Los7.4 Define cay 2 OBJECTIVES LO $7.3 | Perform’ LO $7.4 | Compute break-even LO S7.5 | Determine ex LO $7.6 | Compute net present value 347 ted monetary value of a capacity decision 345 chieved, and supplemental funding be obtained. Capacity LO $7.1 Define What should be the seating capacity of a concert hall? How many customers per day should an Olive Garden or a Hard Rock Cafe be able to serve? How large should @ Frito-Lay plant be to produce 75,000 bags of Ruffles in an 8-hour shift? In this supple- ment we look at tools that help a manager make these decisions, After selection of a production process (Chapter 7), managers need to determine capacity. Capacity is the “throughput,” or the number of units a fa acility can hold, receiv® The “throughput” o store, oF produce in a given time, Capacity decisions often determine capital requit®” tac F ments and therefore a large portion of fixed cost. Capacity also determines whether store, o produce demand will be satistied or whether facilities will be idle If of ne tions of it will sit unused and add cost to existing production, If a facility is too small customers—and perhaps entire markets—will be lost. Determining facility size, With an objective of achieving high levels of utilization and a high return on investments 8 critical. = a facility is too large, por 328SUPPLEMENT 7 CAPACITY AND CONSTRAINT MANAGEMENT 329 Options for Adjusting Capacity Tine Haran Design new production processes, Longrange Planning Aga (or sll exsing) longlecatine | equipment Acquie or sel facts, Acquire competitors, vemedateange planning fete long) ‘Schedule jobs. aH ‘Schedule personnel. Allocate machinery. Use capacity shortange planning (sealing) Modify capacity * Difficult to adjust capacity, as limited options exist Figure $7.1 Time Horizons and Capacity Options Capacity planning can be viewed in three time horizons. In Figure $7.1 we note @STUDENT TIP that long-range capacity (generally greater than 3 years) is a function of adding facili- tis and equipment that have a long lead time. In the intermediate range (usually 3 to 4% months), we can add equipment, personnel, and shifts; we can subcontract; and uecan build or use inventory. This is the “aggregate planning” task. In the short run (usually up to 3 months), we are primarily concerned with scheduling jobs and people, ‘vel as allocating machinery. Modifying capacity in the short run is difficult, as we areusually constrained by existing capacity, Design and Effective Capacity Design capacity is the maximum theoretical output of a system in a given period under ideal conditions. It is normally expressed as a rate, such as the number of tons of stee! ‘that can be produced per week, per month, or per year. For many companies, measur- lngcapacity can be straightforward: itis the maximum number of units the company is ‘apable of producing in a specific time. However, for some organizations, determining ‘apacity can be more difficult, Capacity can be measured in terms of beds (a hospital), active members (a church), or billable hours (a CPA firm). Other organizations use ‘otal work time available as a measure of overall capacity. Most organizations operate their facilities at a rate less than the design capacity. They do so because they have found that they can operate more efficiently when their ‘sources are not stretched to the limit. For example, Ian's Bistro has tables set with 2 64 chairs seating a total of 270 guests. But the tables are never filled that way. Some lables will have 1 or 3 guests; tables can be pulled together for parties of 6 or 8. There re always unused chairs. Design capacity is 270, but effective capacity is often closer to 0, which is 81% of design capacity. ‘ j capacity is the capacity a firm expects to achieve given the current operating Sonstraints. Effective capacity is often lower than design capacity because the facility "ay have been designed for an earlier version of the product or a different product mix "nis currently being produced. Table $7.1 tig illustrates the relationship between ign capa ive capacity, and actual output. ro eee a ae UC eee Sticiency. uttization is simply the percent of design capacity actually achieved. Etficiency the percent of effective capacity actually achieved. Depending on how facilities *® used and managed, it may be difficult or impossible to reach 100% efficiency. Too ite capacty ses customers ‘and too much capac is expensive Like Godlocks's poridge, capacty needs to be jst ight, Design capacity The theoretical maximum output ofa system ina given period ‘under ideal condition, Effective capacity ‘The capacity a fim can expect. toachieve, gen its product mix, methods of scheduling, mainte hance, and standards of quality Utilization. Actual output as a percent of design capacity. Efficiency Actual output as a percent Of elective capacity,330 PART 2 DESIGNING © MITE, 6200119, tresuroments Ce (Ea Es ondltons exist during the time If machines at Ft gned to produce 1,000 bag zit and the plant operates 16 hrs /day (000 bags/hr. x 16 hrs. 16,000 bags/day Design Capacity x loses 3 hours of output per day (namely 0.5 hrs /day Design capacty minus lost output If Frito: because of planned resource "on preventive maintenance + 1 h/day on employee breaks + 1§ Unavallabiity (e.g, preventive frsiday setting up machines for different products) maintenance, machine setups! Effective Capacity = 16,000 bags/day ~ (1,000 bags/hr) changeovers, changes in product mi eet G hrs/day) 5 scheduled breaks) = 16,000 bags/day ~ 3,000 bags/day = 13,000 bags/day Effective capacity minus lost On average, if machines at Frito-Lay are not running 1 he day output dunng unplanned resource due to late parts and machine breakdowns BH __ialeness (e.g, absenteeism, machine etyal Output = 13,000 bags/day ~ (1,000 bags/hr.) breakdowns, unavailable parts, ('hr./day) 13,000 bags/day — 1,000 bags/day | 3,000 bags/day quality probiems) Operations managers tend to be evaluated on efficiency. The key to improving efficiency is often found in correcting quality problems and