r w i t h a se ri es o f Co mmi ss i o n Pc I icy Stat e me n*ts 0 n ul abo n 0 f adva n.ced reacID [s. Th e l ast in th e s e d es p ub I i s h ed in OcID'be r o f 2008 l UI sta l25: T he C o mmi ss i o n ex pttt:s. as a minimum , at l east th e same deg r ee 0 f protectio n o f th e I!Ilv i ro n m ent aoo pub I ic hea l t h a n d safety a n d th e m mm o n de fe n se and secu ri ty that i s req ui red f o r cu r re n t g e n e rat i o n l ig h t-wate r react o rs. l'u rthe rm o r e, the C o mmi ss i o n ex p ects that adva n.ced r eacto r s w ill pm v i dl! e n h anced m argi ns 0 f s afety a n d/o r us e s i m p I iEi ed. inh e r e n t:, pass i ve. o r o th e rinn O!lativ e m ea n s m acm mpli s h th e ir sa f e ty a n d secu ri*ty fun ctio n s. The in.m rpo ra ti o n o f e n.han c ed sa f e ty m a rgi IlS m ay h e l l' o ff se t th e e ffects 0 f added un ce rtai nti es in t h e P RII m o de l a n d/O r in acdd e nt a n a l y ses a ris ing fro m the n Ollei ty o f adva n ced rea c1D r des i gns. l E lsew h e r e other attrb u tes o f adva n ced d es ign s a r e d escr ib ed as: r e i i ab l e a n d l ess c o mpl Bt: s hu tDow n heat re m O!l a l system s; l o n ge r t ime ms ta n ts a n d s uf oo e nt i nstr um e n ta ti o n; s impliEi ed safety sy st e m s; minim ize po ten ti a l f o r seve r e a c dd e n ts 'by in m r po r at in g red u nda n qr. d i ver s i ty. sa f ety syst.em. i n d e pe nd e: nc e; i n mr at e T he importan t aspects o f th i s P o li qr S ta ll!m ent (a) i t m n ta i ns o nl y q uali ta ti v'e pri nd pl es f o r e n han ced sa fe ty o f nu clear r ea.cto r d es i g n s.. a n d (h) i tspec ili ca ll y l acks a n y num e ri ca l cr i ter i a. Beca u se o f l a rg e un certa in t i es o f d s k-based nu m er i ca l r es u Its. ris k a n a l ysts ty p i ca ll y d o not m ns ill e r va ri<1l:i o n s o f les s th a n fact o r s 0 f 10. o r 50 in su c h nu m'be r s as m ea ni ngfu I in c r e me n ts. R is k Bt: perts ma y co n ve rt the abO!l e po l.i qr sta te m e nt in t o a nu me ri c.al cr i ta:io n 'by pr O!l i d ing a n o r de r o f magnitoo e as th e s m a li end i sc dmin alD r f o r dec i d ing h ow mu ch sa f e r adv an ced r eact o rs sho ul d 'be fr o m cur re nt r ea cto r s. Th i s.. ho weve r. i s a n o n-seq ui tur a n d a prob l em inh e r ent to ri s k.:tJ ased cal c ulati o n s. An ord e r o f m ag nitu d e is a ve ry l a'll e i n cm m en t in th.e rea l w o ri ll. a n d cu r rmt nu cJea r are a l ready mu c h safe r tha n a n y o t h er m mpa:r ab l e i ndustria l fad I i t ies a n d haza r oous hum a n ac ti v i ti es. Ultr a-m serva ti s m in d es i g n h as a pri ce, b o th @co n o mi ca ll y a n d o p e ra ti o n a ll y. As d i sc u ssed in Sect io n 3. th.e p roposed n ew s urr oga te num e ri ca l ri sk-based c ri te ri a ca n b e f a r mo re rest d cti ve th a n the QHIOs. They are a l so q u a ntitati vely unp rm i ctable in ri s k s pa ce" a n d not co m pa rab I e w ith Ql R Os as th ey ar e no n-i n mg ra l meas u res o f ri s k. They are m o r e rest ri ct i ve in the sense t hat a r ea.c to r that in a hy po th e ti ca l cas e m ay f a il lD m eet so m e 0 f the new cr i te d a (d esc ri bed in Sect io n 3) ca n s til l m ee t the QflOs 'by o rder s o f m agn i tu d e. fn s pit e o f the abO!l e d i s cuss i o ns a n d th e b r oad po l i qr gu i Cla n Cl! 'by t h.e N R C Co mmi ss i o n e r s, thi s pa pe r's observat i o n is t h at th ro u g h o u t m a n y ptibli ation.s o f the n a ti o nal a n d in ternationa l age n d es a n d co mm er c i a l e n tit i es, there i s a n i ncreas i ng tr e nd mward m o r e pr ev a l e n t u se o f ri s k.based c o n ce pts in g e n e r a l , a n d t h e 'u se o f so m e fo rm o f nu m e ri ca l ri s k t h rI!5 ho I ds as il.CD!p tan Cl! c r i te ri a v i s-a-v i s sa f e ty g oa l s.. in p a rti c ul ar. F o r exa mpl e. a nurrib e r o f N R C s ta ff oo cu m e n ts (e.g.. 1.1 4.1 51). as we ll as in d u s try aoo in ter n atio n.al pubU ca ti o ns {e.g.. I. 1 6-23n h a ve e mp lo yed va riou.s types o f r iska cce ptanc e cr i te ri a (co n s i ste nt w ith th e te rmi no l ogy em pl ay ed w ith in the oocu m ents) wh i ch. i oy o l ve s o m e f o rm o f a freq u e n qr ve r s u s c o n seq u e n Cl! (F C) c urve, or FC a n ch o r po i n'ts o r r eg i o n s. I t ca n 'be s h ow n th at thes e a p pr oac h es gene r a ll y estab l i s h mu c h me re restr i ct i ve nu m e r ica l th r es h o l ds t h a n th.e Qi R Os, a n d a ppli ed as n o n-int eg r a l q u a nti ties. W hi le t h e in te nti o ns 'be hin d th i s tr e n d a r e n db Ie and mo ti vat ed in pa rt: flO m a d es i re t o

