Measurable economic consequences of investments inflexible capacityCitation for published version (APA):Corbeij, M. H. (1991). Measurable economic consequences of investments in flexible capacity. InternationalJournal of Production Economics, 23, 47-57.

Document status and date:Published: 01/01/1991

Document Version:Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can beimportant differences between the submitted version and the official published version of record. Peopleinterested in the research are advised to contact the author for the final version of the publication, or visit theDOI to the publisher's website.• The final author version and the galley proof are versions of the publication after peer review.• The final published version features the final layout of the paper including the volume, issue and pagenumbers.Link to publication

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, pleasefollow below link for the End User Agreement:

www.tue.nl/taverne

Take down policyIf you believe that this document breaches copyright please contact us at:

openaccess@tue.nl

providing details and we will investigate your claim.

Download date: 08. Feb. 2020

International Journal of Production Economws, 23 ( 1991 ) 47-57 47 Elsevier

Measurable economic consequences of nvestments in flexible capacity

Michael Corbey*

~acultv oflnduso ml Eng, nee, mg and Management ~ctcnce Et nd t,o ~ en I/nn t.,~tl o/ Technology P 0 BoA 513, 5600 MB Emdhoven, 7lie Netherlands

Abstract

Investments in flexible cqpacltl2¢ are large It is not surprising that a sohd econom,e ,nfrastrueture Is a precondition From the straleglc prospect,re the structural support will rest on c aahtative grounds Yt t, the central problem posed m the present article ts how the economic consequences of mves),~qcnts in flex~b!e :apac,hco c--.r, actual!3 be ~eas_wed h ~s worked out m tlaree steps 1"o begin wm~, tnc j.~t-'.e-T~me concept ~s placed m an integral framework ofgoods-flow control Th~s is essential because ~t will be found that, w~thout th,s anab~ ~ there is a dang, er that the esUmate of the posmve effects will be too optimistic Subsequently, we introduce a concrete proot.c',on situation Lastly, for the selected production $1to:atlorl the concrete economic consequences of a flexible capacity investment o.l: bc c _ _^.~ c-,,. ¢t~r~,no nnmt t~ that the investment will m fact lead to an adjustment ofzhe log~stics concept Foa that reason, the maximum Just-In-T~me advantages can be reahsed At the same time the possible misrepresentation resulting from "'trad~t)onal" eost~ calculation will be examined

I . I n w ~ , d n e t i o n

In recent years, many enterprises have been confronted with problems generated by the growing dyqamws of the market. The features of the dynamic market are expressed in - growth tn customer-specific products (going to segmented ma;kets, mdlvldual customers), - h i g h degree of rood,float,on m products or component parts (short hfe cycles)

Compames failing to come to grips with these developments wl|l soon suffer the consequences of greatly intensified competit ion A high degree of adaptablhty is called for m selhng, production and purchasing A constant process of adapta- tion of these tunctton~ is called for Adaptation

*lr Michael L' Corbey works as a sclenUfic researcher in the subject group industrial Economy, Faculty of industrial En- gineering and M~'~agement Science, Elndhovea umverslty of Technology He ~s also active as advisor to the |nstttuut ln- formatletechnologie voor Produktautomatlsenng ( ITP/TUE- TNO~ ,~. E,ndt'~'.~n

problems internally and those that exist between the various functions ,,n industry assume the form o f l a r g e d i v e r s i f i e d M o c k . ( u n s a l e a b t h t y r i s k ) a n d

long dellver~ limes. Some examples can be found m Brmkman [ 1 ].

By lntroducmg new working mett~,ods, pro- cesses and adapted logistics management, com~ panies try ',o ward off the abovemenuoned ef- fects Lot size plays an important role in all of this. The last decade has seen the emergence of new logistic management concepts for which the traditional manner of determmmg let sizes, e.g Camp (Harr is-Wilson) ECQ formula, ts not op- timal Namel~¢, logistic concepts such as Just-In- Time (JIT) and OPT (Optimized Production Technology), which emphatically endorse that point of view, state unequivocally that the aim has to be the reduction and, if posz'.ble, the ehm- matron of lot sizes. In other words, the opttroum I ~t slze IS o~.~e The magic word !s Technical Flex- lbthty Machines with short reset times do m- cleed citable lot sizes to be reduced without any great loss m effic~enc~ and dust-in-Time hun-

0925-5273/01/$03 50 © 1991 Elsevier Soence Publishers B V All r,'ghts reserved

48

dhng of orders to be mstztuted Th~ JIT concept can be a powerful weapon in the hands of a com- pany determined to stay competitive in a dy- namic market Schonberger [2 ] speaks ot

" . JIT attempts to control costly resources of ~ d ~ e , SU~n a a

- idle inventories, which constitute waste of scarce material, resources and, indirectly, en- ergy for basic material conversion and refining. - storage of idle inventories, which wastes limited space, -defect:ve parts, subassemblies and final products, which are a waste of materials and

energy" With this last point, Schonberger means the

Total Quality Control reqmred for the proper op- eration of the Just-in-Time method The positive effects of improved product quality is not part of the present article

There are also drawbacks to the advantages just mentioned Apart 7ram all #'_c orgzr.~,~.t iona ~ pr,~blems following tl , , introduction of a JIT co~_cept ~_l~_to a (Western) company (for :n- stance, see Durhnger and Wortmalm [ 3 ] ). cono siderabie amounts of money have to be invested in machines capable of flexlbiy processing a wide range of products To keep the JIT concept op- erable resetting times have in fact to be drasti- cally reduced Authors like Shmgo [4] have shown that, in given manufacturing environ- ments, a certain amount of reduction of setting time can be attained with relatwely 5~mple, low- cost techniques and organlsahonal measures But the posslbflitms offered by such low-cost meas- ures are not mfimte In this connectmn it should not be forgotten that the success story of the set- ting-time leduction of the big presses at the TOYOTA factory, from, 8 hours to 2 minutes, is ~pread over a period of fifteen years. If results with the introduction of the JIT concept have to be successful in less time, then structural invest- ments m flexible capeeity are a necessity.