in effective scheduling, training, and maintenance. Utilization and efficiency are computed below: Utilization = Actual output /Design capacity (S7-l) LO $7.2 Determine design capacity, effective capacity, and utilization In Example $1 we determine these values. Efficiency = Actual output Effective capacity (S72) Example S1_ | peTERMINING CAPACITY UTILIZATION AND EFFICIENCY ‘Sara James Bakery has a plant for processing De/uxe breakfast rolls and wants to better under- stand its eapabiliy. Last week the fait produced 148,000 rolls. The elfective capacity 175,000 rolls. The production line operates 7 days per week, with three &-hour shifts per day. | he ine was designed to process the hued, cianamon-Navored Delore roll at arate of 1.20 | per hour. Determine the design eapacity, utilization, and effcieney for this plant when producing this Deluxe rol APPROACH > First compute the design capacity and then wse Equation (S71) to determine | uttization and Equation ($7-2) to determine liiney. SOLUTION > Design capacity = (7 days X 3 shifts X 8 hours) x (1,200 rolls per hour) = 201,600 rolls. | Utilization = Actual output/Design capacity = 148,000/201,600 = 73.4% | Efficiency = Actual output/Effective capacity 148,000/175,000 = 84.6%| SUPPLEMENT 7 | CAPACITY AND CONSTRAINT MANAGEMENT 337 LEARNING EXERCI: i eee SEP Ifthe actual output is 150,000, what isthe efficiency? | | [ * RELATED PROBLEMS > $7.1, $72, 57:3, 7.4,875,576,571 S78 jnExample S2 we see how the effectiveness of new nf new capacity additions depends on how vellmanagement can perform on the utilization an id efficiency of those additions, en Example S2 | expanpine capacrry ‘The manager of Sara James Bakery see Example $1) now needs to increase production of the increasingly popular Deluxe roll. To meet this demand, she will be adding a second production line, The second line has the same design capacity (201,600) and effective capacity (175,000) as the first line; however, new workers will be operating the second line. Quality problems and other inefficiencies stemming from the inexperienced workers are expected to reduce output on the second line to 130,000 (compared to 148,000 on the first). The utilization and efficiency were 73.4% and 84.6%, respectively, on the frst lin, Determine the new utilization and efficiency for the Deluxe roll operation after adding the second line. APPROACH P-_ First, determine the new design capacity, effective capacity, and actual output afier adding the second line. Then, use Equation (S7-1) to determine utilization and Equation (S7-2) to determine efficiency. SOLUTION ® Design capacity = 201,600 x 2 = 403,200 rolls Effective capacity = 175,000 x 2 = 350,000 rolls. Actual output = 148,000 + 130,000 = 278,000 rolls Utilization = Actual output/Design capacity = 278,000 / 403,200 = 68.95% Efficiency = Actual output/Effective capacity = 278,000/ 350,000 = 79.43% INSIGHT > Although adding equipment increases capacity, that equipment may not be oper= ated as efficiently with new employees as might be the case with experienced employees. For Sara James Bakery, a doubling of equipment investment did not result in a doubling of output; other variables drove both utilization and efficiency lower. LEARNING EXERCISE P Suppose that Sara James reduces changeover time (setup time) by three fewer hours per week.What will be the new values of utilization and efficiency? [Answer: Utilization is stil 68.95%, efficiency now increases to 81.10%) RELATED PROBLEMS 7.1, S7.2, $7.3, $7.4, S7.5, $7.6, $7.7, $7.8 Actual output, as used in Equation (S7-2), represents current conditions. Alter: ‘hely, with a knowledge of effective capacity and a current or target value for efficiency, ‘the future expected output can be computed by reversing Equation (S7-2) : Expected output = Effective capacity x Efficiency the ted output is inadequate, additional capacity may be needed. Much of th remainder of this De ieee how to effectively and efficiently add that pacity,332 ParT2 D STUDENT TIP® 3. ach industry and technology has an ‘optimum size. SNING OPERATIONS Capacity and Strate mpettve aan8s Do jut fy Saxtained profits come fom buildIME TET. decisions must be inte Om a cific process ToT ot to be made as wor at lated ey te financial return on a spec : n and strategy: Tavestments are not to be r strategy: Inv" that will place the firm in an agy vant t. organization's missior es dlitures, but as part of a coordi lan that ese investments even 3 position, The questions {0 be asked are, “Will t tally win ey ae arcampee able customers?” and ‘What compel at i jy, and soon) 22 ve eas other og sell as other orga ay anizational delivery, improved qual ! 's we discuss in this tex sol 5 vi oe : sng and finance, ss fected by changes in capacity. Change in gg such as marketing just as capacity changes have quatiy iy implications, i aplications. All must be considered "PMs will have sales and cash flow intenance it chain, human resource, and mai Capacity Consideratio! In addition to tight integration o considerations for a good capacity decision: 1. Forecast demand accurately: Product addition’ and deletions, competition ai product life cycle, and unknown sales volumes all add challenge to acorig forecasting. 