. E+O 1 E-1 1 E-2 *1 E-3 I ill:' £ -1E-4 "" I" u.. 1E-5 5() rom _",'1llr8!lfb' Bill !!tf ocl!l (lftillla r",nllJll_eBrtjllol ol1ll .. .... 'po ** I>,., -------'JO.

I E-6 1E*7 1.......----------------!I..-

1 1001 O.(n 0.1 1.0 1 0.0 1 0 0.0 100 0.0 Dose , rom . 1_ I'requenq>

. C<I11=fU

Gln-e .* 62 ) af lJR l!C.1 SO[L a nal yt i ca l m ode l s a n d m a n y ot h e r f aclDr:s (e.g.. im pact o f s af e:y E ve n the n, t h e u se o f re pr ese n ta ti ve pa r a m etl!r s (s u ch as th e m ea n) iI.55iOwb!rl w ith. t h e fr eq u encies a n d m n.seq u e n ms o f in d i v i dua l o r in t eg r a12d aa::ide n t SlE n a ri os has I i rni ta: i ons o f i ts ow n , as th.e types and w i dths o f the un der l yi ng distr i b uti o n s o f the inpu t r a n do m va ria bles a r e gene r a l ly ass ign ed by s u b j ect: i ve j udg m mt r t i s clea r t ha t these i ssues bem rTll! m ore do min a nt in a nal yses o ffutu rqal'lva n Cl!!d r eacto r des ign s w ith l ess kn ow l edge abo u t severa l key aspects o f the safety o f t h e des ig n. s u ch as the fi de l it y o f a nal yses i.n ui ds, n e ut ro ni cs, 'fis s i o n prod u ct t r a ns port m atl!rial prope rti es at high 'II';m pe r atu r es , co mpon e nt r e lia bil i ties. and the " unk now n u nk *. Th.e Q H Os h ae a l ogica l r e l a ti o n s hi p w i t h th e ri sk th at th e m embe r s o f the pub I i c a r e o th erw i se ex posed to as a rti cu l ab!rl in th e Q u a l i tat i ve h m J t h o b j ectives. They estab l i s h th e ri sks o f nu dea r power pl a n t 0 peratio ns at a s m a ll f r act i o n o f t h e ri s ks that t h e m embe r s o f th.e pu b li c. not t h e g ene r a l p u b l i c at l arge, b u t t h ose l iv ing in the v i d nit y o f tbe pl a n t a r e a l ready exposed t o. A r ed u ctio n in th ese r is ks fo r 'fu tu r e r eacto r s p ro p osed b y a n y stake h o l der (w hi ch wo ul d b e mns i.sll';n t w ith th.e stall!!d Qua li ta ti ve goa l o f t h e Co mmi ss i o n), s h o ul d be wi thin. r easo n a n d n ot so d r ast i c as to de pri ve the sa m e p o pul a ti o n f ro m th e b ene fi ts th at t h ey ma y o th erw i se real i ze f ro m operat i o n. o f these r eacto r s, *. P Es and l eEs are ca l c ul ated for m mp a ri.so n aga in st t h e Q H Os, Both o f t h ese, as we ll as t h e mor e wide l y u sed s ur roga'lI'; m e tri cs tD Q H Os.. suc h as OJF a n d lRF f o r lW R a p pI.i ca ti o n s, a r e in ll';gra l Qua n t iti es th at a r e der i ved f ro m th e co ntri b u t i o n s o f a ll acdoo n t sce n a ri os th at a r e m n s i oored in the ri s k mo oo L rn tegla l ri sk measures in m rr all! at l east th r ee i m po rtant prope rti es: L De fi.n:iti o n o r c har acte ri za ti o n o f in div i d u a l aa: i oo n ts am ari os i s depende n t o n bo th th e spec ifi c P RA m odI:!! (e.g.. larg e f a'll I t trel!!' s mall eve n t tree ve r s u s s m a U f a ul t t r t!l!,l'la rg e eve nt tr ee) a n d the sped Fi c plant d es ign (e,g.. co mpl ex w ith m o r e act i ve safe;ty sy.sll';m s __ s u s l ess c.o mp lex w ith. m ore passive sa f ety In tEgra12d r isk are not s u b j ect: to s u ch depenoo n des o n the ca l cu lat i o n m o&l o r plant des ign. o I'tw iU b e a cha U e ng e to estab l is h c ri tl!ria to t h at i nd i v i d u a l aa: i dent scena ri os de fin ed o r c h a r acteri:md at the sa m e l evel 0 f across d i ffu r e n t pl a n t dles i gns a n d iI.55iOda12d P RA mo dels fo r u se w i t h. thi s type o f F C cu rve co n st T ucL Th e syst'e m wo ul d be inh e r e n t l y un stab l e a n d depe n de n t 0 n s u b j ective i n terpreta ti o.n s by a ll sides in a d i sp u te. 2 R e lati ve UIl ce rtain*ti es decrease w h.en t h e assodated do m. va ri ables a r e s umm ed, a n d t h ey in crease w h e n the r a n do m va ri ab l es a r e mul t ipli ed_ Therefore, th e effects o f u ncerta in*ti es are minim ized whe n in teg r all';d ri sk measu r es a r e used as opposed t o w h.e n i.n: II';T m edia:te and product Q u a ntiti es, s u c h as f reque n cies and co n.seq u e n ms o f in d i v i d u a l acdde n t sce n a r ios are u sed. 3. Co m pa ri so n o f a n y pa rti a l l eve l o.f p l a n t ri sk. s u ch as t h.ose t hat are based o n in d i v i d u a l aa: i oo n t SlE n a ri os, agai nst SOrTll! Q u a nti ta ti ve cr i te ri a ca n mi s inf o rm. o r eve n. m islead. Th e pote n*ti a l fo r m is i nfo rma tio.n i s la rg e because i t wo u I d n ot b e kn ow n as to w h at fr actio n (is it 0.00 1% or Ha l o f the ove r a ll in teg ral ri sk (we n. w ith i n. th e s am e catl!gory , suc h. as i n ll';rn a l eIIoe n ts l i s be in g co m pared aga i n st th e c ri'lI';ri a o Thus. the ri skof a n i nd i v i d u a l sce n a ri o wo ul d/s h o ul d not necessa ril y be UIl ac:cep tabl e if it f a U s in. th.e "una cxe jr reg i o n o f a n. F C cu rv e. bec au se the Q H CE (as sa f ety goa l s) mi g h t st ill be m et w i t h l a r ge m a rgin o A mnve r se coro ll ary i s that th.e ri sk o f in d i v i d u a l sce n a ri os s h.o u.1Il n o t n ecessa ri I y be v i ewed as " ac:r:e jr in. th e o th e r regio n e ith e r. as a pru oo n t approac h 'ID safety as.su r a n ce a l ways seeks tD in co rp o r all! reaso ab l e additio nal co n tro l s eve r a prope r Q u a li ta ti ve e ngi n eer i ng j u dgrTll!n t 0 r a Q u a nti ta ti ve a nal ys i s so d i ctates, F a U i ng w ith i n the aa:eptab l e reg io n co u I d den.y t h e des i g n e r s and o t h ers fro m t h oro u g h e n g in ee rin g th.i n ki ng i n.th.e safety des i g n process.