Investments In flexible capacities are volumi- nous It is not surprising that a sohd economic infrastructure is a precondmon. From the stra- tegic prospective the structural support will rest on quahtative grounds. The continuity of the company is at ~ssue because, without flexible ca- pacity the competitive power of the company will be Insufficient.

When a flexible capacity investment has to be just'~fied, it :.s often done on no more that ~tva- teglc-ouahtatlve supporting grounds It ,~s ~a a-- gument which calls loudly to the entrepr~ l~eur to take tl~e risk, in fact the lack of concrete financial savings m hard fignre~ l~ replaced by an ~tu~- tlve, strategic vision which certainly cannot af- ford to be wrong The primary task of the con- troller is, however, to reduce the company's financial risk to reasonable proportions and keep them there It is thus for good reasons that the controller will draw up an estimate of the con- c-ete savings that can be made with these investments

The central problem posed in the present arti- cle is how the ecoqomlc consequences of invest- ment m flexible capacities can be measmed it is worked out m three steps

To begin with, we placed the Just-In-Time concept in an integral framework of goods-fow co,~trol. This is essentml, because it will be found that, without this analysis, there is a danger that the estimate ~,,i the positive effects will be too op- timistic (Sect:on 2 ) In Sect~cn 2 ;'-.e ,~: ~':,,duce a concrete ~.2oductloI~ SltuatJon, namely produc- tion of component parts folio'~ed by an assembly line Customers' orders are assembled absolutely flexibly, but what we have m the manufacture of the component parts ~s "tradltmnal" production m lots It is assumed that, by investing in flexible caDacltv, component parts can be made directly to customers' order It will be shown that, with- out adjustment of the integral log s:,es concept, all of the potential advantages o' toe investment in flexible capacity cannot be ext.'o:,te,~.

Lastly, for the selected product)on situation, the concrete economic consequences of a flexible capacity investment wll! be examined (Section 4). Our starting point will be that the investment will in fact lead to an adjustment of the logistics concept and that, for that reason, the maximum Just-In-Time advantages can be reahsed At the same time the m~srepresentatmn result,ng from "traditional" costs calculation will be examined.

2. J u s t - I n - T i m e in an integral l og i s t i c s concept

When an a~:empt is made to improve stock policy, great lrlpormnce IS attached to an anal~-

s~s of the lntegra! goods flow (for exa_m__p!e~ see Bertrand and W0ngaard [ 5 ] )

A simple, yet very percept ive example of such an integral concept IS the so< ailed Customer Or- der Decoupllng Point ( C O D P ) control concept [6]

F~gure 1 shows an integral goods flow Five possible decoupling points are marked in. They separate the customer-order part o f the activities (on the right hand s~de o f the decouphng point) from the ac twmes based on forecast ing/plan- rang ( l e f t -hands ldeo f thedecouphngpo in t ) The customei order moves along the goods flow up to the decoupling point, whence ~t is delivered di- rectly to the customer

In general, the decouphng point will coincide with the main storage point. It likewise applies that, in principle, lhere are no stocks down- stream of the decouphng point, and that up- stream there are stouks only when it ~s economi- cally just if ied Thus the first step ~s the d~c~_smn as to where there will be stock points !n the mte- 5~al ~,l~aiii T~,c =;lo~c uf the Vo~,,.,..:, oi :he de- ~c,o,ei;-3 pn~:,, ~s "" 2he ,qr~* mstaqce, a process ,~ :~ghmg market r eqmremems t~mong other

things, delivery dates demanded by the cus-

, a ¢ m a l e , , a l ~

- - , ~a, ls

, I

49

tomer ) ag.nmst throu~.:put O_mes m the produc- t ion and d~stnbutlen process These, :n turn, are decisive for investments m stocks and m the pro- duct lon process. The hard dlStlnctJon that this control concept makes between customer-on- enied acttvltles on the one , a u u and activities based on planning on the other, grea0y simphfies the task of illustrating how the J IT pnnc~ple works

According to the J IT control concept, there should only be productlon when there is an ac- tual demand It is m fact assumed m the present artmle, that the demand process m the chain is not intended f ~r the purpose of building up stocks, but actually for bandhng customers ' or- ders. This observation lmpiles that, after the cus- tomer-order decoupl.mg point, the Just-In-Time product ion has to be ~mpiement~d up to, and in- cluding de l Ive~ to the customer for instance also the physmal distribution! This is an ex, remely ~mportant marginal condi t ion for ,.nvestment m flexible capacmes and will be con, !tiered m mo~- deta,A *_.,., S~et,.on 3

The capacme~ after the customer-order deeou- pling point have to 0e very f ie~oie m order to absorb the changes m m~x a n d / o r volume which

©

&

L

2

/

d

. u

C ~

I-'tg 1 Ftve posmons for the decouphng point (DP) (Source Van Hees [ 6 ] )

50

are Inseparable from customer order~ in a dy- namic market, wtlhoul disrupting the agleed de- livery times Furthermore, m this concept, the order-acceptance function of mediation can also be apphed to deal with limited flex~h,,hty beyond the customer-order decouphng point