2. Match technology strained by technology. Some capaci power plants), while others may be s Large eapacity increments complicate capacity to sales. ; Si sina operating size (volume): Economies and diseconomies of xa aan neem optaval size for a faiity. Economies of scale exist when aery orn eines a size increases, whereas diseconomies of cae occur when larger a con ne rage cost. As Figure 7.2 suggests, most businesses have an opin aes vcaat until someone comes along with a new business model. For decide, ‘dered optimal. Then along came Nucor, very large integrated stee! mills were consi CMC. and other minimills, with a new process and a new business model tht rai. cally reduced the optimum size of a steel mill. 4. Build for change: Managers build flexibility into facilities and equipment; changss will occur in processes, as well s products, product volume, and product mix, Next, we note that rather than strategically manage capacity, managers may taci- cally manage demand. Managing Demand Even with good forecasting and facilities built to accomodate that forecast, there may bea poormatch between the actual demand that occurs and available capacity. A poor ‘match may mean demand exceeds capacity or capacity exceeds demand. However, in both cases, firms have options. Berne Drea Capacity When demand exceeds capacity, the firm may be pias ae lemand simply by raising prices, scheduling long lead times (which ms) & tees le), and discouraging marginally profitable business. However, because a na ete Er rea below what is possible, the long-term solution sus" . (as we see in the OM in Acti “World's Power Plant in Tamil Nadu, India”) ee i Ns «strategy and investments, there are four gp. Sal increments and sales volume: Capacity options are often ty increments may be large (eg, stel mile small (hand-crafted Louis Vuitton handbag the difficult but necessary job of matchsi UPPLEMENT 7 | CAPACITY AND CONSTRAINT MANAGEMENT 333 Capacity Considerations for Krispy Kreme Stores Figure $7.2 Economies and Diseconomies of Scale Krispy Kreme orginally had £8,000-square-fot stores but found them to large and too expensive for many markets. Then they tried tiny 1,300-square foot stores, which required less investment, but such stores were too small to provide the mystique of seeing and smeling Krispy Kreme doughnuts being made. Krispy Kreme finally got it right witha 2,600-foot-store he Exceeds Demand = When capacity exceeds demand, the firm may want jpsimulate demand through price reductions or aggressive marketing, ot it may ac- cqnmodate the market through product changes. When decreasing customer demand ‘soobined with old and inflexible processes, layoffs and plant closings may be neces- ‘ayto bring capacity in line with demand. Adjusting to Seasonal Demands A seasonal or cyclical pattern of demand isanother capacity challenge. In such cases, management may find it helpful to of- fe products with complementary demand patterns—that is, products for which the demand is high for one when low for the other. For example, in Figure $7.3 the firm isadting a line of snowmobile motors to its line of jet skis to smooth demand. With ‘ppropriate complementing of products, perhaps the utilization of facility, equipment, and personnel can be smoothed. Tactics for Matching Capacity to Demand Various tactics for adjusting capacity to demand include: 1. Making staffing changes (increasing or decreasing the number of employees or shifts) 2 Adjusting equipment (purchasing additional machinery or selling or leasing out existing equipment) Fe a ya ‘Combining the two axe 87-8 Ape ‘i y demand patterns By Combining Products That i F * reduces the Have Complementary Seasonal 3,000 gaint Patterns, Capacity Can Be \*-— Snowmobile Better Utilized 5 JFMAMJJASOND JFMAMJ JASONDJ Time (months)334 PART 2 | DESIGNING OPERATIONS Cit World's Largest Solar Pawer Pantin Tamil Nadu, India The largest solar power pant i the wkd has now stated in Tami Nadu in India. With this 648 MW (megawatts) site geting actated, a step ahead forthe aspiring renewable energy objectives ofthe Indian government has been indicated. The 2500-acre facil is located at Kami in Tami Nad Itis made up of about 2.5 milion PV (photovatac) solar modes, which are spread across 10 sq-km. Adani Power spent $679 milion total) for this project and hired more than 8000 labourers to construct te fait. Even with floods and monsoons in the area, this project was completed ‘within eight months. n ful sing, thefacity has the capability to produce electrily for almost 150,000 houses and it also aims at reducing ar Pollution. A robatic system is used for cleaning the solar panels ona daly basis, This system also ges its power from the solar panels The Kamuthi plant's 648 MW capacity tops the earlier 550 MW record, which was at par between two plans in California —the Desert Sunlight Solar Farm near Joshua Tree National Park (a project developed by First Solr) and the Topaz Solar Farm. This power facili instalation loser has taken India little ck renewable government of nda. The Or inal target put forward by the Fes was 20 GW (gaat) of ser ener Defoe 2027 yaa the year 2016 tat objective as rased to 100 GW. With the nor Indias confident of becoing the wots thi 2017 led by China andthe United States. And targets to power at east sity rl lion houses W Group's Chaiman Gautam Adan reveled thatthe company plans fr producing 11,000MW solar energy nthe net ears Ns ing toa study by Bridge to Inca (a esearch company), the tt nga, ela capacity ofthe cous has just crossed the 106W mark vi py ‘new power plant. ter mn ba arte eee encase 2017 at 4.30pm. S _——————___————————————— | 3. Improving processes to increase throughput (e.g., reducing setup times at yy Global Technology added the equivalent of 17 shifts of capacity) i 4, Redesigning products to facilitate more throughput 5. Adding process flexibility to better meet changing product preferences 6. Closing facilities The foregoing tactics can be used to adjust demand to existing facilities. The stratege issue is, of course, how to have a facility of the correct size. Service-Sector Demand and Capacity Management In the service sector, scheduling customers is demand management, and scheduling ke workforce is capacity management, Recessions (e9., 2008-2010) and terrorist attacks {eg., September 11, 2001) can make even the best capacity decision for an aiine lok bad. And excess capacty for an airline can be very expensive, ‘with storage costs running as high as $60,000 per ‘month per aircraft. Here, as testimonial to excess capacity, aircraft sit idle in the Mojave Desert.and Management When demand and capacity are fairly well matched, de- Detimanagement can often be handled with appointments, reservations, ora frst fistserved rule. In some businesses, such as doctors’ and lawyers’ offices, an ment system is the schedule and is adequate. Reservations systems