FO'SChU"s dk!lil[ und Entwicklung. B6T Cigarettenfabriken cmbH~ Weitergabe an Dritte nur mit Genehmigung, Q Vertraulich, CONFIDENTIAL . Copy No.,.,. R & D · Report No, 126 E CONSUMER DEMAND RESPONSIVENESS Author: Dr, Wemer Schneider December 1992 Issued by: Dr, Emin Kauseh Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 ·1 841n the :blS~Ei D~D ~S701SIV~ESS 311 c:gartr;a :s dcrlntd ~t tho il~;~s 3f tte clF~~!iit~ of a C!PEII~~C. ~p PrOV:de ;ne snoko; u!~~ a sa~ipiyin~ anounr 31 snakl ~i~hou~ unp!laJan~ effects, The Pltipatl objee:lpe of tbl CDIS~p DEMAIID IEgPOlsiP~ISS approach is tbe support of the development and noditicat!on of products specific to consuners' 'cedback, Th~ iaaadi~ts objoc:ive is t~o eanslacion oi sPokers' dls~:le- r:sns Inrc phvsicai and cnorical da;a, tha~ can be utilited by t~s er3due: developer. The :015~fP DEMIWD 1EBTOISIIBPSS Ipproaca a;!l!fea Icaaaa and cite:;" tc!arionships (bpyond nors s;atis.,!cai relaeiPnsh!TJi based on tta knowicdaa on anbkP e3~~rol Sy cicarer.e deslcn and sno~:36 behavioar, jevEra: examples for different aspec:s of t~B ~pplia~!on of ~ho C)IB·EP 0~0 ~SPOICIVWESS approach are prelcarcb: - !he ~c~pability curtPs' of several lov and ultra low 'Lar brands show the tar and nicorina ~l~?vlric~ ~a~ can be Fcr;aivtl by:~a sa3ktrs, wbl~ dit:arln~ rlsr:ic:isns are assumed, - a s3~c!al way of evaluation and :IPrasentation ~arac- :eriles slv~lal 10W tar brands wi~ regard to assunsd car- ~a~ d~~ivlri~o (van~lG by the smokers! and laxi~a~ or nininu;l values for Ilvlr;l rls~ric:!ve Properties, I Ii ~is type of evaluation is aPBlied for the caparlson of auccesetul and Pn~Pccl~lful brands for a longer tlra Period, it lay grovide a feed bac~ about the relevaace of the differeot target and restriotive values. SPnctinPs ranking orders of several brands wi3 regard to strength factors can bl explained ntililing the eOXS~ I~D PESP015IvEIE85 approach, wbich otherwise cannot be understood. los;gn solutions for a low tar and for an ultra lov tar ;lrand have been selected and realilpd according to dii- fcrelr target deliveries and restr:erlve prop~r:les. ClibPDF - v~~fastio.soni -2 Content i, Meaning 2. Objectives 3. Splcil! fcltu:19 of the mIOI ppyW0 ~BWDIPIEIS approach 3.1 Li~ of con~cner dlserip~ion~ 4~ CiglTPtfl data 3.2 "Cause and oifact" ralationships 3.3 Additional io~cnations 3~( Diiior~nt siDoking paraaltars 3.5 Diiieront types of sroting behaviour 3.6 auality oi,target delivery 3.7 qPant:ty of target delivery 3.8 Ptssr!eti~s properties '· gxprasp!ons related to CONSUMER O~D RESPCNSiVENESj 5. Areas of appiiearions I Exanples Ida~ti:ieation or potantial produe: advantages 5,2 Eplanat!m al and !(ld back ira conlPer assesslcnts 5,; Facd back of infcrmdtian from sucessaiul brands 5.1 araPC development Appendix A: Definition of 'sia~Flc:ty' Appendix 8: Suppo~fng tools ClibPDF - v~~fastio.soni .3. ~l~~dP~ The CONSi?l9 ODLAND IIESPCNSIlrMESS of a cigarects :s def:ned as ~~e de~ree of the capability of a cigarette, to provide the smoker with / a larisfy:ng amPllnt 3[ snoks without unpleasant effects, 2, ~ Obiaetivas ~Ltima~e eS!ec~ve: -- to develop and modify Producs j sore specific to consumer.F' feedback,j I~ze~~2te 0Siectives: i to translate smokers' descriptions into useful data, / the:. can utilized by the product developer, Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 - 4 3, SPecia! features of the eONSUIIER~DMAND RESPONBTVENESS apwoaeh 3,r4ink of consumer dcseriptFons~_agd cioare_tte data The CONS~E~~ DE~Af~C RESPONSIVENESS approach is to interpret consumer descriptions of sensory attributes into useful data physical and. chemical data of cigarette and smoke), which a product developer can correlate and apply, For this c~jective "cause and effect" relationships (beyond mere statistical rela:ions~ips; are utilized (see 3,2) together with addi-,ionzl informations from several areas (see 3,3): 3,2 "eause and effect" Jllsticnships The 'cause and effect" relationships describe - the smoke content and amount as a function of the puff volume ~at a smoker takes, · the deteninaticn of the puff volume by physiological condit:cns. by the amount of smoke, which is satisfying i = "ttr~et delivery"): and by limiting values of several 'restric:ive" properties such as pressure drop, _3,3 Additional informations The "cause and effect" relationships incorporate knowledge - on the dependences of t~e amount of several smoke consti- tuenrs on c:caret:e design "factors" (such as degree of ventiia::on, degree of filter retention), and how these dependences are changed by varying puff voli~ples, · on the smokers' behaviour (see 3,5), Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 - 5 - on the levels of smoke amount wanted by the smokers (see 3,6. - on the acceptable levels of several prope~:ies· and 50~ these are perceived by the smokers (i.e, perceived pressure drop, perceived amount of irritants etc,! (see ;.~'· 3,6 Different emokino parameters The approach of the CONS113EI1 I~AND RESPONSIVENESS must cover the smoking paranaters, that are realized by the smokers and have influence on r~e smoke content ad amount per puff and an the puff number. These are [in principle): the puff vo:uae. the shape of the puff, the puff duration and the inter~uff duration, The present re~~rt covers only the probably most important action of the smoker: the puff volume, 3.5 Different types of smoki~~ behavio~f The folloG,ino:;;o fundamental types of aokincr behaviour are conslcerec: (I) a saokinf behaviour, which is an individual's way of s~Pking~ and is nor: influenced by the properties of the cigarette and the smoke, (IIi a ssokine behaviour, which is influenced by the properties of the cigarette and the smoke, The ncker:esulates the puff volume to obtain a satisfying amount of smoke and oxaano!eptlc sat:~faction, 3,6 Quality of taroct delivery The --aansaatorn of "sar:sfying amount of smoke" into chemical constituents sill remain a problem, As a preliminary auxiliary Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 -6 criterion only the condensate delivery per puif is used as targer delivery. Current investigations will provide more specific chemical complementary informations (which can be treated by the iorma:ism in the same way as the condensate delivery per puff). The comparing of two brands, based on the condensate deli- veries, can only be mads, if the types of epndansates of the two brands are not toe different Ji.e. the blends of the two brands must be similar). 3,7 0uantitY of tarPet delivery The ~an;itative aspect is described by the equation i u:i,ulipndl. nli·Y0ii~nrvol' it Iv-VOi (I) M(i,O): delivery of smoke constituent i per cigarette at: puff volume V pn(v): puff number at puff volume V v: actual puff volume (mll vD: standard puff volume ( 35 mi) E: mean elasticity of the cigarette (definition see appendix) In this category an average on the deliveries of all puffs of a ci~aret:e is epnsidore2 In extended approaches deliveries of "early" and "late" puffs are discriminatad· 3,8 Restrictive prooartiea The increase of t~e puff volume can be restricted by several properties: (l) The puff volume itselL is a restrictive property because of the pbysiPlo4ical conditions of the smoker. Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 -7- ;2! ;i~e:lknbwn restrictive ?rPFer:l' is the pressure d::F of the snoi(ed z~4~re~e, vhicb app;oxlate!y linearly increases uit~ the puff volume tar constant puff duratidn)~ i!) ~ier restrictive praper.ier are e.g. the amount of ~5:i:ating substances delivered, (It is obvious, ti~at all restrictions and targer values ale randwll distributed, $;I ehe ass;;nption of the existence of irequealy occurinq values is:cascnahle, which then mav be used as upper or lover 1Li:s·) (, EXOr!ssions related to CONSUhERPMAWD RESWWSIVEHESS; "Ranae of aleusure". "2oFe of ccriort", "zPne of ar=Parance~ Excrassicns such as ''ranca of pleasure", "tcce of COmfOr."l I~zo~e of acceptance" an 5yncn@ous for the same perceptions and car, be IcrSer deiinec: by r.4e;arqer dellverl, and ;.ie restrictive properties· ''A;i~:d re: effort" This expression is t~e rat:P of target delivery relative to tne pressure drop that has to be aepiied. to achieve the recuirai amount of the tarcler del:va~~ This ratio may be used,:f ?ress~re drop is the only restr:c- tive ?r3pertl considered, ad If cigarettes with si~ilbr s~o~e deliveries tar standard c;niitions; are ccmparhd. Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 ·8 5, Areas of apolieatieas i hamPles 5.::deneitieation of p0teniis! Irodllet advantages CONsWG DE3W: RESilONSIVZNSSS can be applied:c identify of single brands in a given tar pcten~ia! I;:odue: advantages segment. Is: this study several criteria and several types or comparisons are illustrated in the tollewing examples. 5.1,1 eonde,ate·nicetins "eapabilit~-curvell :f only ePr.densate and nin~ine de!iveriel at standard ~lakincl conditions are considered· a cigarette is cbardcrerizd by ,,, e~ inLXY diagram (X: condensate, Y: nicotinei· in the CONS~I DMRND RESP6NS;VENTSS approach eacS cigarette is Janr.erizeci by a speclfic~i~ ("cagahility cutve")· the e"~7E represents all nicotine and condensate deliveries, w~ie~:~e nokers can obtain frcn the cigarette (per puff). The rapce of tte curve corresponds to the range of the puff v~ime, which is Ilitod tY the aiore~entionad c~estrictions. The u:?er limit of the capability curve is determined by:he lovest:i the nicotine and oondensare values, which result from re different restrictions ie~g~ from the maximum puf[ volune (Vnaxi Or the maxirplco pressule drop (POmax)). ay aear.r of Fig. Exl some of the applications of capability curves ~n Illustrated: i The capability curves of LOw (I~) and VLTWI LOW TRB IULTI brands shov that smokers can (pr cannot/ get the nifor:Ee or condensate values jat standard conditionsl of FC~LL ~LXOOfi brands (OKSIUSKS): Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 .g. Pig, Rcnoe d ar and dr ddkedos per p~d dlnnt ~ and UC ::are ~aulbo Vma= 70 mt nd PDmw = 180 mm ~ ~16 · Q!4~-----'-------~-. I--------------------~--~~ - 0.;2----· -· -----;-T~I,-----. /USKS I~r .rPDma : P Q1~-----rc~-~ ~i~i~ -~~~*--r~ i aos ~~m~ d .~:·· · /,~_r. ,4. --------------------------· ----- .n O,OE ----~ ----- 0.02~ o' · · aa as ~B ii 14 16 1.a PMWNF pl~ pd Cg) ~ ~sul~ s~ WPU I !~uIl ti ClibPDF - v~~fastio.soni '10 oniv ue LnT brands I~n~UL,BI do not provide the reference nicotine level (Ff), none of the six LT and ULT brands ?r:e vides the reference #ndensa~~ level (FF). with regard to t~ei: maxiiPun nicotine values the brands as be ranked jusr;i, DLi, P?r~Li, A) > (PMITZ) > (8] I with regard to their maximium condensate values (USUL) > (D~i, PMLI, A) > (PM[TL, 8! · !21 i6r any condensate carqPr quantity assumed (fixec! value on me X-ax:s1 the corresponding (different) nicotine deli- veries are qiven~ Vice versa: if nininum nicotine quantities (fixed value on the Y-axisl are required, the corresponding condensate deliveries are shown. (:I TSe slopes of the curves represent the nicorinalconden- sate ratios as a function of the target deliveries (i.e, also as a function of the puff volumes). lo:aai!y the nico~inelc3ndensate ratio is decreased by ar, :3c;easing puff volume. (5:ight dee:eases are exhibited by I1P, P~LI,PMtn,B.I (41 The reason for the limitation of each capability curve is identified. (Xere: either pressure drop (PDmax . 