Supplement 7 Capacity and Constraint Management Background Even though this is a supplement, it contains enough information to be treated as an independent chapter. The bottleneck analysis material itself could easily be expanded to cover a full class session, particularly if the dice game simulation see !ctive Classroom "earning Exercises# is played. Taken together, the tools and concepts in this chapter provide the future operations manager $ith an excellent starting point for thinking about capacity issues.
Class Discussion Ideas %. &ne interesting interesting $ay to demonstrate design design and effective effective capacity capacity is to use the school's school's lecture halls halls as an example. (hile the design capacity is every seat full for all classes, the effective capacity is considerably less. This can be compared to other services such as airlines or restaurants. ). ! simple simple *uestion *uestion that that can generate generate a long list list of ideas is to ask the studen students ts for $ays in $hich $hich firms can increase capacity in the short run and the long run. (hile the distinction bet$een the short and long run is not al$ays clear and not necessarily that important any$ay#, the follo$ing types of ans$ers might be provided. Short run+ using longer production runs implying fe$er setups, using overtime, subcontracting, adding shifts, hiring temporary $orkers, leasing e*uipment, renting more space, using better or faster $orkers, eliminating inefficiencies, increasing productivity, improving bottleneck capacity, capacity, and decreasing maintenance andor eliminating inspections andor elimi eliminat nating ingre reduc ducing ing employ employee ee breaks breaks these these last last three three ideas ideas may have have detrim detriment ental al longlong-run run implications#. "ong run+ purchase ne$ e*uipment, automate processes, design ne$ processes that are more product focused, expand the permanent $orkforce, expand facilities, enter into oint ventures, and purchase merge $ith# the competition. /. 0f the I Love Lucy video is sho$n see Cinematic Ticklers belo$#, 1-%2 minutes of discussion can follo$. This clip is as much about poor management techni*ues as it is about not matching the process time of a station $ith the speed of incoming product. !sk !sk students $hat $ent $rong. They'll likely identify some of the follo$ing. There $as no training of the ne$ employees. There $as no supervision. (e observe 3management by threat.4 The incentive $as only to $rap every candy $ithout letting any get through un$rapped, implying no *uality control and encouraging the hiding of mistakes. There $as no employee involvement. Communication $as one $ay from the boss to the $orkers. Communication $ithin the production line $as non-existent5if the upstream station kne$ $hat $as happening, it might have slo$ed do$n the line.
Active Classroom Learning Exercises %. 6ands6ands-&n &n Simu Simulat lation ion of of the ice ice 8ame 8ame from from The Goal The dice game in Chapter %9 in The Goal (Goldratt and Cox, Third Revised Edition, North River Press, 2004 can be played in the classroom $ith groups of five people. The only supplies needed are dice and 3products4 matches or pennies, for example#. The game as described during the :oy Scout hike in The Goal demonstrates demonstrates the detrimental effects of dependent events and statistical fluctuations on throughput. %1/ Copyright ;)2%/
%19
Ca!acity and Constraint "ana#e$ent
The game places five players in a ro$. The first player has an unlimited supply of ra$ materials. uring each round, player % rolls one die and moves that number of products into the ra$ materials bo$l of player ). Then player ) rolls the die and moves the minimum of the roll and the number of products in the bo$l into the ra$ materials bo$l of player /. This continues do$n the line.
. !fter$ards, each group can be allo$ed to 3purchase4 an option or combination of options. Then let each group play the game for, say, )2 rounds and compare profit results. ?evenue is based on throughput times a sales price, and cost depends upon the option purchased. This game can be automated in Excel and played for hundreds of rounds as $ell see !dditional !ssignment 0deas belo$#. ). (hile the sno$mobileet ski example clearly illustrates the concept of complementary demand, both markets already existed so entry $as a relatively straightfor$ard exercise. 0t can be a different proposition $hen there is no market and the organi@ation has to create one. 0n the %AB2's 6ead had a successful business making metal sno$ skis. 6o$ever, $ith no obvious complementary product, they had to create a market for metal tennis rackets in order to develop a complement to their ski production. 6ave the students investigate the product lines of a fe$ local firms to see if they are using complementary products to balance demand. 0f they are not, have the students suggest possible complementary products that might be developed or adopted.