work well in faa car agencies, hotels, and some restaurants as a means of minimizing customer ‘eg time and avoiding disappointment over unfilled service. In retail shops, a post Mp, ora fast-food restaurant, a first-come, fist-served rule for serving customers may ifs, Each industry develops its own approaches to matching demand and capacity. ‘hier more ageressive approaches to demand management include many variations of fpeounts: “early bird” specials in restaurants, discounts for matinee performances or fee wats at odd hours on an airline, and cheap weekend hotel rooms. opacity Management When managing demand is not feasible, then managing tansy through changes in full-time, temporary, or part-time staff may be an option, Thsis the approach in many services. For instance, hospitals may find capacity lim- id bya shortage of board-certified radiologists willing to cover the graveyard shifts. Gating fast and reliable radiology readings can be the difference between life and death foranemergency room patient. As the photo above illustrates, when an overnight read- ings required (and 40% of CT scans are done between 8 p.m. and § 4.M.), the image can tesent by e-mail to a doctor in Europe or Australia for immediate analysis. Battleneck Analysis and the Theory of Constraints Asmanagers seek to match capacity to demand, decisions must be made about the size «f specific operations or work areas in the larger system. Each of the interdependent ‘ork areas can be expected to have its own unique capacity, Capacity analysis involves temining the throughput capacity of workstations in a system and ultimately the apacity of the entire system. A key concept in capacity analysis sis the role of a constraint or botleneck. A bottle Xeck isan operation that is the limiting factor or constraint. The term bottleneck refers ‘otheliteral neck of a bottle that constrains flow of, in the case of a production system, ‘onstains throughput. A bottleneck has the lowest effective capacity of any operation inthe system and thus limits the system’s output. Bottlenecks occur in all facets of Many US. hospitals use sev capacty for radoogists during night shits Haw, an kdano-based service in Zurich and Sydney, contrac with (20% ofa US. hospitals). These trained experts, vide awake and alin their cayight hours, usually urna Giagnoss in 10 t0 20 minutes, wth 8 quarantee of 30 minutes Oe SI SUPPLEMENT 7 | CAPACITY AND CONSTRAINT MANAGEMENT 335 4 abroad to manage Night n 60 radiologists 00 faites Capacity analysis ‘Armeans of determining through past capacty of workstations or an entire production system, Bottleneck ‘The limiting factor or constraint in a gystem,336 PART 2 DESIGNING OPERATIONS mon Eas Assembly Line a mining A box represents an operation, 2mnnnt straining the work flow to highway ‘a machine is €0! a rone inadequate ane, resulting traffic congestion a triangle represents inventory, ime to produce a unit the time (0r a specig and arrows represent recedence _life—from job shops where relationships where two lanes converge into one inacmt We define the process time of a station 1 ff 16 customers ca Process time batch size of units) at that workstation. For nea aon a. be ch The time to produce aunt rat 2% permarke in every 60 minutes en rae of conc (or specie atch of units) at a customer (= 60/16). (Process tine * i ity, fete minutes per customer (= 60/16). ( 75 minutes per customer = 16 customers per in this case is 60 minutes per hour/3. om tion system, simply identify the stati 7 To determine the bottleneck in a production ft ve me Bottleneck time the slowest process time. The bottleneck tim® the eee fi ie orks Faeenee ese cayt tthe one that takes the ongest) in a rogue To process times al (slowest) process, ie, the Figure 87.4 shows a simple assembly line InS4 tes, This is because sta nag bottleneck 3 tainutes, respectively. The bottleneck time is 4 minvics station Bis to speed up station A, the entire production pr in front of station B even slowest station. Even if we were would not be faster. Inventory wow than now. Likewise, if station C coul id simply pile up id work faster, we could not tap its excess capaci be able to feed products to it any faster than I every 4 ming Throughput time re ee ates eee ae me unit ho Bo through, The tne it ties ora product duction from start to end, with no waiting: (Throughput time describes the behavior fan empty system. In contrast, flow time describes the time 'o 80 through a product {0 through the production process descr siren he te ine process from beginning to end, including ide time waitin for stations to finish workin «time to produce a new completed unit in Figure $7.4 Reamer. eo ates Be anes d " 9 minutes (= 2 minutes + 4 minutes +3 minutes). y be quite different. For example, a Fond Bottleneck time and throughput time mas assembly line may roll out a new car every minute (bottleneck time), but it may take 2 ish (throughput time). This is because the hours to actually make a car from start to fink i assembly line has many workstations, with each station contributing to the complete car Thus, bottleneck time determines the system's capacity (one car per minute), while its throughput time determines potential ability to produce a newly ordered product from scratch in 25 hours. ‘The following two examples illustrate capacity analysis for slightly more complex systems. Example S3 introduces the concept of parallel processes, and Example S4in- troduces the concept of simultaneous processing. because station B will not Example S3 | CAPACITY ANALYSIS WITH PARALLEL PROCESSES | Howard Kraye’s sandwich shop provides healthy sandwiches for customers. Howard has two | {identical sandwich assembly lines. A customer first places an order, which takes 30 seconds. The order is then sent to one of the two assembly lines. Each assembly line has two workers and three operations: (1) assembly worker | retrieves and cuts the bread (15 seconds/sandwich), (2) asset bly worker 2 adds ingredients and places the sandwich onto the toaster conveyor belt (20 onds/sandwich), and (3) the toaster heats the sandwich (40 seconds/sandwich). Finally, noth employee wraps the heated sandwich coming out of the toaster and delivers it to the cust (37.5 seconds/sandwich). A flowchart of the process is shown below. Howard wants to determi? the bottleneck time and throughput time of this process.SUPPLEMENT 7 CAPACITY AND CONSTRAINT M First assembly line - 5 - Ee - ee { ae | 18 secisandwich 20 sec/sandwich 40 sec/sandwich 30 sec/sandwich Bread | -| Fill - roaser_| 18 sec/sandwich 20 sec/sandwich 40 sec/sandwich + Second assembly ine ———— | APPROACH P Clearly the toaster is the single-slowest resou s it the bottleneck? Howard should first determine the bottlenec! sembly lines separately, then the bottleneck time of the combined assem | bottleneck time of the entire operation. For throughput time, eac Howard should just sum the process times for all five operations. | SOLUTION D | Because cach ofthe thee assembly-line operation: (worker of machine), different partially completed sandwiches can be at each station. Thus, the bottleneck time of each assembly line is the longe: of the three operations. In this case, the 40-second toasting time represe of each assembly line. Next, the bottleneck time of the combined assembly seconds per two sandwiches, or 20 seconds per sandwich. Therefo ing operation, with a process time of 37 operation. The capacity per hour equals 3,60 sandwiches per hour, The throughput time equal (or 2 minutes and 22.5 seconds), assuming no wait INSIGHT Doubling the resources at a workstation effectively tion operation will equal 1/n times the original process time.) | LEARNING EXERCISE P If Howard hires an additional hourly capacity? [Answer: The new bottleneck is now the o 3,600 seconds per hout/30 seconds per sandwich = RELATED PROBLEMS $7.9, $7.10, $7.11, S712, $7.13 In Example $3, how could we claim that the process time of the toaster was ‘Vsconds per sandwich when it takes 40 seconds to toast sandwich? The reason is wehad two toasters; thus, two sandwiches could be toasted every 40 seconds, for an rige of one sandwich every 20 seconds, And that time for a toaster can actually be dif the start times for the two are staggered (i.e., a NeW sandwich is placed in a ‘every 20 seconds). In that case, even though each sandwich will sit in the toaster +40 seconds a sandwich could emerge from one of the two toasters EVETY 20 seconds. Ne se, doubling the number of resources effectively cuts the process time at that in hal, resulting in a doubling of the capacity of those resources. ce in the five-steP xk time of each of bly lines, and finally the fh assembly line is i s uses a separate ‘worked on simultaneously st process time of each ts the bottleneck time Tine operations is 40 re, the wrapping and deliver~ 5 seconds, appears to be the bottlene \) seconds per hour/37.5 seconds per s is 30 + 15 + 20 + 40 + 37.5 = t time in line to begin with ccuts the time at that station in half (ifn parallel redundant] operations are added, the process time of the com! ANAGEMENT 337 97.5 sec/sandwich process, but f the two as- identical, SO resource ck for the entire andwich = 96 142.5 seconds bined worksta~ | wrapper, what will be the new rder-taking station: Capacity = 120 sandwiches per hour]338 PART 2 | DESIGNING OPERATIONS Example S4 | capaciry ANALYSIS WITH SIMULTANEOUS PROCESSES | De Catia tts en rc eG he cstoner eT cates atab technician takesand develops X-rays (2 and 4 minutes, respectively, 12mg | aa eeand examines the X-rays (5 minutes) while the hygienist cleans the tet (4 My i 2 Uo) the dentist meets with the patient to poke ata few teeth, explain the X-ray results gi rte floss more often (8 minutes); and (5) the customer pays and books her neyq hy flowchart of the customer visit is shown below. Dr. Knott want, “Pi "50 ae | patier | ment (6 minutes). A mine the bottleneck time and throughput time of this process. | Hygienist | | = | | | [ ereckin b=] Seay ee @minunt | 2minvunt 2 mintunit 4 mint fe cy 5 mirvunit Tran, APPROACH P With simultaneous processes, an order or a product is essentially diferent paths to be rejoined later on. To find the bottleneck time, each operation sling arcatey just as though all operations were on a sequential path. To find the thrgg td time, the time over all paths must be computed, and the throughput time is the time Jongest path. Of he SOLUTION ® The bottleneck in this system is the hygienist cleaning operation at 24 pj per patient, resulting in an hourly system capacity of 60 minutes/24 minutes per patien patients, The throughput time is the maximum of the two paths through the system, The 5. through the X-ray exam is 2+ 2+4 +5 +8 + 6 = 27 minutes, while the path throug 4 hygienist cleaning operation is 2 + 2 + 4 + 24 + 8 + 6 = 46 minutes. Thus a patient aad be out the door after 46 minutes (i.c., the maximum of 27 and 46). INSIGHT & With simultaneous processing, all operation times in the entire system ate pa simply added together to compute throughput time because some operations are oc simultaneously. Instead, the time of the longest path through the system is deemed the throug time. LEARNING EXERCISE Suppose that the same technician now has the hygienis sat immediately after the X-rays are taken (allowing the hygienist to start 4 minutes soon The technician then develops the X-rays while the hygienist is cleaning teeth. The dentist si Jeaning is occurring. What would be the new sysen examines the X-rays while the teeth ver: The X-ray development operation is now on the parle capacity and throughput time? [Answe path with cleaning and X-ray exam, reducing the total patient visit duration by 4 minutes 2 throughput time of 42 minutes (the maximum of 27 and 42). However, the hygienist canis “operation is still the bottleneck, so the capacity remains 2.5 patients per hour.] RELATED PROBLEMS P S7.14, $7.15 To summarize: (1) the bottleneck is the operation with the longest (slowest) procs time, after dividing by the number of parallel (redundant) operations, (2) the ster capacity is the inverse of the bottleneck time, and (3) the throughput time is the tot time through the longest path in the system, assuming no waiting.of Constraints nyo coats (706) has been popularized by the book The Goal: A Process of Pe ing 2 ovement, by Goldratt and Cox.! TOC is a body of knowledge that deals vis ‘ying that limits oF constrains an organization's ability to achieve its goals. a ats CaM DE physical (e.g., process or personnel availability, raw materials, oF Gani) of nonphysical (C8. procedures, morale, and training). Recognizing and aie these imitations through a five-step process is the basis of TOC. 1; Identify the constraints. f, Develop & plan for overcoming the iden fz Focus resources on accomplishing Step 2. 