• r f it i s ass u med t ha t a fu t ur e des i g n. o f an. HTG R o r a n. S MR m eets th e F C curve. th e n t h e N R C w ill be o n r em rtlfo r m rt i.fy i n g th at t h e l eve l o f ri s k -bil!il!ld safe!ty o f t hi.s des ign i s " a a:e p tilb l e, -a n d in (JJ n trast; a n y d es ign th.at does n ot meet th i s l evel o f saf e ty. eve n Fe r a s i ng l e a a: i de n t s a ma ri o w i th a ll th e attmda n t un m rta in ty. i s un safe. Th e sa m e p rob l e m i s en (JJ u n b!r ed eve n if t h e g ove rning docume n t i s fro m th e i nd u stry. w h e the r o r n ot i t i s I!X pli c itl y md o rsed b y t h e N: R C, s u ch as a n A S ME o r ANiS s t m dard as in 1 1 8 HOw (JJ ul d th e r egulat o r a a:e p ta des i g n wit h o n e o r rno r e ac cident sce n a ri os in t h.e *'u.nac m ptab l e-r eg i o n whe n t h e g ove rnin g in d u s try sta nda r d itse lf h as l ab e l ed i t as s u c h?

• So m e cu r re n t L .... 1Rs w ill li kely n ot m ee t thi s F C CUIV e. A mi s un der sta n d in g o f th e in tmt o f thi s culV e and t h e ro l e th at N: URE C r e porn pl ay at N:RC (JJ ul d l ead so m e t o in co rr ect (JJ oc l u s i o n s c o n m rn i n g th e ad eq u acy o f sa fe!ty o f cu r re n t pl a n ts, becaus e t h e N: R C m d Jo r t he n uc l ea r in dustry the mse l ves (as. e.g. in [1 5 , 1 8 1) ha v e l abe l ed pl a n ts th at d o n ot meet th i s c u.rve as " un acm p tab I e.-* Th e F C curve i s. in fact. i ntr od u ci n g n ew a n d m ore r estr i ct i ve a a:e pt a n m c ri b!ri a tha n the QH O sa f e ty goa l s as ev i d e n t by i ns pecti o n a n d as m e n t i o n ed in 1 1 5 in co n t r ad i cti o n ID th e ACRS guida n m m e nti o ned ab ove.

