-
7/27/2019 33960864
1/84
PRODUCTION PLANNING AND PRODUCTIVITYMETHODS FOR A MOLDING MANUFACTURING FACILITYby
Mary Elaine JohnsonB.S. Chemical Engineering, Massachusetts Institute of Technology (1993)
Submitted to theDepartment of ChemicalEngineeringand the
Sloan School of Managementin Partial Fulfillment of theRequirements for the Degrees of
MASTER OF SCIENCE IN CHEMICAL ENGINEERINGand
MASTER OF SCIENCE IN MANAGEMENTat theMASSACHUSETTS INSTITUTE OF TECHNOLOGY
June, 1995C Massachusetts Institute of Technology 1995. All rights reserved
Signature of Author 2 I MIT Department of Climical EngineeringMIT Sloan School of ManagementMay 12, 1995
Certified bye r 'erick J. McGarryProfessor o g neering and Polymer Engineering
Certified by_ _ / -.... James Utterbackssor of Management and EngineeringAccepted by
MASSACHi ErTS INSTITUTE Robert E. CohenChairman, Committee for Graduate StudentsJUL 12 1995
-
7/27/2019 33960864
2/84
PRODUCTION PLANNING AND PRODUCTIVITYMETHODS FOR A MOLDING MANUFACTURING FACILITYby
Mary Elaine JohnsonSubmitted to the Department of Chemical Engineering
and the Sloan School of Management in partialfulfillment of the requirements for the Degrees ofMaster of Science in Chemical Engineering
and Master of Science in Management
AbstractThis thesis focuses on production planning and productivity improvement efforts for amolding operation. The goal of these efforts is to reduce variation in quality measurementand control at its facilities and increase capacity. The facilities that are a part of themolding division are experiencing material shortages, seasonality in demand, and capacityshortages. Data collection and measurement inconsistencies provide an additionalchallenge. The limited testing of raw material inputs adds to process variation. Qualityimprovement efforts, production system yieldmodel and production planning model arethree strategic tools used to address the issues facing the facilities.Quality improvement efforts include increasing incoming inspection of raw materials,improving measurement of material losses through the process, and establishing heappropriate measures and procedures to determine and eliminate root cause of materialloss. Concentrating on improved product quality will reduce customer returns, provideraw material savings, and increase capacity through productivity gains. As the productreceives more processing and increases in value, the cost implicationsof scrapping itincrease.The production system yield model, a facility asset utilization model, was applied toseveral facilities. This model calculates an efficiency metric for each facility based on fiveparameters: Rate of Quality Products, Run Speed Efficiency, Scheduled Time, NetProduction Rate, and Molder Run Hours. This model is a tool for management todetermine where to focus improvement efforts and to assist in standardizing productionmeasures at various facilities.The purposes of the production planning model are to assist management in makingdecisions about the business unit on a regional level, and in establishing a strongercommunication link with all functions of the organization. The production planning modelis a spreadsheet tool that incorporates information from the Production System Yield
2
-
7/27/2019 33960864
3/84
model to establish demonstrated capacity and to analyze the ability to meet variousdemand scenarios. A key finding was that a 67 percent decrease in internal process scrapwas shown to increase regional capacity five percent. For effective use andimplementation of the production planning model, production system yield model andquality improvement efforts the following is recommended: increasing transfer ofmanufacturing operating knowledge between facilities,determining appropriate levels ofexcess demonstrated capacity and levels of product inventory, and improvingforecastingtools and techniques.
Thesis Supervisors:Professor Fred McGarry, Professor of Civil Engineering and Polymer EngineeringProfessor James Utterback, LFM Professor of Management and Engineering
3
-
7/27/2019 33960864
4/84
Acknowledgements
The author gratefully acknowledges the support and resources made available to herthrough the Leaders for Manufacturing Program, a partnership between MIT and majorU.S. manufacturing companies.I would like to thank my thesis advisors, Fred McGarry and Jim Utterback. Also, aspecial thanks to Dick Garrett. The author is also grateful for the support and guidanceshe received from several members of the host company. I would like to thank JimCiccarelli, Ed Selby, Mike Zurawski, Dave Pritchett, Chip Brown, Gary Petee and theentire divisionfor their assistance throughout the internship.In addition, I would like to thank my family for their encouragement and support.