The lmpor:ance of an integral analysis ~s thus demonstrated Shifting the CODP from DP-2 to DP-3, for example (see FJg 1) wdl at once af- fect component-part production before assem- bly. That d e p ~ m c n t ~,dt then .aot be m a pos~- tion to paCs parts to assembly at will, but only when there is demand from assembly This means that ec-npo~ent-part manufacture can be faced with a build-up of stocks if the production r'la.,:- nmg is not adjasted if the mtegrm cnam is not kept a : ]er constant observation, this effect could easily be neglected and even lead to the ~umpo- nents lhctory being negatively judged on account of ~t~ b ~n, ' ,o excessive stocks v In assessma ~he situation, one should therefore be very careful to reahse that ~"en J!T ~s aoFhed, the main s~,~ck point has It, Hlove upstream m the chain. ~ar~ Hees says that this is related to the entrepre- neur's risk mentioned earlier It ,s illustrated lC~ Fig 2

Considerable amounts of money are invested in DP-1 in stocks of end products spread over a number ofdlstrlbutmn points This implies a high risk of being left with unsaleable goad: -~t D o-~ on the otne~ hand, the risk lies mainly in faliule to meet delivery times and in overrun of precal- culation costs of the project Finally, the entre-

F~g 2 Entrepreneurs risk and the locatmn oethe decouphng po:vt (Source Van Hees [6] )

preneur's risk related to the fixed assets w,,.i! also be increased because of investments in (costly) flexible capaotles when the decouphng point is shifted upstream

The strategic appeal to taking the entrepre- nea:sh~p risk referred to lp d~e introduction is mus certainly justified The question is however, whether concrete effects can be discovered which support such strategic intuition It is for this ~ur- pose that ~ concrete production sm.atmn ~s examined

3. Concrete production situation

An industrial enterprise manufactures a ~reat variety of end products it consists of a namber of component part factories and an assembly line In ~he interests of slmphclty we shall concentrate on the combination of one components factory. plus an assembly line The customer-order de- coupling point is positioned beyond the compo- nents factory

In the components factory In question, work includes heavy pressing on a traditional rougiaiy 2000-ton press. Hundreds of dlfferent semlfin- lshed products are pressed from about thirty dif- ferent raw materials. In addltmn, the factory car- rles out various cutting actlvlt:~s with conslderabl ,~ overca~ac~tv The ~ress*~d semifin- ished pakt~ a~c a~oembled on the assembly line together with several parts from other compo- nents factor,.es The possible combinat'o~.s are practically unhmited It could be said that the heavy pressing work of the components factory &vldes the product range rata 100 main types (semlfinlshed product) Depending on cus- tomer orders, the main type is processed on the assembly line to a practically unique, customer- specific product There are m fact ~_housands of types of end products

The components factory centres on the big press As the resetting tames are cons!derab!e, a lot size rule is followed, based on Carop s EOQ tormma, w m ~ n I . l~ tCl l l l l l lCa [h(~ Iot ~lZe i.'y nora- paring the costs of holding stocks of semlfinlshed 7rc~t:ct with the costs revolved m resetting the press Green this lot size, t~ae internal co~t price of the semlfimshed product can be determined The structure of tht~ cost pnc". !s as follows"

51

- The material value (to be paid to suppliers): 30% - The fixed part of the added value (fixed man capacity, deprecmtion, etc. )" 60%. - The varmble part o f the added value (energy): 10%

Sales have greatly increased m past years thanks in part to extensmn of the already wide range of end products F u ~ h e r growth ~s, how- ever, impossible without budding a new assem- bly line, as assembly is already being done in three shifts. How :ver, no substantial further growth is expect td In the market that IS served, so that the company has reconciled itself to the fact that ~t wall not qerve a (relatively ~,nall) part of that market

The growth o f sales has also had its effect on the component-par ts factory Capacity problems arose with regard to the 2OO0-ton ~;ess T~Is has led to contracting part o f the wo-k out (10%). The price charged by the suppher charges amounts of 110% of the internal price. Growth has also brought the regular need to work over- t~me m the event o f breakdowns.

The delivery t~me agreed with tree customer is longer than the throughput t~me of the assembly process, but not so long that one can simply shift the de~ouphng point to DP-4 (see Fig l ) Th~s would mean that the Camp tot s~zc osed for the 20O0-t(,:a, :ess would have to be clropped m fa- vour of working with flex:ble customer-order lot s~zes (lot s~ze 1!) The ratio of setting t ime ~o product ion t :me would then become so unfa-

durable, that ser,.ot~,s efficiency losses would oc- cur In addit ion, the work contracted o , t would ha,. e to increase c o ~ d c r a b i ~

| v order to be sure that every customer order c~ , he ~andied m the present-day situation con sloe: ~b!e safety-buffer stocks of every mare type (se ' .mfin:shed) are held a: the main stock point TL. aim is reducing as far as possible the risk of being unable to start assembiv because of waiting for product ion o f components . Such a s~tuatlon could endanger dehveD t imes as agreed w~th the customers ba~c:y-buffer stocks ~s the standard means o f compensat ing for uncertainty

The ct 'aractemsttc problem o f th~s product ion s~tuat~o,~ is that the total number of main types that will be bought is known (namely the maxIo

mtlm number that assembly can process), but not how many of what type In other words, compo- nent-part manufac:ure has no preoiem w~th vol- ume flexiblh~y, but cloes have one with flexlblhty o f the mix. Prebiems with reference to volume flexlblhty (for example, see Bertrand [7] ) w i l l not be considered m the present article.