• Th e co m b i n ed e ff ect 0 fu s i n g ri sk m e t r ic s as a a:e pta n m c ri b!ri a a n d a p p l yi ng th em o n th e l eve l o f in d i v i d u a l accid e n t sce n a ri os ca n lead to othe run d es i r ab I e o u t (JJ m es. Fu ture ma c1D r d es i gns lowe r tota l (in teg r ab!d) ri sk th.a n c urr e n t o pera tin g reacto r s m ay b e e r ro n eo u s l y l abe l ed as u nsa f e-a n d not b e pu r s u ed. 0 r b e b u r de n ed w i t h c ost l y a n d un n emss ary d es i g n m od ifi ca ti o ns. o An exa m p Ie o f the abo ve (i nvo I v i 'Ill a po b!n ti a ll y safer fu t u re r eact o r des i g n) is a react o r aJO l a n t break !b r a hi g h-12 m p e r atu r e gas-coo l ed r eact o r (HTG R). [n a by po thti c al c.ase, i t ca n b e ass u m ed tha t a n a p pi i ean t c alc: u l ab!s th e f r eq u e ncy a n d the co nseq ue n c es o f th e sce nari o in a way th at a ll ow s th e m t o s h ow t h at i t i s " ac ceptah l e.-A n yo n e in d in ed to q u est:i o nth.e i ty 0 f t h e c a l cu.l a ti o ns c a n: (a) lXI in t t o th e deg r ee o f u nc.erta in ty in th e pip e b r ea k f r eq u e ncy b eca u se of v e ry m i ted nu mb e r o f year s 0 f ope-at i ng expe ri e n m w ith thes e r eact o r s; ) po i n tln (JJ n d iti o n s s u c h as hi g h o p e r a tin g t e m pe r at ur es as add iti o n a l r&so ns fo r mu ch hi g h e r fa i lur e fr eq ue n cy lXl't e n ti a l th.a n in t h e l W R ex pe ri e n m; a n d (c) c ha ll e n g e t h e ass um ed ra d i o nu d i d e air b o rn e fracti o n s p rod u ced by un m rta in t i es in so ur ce t e r m s {e.g. l o n g-b!r m d i ffusio n o f rad i o nu d i d es th ro u g h aJilteC f uel IJ'il rti d es. pens i o n causul by v i b r a ti o n effects, hi g h er t e m pera tu res. l owe r pl ab!o u t, T hese c hall e n g es ca n l ea d ID a m n clusio n t ha t th e .sm n a ri o f a ll s in the "u n acce ptab l e" r egi o nin stea d

• S i mp I e a n d Jo r pas.siv'e rea c l m d es i g n s wo u III ha ve fe!we r num b e r:s o f ac cid e n t sce n a ri o s t han co m pi ex a n d act i ve des i g ns at th e s am e Iev'e l o f a c cid e n t s mna ri o d e lini t i o n (e.g. sys b!m l eve l) a n d w ithin th e s ame P RA m od e l. Th e in th e num b e r o f ac:c id e n t sce n a ri o s (JJ ul d be in multipl es o f 1 0 r athe r than in al geb r a i c fr actions. As a h ypo th.et ic.a l exa mp le. two react o r s may h a ve th e s ame ri s k pr o Ii Ie. b u t th e Ii r:st h as 1 0 seq ue n ces w ith JO r e m at 1 E -61 y r. a n d th e seco n d h as o n e seq u e nc e w ith a co nseque n ce o f JO r e m at l E-5/Yr. Un d e r t h e F C curve (JJ nstru ct, o n e i s dee m ed acre pt.ab l e a n d the 0 th er i s not; wh i c h d oes n o t m ake se n se in " r ea l ri s k s pace.'* o Th u s. th e use o f ri s k-based a<E:eptance 0 nth e leve l 0 f in d i v id ua l a a: ident .sm n ar i os (as o pposed t o in teg r a l q uantiti es) m<l1( be v i ewed as pena l iz: i ng s impl e and pass i ve des ign s in favo r o f acti ve a n d (JJ m pi ex des ign s. in vio lati o n 0 f the Co mm issio n Po l icy Stateme n t o n Mva n md R mct o r s l U I.