4 Reduce the effects of the constraints by officading work or by expanding capability. Make sure that the constraints are recognized by all those who can have an impact on them. 5 When one set of constraints is overcome, go back to Step | and i constraints. ied constraints, ntify new Bottleneck Management srattention on it, We present four prineiples of bottleneck management: 1. Release work orders to the system at the pace set by the bottlenecks capacity: The theory of constraints utilizes the concept of drum, buffer, rope to aid in the imple- mentation of bottleneck and nonbottleneck scheduling. In brief, the drum is the beat of the system. It provides the schedule—the pace of production. The buffer jg the resource, usually inventory, which may be helpful to keep the bottleneck operating at the pace of the drum. Finally, the rope provides the synchronization ‘or communication necessary to pull units through the system. The rope can be thought of as signals between workstations. Lost time at the bottleneck represents lost capacity for the whole system: This princi pleimplies that the bottleneck should always be kept busy with work. Well-trained and cross-trained employees and inspections prior to the bottleneck can reduce lost capacity at a bottleneck. Increasing the capacity of a nonbortleneck station is a mirage: Increasing the capacity of nonbortleneck stations has no impact on the system’s overall capacity. Working faster on a nonbottleneck station may just create extra inventory, with all of its adverse effects. This implies that nonbottlenecks should have planned idle time. Extra work or setups at ‘nonbottleneck stations will not cause delay, which allows for smaller batch sizes and more frequent product changeovers at nonbottleneck Stations, sck increases capacity for the whole system: Increasing the capacity of the bottlene Managers should focus improvement efforts on the bottleneck, Bottleneck capac- ity may be improved by various means, including offloading some of the bottle- eck operations to another workstation (¢.g., let the beer foam settle next to the ‘ap at the bar, not under it, so the next beer can be poured), increasing capacity of the bottleneck (adding resources, working longer or working faster), subcontract- ing, developing alternative routings, and reducing setup times, SUPPLEMENT 7 CAPACITY AND CONSTRAINT MANAGEMENT 339 Theory of constraints (TOC) jedge that deals A body of ko with anything that limits an ‘organization's ability to achieve its goals Acacia constraint in any system is the bottleneck, and managers must focus signifi- € STUDENT TIP ‘There are always bottlenecks ‘a manager must identity and ‘manage them.340 PART2 DESIGNI Break-even analysis ‘Armeans of finding the pint, in dblars and units, at wich costs equal revenues. Figure $7.5 Basic Break-Even Point ING OPERATIONS Even when managers have process and quality variability under cont technology, personnel, products, product mixes and volumes can create mt shifting bottlenecks. Identifying and managin® bottlenecks is a require task, but by definition, bottlenecks cannot be “eliminated.” A system will °° at least one. always 101, Break-Even Analysis he critical tool for determining the capacity a facili The objective of break-even analysis is to find the pe Thay revenue. This point is the break-even point fre hieve profitability. As shown in Figure $75, jy f fixed costs, variable costs, and revenue. Teak, He pois ave costs that continue even if no units are produced, Exampl depreciation, taxes, debt, and mortgage payments. Variable costs are ‘bon ae with the volume of units produced. The major components of variable =a ay wad materials, However, other costs, such as the portion of the utilities that oe ante, are also variable costs, The difference between selling price and mee Yotuar a aly athen total contribution exceeds total fixed cost il thee be ‘nother element in break-even analysis isthe revenue function. In Figure oe enue begins at the origin and proceeds upward to the right, increasing by ne price of each unit. Where the revenue function crosses the total cost i (hea Prt and variable costs) isthe break-even point, with a profit corridor tothe and loss corridor to the left. Motel ‘Assumptions A number of assumptions underlie the basic break-even Notably, costs and revenue are shown as straight lines. They are shown to a arly_-that i, in direct proportion to the volume of units being produced Taeee a xed costs nor variable costs (nor, for that matter, the revenue function) ml Break-even analysis is t to achieve profitability and units, at which costs eq operate above this level to ac analysis requires an estimation o Total revenue ine Total cost ine ag Break-even point: Total cost = Total revenue s g Variable cost Cost (dollars) 8 é Fixed cost © 100 200 300 400 500 600 700 800 900 10001100 Volume (units per period)SUPPLEMENT 7 | CAPACITY AND CONSTRAINT MANAGEMENT 341 jne, For example, fixed costs change as more capital equipment or warehouse ight ed; labor costs change with overtime or as marginally skilled workers are 5s v8 func 3 the revenue function may +hange with such factors as volume discounts. joy eit je-Product Case alas for the break-even point in units and dollars for a single product are LO $7.4 Compute 1 re ow. Let bresk-even___— . : e break-even point in units TR = total