4
-
7/27/2019 33960864
5/84
Table Of ContentsAbstract .................................... ................... 2...........A cknow edgements ......................................................................................................... 4List of Figures .............................. ........................................................ ....................7L st of T ables .................................................................................................................. 9
Chapter 1: Introduction ........................................................................................... 101.1 Company Background ..................................................................... 101.1.1 Manufacturing Process Flow ........ 11................. l...111.2 Highlights of Research ............................. 121.3 Motivation for Research ............................ .......................... 31.3.1 Measurement of Process .............................. 151.3.2 Raw Material ........................................ 151.3.3 Scrap Rates .................................. 19
Chapter 2: Method for Improvement ...................................................................... 232.1 Quality Definition .............................. 242.2 Benefits of Quality .............................. 252.2.1 Relationship With Vendors ........................................ 26Chapter 3: Production System Yield .................................. 303.1 PSY: Asset Utilization Model ................................... 313.1.1 Components Of The PSY Model ............................... 323.2 Analysis of PSY Model ................................. 373.3 Applications of PSY ................................. 413.4 Limitations of PSY .......................................................................... 43Chapter 4: Production Planning Model .................................. 454.1 Components of Production Planning Model ....................... 46
4.1.1 Aggregate Sales . ....................................................................... 464.1.2 Theoretical Capacity ............................... 474.1.3 Speed Adjustments ............................. 524.1.4 Capacity Adjustments ............................ 524.1.5 Demonstrated Capacity .................. ............................... 534.1.6 Production/ Inventory Plan ....................................................... 534.1.7 Materials Requirements............................................................ 544.2 Facilitating Communication ......................... 54
4.3 B ase Case ........................................................................................ 554.3.1 Comparison of 1994 Data ................................ 57
5
-
7/27/2019 33960864
6/84
4.4 15% Increase in Demand ........................................................... 594.4.1 Adding Machinery ................................. 594.4.2 Productivity Improvements .................................. 674.4.3 Comparison of Alternatives ...................................................... 684.5 Other Production Planning Scenarios ............................... 694.5.1 Implementation of New Product Design ................................. 694.6 Limitations of Production Planning.......................................... 70
Chapter 5: Conclusions and Recommendations.................................. 725.1 Conclusions................... ............................... 725.2 Recommendations for Future Work ................................... 76Bibliography ........................... ........... ............................................................................78Appendix ................. ................................... ...................................... 79
-
7/27/2019 33960864
7/84
List Of FiguresFigure 1-1: The production process is designed for molding to be the bottleneck ........... 12Figure 1-2: There are several reasons why meeting customer orders is a challenge,and the boldface issues will be discussed in more detail .......... ...............14Figure 1-3: Comparing Facility 1 and Supplier Melt Flow Reveals Wide Variabilityand 4% of shipments are out of specification ......................................... 17Figure 1-4: No polymer shipments have been rejected, but there are large variationsin melt flow .................. ......... ......................................... 18Figure 1-5: Due to estimation techniques it is difficult to determine exact scrap rate ..... 21
Figure 1-6: Manufacturing cost and value added steps related to manufacturingprocess ............................... ............................ 22Figure 2-1: There are several factors within manufacturing that can influence theorganization's ability o produce qualityproduct profitably.................... 24Figure 3-1: The continuous process of updating manufacturingstrategy dependsheavily on accurate data reporting and accountability .............................. 31Figure 3-2: Multiplicationof the five key factors results in Production SystemYield number. Formula for each key factor is shown above .................... 32Figure 3-3: Alternative method of calculating PSY number ......... .............................. 34Figure 3-4: The PSY model . ........................................................... 35Figure 3-5: The PSY review process ............................................................ 38Figure 3-6: PSY components of Facility 1 ................................................. 40Figure 3-7: Product specificPSY comparison of two facilities ...................................... 42Figure 4-1: The ability to produce customer orders on time affects customerloyaltyand goodwill ........................................................... 45Figure 4-2: The production planning model links forecasted demand with.demonstrated capacity to determine a production plan and materialrequirem ents ......... .............................................................................................. 48
7
-
7/27/2019 33960864
8/84
Figure 4-3: Cumulative production information shows that if demandincreases 15% every month that after May demand will not be met ..................... 58
Figure 4-4: In order to meet a demand 15% stronger than expected,given no changes in productivity, 13% more machines wouldhave to be added(total of 1.13Y machines) to the base case toensure no product shortfalls .......................................................... 61
Figure 4-5: The internal rate of return assuming existing machine throughputvaries depending on the life of the machines ........................................................ 64
Figure 4-6: The internal rate of return given a percentage change in product costfor a three, four, and five year operational period ........................................ 64Figure 4-7: An increase in throughput increases the internal rate of return ....................66Figure 4-8: A decrease in machinery utilizationlowers internal rate of return ................66Figure 5-1: A proactive improvement effort enhances a company's abilityto meet customer requirements .......................................................... 73Figure A-1: Garvin's analysis of quality and profitability ................................................ 80Appendix 1: This production planning model examines the impact of a new productdesign ..................... ................................... ...................................... .. 81
8
-
7/27/2019 33960864
9/84
List Of TablesTable 4-1: Capacity adjustment categories with corresponding product losses .............. 56Table 4-2: With 1.13Y machines the region will be able to meet all orders
without delay if demand increases 15 percent ...................................................... 62Table 4-3: Increasing number of machines and maintaining productivity
levels increases scrapped material while orders will still be missedif productivity improvementsin scrap occur but scrapped materialwill be reduced ....................................................... 68
9
-
7/27/2019 33960864
10/84
Chapter 1: Introduction1.1 Company BackgroundThis thesis focuses on production planning and productivity improvement efforts for amolding operation. The molding operation is part of a division of a diversifiedmultinational corporation. The molding division has numerous facilities manufacturingsimilar products on machines at various locations globally. Some facilities are smalloperations with a handful of machines while others are much larger and operate numerousmachines. In addition, a group within the division is an original equipment manufacturerof the machinery that produces the product. Some of their competitors buy equipmentfrom them, while other competitors purchase an alternative technology. Research anddevelopment, product design, and other support functions are located at headquarters.Depending on regional markets, the intensity of competition and number of competitorsvary. Product mix varies depending on facility. Also, customer order quantity varies atlocations, with some locations having a higher product mix with many smaller orders. Thedivision has a large market share domestically, as well as in certain global markets. All ofthe facilities combined produce billions of a similar product per year and this is on thesame scale as pill, disposable cup or paper clip production. Facilities are often regionallylocated close to the customer to reduce shipping costs. In some markets, the product is acommodity, while in others it is a specialty product. The product is generic in size, butstyle, color, and appearance can change based on customer requirements.