Because of its .ong rcsettlng (setup) t'.mes, the tradit ional press is a thorn m the flesh of produc- tion management , which suggests replacing ~t by a very flexible heavy press. Managemeet clmms that, after the investment, Just - ln-Tlme "an de- l lvef on call by assembly In other wares, after making the investment there ~s no need to work In the component-parts factory on the basis of the fixed Camp lot size it also clmms that contract- ing out will no longer be necessary and that over- t ime will be a thing of the past. This cotfld br)ng savings The number ofopera~or~ needed ~t,, th~ new press ~s the same as for the old one. The space the new one will occupy remains unaltered. There are no savings here

The flexible heavy press is an investment ra- ~, o.)x ...~,-- ~,nlOnS,.., The ,-n,,*--"~r__.,~, hz~. h,s goabts The concrete expendztures involved in the acqm- s~tion of the fle^,o,~ p.~oo . .~ ~,,la,,n) fo him But how does one clctermlne the concrete yield from such an investment in flcxiblhtv 9

Thv cancretc ;avi~tgs ~: valved in this invest- ment will be considered m detail m Section 4 However , before we can do that we must first make an empha),.c distraction between an invest- ment :n fie>~!ble capacity wtthout and one w~th adjustment in the logistics concept. Let us as- sume that the stock m the lmtlal situation, that is be~ere the investment m flexible capacity, cam- prises 3 kinds - a quanti ty of raw material :a stock due to pur- chasing-order sizes, - a quant i ty of work in process stock ~n comp_~- nent-parts p~oductlon Th~s stock depends on the throughput t~me of parts production, - a quanti ty o f semlfinished product stock at a decouphng poirA where, on the one hand its function is that of a buffer stock and. on tbe other. is ltseif caused by the lot size :n component-p_art pi oductlon.

After investment m flexible capacity wtthout adjustment o/ ' ,h~ l~gi~tzc ce~ce0t, the customer-

52

order dot-onpim~.nolnt remains located bekond the component-parts factory, which means that a safety-buffer stock is stili required -~ *~'~ -~ . . . . phng point The safety stock can perhaps be cut d o ~ ' n "~ h i t her,~llSe "~ho thral~ohr~ll t t i m e I~ the

component-parts factory is reduced We will not continue with th~s line of enquiry here In any case the decoupl,ng-pomt stock will shrink be- cause the lot size stock will disappear I'he quan- t r y of work :n ~orocess m the compenent-par ts factory w,Al also decrease daanks to the shorten- mg of the throughput t,m~ hrcn,ght ~hc,,,* h, ~h,~ investment m flexible capacity, ttms in lact yielding an economic benefit On the other ha~d, the quantlly o f raw material wlii increase as t~:e offtake from me components f a c t o ~ has become smaller, causing bulld-uv of raw material stocks This can, of course, be avoided by adjus mg the purchasing order quantities, but th,.s posslbihty is independent of the lnves tmem in flexible ca- pao ty (Observauon even ~¢the s,.Te of purchas- ing order quantities can be cut down, the upshot can be a . . i . . ,~,. , . ,~.0. , . . quantity, . h,Jg, al,v~. ec,,,, , , ,-,o con~oo,,onoo in th,= o.~o

Let us now assume that the product ion man- agement guarantees that the total throughput t~me a f component-part product ion and assembly, lr-

. . . . c., . . . . . . . . . n ~,~ ~hr,,'~r than dehv- r e b l O e C t l V C ~ t t n ~ ~ H A , . . . . . ~ . . . . . . .

cry t~me to the customer In other words, thanks to the investment, the Iog,,st:cs co.ncept can be adjusted by shifting the C O D P to DP-3 (assem- bly to customer order ~, to Dp-a (producing and db~ClllUllllt$ ~,U~I.U][,.~I order, nee also Fig I) . What ~s the effect on stock size in that case9

The old decouphng-pomt stock of semlfin- ~hed product now disappears entirely The quanti ty of work m process is reduced in the component-parts factory just as in the preceding Sl~U,,,on Here, too, stocks of raw materials ,n- crease, a process which canpot however be com- pletely prevented by adjustment of purctL~smg order quantit ies This is because, just as IL the old situatlon, buffer stocks have to be me, n- tamed at the decouphng point, as the compo- nent-parts factory must be able to start when the customer order comes m The conclusion is that a flexible investment that g o ~ ~'.ong with an ad- Jested logistic concept yields the greatest saving In tbe next section, in which the y~eld from the

s~t,,at,on will be &scussed m detail, the starting penl t will be that the investment does lead to the adjusted s,.tuat,.on ir. winch assembly to cus- tomer order is further adjusted to a sRuaUon wl:h both component-part productmn and assembly to customer order

4. Economic covsequences of a flexible capacity investment

In our case, seven economic consequences can ~.~ "-~,.,~,~a~u.'~ First of all the decouphng-pomt stock of semlfimshed products disappears, It was, on the one hand, set up from buffer stocks m or- der to be able to carry out customer orders in as- sembly at any t ime and, on the other, because components manufactare supplied pans m fixed lot ~zes The~e effects are dealt w~th m Sections 4 1 and 4.2, respectlveb

We have also seen that the quanUty of work m process had dropped m the components factory owing to the cut in throughput t~me. Th~s will be dealt with m Section 4.3 Since stock,, &sappear, there ~s less need tbr storage spa~c, see Sectmn 4.4. The productmn management guaranteed that over t ime and contracting out wo,qd no lohger be needed in the new sltuat~Jn Fhese effects will be dealt wRh m Sectmn~ 4 5 and 4 6 i_ast of all, l~. must be assumed tb_at smaller stocks represent less risk of unsaleables and this will be discussed m Section 4.7