• Aga in. bo::a u se in tegra l m llB.S ur es o f ri s k a re n o t Obta in ed in thi s m odel a ppli cat i o ns o f thes e s mna r io*-I ..... e l a n d ri sk-based ac m p ta n Cl! cr i t e r ia w ill b e va ri ab l e !b r ea c h d es i g n , s pec:i fi.c P RA m o de l , a n d mac1D r s i lil!. The ca n b e s u bsta nti a l in so m e GlSe.S.. It i s i m lXIrta.n t that t h e N: R C sta ff be mg ni z.a.n t o f t h e ab ove i ss ues in co mpl yi n g w i t h t h e Co m m i ss i o n d i roct:io n in b!sti n g th e co n c epts e mb od i ed in N: URE&1 860 in a n actu al li m n.si n g a pp rova l p roc ess f o r a futur e pl a n t Th e .<ta.ff s h o ul d e n s u re th.at t h e ir r ev i ew w i II not d ev i a b! f ro m t h.e 1 0 ng-s ta n d i n g Co m mi ss i o n pr emae nt s in t h e man y elem e n ts o f a r is k-inf o rm ul a p p roach. W hLl e th i s pa pe r h as IDuc h ul u lXI n o n I y a r..w ID pi es. fum A! I>aper:s ca n d i scu.ss th e use o f P RA. in cl u d ing th e intr od u ct i o n o f a p ro iJ(lSl>d tech n o l ogy-oo utr a l generi c ri s k m eas u re t h.at w i II a ll ow !b r c r oss-co m IJ'il ri so n 0 f th e l eve l o f sa fe!ty !b r d i lfe r e n t pl a n t d es i gns in d e pe n d e nt of s i t e-spec i fi c c har acb!r i st i es; a l>" p roac h t o de fe!n.se-in-de:p th; s electi o n o f t h e So-Gl U ed l.i m n s in gbasis eve n ts; a n d se l ert i o n o f safe!ty SSC:s in a ri s k-i n.f o rm ed and pe r!b rm.an<p-bas ed f ra m e:wo 11<. I t s h o ul d be added t h at a l t e rn at i ve a n d co mpl eme n tary ri s k m e t r ies ID Q H Os ca n be us e fu l ID a IXIt e nti a l a p pi i ca n t f o r a d es ign ce rti.li c a ti o n o r (JJ m b in ed l ic e n se, f o r exa m plelD ass i s t in d e m min.a ti o n o f h.av in g reach ed a s uffi c i e n t mi x o f p reve n*ti ve a n d miti ga ti ve f ea tu r es in a new des ign (i.e.. saf e ty des i g n t ro ff d ecis i o ns l arID (JJ m lJ'il re r elati ve sa fe!ty o f d i ffere n t d es i gns. T he tech n o l ogy -n e utr a l g e n e ri c ri s k meas ur e m e nt i o n ed ab ove w ill sa ti s f y t h e l att e r n eed!b r futu re react o r d es i gns !b r w hi c h th e CD P a n d lRF met ri es m ay n o t b e ful ly a ppli cab l e. An exa mpl e o f a n a l b!m a ti ve F C c urve tha t ca n b e effe!ct i ve l y us ed f o r saf e ty d es i gn t r ad e-o ff dec i s i o ns i s d i sc u ssed in Sec tio n 6. Th e im lJ'ilct o f th e a f o r e m e nti oned i ss u es ma y not be as g r ea t in p r acti Cl! w h e n t h e F C cu rve o f N: U RE& 1 860 o r a s i m i la r (JJ n s tru ct i s used o nl y by t h e d es i gn e r as o p iJ(lSl>d ID the r egulat o r. Th e d es i g n e r c a n u se such co n structs 0 r co n.CI!p ts as (JJ m pl e m e n tary in!b rm at i o n in a n ite r a ti ve m a nn e r th ro u g h o u t th e d es i g n p ro mss. A pr Ob l e m t ha t m ay be e n (JJ un t e A!d in that pr o mss i s t h at a p roper int e rpr e ta ti o n 0 f so m e ri sk-bil!il!ld (JJ nc e p ts m a y n ot b e as in t ui t i ve f o r the d es i gn er. es pecia ll y f o r t h ose w h o a r e n ot P RA ex perts. as it m ay a p pear at lir s t. rn add i t i o n , m a nu a l s o f p ractic e, s tICh as s ta n dards 0 r gui d es t ha t a A! d eve l o ped b y the in d u s try m ay be mdo r sed o r r ef el1>n ced by th e r egu l aID r s a n d be usul in wa y s th at pr od u Cl! t h e uni.n b!n d ed r es u.l ts (e.g. l ead in g t o r e j ectio n o f s af e r d es i gns). Fo r t hi s reaso n, i t i s s u gg es b!d th.at t h e u se 0 f qua n t it a ti ve P RA r es ul ts in t h e (JJ n text o f des i gn o r regu l alDry ri s k-ac c ep'tanc e c ri b!r i a be a vo i d ed by a ll. In s b!ild. Sectio n 6 pr o v i des a n a l b!r n a ti ve (JJ n s tru ct th at ffi<IY b e u sed by t h e in dustry t h at w ill a a:o mpl is hth e in t e nded pur po se (d esi gn s a fe!ty trad e-o ff d ecis i o ns) wit h o u t the n ega ti ve (JJ nn ota ti o ns th.at a r e ass oc iated w ith N: U RE C-1 860 s ve r s i o n 0 f a n F C cu eve. S. I nmlpmtati<>n of tho! as : R e m cri teri'l>n u....t in lU (FR l00 ,l5 U.34 Th e 15 rem c ri b!r i o n u sed in 1 0 C FR 1 00 a n d 1 0 c m 5 03 4 i s o fb!n us ed as a d e fact o do se ac:c e ptance cr i b!ri o n !b r DBAs by th e N: R C staff. T hi s *u sag e i s. bo weve: r. co n trad i ct o ry to actua l Co mmi ss i o n IXI I i cy a n d gu i danc e as descr i b ed exp l.i cit l y in N: R C regu l a ti o ns. as d i scu ssed in t hi s sectio n. S in Cl! a nu d ea r pl a n t i s d es i gn ed t o adeq u a b!l y r es p o n d t o th e O<E: U r", n ee 0 f Des ig n &a.s i s E ve n ts {D B E in cl ud es Ant ic i pated Ope r a ti o n a l O<E:u n en ces and OIes i gn B as i s t h e ex pecta.t i o ni s t h.at th e associa md o ff s i t eco n.seq ue n c es w ill be s m a U {e.g. fr acti o n s o f 15 re m T ED E l.