In the past the strategy for manufacturing has been "make to order", but due to the needto increase facility asset utilization, it is becoming "make to stock." All facilities have adegree of seasonality for the products they produce. As the demand for the productincreases, in all of the markets they compete in, the level of competition is increasing. Thedivision is rapidly expanding internationally to better serve the increased demand, as wellas to better serve customers requiring in-country manufacture. To maintain highutilization of capacity and ensure that orders are met, when possible or necessary, facilities
10
-
7/27/2019 33960864
11/84
will support each other with product. Price and quality have increasingly been issues.Certain customers are requiring shorter lead times between order placement and productdelivery. Due to the increased competitive environment, cost, quality, increasedproductivity, and capacity are critical issues. Material costs are rapidly increasing due totight supply. Presently, the division relies on multiple suppliers of raw materials.
1.1.1 Manufacturing Process FlowThe process has three basic steps, as shown in Figure 1-1. In the first step, polymer isextruded and molded into the product design shape. The rate limiting step in the processvaries, depending on equipment layout. Inspection of molded product is performed byoperators at regular time intervals to ensure the product has the correct dimensions andmeets specifications. There are numerous reasons why product could be scrapped andthese include machine startups, color changes, off-color product, other visual defects,mechanical defects, incorrect dimensions, and process adjustment, etc.. Blocks ofextruded polymer that are not molded into product shape, polymer purge, are sold.Generally, the amount of polymer purge is small in comparison to scrapped product andmay occur during startup or when a process upset occurs. Product that is scrapped afterthe first step can be reground and the polymer can be reused in the process. During thelater steps of the process polymer cannot be recycled. Depending on the product andcustomer specification, an off-line step may need to be performed or Step 2 and/or Step 3may not be needed.
In Step 2, Post Mold Processing, additional functional requirements of the product areadded, and the product is modified to meet customer specification. A customer specifiedand approved material, such as a customer approved polymer, will be incorporated intothe product. Customer approved polymer purge and scrapped product(molded unit withor without customer approved polymer) are both sold. Product can be scrapped for thesame reasons as in molding. In addition to operator inspection, on-line and off-lineproduct inspection machines evaluate product for defects.
11
-
7/27/2019 33960864
12/84
In Step 3, final finishing of the product is done, but no additional mechanical modificationis performed. This final step includes labeling and packaging the product. The operatorsinspect the product at regular time intervals. Product scrap reasons are similar to thosementioned earlier. After the product is packed into a box it is sent to the warehouse andshipped to the customer.
Step 1Heatedpolymer - 1extrudate
Step 2
Step 3
Warehouse andShip to customer
~- scrapped product ,-- polymer isrecycled-- polymer purge*
customer approvedpolymer purge*- scrapped product*
_. scrapped product*I *scrap and a percent ofpurge are sold- > Customer Returns
The production process is designed for molding to be thebottleneck. During each step inspection determines if productcontains defects. Scrapped product at the molding step isrecycled and reused in the process while other scrap is sold.
1.2 Highlights of Research
The need for standardization and control as well as the present state of material andquality management within the division lead to the thesis focus. The major topics of thethesis are material and quality systems, measurement control and the benefits of
12
Mold
Post-MoldProcessing(add requirements)
Finishing(label and print)
Figure 1-1:
-
-
7/27/2019 33960864
13/84
production planning at a regional aggregate level. The remainder of this chapter explainsthe motivation for the thesis. In Chapter 2, an initialframework for addressing quality,measurement control and the linkage between scrap, speed and capacity as it relates toproduction planning is outlined. The need to define quality is discussed as well as howquality can positively impact activities within the organization. In Chapter 3, ameasurement control model, Production System Yield, is explained and the benefits andlimitations of the model are examined. In Chapter 4, Production Planning expands on thework of Chapter 3 and discusses capacity losses and their implications on manufacturingand material requirements. Chapter 5 re emphasizes the benefits of quality measurementprocesses and control and some of the key learnings.