4 1 Reduction o/safety-buffer stock

In trle tradit ional appraoeh, the safety-buffer costs are calculated by multiplying an interest percentage by the internal cost price of the article in stock. In principle there is always a safety buffer and the standard safety-buffer stock level will be included m our considerations. In our case ~t would mean that the saving; on the buffer stocks of semlf imshed pressmgs per type can be calculated according to the ollewlng formula

[savings = standard safety stock

× internal cost price × interest percentage ]

A number of marginal no:es must be made to this calculatmn For instance, we cannot be cer- tain that the buffer stocks w,Ai disappear entlrzly

53

As the main stock point is shifted, the b,.~_. stocks o f pressed sem~fimshed product are m fact no longer required. However , It may m view of the t~tal throughput t ime from the date of cus- tomer order to delivery, be necessary to lay in buffer stocks of raw material. It must be borne in mind that, i f a customer order comes In and there are no stocks of t a~ n~atcrza!s, the tot~_! thrau~h~ put t ime of the process is extended by the dehv- er~, t ime of that ra~ material Shifting the same buffer stocks uostream does, howe~er, have a posmve effect, as the number o f types contained in It drops from a hundred (semif imshed) to thirty (raw mater ial) . Thus, the safety buffer stock can be reduced per type ofs~ock item, since aggregated patterns of demand arise [ 8 ] The tc,- tal buffer stock (in terms of end products) wlii therefore shrin!

Fhe above ~aVlligS calculation suggests that buffer stock in i .~w materials is cheaper than that in semlf imshed products. The cost of raw mate- rials is o f course I 's~ The quest ion is whether the argument is corre ~ Money (capital) is invested ,.n safety-buffer st ~ck Cap'~ta! :s not gratu, tca; ; the company, fo~ e~.~ample, has built up a ba.-_k aeot ~or its capital and has to pay lntele~t od ,?. The crue,~al point ~s thus the extent to which that d_~b~ will drop If the buffer store is shifted. We are faced here ,~,,.h a dec!sion calculation In a going-concera situation We therefore try to cal- culate the ,~onoete savings This means that we must ask ourselves how much the expenditure and prouucnun . . . . . . . . . . . . . w . , tea[/), be am:mo-,,_.ca . . . . . . . . If the buff r stocks o f semffimshed product d~appear Va,, der Veeken [ 9 ] observes that, v, a h regard to labour, machines, overheads, e t c , a lower bank debt cannot really be obtained as the result of lowering stocks His conclusion is unequivocal . the real effect is on the material value and, pel- haps, energy All other expenditures remain un- affected ff buffer stocks dnsappear On the one hand. this can be traced back to expenditure re- curred m the past (a ;qmsi t lon o f machine) and hence no longer subj -ct to influence, on the other, because savings on hours worked will be so mar- gmal that no actual reduction, tot example, in personnel can be achieved The real expendi- tures on salaries thus remain unchanged and the savings are calculaUve fictions It could be ob-

setx ed at this point that real savings can indeed c" achlev, d by working variable hours of over- time. l'hls, ~ ~,aever, IS no~ the case As the work is in fixed k,, sizes and one is always obhged to supply as many items as assembly can process at a max imum (purely mix flexlbllIty~), over t ime and contracting out is inevitable in the stable sit- uation In the new. pa~t-mvcstmcnt situation, over t ime and contracting out will be things of the past as the result of pure Increase in capacity thanks to decreasing the resetting t ime and not because of reduction in safety-buffer stock Sav- ings en over t ime and contracting out are dealt with m Sections 4 5 and 4 6 The real savings on interest costs will have to be calculated by the fol- lowing formula

[savings = difference m buffer-store level

× (materials value + energy ) ×

interest percentage ]

It should be noted that, as regards the expend- itures which can be Influenced. it makes little dif- ference whethel l: is raw material or semifin- t e e* ,o" e l h c u m y u n u g W h i C h _Sll.Cl o:u~x~ tha~ ale kept "~ . . . . . . . . can be influenced is, of course, energyV The real savings are thus mainly brought about by reduc- nan o f buffer stoc/~'~ m the chmn, but not by shift- lng them

In addmon, there is a further observation tc be made about the interest percentage applied in these calculations The assumption is that the l iquid means made a~ailable will be used to dJ- romish the (short- term) bank debt fhls is a real assumption, though one could also consider us- ing this capital fu~ uther purposes Once cannot al, ways be certain in advance of getting an inter- est percentage on short-term bank debts~ l he company 's financia! management ,vlll pron- ounce as to now the released liquid means are to be used It is on this basis that the co~erete sav- ings that can be obtained with said available means will be de te rmmea Yms article will now restrict Itself to the quesnon of how roach liquid means will become avadable by investment in flexible capacity

The above calculation :s an example of so- called ~maney-flow ca!culatlon, see [ 9 - ! ! ] In general, sceptics note that, "al though the costs e f labour, machines, e t c , cannot be influenced im-

54

mediately, the effect mus t come in the long run " This a rgument is c o n e c t in principle, bu t it is precisely this long- term s i tua t ion which is so un- predic table In fact, the wish to inves t in flexible capacit ies is because future offtake is unce r t a in and difficult to forecast in a dynam i c marke t It is thus extremely doubtful , in antw~patton, and whether the " long- t e rm" s i tuat ion in which the ca lculanve savings are to be ob ta ined will ever OCCUr

Tha t all costs are var iable in the long t e rm is cer tain But was it not J M. Keynes who in th~s context once tc, , ,arked, " i n zhe ?o,,g ru,t, - ' e will all be dead ''~ In view of the i nves tmen t of mll- hons of guilders in flexible capacity, t ha t is a s ta tement which should make one thoughtful , for more than one reason

Conclusion. Two mis represen ta t ions in calcu- lat ing the proposals to out buffer stocks are - valua t ion of in teres t expendi tu re at the full in- t - : n a l cost price ins tead of value of mater ia ls and variable energy costs, because of which the esti- ma ted savings are too opt imist ic , - failure to reah~e tha t it can be necessary, to ma in t a in buffer stocks at the new ma in stock po in t Because o f the a b m e valuat ion , the differ- ence between buffer stock in the old ann new main ¢:ock po,nts may be less favourable tha ~ :he costs approach would lead one to believe.