Company Videos %. Ca!acity Plannin# at %rnold Pal$er &os!ital (')* The !rnold
Copyright ;)2%/
%11
Ca!acity and Constraint "ana#e$ent
%2 years, and, if so, should building space even if empty for the time being# be built at one time Moreover, is bigger al$ays better 6o$ many births should this hospital perform before it begins to turn customers a$ay
Cinematic Ticklers %. I Love Lucy, The Co$!lete +econd +eason -o. +/itchin# (Lucille 1all and ivian ance, Para$ount, ori#inal airdate +e!3 5, *52 This classic clip sho$s "ucy and Ethel $rapping candies on an automatic assembly line. The clip never fails to generate laughter in the classroom. Soon after the line starts up, the $omen fall behind and start to store, hide, and eat the product to avoid letting any get through un$rapped because they'll be fired if that happens#. The situation only $orsens $hen the supervisor returns to observe no problems due to hidden and destroyed candy#, so she yells out, 3Speed it up a littleD4
Jay and Barry’s ! Blog %. 6" in the Ne/s 6verca!acity &its the Euro!ean %uto "a7ers European auto makers are no$ facing the dramatic issue of too much manufacturing capacity. !bout /2 of the A European auto-assembly plants are operating belo$ 72F of their capacity. http+hei@errenderom.$ordpress.com)2%)2B)/om-in-the-ne$s-overcapacity-hits-the-european-automakers ). 6" in the Ne/s &o/ %irlines "atch Ca!acity to 8e$and Many fe$er people fly in the $inter than during school breaks, maor holidays, and summer vacations. 6ere $e see four suggested &M strategies that airlines can implement to break the cycle. http+hei@errenderom.$ordpress.com)2%)2/2)om-in-the-ne$s-ho$-airlines-match-capacity-todemand /. 6" in the Ne/s 9hat &a!!ens 9hen 8isney, :niversal, and Le#oland Reach Ca!acity This is an interesting story of ho$ different companies deal $ith capacity constraints. Gniversal and isney's tactics for managing operations at full capacity are described here. http+hei@errenderom.$ordpress.com)2%)2%2%om-in-the-ne$s-$hat-happens-$hen-disney-universaland-legoland-reach-capacity 9. 6" in the Ne/s :P+ 8rivers Pic7 :! the Pace The 9+- Sept. %B, )2%%# describes ho$ G
"resentation #lides
Copyright ;)2%/
%1B +u!!le$ent ;
0IT?&GCT0&I S7-% through S7-1#
#$%&
#$%'
#$%)
#$%*
#$%(
C!
Copyright ;)2%/
%17
Ca!acity and Constraint "ana#e$ent
$eek period in the latter half of the month#, or even a year e.g., deep discounts at ski resorts during summertime#.
#$%+
#$%$
#$%,
#$%-
#$%&.
#$%&&
#$%&'
#$%&(
#$%&)
#$%&*
#$%&+
#$%&$
Copyright ;)2%/
%1 +u!!le$ent ;
#$%&,
#$%&-
#$%'.
#$%'&
#$%''
#$%'(
:&TT"EIEC !I!"JS0S !I T6E T6E&?J &L C&IST?!0ITS S7-)9 through S7-/%# Slides )9-)1+ This section covers a topic often ignored in introductory operations management books but that is crucial for anyone analy@ing processes5bottleneck analysis. Kust as a chain is not stronger than its $eakest link, a production system can produce no faster than its slo$est station bottleneck#. Therefore, operations managers must learn to identify the bottlenecks# in a process and focus on improving its their# capacity to increase the throughput of the entire system. 8oldratt's and Cox's popular book, The Goal % Process o< 6n#oin# I$!rove$ent , /rd rev. ed. 8reat :arrington, M!+ Iorth ?iver
Copyright ;)2%/
%1A
Ca!acity and Constraint "ana#e$ent
Slides )-)A+ These slides summari@e Example S9, $hich incorporates simultaneous processes. Slide )A includes all of the calculations, and helpfully puts the process flo$ chart at the top of the slide. The important point in this example is that the computation of throughput time does not involve simply adding all times in the system together. ue to simultaneous processing, the time of each path through the system must be computed5$ith the longest path representing the throughput time. Slide /2+ This slide identifies the five steps of the theory of constraints, populari@ed in The Goal . This approach applies to managing bottlenecks and more broadly to anything that limits or constrains an organi@ation's ability to achieve its goals. Slide /%+ This slide identifies four principles of bottleneck management.