1..3 Motivation For Research
As the business unit has grown in size many issues have arisen. Figure 1-2 shows severalmajor reasons why meeting customer orders has become an increasingly difficult task. Asthe division begins to grow, linking demand forecast, production planning, productivityand materials requirements become essential. Raw material shortages in 1994 and 1995and limited capacity are major issues facing most locations. Planning production levels isa challenge, since marketing could be off as much as 20 percent on forecasted demand.Marketing states that the customers do not have a good feel for what their demand is andthis makes it difficult to determine. In the past, the strategy of manufacturing has beenmake to order, but as demand increases beyond capacity in peak months it is difficult tomaintain this focus without adding capacity. To counteract the seasonality of the business,inventory leveling is being attempted. However, excess capacity targets to provideflexibility and inventory targets have not been firmly established. Product mix, toolshortages, and product changeovers are major scheduling issues. Also, an effort tostandardize manufacturing performance measures is beginning since information isrequired on issues such as downtime, scrap rates, and promised performance.
13
-
7/27/2019 33960864
14/84
The organization has a very lean management structure. Being geographically spread outincreases the challenges of managing the business. The rapid growth and expansion hascreated a reactionary environmentwhere individuals are respondingto day-to-day crises.Therefore, a limited number of people are involved in planning, competitivebenchmarking, and sharing of knowledge between facilities. New customers are beingacquired, but not as quickly as they could be in specific markets since there is not enoughcapacity to meet demand. In certain markets the facilities do not have space to add newmachines and the market is mature, therefore, the business unit does not want to addadditional capital. These more mature facilities are supporting demand in emergingmarkets while plants are completed and expanded to support local demand. Resources atthe mature facilities are also being assigned to assist in plant start ups. Expansion has notslowed, so the problem of mature markets supporting new markets will continue.
Limited Material Qualityand Quantity Result in Wasteand Shutdowns
Customer ReturnsProduct Bottlenecks,Tool Shortages, andProduct ChangeoversAffect Scheduling
L ,
Present MeasuresDo Not Lead ToBusiness Accountability
Better Tactical Data Expansion To AddTo Identify Causes of Plants and MachinesProblems is Needed Causes Delay
Wide Variations inAccuracy of ForecastedDemand
Figure 1-2: There are several reasons why meeting customer orders is achallenge, and the boldface issues will be discussed in moredetail.
14
Insufficient Capacityat Certain Facilities
Difficulty MeetingCustomer Orders- Ii II i iA
-
7/27/2019 33960864
15/84
1.3.1 Measurement Of ProcessThe facilities collect a large amount of data. Data collection measures were implementedbecause the information may be helpful in the future. When a new measurement systemwas created, the old system was not modified or eliminated, since certain individuals stilluse the old system. There is no agreement or standard on one measurement system. Also,the data that is recorded does not always reveal the root cause of problems. When anequipment failure occurs and results in the shutdown of a machine there may be severaloptions for the operator to record downtime for the same failure. For the first step in theprocess there are 100 downtime codes for why a molding machine that is scheduled to runis not operating. Facilities also have the ability to customize and add their own codes.Oftentimes,there is no agreement on downtime definitionsby data collectors and all usersof the data. Therefore, data accuracy is not consistent and becomes questionable whenmaking decisions. Also, the structure for people to be accountable for their data is not inplace. When a facility runs out of material unexpectedly it is not always clear who isresponsible for ensuring the situation is resolved.
1.3.2 Raw MaterialThere is a limited amount of raw material available to make the product. Increaseddemand for polymer, natural disasters that have closed down polymer facilities, and thedelay of new plants producing the polymer all have contributed to creating a worldwideshortage. Some suppliers will not guarantee that they can provide material. One supplierhas an exclusive contract with the competition. Also, the price of polymer is rapidlyincreasing, escalating manufacturing costs. Although increases in the price of the producthave been announced, these increases have not kept up with the escalating material coststo the facilities, and the customers have not always agreed to the increased price. Onefacility shut down for several days since they could not get additional shipments ofpolymer material. If a facility has 10 molding machines, shutting down one moldingmachine for a day results in a ten percent average daily revenue loss. Based on facilitysize, shutdowns due to material shortages can be extremely costly. Although some
15
-
7/27/2019 33960864
16/84
recycling of material is done after the first process step, not all of the facilities arecommitted to doing it. One facility has the equipment in place to recycle material, butover time, recycling was stopped. Assuming that this one facility could recycle 70 percentof scrapped material after the first process over an eight month period it could have savedapproximatelyone percent of material costs(this number is based on material cost duringthe seven month period, and material costs have increased 20 percent since then).