180 mm WC Ic?.'l Ins.WGi) pr puff volume (Vmax · 70 mil ), Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 - ti 5,!,2 ~aPlatieos ttem tnlcrance limits ~nn:el vay ;e eet a csmpac'- inf~native su:7ey on -,~e C3NSi?l3 D~NI! ;1ES7ONSZVEY3SSES or' several eigarer;es in a given tar segment is il!us-.:ateC by Fig~ Ex2: Six Icz,:ar brands (A-f) are a;pared. with IKWNF i"partic~lat; aa:rar ;iater nicotine free") deliveries ITOFi 5,9 to 9·9 =F~ (The respective PSNF values are indicated above the bars' , fSs *sss~d target quantity is the condensate de!ive~l per lavor brand a: !Z 31 ccif :f a full f' the Issu;erl restrictive properties are puff volume, p:aj;~rfi 2:32 and nicotine per Tufi· The assumed reqect!ve aa~:~:~a o~;icinuJ valuer are vmax 1 70 nl, 3Dmax · 7 ins.we, NI:Cr,i3 = '3,1 mg, The actual values of the res;r:c:flre properties result fr~; the actual puff volume ac~l!ed. to get the condensate :a:eer de!:very, The actual values of the restrictive pr"' ?ert:es are divided by the :especive maximun or minilc:'l~lces and 3:ver: in relative units IJ;~ T~e brands in which the restrictive properties exceed the naxi~uc values or do not reach the mir~imun values, can easillr be identified by the "',-iple bar dla4ral3" (fi~. Ex2' and are assumed to be less consumer demand responsive, :3r example ~U~e followinP. conclusions follow fi~om fig, Ex 2: ,,: 3:ant ~ is insufficient because 31 its excessively ~;P pressure drop and its low nic3t!ne de!~ver~. ~s~ 3:;nd C vie!ds an excess:ve Encunr of nicotine, >,: Brands A,D,E,: are more 3: less "balancec~" wit slic~: a~vanraga ct brands I and F ialehou~n ' nas t,LD Icues~ P?T~Nf 'talue tee: siga:e~;el ~ Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 ·12· Fig~ Ex 2 C~Pv;atlr~n~ Irom tolerance limltn [F.pas=:3. T3mils=150. ~iCain=3,1 ~h',SF:: Pl~rY~;/J~ii =1,5 mg 165-1 - Is! ;201 11 5,P "100; 4 eal 2r,- A`B'CD E F -- p,voiii0 ~ p.drop/lB0 I nicl0,l Clit; PDF -!::!!::!!::!.f3 StlC,i: 0111 ·13 - 5,2 E~!aaation of and feed hack free eeomnv aste~ ~C:e::he orqar,olep~ic proparr.le~ of several brands ~a?o tee: assessed by a cpnsu~er or an expert panel, usually tcsa taste itezs, whic~ are related to Is;reng:h" impre55Lens, (Be soFa!!ed "s::enqt~ factors") are associated wi:h the condensate or nicotine deliveries of the brands, One prscedare for co~pa:isor, is the co~parison of the rank crier of t~e brands according to the strength factor ui~l the ranj( order of the brands according to the smoke de!iver"· The sa~oke delivery can be expressed by different reference values: (li occasionally t.i.e total smoke delivery per eiga:e-.:2:s applied, (i) eenera:!11 the noke delivery per puff is appliec, ve:~ often this reference value is sufficient:o ehain sini!ar rank orders for the strength rac:x and fc: ~6 smoke de!ive~r. In these cases there is no nesd for other reference values~ (~) Specifically for the LT and UI~T segment a smoke d,pilerrl per puff of 70 mi is appliod~ 14i Within the CONStllER DfMRNO RESSONSIBENESS approach the smoke deliveries which result from the assumptions or :ar:e: ~el;veriel Ind rerricior, values can bs applle0, ~ example is given in tig. b3: Four lr brands are compared successively using the reference va!Ees ill to iC). nio of the far brands (li,c) ge~ disrinc:li' differentranje positions by the different reference values· Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 .1C Pie, Ex 3 Rbn~ina 3( the brands A, B. C. D f~r ~~ier~nt refer~nce vbiues (1) t~ :~I 1:0 - ·'~" A A C C 10G- i B B 9c- 50- " j~rbp ~nL~n ~~~n ~~~~ ClibPDF - v~~fastio.soni -15 1 For example: .A s::aqt~ factor ranking order with C, i can only be exPlained by reference values (3) or (s). (For (l) tie following asssptior,s have Seen made: target delivery: 1.5 mo PMWNf/Puff, restrictions: maximum puff volume Vpax ·10 mi, maximum prassur~· d:oo P~maX. 180 mm wC (ca, 1 ins. WG)I · I strenoth factor ranking order with B , R can on!Y be explained by reference value (1)· - A strength factor ranking order with C cr.. D could have been pxplainPd only by reference value (1), There are two aspects of;his proceeding: (Ai 5ome;;mes ranking orders If 5;:eng-,h factors can be ex- p:ained, which otherwise cannot be understood. (Bi The aqrecPcl~ of the st;engttl facrlr rank order with the rank order of one of t"ie four typer of ~e smoke deli· ve:3 reference values suPPOI: tie asn~ap~ians behind the res:ec:ia reference value ("feed back" aspect) · Fc: example: Ii, for a special class of brands, the bes; agreement with the 5::Pnqth factor rank order is found by the reference value;'ipa (2), this indicates, that:nis class of brands is preferentially smoked with an average puff volume of 35 mi. · It the best agreement is round by the reference value type (]),:his indicates, that this class of brands is prefaren· tialll smoked with an average puff volume pt 70 R1, · lf ~le best agreement is found by the reference value type (ii, tais indicates, t~at:hi5 class of brands is smoked Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 -16- vith a ?utf voluJe higher er,an 35 mi anti rhar the asmne~ ;est:ic=idns are valid~ 1,3 Feed baci~ of ipionation iron auecesrl~! brands Reather feed back of information Ean.origiiate from eva:nat:aes of successful brands, which are presented in tig, Ex2: The values If the assumed upper and lover limits of the restrictive p'Dpe'~ies CC? be varied, Those values, vhic* Ir~vide ~halanred triple harr" (see 1,1.2 anE Ii4· Exll for ;Se mps; succosaful brands. ale acceptable values. 5,~ Brand development ExamPle i: foX tar Droa~ W1"!'~·~slLWL=··""--- The upper limit of the pui( volume realized by some Lba TAI1 nokers is asnaaC;o be about ?O mi, Ces;Pr, and blend of sampic ii are chosen vith de falloving objectives: - P~i~u~; figure as brand, L1, ,p~tirlflnie3tine ratio not less t3an 9 (at standard conditionsl· - nicotine and condensate de:lver! per puff at iO mi ~iseinc:!y more than that of II, - pressure dI3e It 70 mi comparable to that of Pi at 35 mi. The results are shdun in tab, E!: Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 ·17· Tab. 51 FHWNi Inicdtme I~MWNr ( nicotinel p.droy! ;Y .i per per I per per cig, cic. puff puff scr:ie (]Pg) (mq) (mg) (mg) Fl at 35 mi 13 0,82 1.6 a,io 3,6 I: at 35 mi 6.7 O,SO 0,79 0,059 ~ at 35 mil 6,2 0,68 0,71 0,078 ii a; 10 mil i 1 1,6 0,105 1,9 L2 at 70 mi 1,9 0,189 1,2 ; i I I I I Te design a high potential delivery per puff it is important to start with the highest da!ive.y per puff at 35 mi, which is possible under the given restrictions, and then to provide an elasticity as high as possible, A corresponding comparison of Zr and 12 is shown in table 12, Tar. EZ irelat. difference Il vs, I; (~) ~!n~b~TiFlpuff at 35 mi I - 10 nicotinelpuff a: 35 mi ,32 ~csndensata elasticity ' 47 jn~Pt:na alaclcici tl~l Ex~ols~Z: Oltn low tar brand with_hioh condensate elastieitY ~he upper limit of the puff volume realized by some UUT~W. L0W TXFi smokers is assumed to be about 70 mi, Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 -18· Brand ~1~3 cl3pared to brand UL1 gives an example, that an actual P~HF delivery per puff (at 70 mi) can be maintainci, even if the PMWNF delivery par cigarette land. Cu~B P~WNF delivery per put:) at 35 mi is decreased by more than 501 Itab, EZ), Tab, 85 at OLZ P~NF per cigarette (at 35 mi) (mg) 2,3 l,b RRII per puff (at 35 ~) (mgl 0.34 0.11 PMWNF per puff (at 10 mi) (mq) 0,83 O.B1 i The high elasticity of UL2 is