Add it i onalco r'l trols. may be co n*sldered Ad d l tlona l l controls need no t be co n sidered l o g Consequ e ooe i s-A f;requ_erq' "eJ'5U!5 am sequ_eTl.a!: Ql:rw! .h t GIll used by .o;pp lic.o.l1I:

dosign procrs .. whet h e r add iti o n a l m n tro l s s h o u I d b e m ns id e r ed f o r th e s(ll!C i Fi c s<l!.n a ri o. (v) Th e two reg i o n s are se pa la ted bya ba n d 0 f p e rh aps a n 0 l1l e r o f m ag nitu d e va ri at i o n w ith d iffu sed bou n da ri es (s uch as in Regu l atory Gu i d e 1.174) o n freq u ency a n d co n.seq u e n ce. la th er t h.a n Firm bo un d ari es. T h is is becau se a n y s ingl e par a m e tl>r o f sa ma r io fr eque n cy or m ns l'fl u e nce (th e m ea n is typ i ca ll y u..oo fo r a ll) i s i ts e lf sub j ect to un<l!.rtain ty a n d eJ1 s uin g ch.a ll eng es, as th e r a ng es o f va ri ab ili ty a n d th e un derly in g d i strib uti ons a r e g e n e rall y ass ign ol s u bjectively. (v i) Th e m n.seq u e ne e sc ale m ay b e re la ted ID a pp ro pri at e hea l t h meas u r es a n d/o r cos t-beJ1 e Fi t fo r t h e i ne l u s i o n o f th e add iti o nal co ntr o l un d e r m nsi d e r a ti o n.. (v ii) Si n ce t hi s cu."'" i s u..oo as a desig n a i d fo r t h e a ppli ca n t, s ta ff wo ul d ha v-e no pos iti o n abo ut t h e acc:e ptab i I i ty o r th e l ack thereof assoc i ated w i t h a n y (>a rt of i ts m n str u ct. in d u dL n g the a n c bo r The r egu l aID r mu st us e th e to tal i ty o f th e sa f ety in f o rm atio n de l i v-e r ed by th e des ign a n d t h e pro p osed operatio nal pl a n t h at in d u d es th e tra d iti o n a l det e r mini stic r eq ui re me n ts alo n g wit h th e su ppl e me n ta l P M in fo rm a ti o n in m n cJ u d in g tha t th e ppo sed pl a n t i s saf e. Not e th at the bo un da'Y r egi o n 0 f esse n t i a l ly co nstan t r isk i s o n 1'1 m n ce p t u a L Th e d es i g n e r m ay dec ide th at in ce rta i 0. s u b-Jt>gio ns a n d boca u se o f spec ifi c co ns i deratio n s. s u ch as ev-e n ts w i t h pa rt ic ul a d y hi g h or l ow f lElq u e nc i es a n d/o r co n.seq u e n Cl!5. a n d in t h ose areas g.ov-e rn ed b y ex i s ti n g r eg u l at i o n s. dev i.atio n s f ro m th e bo un dary reg i o n are wa rr a nt ed 7. Summ ary and. c o nc: I'us im lS R is k-inf o rm ed r egu l a ti o n is b uil t aro und t h e m ac e p t o f u si n g tr ad i tio n a l d e tl>r mi.n.is ti c tech ni q u es o f safety assu r ane e s u p pl e m e n lEd b y P M inf o r m a ti o n a n d in s i g h ts. Trad iti o n a l d eb>rmini s ti c tech ni q u es in clud e co n ce pts s u ch as in m rpola ti o n o f r ed un dancy and d i v-e r s i ty. in co r po r a ti o n o f sa fe ty m a r g ins. a p pi i ca ti o n 0 f d e f e n se in d e pth. a p pi i ca ti o n o f ty as s u lan ce. e t c P RA res ul ts s h o ul d pl a y a a n d s upp ort i ve ro l e in m aki n g decis i o n s about ad eq u aq o f sa fe ty i.n a ri s k-in.fo tm ed r egu l alD 'Y fr amew o r k. Hcweve r. r e<l!.n t trends i nth e d ev-e l o p m e n t o f n ew ri sk-r e l alEd ap!>rlladl es, w h e th e r they are pe r fo on ed b y th e in d u stry. N: R C sta ff 0 r ot h er do m est i c o r i n b>m a ti o n a l bod i es, a r e towa r ds h ea v i e r e m pb as i s in u se o f q u a n ti ta t i v-e PM "", til ts. T hes e ri s k m eas u res are so m e t i mes co mpa red to r isk t h res hol d va lu es t ha t h a v-e attai n ed a n actua l. o r ev-e n a d e f a c to. r egu l alDry stat ur e o f " ri sk a cce p ta n ce c ri t e ria" i.n CE rta in inst a n C1!5. S uch a p pi i c a ti ons o f ri s k m eas u"", fo r a n ucl ea r r eact o r d es i g n 0 r a spec ifi c pl a nt a r e n ot a l way s in kee pi ng w i t h th e t e"" ts 0 f ri s k-i n f o rm ed r<'g ul at i o n s. w h.ic.h c a Li fo r co m pa ri n g (i.n tegra l) m eas u r es o f th e ca l cu l alEd ri s k (e.g.. PE s a n d L C F s or t h e ir s ui tab l e s ur rogat es s uch as t h e CO P o r th e LRP) aga in st Q H Os (o r th e ir s'urr ogab> targets. e.g.. 1 E -4Jyr fo r CD r) o n 1'1 as safe ty g o als." In add iti o n, u s in g nu me ri ca l P RA r es u lts. pa rti cllla rl y tho se th at are n o t in tegr al q u a n*ti ti es, in a ris k-a a:e pta ne e m n text, ev-e n b y th e nu clm r in d u stry (as o p po sed to th e reglIlat o rs) ca n h a ve num ero u s u nd es ir ab l e co n.seq u m ces. E xa mpl es o f th ese a m o n g m a n y d i.scussed in th e text i n d u d e: th e t e nd e n cy to s impl e. p ass i ve safety syst e m des i gns in f avo r o f co mp lex. a c ti ve d es ign s; a n d futu r e r ea::t o r des igns 0 ff e ri n g lowe r i nt eg r alEd ri sk t h a n t h ose 0 f the c u m mt a n d hi g h 1'1 sa f e 0 pe r at i '1!l reacID r s m ay b e e r ro n eo u s l y l abeled as un safe and not b e pur s u ed. o r be b u rden ed w i t h m stly b u t unn ec essary des ig n m od i li catio n s. Thes e i ss u es ca n l ea d ID ser i o u s un i n b>n d ed m n.seq u e ne es in li ce ns in g o f fu t u re r mc:t: o r s o r creat ing new cha ll e ng es r ega r di n g t h e sa fe t y adeq u acy o f ex is tin g pl a n ts. Th e p aper al so 0 ffered a n a l t e rna t i v-e u se f o r a f req u ency v-e r sus co n.seque n cec u rv-e as ad es i g n o r o pe r a ti o n a l safety o p t imi za ti o n too l f o r u se by the reaclD r des ig ne r o r pl a n t operato r. Diisdai mo<Y Th e wo rk r e l alEd t o t h e d evel o p me n t o f thi s (>aper was co n dUCb>d at t h e IElq u est 0 f th.e Oi rector o f the Adva n c ol React o r P rog r ams at t h e O ffi ce o f Rea clll r s (n ow retired) in th e l ast q u a rt e r 0 flOOS at the U S N: R C. w hi I e t h e a u th o r was o n l oa n f ro m t h e U S Dl!pa rtm ent o f Nei th e r th e aut h o r. no r t h e Uni ted States G ove rnm e nt; a n y agency th e reof. 0 r a ny 0 f th e i r em pl oyees ma k es any wa r la n ty. ex pressed o r impli ol. o r ass um es a n y l e ga l li ab ili ty o r r espons ib ili ty fo r th e a cc ur acy. m mpl ete n ess. o r a n y thi rd part y's u se o f t h e r es ul ts o f such u se o f a n y in fo rm atio n. p lUd u ct, o r p roc ess d i sc l osed. 0 r r e pr ese n ts t h at its us e wo ul d n ot infri ng e pri vat e l y o w n ed ri g h ts. R e fe re n ce h ere i 0. to an y s ped li c co m m e r ea l pr od u ct, pr ec es s. 0 r se rv i ce by tr ad e n a m e. tlad e m a r k. manu factur er. o r o therw i s e. d oos not n eCl!5 sa ri ly m nsti t ut e o r impl y i ts e n d o r se m e n t; rem m m endati o n. 0 r f a vo rin g by the Uoi t ed S ta tes Gov e m m e n t, 0 r any ag e n cy th e reoE Th e v i ew s a n d 0 pi 0. i o n s o f th e a uth.o r ex p res sed her e in d o n ot n ecessa ril y stat e o r r e f lect t h ose o f t h e Uni lEd States g.ov-e rnm e nt o r a ny ag e n cy t h ereo E A.d lat ow Ied grDl!n1!5 Th e a u t hor w i s h es to th a nk O r. Do n Du b e (U S N: R C) w h o was t h e Ii r st ex pert t o l1!V i ew th e ea d y ver s i o ns o f t hi s pape r and offe r ed hi s b r oad and in-de pth kn ow l ed g e in s u pp ort o f i ts d ev-e l o pm e n t. Mr. A l a n K uri tz.ky a n d D rs. M o hs eJ1 Kha t i lr-Ra h ba r a n d Do u g T ru e p r ov i dol m a n y u se fu I in s i g h ts. (1 1 ChojllllO n J, H= .. oo 1ia=i'18 f.,.",.,.".n ilr n .... tJll'SA during /he .... .. 12 1 US c. R egis .... j;Jt l W2fL s..fety goo.t ls f or /he oporotio", of m dur paw er plo n ts" Augu st 2 1. 1 9!1 1l II I lIS NJC. le:teQ l ll.<8ister. ;q. _ an u .. 01 prdJ.m lis;;c ris k _=men t metI1oo. in mdo.oT regulrnl l)' Micy August 1 6. vo t 60; US). l!l!1!!i. P. 42622-!L 1<1 1 lIS N:R C Ister CUIIlO2<i.