There is limited or no quality inspection of incoming material. Raw material suppliers arecertified to ensure that material specifications can be met. The philosophy of the facilitieshas been to trust the suppliers to test material in their performance laboratories.Additional testing equipment and training are considered to be too costly. Also, laborwould need to be trained to do these tests. There are some questions regarding theappropriate tests to perform, since correlations on which material properties will result inproduct defects or increased process waste are not well defined. One facility does a meltflow index as their check and it had rejected four percent of shipments in 1994 as shown inFigure 1-3. Melt flow is a measure of polymer molecular weight or chain length, reportedin grams. The target melt flow is X grams and the upper and lower specifications are1.1 8X grams and 0.82X grams respectively. The suppliers and facilities melt flowmeasures cover the whole specification range during the given time period. Figure 1-4shows that this facility has not rejected material in 1995, but the variation in melt flow isconsiderable. Other facilities do not do this test, so there is no basis for them to rejectincoming raw material shipments.
16
-
7/27/2019 33960864
17/84
c r.wD '[ kL; o U- I+1', 1Et i 5 .
(swer.i) .ia!lddnS pur l!!am Sq luauainsalN MoIj la]W
0
o
C,
0
tr6/6Z/
t16/16,1L 'C
t6/lZ/O Lt?6/L/O Lt,6/Z/OV6//6 0t'6/l6 0t6/1/L Ctr6/6ZL/Zt61/9/
tb6// I/S6191/ r
b6/ZL/Lt61,L -
17
-
7/27/2019 33960864
18/84
x x x0 X X- o
(swve) .ia!lddnS pue iatuosn Xq sluauwainsuaN MOi laWE
0lo
,AI::.C
S6191 >S6/l,
S61/GZIZ co
S61I . , -
I~S6I6/Z/Zm961/LZ -)
96/OZILS6/91W
00zS6SZIL l96/gLS6// L
C C:0 W =
S [+t=K&u = Pd2IIHIEn~:
18
-
7/27/2019 33960864
19/84
1.3.3 Scrap RatesIn Figure 1-1 the various types of scrap that are measured are given. The definition ofscrap varies and all definitions have some component of estimation in them. ReportedTotal Process Scrap is the daily scrap occurring duringthese three steps which direct laboridentifies, estimates and records as having been generated. Estimation techniques involvedirect labor visually determining how full a given container of scrap is and converting thisvalue to number of units scrapped. A gap between reported and calculated scrap ratesexists, and Figure 1-5 illustrates this gap over a ten month period. For the ten monthperiod, if calculated scrap was X percent, reported scrapwas 0.3X percent. Part of directlabor's performance evaluation is tied to individual scrap rates, and is one possibleindicator of why reported scrap rates are lower than calculated. Calculated Total ProcessScrap is a measure used in the Production System Yield model; the latter is defined indetail in Chapter 3. Calculated total process scrap is a measure of scrap between the moldprocess and completion of finishingprocess. There is an actual count of the product at theFinishingstep that is boxed and ready for shipment (packed), but the amount of productmolded is often estimated, which may cause some error in the calculated scrap number.The definitionof calculated total process scrap is one minus the quantity of productpacked divided by total product molded.
Since the calculatedtotal process scrap measure relies on estimation, a check of theassumptions suggests that this number overstates the scrap levels. Given the reportedtotal molded number, machine speeds can be back calculated. Operators record machinespeeds(this includes molder speeds) once a day and they are generally constant, withminor fluctuations. The total product molded measure should approximatelyequal thereported molding machine speed multiplied by molding machine run hours. Moldingmachinerun hours are recorded by the operator fromthe machinecontroller (electroniccontrol system that operator resets at given intervals). Molding machineswould have tooperate between four to fivepercent faster than reported for the calculated total scrapmeasure to be valid. Assuming the reported molder speed is correct, and using moldingmachine run hours, a revised number for scrap, the Estimated Total Scrap Rate, can be
19
-
7/27/2019 33960864
20/84
determined. This estimated scrap rate is 0.6X percent for the ten months (versus Xpercent for calculated scrap) and this number is between reported and calculated. Theestimated scrap rate is one minus the quantity packed divided by the quantity moldingmachine run hours multiplied by reported molder run speed. Although there is potentialfor machinerun hours to be understated, it is a more accurate number to use to calculatetotal product molded than using a total molded numberbased on visual estimation. Ifestimated total scrap rate is assumed to be the actual scrap rate, then approximately sixpercent in potential revenue was foregone at Facility 1. Another way to state this is thatproducing this level of scrap during the given ten month period is the same as having twomachines unable to operate for eight months.