caused by the specific dependencies of the tobacco rod delivery, the filter retention and the filter ventilation on the puff volume. In Fig~ EZ the successive (not purely additive) contributions ef these thral factors to the per puff delivery at 70 mi are illustrated, Abbreviations : Pp35: PMWNF per puff at 35 lpl Pp7o(:]: PMWNF per puff at 70 mi, if only the increase of the PMWNF delivery from the rod would contribute to the increased PH~NF delivery of the cisareae (hypothetieal value) Pp'lO(r,:ctJ: P~WN~ per puff at 70 nl, if the increase of the PMWNF delivery from the rod and the decrease of the tilter retention (at constant degree of ventilation) would be the only contributions to the increased P~IWNF delivery of the cigarette (hypothetical value) Ip70(r,ret,vf)l PMINT per puff at 10 nl, caused by the increase of the PBWNF delivery from the rod, the decrease of the filter retention and the decrease of the ventilation tactual value) Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 ·19· ~p~Z~dll~Pf who revisved ThB first I would like to thmk p,g.?lulhollen version ot this paper and preposed many hal~ful im~roven~enrs DB co~isrma oMINn I(ES~ONSIYEN~Si Ipp'ole~ *"el!to fro~ the of the toxt· ~qrimls "t ": p~dYit davtig~' LT1:U5 ~ cf nollh~ ~hlYi~~V IIUdi(ll thL *Ya been periarme~~~ Sevetal iaboratPriss duTina ~s pas: decades~ The ~rr·r---- ta bank roierences is larg;, and tharafete we refer to the de of the library I 1 Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 U1 ri ·20 · ~pnrpsi~u; n~ni~i~bnc~ na~d~r CI Cll*~"i an b rd~ ~ ·'4 nlslL ,i ~( msnt llb o~ !Ilonlt!oP ~ts co~lti~~nt. Ilcaall ,~Uld· r Ld·at pr 911·, ads b· al~·o rd darn ~U*d·~ 'g cm~~l ·d d~lhe 4au· Irddt· *1 ·' 'I"~o 1*~·4 *' ~I *de It B' ~*I~h41~~L1 *r I·'YI I~~a hs~bl ol ~( dr dli~a hlmd ~~ ,D ~1 potmtf~l Lcna~~ s;~ndlrd pPii tol~c: condlnlatl ~~n P~ ~Uii Ilastiei~: ,.~~na)k~ai- Prfll~I~'~V~~ i P~l(vl): ~r dlli~lV Ir loii rel~e pi Pn(v~: pli a~~ It ppii 10101 91 tO Itandard ~Ii ~1~(· ]~ 1112~ y:: r~llrmcl puii vol~~ BliltlDel Pif ~el~: ~ ~r~ dS~d·0 ~ ~''P~ ~llrLI m at pb al·l ,,~~ ~c~L Icr~cllh pL~~ 01 PsddB ~,Y * ~.nl~ DI ·'d Idi ·b·:· e~e dr cq~ri~~nt~l ~iiort a linsar ,ppr~intion em be 151Pd. ii dr aic·· Pia~ ~o· dl"P1 *dh(~ r~l rr oi Ipglitlti0D· ClibPDF - v~~fastio.soni ·211 and equation (I:) (senion:·:) :Ic~a: Aldoudh equations I::) look very sini:ar, de me~nings are comeierely different' · equation IIII specifies Ici9"erte proP''tY (olastieiY)I vhicl is constant for a given reierence puf f volume I of the actual · eaa:ion (Ii besE~i~e9 the dependence ,iunction of target and restrictive smoke deliverl aS a function of cigarette criteria of the s;oklrs and as properties ii"cl~ a!asricitY)· The definition pflBi a~pharilor thp dytstn~e of Lie ac:uJ increase of the SP0k( de!iveri PB' increase of the puff volume t:om an assumed ;inear increase vi'h pui~ ygiUDL: definitions is given by The relation between the ts;o (IV) ,...cr*C' "'~ Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111 -22. IsPe?d:~ 5~iPOP~'~IL1~99~~ The ~roslem of :inXlna coar~r! desCtiptionJ and elgarecl dl:1 :I c~aracati2d by incDmplete and vague ("fulry'i knovlel;?· G~nerailv analytical statistical ae~lods are applied in air arts During the las: years on the area of Ilknavl,dqe roprosg"l'l'~'' and nlncuiedge pf0ceainq" malhds and tools have been devrlc?$. uticl seen io be useiu! tor the problem above, Ce nin benefits are not r~e avai~hla nov soitvare tools the handling cf the problem essaa- (altlcugt their facilitate hu:tle methodical ap9:)lchea IO' IClila:ive simulatiossn · The scf:i;are tools, vhifh are Enrren~!y applied by us, are a sprellr$ot progra~ vith Ic:lve graphics and a knavlodge beset ~vs:es shell iXi Plusl, Clit; PDF -!::!!::!!::!.f3 StlC.i: 0111