..... min-. i" ..ment vulner;lb ili ties,.}lavembor

n. 19sa lIS :RC SOC'M I!l-144. White popor .., ris!c-in:b!:mod regu1oticm. 22. 1 :Il SQr f ""1uiT<:mO:Ill5 mo:mor.oTldum UoTcl1 1. 1 99!L 16 1 lIS JC.

gu.ide 1.1 , 4. An fo , ... in g prdJ.mlistic risk _.....,., n t in risk in:brm<d docisi<Jn.

an plo_pocific dh_ lie tile li*EJ1sm:!:

msis,.

1. moem1Jer 2 -1'7 1 lIS NIl<: NUR1lC-()fIiliJ.

lIS ..a...T l iqu J .. t<JI)' Cammission

.uoom """;ew p1 .. n. limsion l. l Ch.>pter 19 1* lS I lIS c. NU JtrC.l UiJ. Severe ..nctent rim: "",,,,,,,,,,,n t f or five lIS _T _erP:Io_;

0rt.dJer 1 9

• 19 1 lIS JlC.SE c 4!9.1 02.Im1'lemen ta:i.., gm l p<J!i cy;M.>:r.nlll. 199!L 1 1 11 1 lIS C Ml!S I.<;Jbor I!o iNRCCI"" j,:n",n. Ml!S<:<II1llDeI1ts..,