4 2 Reductton o f lot s 'ze stocks

The same va lua t ion p rob lem is encoun te red m the savings calculat ion for a r educ t ion o f lot size stocks in semlf inlshed product . Here too, one will be Inclined to va luate at in terna l cost price m- stead of taking the mater ia l value and the vari- able energy costs There is yet ano t he r p rob lem tha t occurs in lot size stocks which has been de- scribed in detail in an article by Van der Veeken [9] il~ th~ a im is to use noth lnL bu* real eco- nomic effects and not calculaW. ¢ results as the bas,s for the decis ion-taking, then changes in ;,avments in which the outstde world is involved ,~ve to be es t imated

As the lot size stocks of semi f inished goods are to be found in the midd le o f the logistic ch,u,i and r )t at the purchas ing or selling sides o f the firm, one has to de t e rmine whe the r the al tered lot size pollc) ",~ill ~_Lc, lead to changes in the pat-

t e rn o f p a y m e n t ~ on bo th sides Gf the f irm in this case the offtake and p a y m e n t pa t t e rns for fin- ished produc ts do not change We can therefore restr ict ourselves to the purchas ing side o f the company The p a j m e n t - p a t t e r n cbm, ges, due to a delayed Increase in expendi ture , to the a m o u n t of the purchas ing value o f the mater ia l s to be ac- q m r e d are o rdered in smaller quanttttes and paid for tn accotdTnce wtth th is o rder pa t t e rn Van der Vccxen summar, .sed this m Fig 3, in which he assumes tha t the delayed expendi tu res are shown m the , . , ,-- ,va. . /o b , , k d~bL

Note tha t there is no m e n t i o n of smaller put- chas ing order quantltlesV We have, in fact, seen tha t buffer stocks o f raw mater ia l s will cer ta inly bc required, but no a rgumen t is pu t forward in jus t i f i ca t ion of reducing the purchas ing order quant i t ies in the new si tuation. There are no such a rguments in fact The possibili ty of cut t ing down the purchas ing may well exist, bu t tha t is qui te unconnec t ed with i n v e s t m e n t in flexible capac- ity) The real economic effect of the disappear- ance of lot size stocks o f semi f in l shed goods IS

This conclus ion will s tr ike the reader as qui te incredible

Stocks d i sappear and vet there is no economic effect) Cons iderab le effects can be achieved, bu t these have to be a t t r ibu ted to a di f ferent posi t ive effect of the i n v e s t m e n t )n f ex Ib l e eapac~_ty, namely a reduc t ion in t laroughput t ime, a n d wi th it, a reduc t ion in the amoun, ~ o f work in process. i t is namely the case tha t a change in lot size can only lead to a change in the p a y m e n t periods. A change in t h r o u g h p u t t ime, ach leyed as a result of lot size ad jus tment , can lead to an actual re- duc t ion in the , ,olume of work in process For ex- ,*,,,v,,., ,-.,., ,, ,,,,d t , J , . ~ . u o n s , ll~e ~evera! o ther authors , the favourable effect tha t shor ten ing the t h r o u g h p u t u m c can br ing a b o u t This !s die- cussed m 3ect ion 4 3

4 3 Reduci~un m mru,,~hput ,~.me

C o m p u t e r s imula t ion pe rmi t s one to calculate wha t happens when the lot sizes are reduced or, as in the present case, these sizes are d e t e r m i n e d by actual cus tomer d e m a n d . A cond i t ion for this '.s, o f course, tha t a realistic package o f cus tomer

5 5

B a n k d e b t

1 l a r g e s e ~ e !

i 0 t t l r r~

B a n k d e b t ]

I

I i

4 s m a l l b ~ ) J ~

O t ~ t 2 t 3 [ 4 tLrne ~.

Pd~,t l l~l l~ IJct)ud5 to I~l-p, hepb

DI f fo)onco in bank d~ht on which I~ss btltO,:St I lab l ,

Fig 3 Order Quantities (series) reducuon and payr le:,t pe- riods (Source Van der Veeken [ 9] )

orders and rehab!e resetting and processing t imes are available.

With the aid of such a simulat ion, the length of the new throughput t ime in the component-par ts factoLv can be calculated and f rom that mfor- matron the amount o f work in process can be de- te rmined as well. The average throughput t ime is also dependent on the ,~-'~'; . . . . . . io , ~ . . , , , , , ~ ,u , , . s .hat are ap- plied. We assume that these will remain unai-

(~)~)UlII%~LI [ n a t tnls tered Up to now ,~e .'.,ave . . . . . . . . . . . . . work in process will be less than in the old s l tuauon, but we cannot be cer~tain o f that. The effect on the work in process is namely de termined for the most part by the loading degree of the new ma- chine (resetting pl.us productior, in relation to the full t ime avai lable) In )he new situation there

_t 1 are two cnang ng variables which are related to the loading degree, hence the resetting time, on the one hand, is dm*hcal ly reduced, thus lower- mg the loading degree and, on the cther, the work

is done with small, customer-order-dependent lot sizes, which increases the degree of loading

For the sake o f simplicity, let us assume that we have found from the simulation that the work in process will decrease. Again we are faced with the question, "Wha t are the concrete sawngs which are concerned in this reduction of work in process 9''

The total o f slocks held by the company be- comes ~,naller, which means that the company 's requn-eta.,ras ofcapl ta l w','_, be ,'educed Here too, internal cost prices does not provide precise in- sight Just as in the case of the buffer stocks, om point of departure wall have to be ~hat only the expenditures on energy and material can ensure less need for capital. The costs a,3proach in this t.ase too, suggests a higher calcu!ati,,ely fictitious economic consequence.