Material usage is important to other functions within the organization outside ofmanufacturing. The financedepartment keeps track of how much material was used forthe month based on silo inventory and compares this to the amount of qualityproductproduced and Reported Total Process Scrap Rates. A variance suggests that there is adiscrepancybetween the amount of material used in the process and the amount ofmaterialpurchased for use. During a ten month period Facility 1 was able to account for98 percent of purchased material that entered the process, with 97 percent endingup asshipped product, and one percent ending up as reported scrap. At Facility 1, two percentof material that entered the process could not be accounted for. Some informationonmaterial usage is not recorded, since there is no process or procedure in place to capturethis information. An example is that research and development performs designedexperiments on production equipment, and may or may not use their own materials.Generally, it is assumed that information that is not recorded is a small fraction of theoverallmaterial used.
20
-
7/27/2019 33960864
21/84
Month Calculated Total Reported Total Estimated TotalProcess Scrap Process Scrap Process ScrapJanuary Xl% 0.33 X1% 0.59 X1%February X2 % 0.41 X 2 % 0.67 X2 %March X3% 0.34 X3 % 0.63 X 3%April X4% 0.34 X 4 % 0.68 X 4%May X5% 0.28 Xs% 0.63 Xs%June X6% 0.23 X6% 0.56 X6%July X7% 0.23 X7% 0.47 X 7%August X8% 0.23 Xs% 0.60 Xs%September X9% 0.24 X9% 0.58 X9%October Xlo% 0.20 Xlo% 0.46 Xlo%Year to Date 1994 X% 0.3 X % 0.6 X %
CALCULATED = 1 - (PACKED/TOTAL MOLDED)REPORTED = VISUAL ESTIMATION BY OPERATORESTIMATED = 1 - (PACKED/(RUN HRS. x "REPORTED" MOLDER SPEED))Figure 1-5: Due to estimation techniques it is difficult to determine the exact scrap rate.
Customer returns are also a part of scrap. At a minimum, when a customer returnsproduct, a partial or full product price credit is given, depending on the reason for thereturn. Over an eight month period of time Facility 1 had 0.2 percent of productshipments result in returns and Facility 2 had 1.3 percent of product shipments result inreturns. After the product goes through the finishing process, no additional value isadded to it but shipping product to the customer does add cost. Figure 1-6 show themajor cost steps and the value added steps that the product undergoes prior to leaving thefacility. The value added steps are during product processing. An additional cost that isnot shown is inspection cost incurred after each manufacturing step. There are more coststeps in the process than value added steps.
21
-
7/27/2019 33960864
22/84
MANUFACTURING STEPS COST STEPS
extrude polymerPolymer Value
MoRegrind processscrapadditional rawmaterial FunctcRequinprocessscrap 1
sell IFinishing
Processing/ScrapReprocessingCostRaw Material ValueProcessing Cost
Processing Cost
Provide ProductShapeAdd FunctionalRequirementsand Customization
AdditonalCustomizationWarehouse
Ship to Customer
Holding Cost
Shipping CostFigure 1-6: Manufacturing cost and value added steps related to manufacturing process.
22
Silo withPolymer
VALUE-ADDEDSTEPS
Id
inalements
-
7/27/2019 33960864
23/84
Chapter 2: Method For Improvement
A framework for addressingthe issues facingthe facilitieswill provide a systemsapproachto working toward improvement. Included in this framework is an increased qualityfocus, a measurement control plan emphasizing he reduction of disruptions and variationsin manufacturing, and an understanding of linkages between variables that will increasecapacity and throughput. Quality and quality goals need to be defined for theorganization, programs and measures determined to help the organization achieve thesegoals, and documentation of results of quality initiatives provided. Process evaluation,which includes measurement tools, can help determine the status of the process.Monitoring, by measuring the process, assists in providing information to improvetroubled areas. The difficulty is to determine what to measure and what tools to use.
There are several factors that influence the organization's ability to produce qualityproduct profitably, as shown in Figure 2-1. The quantity and quality of raw materialavailable influence the time available to make product, as well as the quality of productproduced. There is a tradeoff between performing preventive maintenanceversus notperforming preventive maintenancein order to continue making product. If preventivemaintenance is not performed, at some point, the machines will fail. Methods to improvequality product are desirable, since quality product affects profitability. A quality initiativeis one method of increasing the amount of product that the customer receives. In laterchapters, other methods that can be implemented in conjunction with the qualitycontrolprogram will be examined.
23
-
7/27/2019 33960864
24/84
Raw Material Availability, Preventive Maintenance Hours,Machine Failure, Mold Speed,Preventive Maintenance Time, Raw Material Consistency,Scheduled Orders Product Design
Figure 2-1: There are several factors within manufacturing that caninfluence the organization's ability to produce qualityproduct profitably.