<JJl p'b.n i

l1lihe uiot

r gm l policy.IiIb)' 1 1. 199'. 1 11 1 lIS JC. SE CI' . .tJ 19fL Stm!s "'JI<>rt o n ri sk';nfmmed dh_ lie tile a;R port 9ll (optian l)w .. ti<Jn.o n ris!c . :b!:mod 10 1 0 a;R port 9a.4<1 (o:nnbustihle g& tembor 14.2Il (1 2 1 lIS -C SEC/' . 1 11 utianol)'

light wo1:<r .-(U'IIi) iss"", w /h eir r&1iomh:ips I!o reguIo t Ol)' requi.-.men Ju no 2Il. 19 113 1 lISNJC.Feder

.. Rqister."" t 7:l (99).N 2IlOS-tlZl

,. oIicy .... t<m en t on regul .. 01 0rt.dJer 14. 2IlOS. P. 11 41 lIS NU

... "'JI<>rtfor

/he m>du.IoT gos-cooled Uorcll 1 9s9. 11:5 1 lIS NJC. NUJtrC.1S61l.

""","".i1i l)' -d)' for .. risk inbmed w I""f<Jrl1l.>TI . ....b&ed regul ..

, lr uctluTe f or hi;tu", plott D icensin<. * ....,""""'" 200'. (16 1 .. Al<lmia.

/he .... Tl dm -IHT CR"OO I<<CR-8!5

• Sept"mlJer 199!L (1'1 1lur0000000 TI (hmn:rissiom.. ...rei)' 'Jlllroo<il1 for mOOuJ .. T Hi' DoaJment na. . .-Sf 4.l.

dism:mution.

Ap ril 1:5" 11 AN S-S.1.2tJrot Nud .... , uiot:r <ria .. ... f<ty do:sign p:roc= for madu:lor bolium-<llalod

......am Jun.e n. 200fL 1 1 9 1 1Ii!:p

Tlts wi/h m>du.IoT high t<:mp<:r.rIlU"" g& C<101ed ... Ap ril. .lIOO:Il I2Il I (Coou!hor). "',..,. D",;;ll1Jte of Nuclo.o T s..fel)'. Ilegu:lor"')' l'iewpaint <JJl inmvative VH1lt in """"'-: 4th "'Pia l meeting .., tempera lll"" __ ""Iogy. Septembor 2!I"()""'er

1. 200!i. 12 1 1 ]. 02-02. N..a...T "'&I' In_*'. A risk . :b!:mod. p".i",,,,,,nce.m.ed regu'b.tOl)'

nmewor k il r paw ... r&ctors" u.y 2001-122 1 -""'TI Jo"'-N.tiom l Nu<lJe..r SouJtb Arne.... SouJtb AfriaTl ...rety ..... ment lnmo.wor k ii , lihe p01tIle bed madulor 4th intermtiom l "'Pial meet:iJ1g OIl high __ mlogy. Septe_ 29-0dJollor 1.200!i. 123 1 m m (J<ty) lt d. risk """""""", t "1'J-<lrll f or the !He modu:lor ,..um. 1.J une 12" llOO l1 12 4 1 lIS NJC. _ r=mmen:l .. mk-in:I:J rmed w """;sion I!o 1 0 a;R p .. rt 9ll r ill 3 U6.AH8 1); Ju ne 14. 2

• St.off requirements mo:mor.oTldum

.. JlP7""ed Sep t em'bor 1 0. lOO7. 125 1 lIS c. SIlCV-iI!Ul(J!;G. SUfi .. ri sk.,i nformod md perfor:n:a n...b&ed ,.".,..., to p.ort 01 ;;t1o 1 0 01 the (h de 01 I'<d_l !i.guloti ....... TI d Policy _ <JJl Def""'e4n 111 for _:reRad<lrs"Ap ril 7.2Il 1l!L I2G I lIS N:RC ACIlS1W2Il'. ",.oopment of tMlno1ogy-"1eutral regu'Llt"')' nmeworl<" Septembor 201.2 12'1 Memor.ondum. Eo .... Iml]7o I!o]..l!. .... Foreign Invel trip rep<Dt ilr the In t ermtion;, l Agency meeting lie !A!lJ\ ... fety gui de on 01 """'''''''' Sf5"""'- corop<l11ellts fnmJ ApriU4 lihrnugh ilp ril 2 2006;),by:5" 200 11 lIS NJC. <='I Regi5'"". 9ll l2 1 3fL J\ol;q. .werr.ienl on ..vere ..nctete row"" d"';gns .. n:I plo n ts. August 8. 19 f1l1. 12 I (h de 01 llegW- .. m 1 0; J .. ,., ,,, I)' 1. 200!i I o...pter 1 ii i. Pil i (h de 01 Regul .. _ S 1-1 99; J"""'l'

1. 200!i 1 00001't ... 1 1l 1. I! 1 1 St.ollin -c. ... from m<d:ic.o l miotion .,,, ,, as.

""Ym SoOety. .. 11tml}; 1Ip d.>ted u.y 2Il. 2Il1l!L P 2 1 P i!.. T be high b rl:grOlJlld mi.>ti<Jn ..,..,. i n b=. iraTI: Ceology. nann" b>o1ogy. 11<< ... TI d possible regulotOl)' fm L In: VIM "W 24.-l!i. 2 .2. P3 1 J.>w<llVWSlci

z. I oni zing W .-..diaod:Mt)'

in !h.e 20th <=tuI}'. In: T eb:r.on. Iran. Ckitl 1 8 0. 2000.