4 4 R e d u c i n g s t o rege space r e q u i r e m e n t

In most companies, part of the accommod-.- t ,on costs are aiiocated to ea~h dcpdrtrqcnt by means of the costs allocation sheet in the desart- mental budget, q'_ h~ is t, sually based on a rate per square or cubic metre of space used by the u~va , ,m~nt

~.ceg_'~_w~__ocl_n_l.~on costs are fixed ill most cases Reduct ion of these costs for a product ,on deoart- ment because it needs less room is a calculatwe sav,,ig The rate per un,t space is not the appro- priate "_nstrument fGr determining the real sav- ings related to accommodat ion

In principle there are two conceivable situa- tions In the one, the company ~s not short of space and any eventual redundancy will hardly be statable for a different purpose, in which case there are no co~crete savings to be expected The situation is different where the company is fa_ed with a shortage o f space. In ou ; case, ho',~eve:. ~t is not a hypothetical s]tnatmn, because sales have risen ~',,,ip;~ ,n re~.ent years A shed may well have to be (or already has been) hired in the present s;tuatlon because the storage capacity has simply become Inadequate. In this case too, the rate per square m e t ' e or cubic metre does not provide a eorrec~ picture of the gain to be ex- pected when the shed r no longer needed. The real gains are namely determined by al l the ex- penditures involved in r~ntmg the shed Tl,e :a-

56

tal of costs invo lved will, in mos t cases, be more t han the space concerned, mul t~phed by the rate, since it is usually impossible to rent the space strictly required

The cos,'s rate per uni t space is m bo th cases inappropr ia te for de t e rmin ing real savings on space Where space is not a bot t leneck factor, the gain suggested is too opt imist ic , where a bot t le- neck does exist, it suggests too low a gam

4 5 Reduction in overtime

In our case. me p roduc t ion m a n a g e m e n t guar- an teed tha t ove r t ime would not be necessary in the new s i tuat ion Ove r t ime is abou t 35% dearer in most cases The effect on the cash flow is, how- ever, more t han is suggested by tha" 35%, as over t ime, in cont ras t to regular hours of work vary, roughly in p r o p o a l o n to p r o d u c u o o vol- ume The real effects can thus be calcula:ed f rom the variable m a n - h o u r rate, plus the extra-al low- ance fact,., %r over t ime Here the s i tuat ion is nn t so mis leading This is becaflse the lrlves* a.,~nt in flexible capacity actuaily provides more p, odu¢- twe hours wi th in regular working t~me, so tha t more expensive, i r regular ove r t ime is no longer necessa~ ' as a ".,.,hole

4 6 Reduction m work contracted out

The p roduc t ion m a n a g e m e n t c la imed, as we knov,, t ha t there would be a reduct ion , not only . . . . . . L m wu~r~ cont rac ted out, bu t in ove r t ime as well "Ihe seml f imshed con t rac ted-ou t p roduc t is 10% dearer than the in terna l cost price This mean~ tha t a sav,_ng a ¢ t t, oL . . . . per ~emlf inlshed par t cvn- t rae ted could ~e ach ieved F r o m the strategic perspect ive tht ~ i-; correct In *.hat case namely', we assume tha t the factory has yet to be put up aad tha t all the expendi tures requi red for rr,:,- ductio'a have still to b ~ incurred. This p rob lem can be cor, ,pared to a m a k e / b u y dccision. Hove- ever, tha t 1~ not the case m the present Instance. The real yield which can be ach ieved wi th con- t rac ted-out work is far h ,gher t han 10%. One should ask onese l f how mucl, e*.tra it would cost to produce tha t work internally 3 : s t a~ in the case of the buffer stocks discnssed in Seztton 4 1, ex- t ra expenditu*e ~s incurred only m re la t ion to mater ia l and energy The outgoing cash ~..,~ ,v for

I a 3 ~ ] , , _ t [

. . . . . L ! i _ _

, ~ r~a l 2< t c r l c~ , n ~ Ce a u~

~ r r , f , r , l bnec ] c c r t r a t ed )U t

F~g 4 internal product ion versus con t lacWigout

work con t r ac t ca out is reduced by far more t h a n 10% In Fig 4 we see at the left-harM side the m- tc rna l cost-price s t ructure At the r igh t -hand side we see the (va r i ab le ) invoic ing of the work con- t i ac ted out. C o m p a r i s o n be tween the two a m o u n t s indica tes t ha t the real effect ( = ~ d u c - t ion of outgoing cash f low) is _+ 63% of every :n-- ' ,oice for ~ o r k cont rac ted out) Here the costs ap- p roach wrongly suggests a m u c h lower effect