2.1 Quality DefinitionFor the gains provided by a focus on quality improvement to be realized, a consistent,unified definition for the division is needed. Quality can be measured by the defect ratesor scrap levels. It can also be measured by determining if the product performs accordingto design when in the market, if delivery times are met, and if prompt, effective service isprovided when product issues arise. Garvin, states five major approaches to the definitionof quality and eight dimensions of quality.' He states that, "both reliability andconformance are closely tied to the manufacturing-basedapproach to quality.Improvements in both measures are normally viewed as translating directly into qualitygains because defects and field failures are regarded as undesirable by virtually allconsumers."' The quality managers at the majority of locations have statistical processcontrol and cost of quality initiatives in progress in the manufacturing area. These are
24
Hours AvailableTo Attempt ToMake ProductProcess Scrap andCustomer Return
Net Qualit Product Sales m rftbltProduct I-r andPrice 'ftbly
-
7/27/2019 33960864
25/84
-
7/27/2019 33960864
26/84
-
7/27/2019 33960864
27/84
-
7/27/2019 33960864
28/84
facility 1 has started to weigh scrap at the end of the shift. When there is a discrepancybetween total weighed scrap at the end of a shift and reported scrap, reported is increasedto reflect the weighed amount. The present technique does not allow the operators topinpoint at which process the discrepancy occurs. In the future the facility plans to havescales at every process and as scrap occurs it will be weighed and the reason for scraprecorded. This method will alleviate the need to estimate scrap. The facility is stillexaminingeffective ways to measure the total amount of product molded at the moldingmachine.
Material is considered to enter into the process as soon as it arrives at the facility.Therefore scrap losses could occur before the materialwas processed and this needs to becaptured. Accurate data need to be obtained on where scrap is occurring. Also, workingwith finance, who keep material variance information, will provide a cross check on dataaccuracy. Other functions, such as purchasing may also be appropriate to work with. Inaddition, procedures for materialusage between groups such as research and developmentand machinery will need to be set up.
Early detection of insufficient product quality is essential since as the material is processedit goes through several steps which add value. Therefore, ensuring early on in the processthat material meets specifications will reduce the need to scrap product that has had valueadded to it. The more processingthat the product receives, the more value it has, andmore costly it is if the product needs to be scrapped. After the molding process thepolymer can be recycled, and this scrap does not add significantly to material costs. Afterpost-mold processing and finishing any scrap is sold and cannot be recycled, thereforeincreasingraw materialcosts. If product is returned, the costs are significantlyhigherthanif the defect had been identified before the product was shipped. Therefore there may be atradeoff between higher internal scrap rates to reduce customer returns. Also, a tradeoffbetween more costly incoming inspection to reduce internal scrap rates can exist. Theupfront, proactive investment to ensure quality when the product is still of very low valuewill reduce costly quality issues later in processing chain. Over time the manufacturing
28
-
7/27/2019 33960864
29/84
learning curve regarding quality increases and it becomes even more efficient to producequality product. Fine states that "economic conformance level analysis (i.e. cost of qualityminimization)provides an accurate model of the strategic quality optimization problem,but that quality improvement enhances learning about the production process so that thecosts of achievinghigh quality decreases over time and zero defects becomes the long-runcost minimizing quality level. 4
29
-
7/27/2019 33960864
30/84
-
7/27/2019 33960864
31/84
-
7/27/2019 33960864
32/84
PSY I SCHEDULED RATE% RUN MOLDER NETTIME % X QUALITY X SPEED X RUN X PRODPRODUCTS EFFICIENCY% HOURS % RATE %SCHEDULED OURS PACKED+ WIPa (TOTALMOLDED/RUN IN) RUNHOURS PACKED RETURNS INVOBSOMOLDERS CALENDARHRS TOTALMOLDED DESIGNSPEED SCHEDULED OURS PACKED
IDLE TIME LOSSES CLOSURESCRAP RATE MOLDERSPEED DOWNTIME%I -A 1 B DESIGNSPEEDXC 1-D
Figure 3-2: Multiplication of the five key factors results in Production SystemYield number. Formula for each key factor is shown above.
3.1.1 Components Of The PSY ModelScheduled Time Percent, component A of the PSY formula in Figure 3-2, measures theamount of time the facilitywas scheduled to operate against calendar hours. Scheduledtime percentage is calculated by dividing scheduled molder hours by the quantity numberof molders times calendar hours. Idle time losses equals 100 percent minus ScheduledTime percent. Preventative maintenance is not an idle time loss, but is considered underdowntime. The model tracks the idle time losses such as seasonality, material shortages,insufficient crewing, and no tooling. Also, sales versus projected production, order leadtimes and equipment shortages are tracked in an effort to focus on what idle time lossesare occurring.
Rate of quality products, component B, measures the amount of quality sellable productbeing boxed (packed) and ready for shipment to the customers. This is calculated bydividing the quantity packed product plus change in Work in Progress(WIP) by total
32
-
7/27/2019 33960864
33/84
product molded. Scrap equals 100 percent minus Rate of Quality Products percent.Pareto charts of scrap by process, product, machine,and scrap cause by process assist inunderstanding the calculated rate of quality products.
Run Speed Efficiency,component C, measures the performance of the molding machinesagainst the manufacturer's design speed. This is calculated by dividing the total quantitymolded product by machine run minutes (minutes machine operating and producingproduct) and by design speed. The molder speed is calculated by multiplying the designspeed by the run speed efficiency. The model provides a Pareto of molder set speed byproduct and color, since the run speed efficiency is based on a weighted product speedaverage.