revenue = Px BER preak-even point in dollars F = fixed costs = price per unit (after all discounts) variable costs per unit Fe number of units produced TC = total costs = F + Vx peal-even point occurs where foal revenue equals total costs Therefore: TR=TC of Px=F + Vx solving for x, We Bet: Break-even point in units (BEP,) = oo aot: F @-n/P_ 1-¥/P Profit = TR - TC = Px — (F + Vx) = Px - F- Vx =(P- Vix - F . F peaeven point in dollars (BER) = BERP = 5 —;P ig these equations, we can solve directly fr break-even point and profitability. The tno break-even formulas of particular interest are: Total fixed cost Break-even in units (BEP.) = p25 — variable cost ~ (s7-3) T fi Break-even in dollars (BER) = Tova eee = eal Price P (87-4) In Example $5, we determine the break-even point in dollars and units for one product. SINGLE-PRODUCT BREAK-EVEN ANALYSIS Stephens, ne, wants to determine the minimum dolar volume and unit volume needed a its new facility to break even. APPROACH > The firm fist determines that it has fixed costs of $10,000 this period. Direct fabor is $1.50 per unit, and material is $.75 per unit. The selling price is $4.00 per unit. SOLUTION The break-even point in dollars is computed as follows: $10,000 $10,000 x eg ings O00 a = 510,000 BER = (7pm ~ T= 1050+ 15.0] — 437s — 92285714$42 PART 2 DESIGNING OPERATIONS. tere The break-even point in units is | ote. $10,000 : f BER. = py = 400 = (1.50 +75) ~ 5714 | Note that we use total variable costs (that is, both labor and material) INSIGHT > The management of Stevens, Inc, now has an estimate in both unin the Volume necessary for the new facility nd dota LEARNING EXERCISE P If Stevens finds tat fied cost wil increase to 19g Pens to the break-even in units and dollars? (Answer: The break-even in wha And break-even in dollars increases to $27,428.57] a) | RELATED PROBLEMS P $7.16-$7.25 Multiproduct Case Most firms, from manufacturers to restaurants, have a variety of of ing may have a different selling price and variable cost. Utilizing we modify Equation ($7-4) to reflect the proportion of sales for e: this by “weighting” each product's contribution by its proportion o is then: Terings, Each wae od Offey. break-even anajar ach Product, Wey oF sales. The fom Break-even point in dollars (BER) = ——— S75 ) where V = variable cost per unit P = price per unit F = fixed cost W = percent each product is of total dollar sales i= each product Example $6 shows how to determine the break-even point for the multiproduct cae at the Le Bistro restaurant. Paper machines such asthe one shown here require a high capital investment. This investment resut ina high fired cost but allows production of paper ata very low variable cost The production manager's jb is to maintain utilization above the break-even point to achieve proftabilty,s UPPLEMENT 7 | CAPACITY AND CONSTRAINT MANAGEMENT 343 F mple S6 — MuttiPropuct BREAK-EVEN ANALYSIS Le Bistro, like mos break-even point in t other restur: ants, makes more than one product and would like to know its dollars. Information for Le Bistro follows. Fixed costs are $3,000 per month. ET : FORECASTED SALES UNITS Sandwich " 9,000 $5.00 Drin 2 9,000 150 Baked potato 1,00 APPROACH & With, single-product cas using Equation ($7-5 4 variety of offerings, we proceed with break-even analysis just as in a except that we weight each of the products by its proportion of total sales | SOLUTION > Multiproduct Break-Even: Determining Contribution orcs rental (Las Feo LS CS mn Team Un td BUTION | CASTED | SALES| CONTRIBUTION Ss I SAC eo ITEM () Sandwich 9,000 $5.00 $3.00 | 60.40. $45,000 621 248 : | Drinks 9,000 150} 050 | 33 | 67 13,500 .186 123 | Baked © |“ T00 "/""200'| “1G0 | 80 | so |” 4000") 58 097 $72,500 1.000470 Note: Revenue for sandwiches is $45,000 (= 5.00 X 9,000), which is 62.1% of the total re enue of $72,500. Therefore, the contribution for sandwiches is “weighted” by .621. The weighted contribution is .621 x 40 = .248. In this manner, its relative contribution is properly reflected. | Using this approach for each product, we find that the total weighted contribution is .47 for | each dollar of sales, and the break-even point in dollars is $76,596 | —_ F = $3000 X 12 _ $36,000 _ 57 so He cop Pal asa A a7 | -(4) xm | = -G) «on The information given in this example implies total daily sales (52 weeks at 6 days each) of: | $76,596 312 days = $245.50 INSIGHT The management of Le Bistro now knows that it must generate average sales of | $045.50 each day to break even. Management also knows that if the forecasted sales of $72,500 | are correct, Le Bistro will lose money, as break-even is $76,596, LEARNING EXERCISE P Ifthe manager of Le Bistro wants to make an additional $1,000 per month in salary, and considers this a fixed cost, what is the new break-even point in average sales per day? [Answer: $327.33] | RELATED PROBLEMS > $7.26,S7.27 errno344 PART 2 DESIGNING OPERATIONS Break-even figures by product provide the manager with added realism of his or her sales forecast. They indicate exactly whi insigh at must be sold to che we illustrate in Example $7. Example S7 | unmsates ar BREAK-EVEN Le Bistro also wants to know the break-even for the number of sandwiches that m | every day. st be | APPROACH ® Using the data in Example $6, we take the forecast sandwic h sales of 6 | times the daily break-even of $245.50 divided by the selling price of each sanduicn go! wich ($5.9), SOLUTION P At break-even, sandwich sales must then be: Set = Number of sandwiches = 30,5~ 31 sandwiches ech dy INSIGHT With knowledge of individual product sales, the manager has a basis for den ing material and labor requirements, “a LEARNING EXERCISE ® At a dollar break-even of $327.33 per day, how many sands ‘must Le Bistro sell each day? (Answer: ~ 41.] 