#$%')
#$%'*
#$%'+
#$%'$
#$%',
#$%'-
#$%(.
#$%(&
:?E!-EPEI !I!"JS0S S7-/) through S7-99# Slides /)-/+ These slides provide the foundation for break-even analysis, $hich is the critical tool for determining the capacity that a facility must have to achieve profitability. Slide /) provides the definition, obective, and re*uirements. Slide // describes the t$o different types of costs considered+ fixed and variable. Slide /9 describes the behavior of the revenue function. The graph in Slide /1 Ligure S7.1# suggests that approximately 971 units must be
Copyright ;)2%/
%B2 +u!!le$ent ;
produced and sold in order to start making a profit. Slide /B identifies limitations of using the simple linear break-even analysis approach. Slides /7 and / provide the necessary formulas, $hich can all be derived $ith some simple algebra. Slides /A-9%+ These slides present Example S1. Iote that the graph in Slide 9% is not presented in the text. Slides 9)-99+ These slides present the multiproduct break-even analysis. Slide 9) presents the formula for the break-even point in dollars. Slides 9/ and 99 present Examples SB and S7 together.
#$%('
#$%((
#$%()
#$%(*
#$%(+
#$%($
#$%(,
#$%(-
#$%).
#$%)&
#$%)'
#$%)(
Copyright ;)2%/
%B%
Ca!acity and Constraint "ana#e$ent
#$%))
?EGC0I8 ?0S (0T6 0IC?EMEIT!" C6!I8ES S7-1) through S7-11# Slides 91-9A+ Slide 91 Ligure S7.B# displays the four approaches to adding ne$ capacity+ lead /ith incre$ental ex!ansion , lead /ith one>ste! ex!ansion , la# , and straddle. The next four slides focus on each approach individually. ! leading strategy captures all demand, but at the cost of excess capacity for certain time periods. "eading $ith incremental expansion Slide 9B# involves less excess capacity and doesn't tie up as much money compared to a one-step expansion Slide 97#N ho$ever, economies of scale in construction and e*uipment purchase may encourage expanding all at once. ! one-time expansion also avoids the effects of inflation and maximi@es flexibility in output rate. ! lag strategy Slide 9# never underutili@es capacity, but it may be unable to capture all demand. ! straddle strategy Slide 9A# attempts to minimi@e the disadvantages of the t$o extreme strategies, $hile still retaining some of the advantages.
#$%)*
#$%)+
#$%),
#$%)-
#$%)$
!<<"J0I8 EQ
Copyright ;)2%/
%B) +u!!le$ent ;
as this $ith only one decision point, a decision tree see Module !# could help visually but is not necessary to conduct the analysis.
#$%*.
#$%*&
!<<"J0I8 0IPESTMEIT !I!"JS0S T& ST?!TE8J-?0PEI 0IPESTMEITS S7-1) through S7-1A# Slide 1)+ The discipline of finance provides us $ith the concept of net present value I
#$%*'
#$%*(
#$%*)
#$%**
#$%*+
#$%*$
Copyright ;)2%/
%B/
Ca!acity and Constraint "ana#e$ent
#$%*,
#$%*-
Additional Assignment Ideas %. Monte Carlo Simulation of the ice 8ame from The Goal The dice game from The Goal see !ctive Classroom "earning Exercises above# can be simulated in Excel rather easily. This exercise can be a nice opportunity for students to $ork on their Excel skills and explore Excel's random number capabilities. The assignment demonstrates the detrimental effects of dependent events and statistical fluctuations on throughput. Extensions include adding inventory buffers to eliminate dependent events, using a less variable die to minimi@e statistical fluctuations, exploring $here to place the bottleneck in the production line, and examining $here to place excess capacity in the production line. ! sample set of instructions is provided in &ther Supplementary Material belo$.