4 7 Reduction risk ofunsaleables

There has been no dlscusslo~ o f the r'.sk o funo saleables m buffer stocks c f s eml f imshed goods and lot size stocks o f s eml f imshed goods One could imagine tha t par t of these a e m s could be- come useless th rough wastage or technical age- lng. In view of a dynamic marke t , the risk o f technica l ageing is always present . In the new sit- ua t ion tb ,s risk no longer exists for seml f in i shed i tems, and zdvan tages are thus to be ga ined here, too. In the light o f the foregoing coi~slderatlons, it will corae as no s u n r i s e t ha t the ,ntc~ hal cost price gives a d is tor ted pic ture of the f inancia l re- tu rns m th is case as well. F r o m a s t ructura l per- spective, all t ha t is invo lved is lower expend i tu re on mater ia l and energy

5. Summary

It was shown tha t there are abou t seven differ- ent economic effects to be d i s tmgmshed , all o f which are invo lved in an i n v e s t m e n t m flexible capacity. We have shown tha t such an invest- m e n t only generates all the suggestea Just-P,,- T i m e advan tages i f the cc s tomer -o rde r decou- p h n g po in t can really be m o v e d ups t r eam as a result of tha t i n v e s t m e n t Otherwise stocks are only m o v e d internal ly in the company , b r ing ing

57

abou t far shghter effects tha t the costs approach suggests a n d do ing less t h a n jus t ice to potent ia l advan tages which can be ach ieved t h a n k s to flex- ible capacity.

An inves tmen t in flexible c a p a o t y which really results in shlft, ng the cus tomer -o rde r d'~ coupl ing po in t ups t r eam has showr~ tha t seven types ~f savings are possible. One can also be m~sled the same n u m b e r o f t lmea aa well. The reader may rightly observe tha t some of the misrepresenta- t ions are m e r e or less t r ivial , for example in prac- tice, space saving will se ldom lead to ma.lorated y:ela calculat ions.

One c a n n o t fall to no te tha t the costs approach promises too op t imis t i c a result in m a n y cases. The " t e n d e n c y to mi s r ep re sen t " will therefore be great for the mos t par t a m o n g the innova t ive forces w~thln the o rgan l sa t lon In fact. it is al- mos t as i f i n v e s t m e n t in flexible capaci ty, as ske tched out here, can practmal iy neve r be just i - fied economical ly! But th~s is no t a l together so If the old press is worn ou t and it is thus correct to speak o f an unavo idab l e r ep lacement invest- ment , the pic ture w,-il be m u c h rosier for flexible capaci ty In addi t ion , a great deal more can be ea rned in the area o f con t rac t ing out, over t ime, etc., t h a n is suggested in the costs approach

W h e n inves t ing in flexible capaci ty it shou ld be r eahsed tha t the " p r o o f of the F, ex lb lh ty pad- d lng" is the market . The cus tomer will make his choice ou t o f m a n y types and ._, rehable , i f possi- ble, shor t delivery, t ime. It will no t ma t t e r to h i m at a / / i f I e gets his goods del ivered on these con- dl t lons, e i ther f rom stock or by way oj~='xlble ca- pac:ty. (Poss ib le i m p r o v e m e n t s in qual i ty ach ieved by flexible capaci ty are no t t aken in to cons idera t ion )

Bet ,er t han that , i f there is a m a n u l a c t u r e r who produces in lots, t ha t is wi th stocks, bu t by good m a n a g e m e n t of these stocks can offer the same del ivery pe r fo rmance as the flexible manufac- turer , the conven t iona l p roduce r cur. p".ss,_b!y be even cheaper . Such good cont ro l of ~lot ~t7e)

stocks, however, requires lnSlght in to the han- dling of an integral logistic concept , as the pres- en t article has il lustrated. Fur the rmore , C a m p ' s Economic Order Quan t i ty is no t a suitable in- s t r u m e n t in this context wi th which to de t e rmlae the lot size, as the costs approach fails because of the calcala t lve mls ieprescn ta t lons which ensue [01

Acknowledgement

The au tho r wishes to express his gra t i tude to t t J M van de r Veeken, J A M Theeuwes. E G J Vosse lman and R A Jansen for the i r c e m m e n - tary on an ea rhe r ve rsmn o f this article

Refcrencea

1 Brine, man, S 1989 FlexlblhtmI, procesbeheer~.ng en beheersmg van processen Ph D l hes.o, ~,.aah,we~ Umvers~t~ of Tecnnology

2 Schonberger R J 1982 Japaucsc Manafa~tunng l'ecn- mqaes Free Press, New YOrK

3 Durhnger, PPJ andWommann, JC 1983 KanbanKan m Japan, maa- hmr n Bedrofsvoermg, 4

4 Shmgo, S, 1985 Sludy of TOYOTA production sys- tems from an industrial engineering *mwpomt JR;2 nese Management/-,tssoclallon

5 Benrand, J W M andW~jngaard J 1985 Deslructuur vaq produeuebeheerssmgsystemen lnformaue, 27 342- 355

6 Van Hees, R N ~ 1984 Orgamsahestructuuren ,nlegraie besturmg als bas'~s voor ,,oorraadbeneersmg Internal PHILIPS Report vp45N, Emdho,,eq

7 Bertrand J W M 1988 Kostenverlaglngdoorloglslmk B&ID Bednjfskupdlg Vakblad, nr 0

8 Brown, R G, 1987 Materials Management a modular hbrary Wdey, New ~tork

9 Van der Veeken, H J M, 1988 Kovenmformaue b:J ~og~stmke beshssmgen Bednjfskunde, 19~8/1 20-30

!0 Van der Enden, C 1975 Beshssmgscalculalms Sam- son, Alphen aan de R.jn

11 Theeuwes, J ~ M, 1988 HeI gelastroomprmclpe eep gezond verstand ben~dermg TeGl--~-c,~e Bedrqfsvoer- mg November-December 190-194