7 RELATED PROBLEMS P S7.26b, $7.27 ‘Once break-even analysis has been prepared, analyzed, and judged to be reasonab, decisions can be made about the type and capacity of equipment needed. Indeed, abt ter judgment of the likelihood of success of the enterprise can now be made. Reducing Risk with Incremental Changes When demand for goods and services can be forecast with a reasonable degre o! precision, determining a break-even point and capacity requirements can be rth straightforward. But, more likely, determining the capacity and how to achieve itwil be complicated, as many factors are difficult to measure and quantify. Factors sw stupent niP@ technology, competitors, building restrictions, cost of capital, human resource 0 cops cesinsrenie acing and regulations make the decision interesting. To complicate matter frthes d= ceraiy tose, ins growth is usually in small units, while capacity additions are likely to be both ise ai feat, taneous and in large units, This contradiction adds to the capacity decision 1 reduce risk, incremental changes that hedge demand forecasts may be a good 0P" Figure $7.6 illustrates four approaches to new capacity. ead Alternative Figure 87.6(a) leads capacity—that is, acquires capacity to stay he demand, with new capacity being acquired at the beginning of period 1. This bee handles increased demand until the beginning of period 2. At the besinine °F 2, new capacity is again acquired, allowing the organization to stay ahead o nto until the beginning of period 3. This process can be continued indefinitely — at future, Here capacity is acquired incremencally—at the beginning of period | beginning of period 2., QL. ee SUPPLEMENT 7 CAPACITY AND CONSTRAINT MANAGEMENT 345 (©) Lagging domana wit (6) Attemots 1 nave an ereg® Sredsin eran er tal expansion ‘capacity that Nes oe ‘Sin incramenal expansion | Expected { Expectes | comand Soman Demand Demand Ths pear) Tee oes) ses graces t Capacity Expansion pe! gut managers can also elect to make a lan Ber increase at t 1 feureS7-6(0)]~an increase that may eee Satisfy expected demand until the beginning ions managers flexibility. For instance, in the the form of rooms can allow a wider variety in room cleanup schedules. In manufacturing, ups, shorten production runs, and drive down ford industry, added (extra) capacity in si om options and perhaps flexibility is fess capacity can be used to do more set inventory costs. : Figure $7.6(c) shows an option that lags capacity, perhaps using overtime or sabvontracting to accommodate excess demand. Finally, Figure $7.6(4) straddles de- mand by building capacity that is “average,” sometimes lagging demand and sometimes facing it. Both the lag and straddle option have the advantage of delaying capital expenditure. In cases where the business climate is stable, deciding between alternatives can be rilatively easy. The total cost of each alternative can be computed, and the alternative with the least total cost can be selected. However, when capacity requirements are sub- to significant unknowns, “probabilistic” models may be appropriate. One technique ¢STUDENT TIP formaking successful capacity planning decisions with an uncertain demand is decision theory including the use of expected monetary value. Applying Expected Monetary Value (EMV) to Capacity Decisions Determining expected monetary value (EMV) requires specifying alternatives and vari- LO $7.5 Determine ous states of nature. For capacity-planning situations, the state of nature usually is expected monetary value future demand or market favorability. By assigning probability values to the various of a capacity decision —_ Sales of nature, we can make decisions that maximize the expected value of the alter- “atives. Example S8 shows how to apply EMV to a capacity decision. Uncertainty in capacity decisions ‘makes EMV a helpful tol,346 PART 2 eae? DESIGNING OPE RATIONS. ui — EMV APPLIED TO CAPACITY DECISION Southern Hospital Supplies, a company that makes hospital £0WNS, is Considering r “a expansion ay APPROACH: Southern’s major alternatives are to do nothing, build a smal Plant, rsdium plant, or build a large plant. The new facility would produce a new type of gy merently the potential or marketability for this product is unknown. Ifa larg plants ps Stavorable market exists a profit of $100,000 could be realized. An unfavorable margart®% yield a $90,000 loss. However, a medium plant would earn a $60,000 profit with jt ‘market. A $10,000 loss would result from an unfavorable market. A small plant, on poe le Fran would return $40,000 with favorable market conditions and lose only $5,000 ge vorable market. Of course, there is always the option of doing nothing. Dw, Recent market research indicates that there is a4 probability of a favorable Market, mena that ther is also a .6 probability of an unfavorable market. With this inform , Whig ati alternative that will result in the highest expects 00, he ed monetary value (EMV) can be selected. ive: SOLUTION > Compute the EMV for each alternat .4) ($100,000) + (6) ($90,000) = —$14,000 EMV (large plant) = EMY (medium plant) = (4) ($60,000) + (6) (~$10,000) = +$18,000 EMV (small plant) = (4) ($40,000) + (.6) (~$5,000) = +$13,000 EMV (do nothing) = $0 Based on EMV criteria, Southern should build a medium plant. INSIGHT PI? Southern makes many decisions like this, then determining the EMV for alternative and selecting the highest EMV is a good decision criterion. aa LEARNING EXERCISE Ifa new estimate of the loss from a medium plant in an unfavora Coa To_ $20,000, whats the now EMY for this alternative? [Answer: 12000 ahs Changes the decision because the small plant EMV is now higher.) se RELATED PROBLEMS & S7.28, S7.29 Applying Investment Analysis to Strategy-Driven Investments Once the strategy implications of potential investments have been considered, Pe ai ae 5 STUDENT TIP® ditional investment analysis is appropriate. We introduce the investment aspects of ‘An operations manager may be hed responsible for return on investment (0, capacity next. Investment, Variable Cost, and Cash Flow pea eae yore alternatives exist, so do options regarding capital ines a la eapail a prec choose from among different financial opto as wall as capacity and proses ltematves, Analysis should show the apt i and cash flows as well as net present value for each alternative