Copyright ;)2%/
%B9 +u!!le$ent ;
Additional Case #tudies 0nternet Case Study $$$.pearsonhighered.comhei@er # o
+outh/estern :niversity?s @ood +ervice (8 + ?e*uires the development of a multi-product breakeven solution.
6arvard Case Studies http+harvardbusinessonline.hbsp.harvard.edu# @resh Connections RB222)), ; )22/# o AristenBs Coo7ie Co3 (% (%.rid#ed RB22/7, ; )227# o +char<
Eli=a.ethBs Country 9ares RA:2A22A, ; )2%2# Pre$ani7 Containers and the 1ottlenec7 Challen#e RA:%)E2%1, ; )2%/# 9arehousin# +trate#y at ol7s/a#en Grou! Canada Inc3 (GC% RA:%)22), ; )2%)#
Internet /esources !merican Council of Engineering Companies !ssociation for Manufacturing Excellence !?
$$$.acec.org $$$.ame.org $$$.!?
t0er #upplementary !aterial &ptions for Capacity 0ncrease &ne concept not presented in the textbook that might enhance the lecture is a comparison of options for capacity increase. &n-site expansion !dvantages+ generally cheapest, does not disperse existing labor force, no product or process separation, and can take advantage of possible economies of scale isadvantages+ de-optimi@es layout, might postpone the introduction of ne$ technology, increased complexity strains management, larger $orkforce may cause deteriorated labor relations possibly unions#, and disaster risk Ie$ :ranch !dvantages+ can use ne$ technology, can design an optimal layout, and can tailor policies and systems effectively isadvantages+ multi-site overhead and cannot solve problems back at the original plant ?elocation !dvantages+ can solve problems at the original plant that is being moved, can use ne$ technology, can design an optimal layout, and can take advantage of possible economies of scale isadvantages+ moving and startup costs, might lose good people, and disaster risk
Copyright ;)2%/
%B1
Ca!acity and Constraint "ana#e$ent
&perational 6edging !nother capacity concept is 3operational hedging,4 $hich refers to the practice of having excess capacity at plants around the $orld that is utili@ed $hen exchange rates shift favorably. !s a *uick example, a company has a plant in both the G.S. and the G.., as $ell as demand in both countries. ! G.S.-based shipper charges H%.22 per unit to ship bet$een the t$o countries. !ssume no taxes. Monthly demand is %2,222 units in the G.S. and ,222 units in the G.. Monthly capacity is %1,222 units in the G.S. and %),222 units in the G.. The production cost is HB.22 in the G.S. and /.22 in the G.. sales prices are irrelevant#. (e examine three cases. Case %+ the exchange rate is H) = %. 6ere the production cost is the same in both countries, so given the shipping cost, each country should produce its o$n demand. Case )+ the exchange rate is H/ = % dollar depreciates#. !sk the students $hat they think might happen. The production cost in the G.. umps to A. Thus, even $ith shipping, production in the G.S. is cheaper. The G.S. plant should maximi@e its production at %1,222 units and ship 1,222 units to the G.. The G.. plant should produce the remaining /,222 units for the rest of the G.. demand.
Copyright ;)2%/
%BB +u!!le$ent ;
#ample !onte Carlo #imulation Excel Assignment 1or t0e Dice 2ame in The Goal
Copyright ;)2%/
%B7
Ca!acity and Constraint "ana#e$ent
eliverables for
Jou have a mistake in your spreadsheet if it contains negative numbers or if any of your throughput columns contain a number greater than B. Iote, ho$ever, that some of the entries in your bo$l columns $ill likely exceed B.
Copyright ;)2%/
%B +u!!le$ent ;
Copyright ;)2%/