Molder run hours, component D of the PSY formula in Figure 3-2, is the percentage ofoperating time that the molding machineis actuallyoperating and producing product.Whileoperating, the product produced is either sellableproduct or scrapped product.Molder run hours is calculated by dividing machine run hours (hours molding machine isoperating and producing product) by the scheduled molder hours(hours molding machineis scheduled to operate). Downtime is the percentage of time a molding machine wasscheduled to operate but is not operating. Downtime equals 100 percent minus MachineRun Hours percent. The PSY model then creates Pareto charts" of the downtime reasonsand tracks specific indicators such as turnover, absenteeism, and training hours to assist inunderstanding why the machine is down.
Net Production Rate, component E, tracks how much product remains sold. It iscalculated by subtracting returns and inventory obsolescence from the quantity packedproduct and then dividingby packed product. The model tracks inventory obsolescenceand customer returns. Other information that is pertinent to the facility's production can
A pareto chart is a bar graph that ranks the biggest problems on the left followed by lesser problems.Arranging problems in this order helps in determining which problem to attack first.
33
-
7/27/2019 33960864
34/84
-
7/27/2019 33960864
35/84
ai;5 nSw Jfo
a owI19US0 0iI2
wI08P A
z
35
,o,
I I As I I B
IIn ' ' i-'8g _ I ,i :
o o g SNo e o X o
C)uscna.enene
z -0t _
_ , a
I". g.1Iz3 o2
Ii;
3,
I 1 94
I, g g g g g Is~ R O
z9 R9
: I > 9 '
-
7/27/2019 33960864
36/84
-W
i 1 , II i - Il - -I
-
I l g z t I
I6V7z.0
I -u 8~~1 IS~5
f .VI C >~~o~_: o%9ffi
ozZ'
jI{c-po
laO ~ - ~- ~ ~z _ _uruZ sol_IC) - c-- LoCL.:
- - o O O o o
4 1-
r II _
CToi6DSo
I I
o
o 8o
oRTo
0L
mL1J0
U0R.z3n
Ioo
83I?
0 0 0C' 4- 7
36
- .Iir I - l rv
15ei 1
t
9029
-.~~~~~~~~~~~~T
I 'zi-2I,
;z
4
-
I g o ee f I ma ma s II9
9
Li.j2 d --
d' g g g g g zit
-
o -
94
'"'"
' -- P.
tuE f
i i -E - t r-~~~~~~~~~~~~~~~ Ii i i - s 6 . o I_ I _
I I . I hii 01.
- I ~~r~~9~ II iS I : i _ , I -- -- -- -- p! C: I 2717 e I1.. V) ~L r, Iz
"9 IhI . . r"IQ C) C? 4
fZ
cB..I .1 I-f. Bl11
1, ITr 1
2
U 1 I ..__ e I
l I I
-
7/27/2019 33960864
37/84
3.2 Analysis of PSY Model
After examining historical data for all facilities, a benchmark level of PSY was determined.The facility that had sustained as a yearly average the highest level in a certain factor wasused as a benchmark for the given factor. Then, the five factors were multipliedtogetherto establish the benchmark PSY number. None of the facilities presently have attained thebenchmarkPSY number, although some are close. A framework for analysisof eachfacility's PSY was established. See PSY Review Process in Figure 3-5. The reviewprocess has a location compare its PSY number and level of performance in each of thefive factors to the benchmark level. In the PSY model the factors and PSY number havesupporting data that would be analyzed to determine how to reach or exceed benchmarklevels in each area. At the end of the review process the facility's improvement action planwould be updated to incorporate projects that increasePSY.
37
-
7/27/2019 33960864
38/84
-
7/27/2019 33960864
39/84
-
7/27/2019 33960864
40/84
-
7/27/2019 33960864
41/84
-
7/27/2019 33960864
42/84
-
7/27/2019 33960864
43/84
This product PSY analysis can be used to evaluate standard operating policies such as onefacility running the same product at a slower speed. Facility one, with a faster machinespeed, provided more product to the customer using one less machineduring the sametime period than facility two. Facility one had a slightly lower rate of quality products,but this could be the tradeoff of running faster. Also, looking at the data, there is asignificant difference in scheduled time between the two facilities. More investigation intothe discrepancy revealed that facility 2 had not staffed for the present demand levels, andother issues such as absenteeismand employeeturnover were causing a lower scheduledtime.
3.4 Limitations of PSYPSY can be used as a high level tool to identify significant improvement opportunities. Toenhance the model's effectivenessit needs to be used in conjuction with other data. Thoseusing the model need to realize how data are being obtained, so that they can assess itsaccuracy. If measures are not in place for areas where data are needed a process to obtaininformation will need to be developed. This model is data intensive, and data entry is anissue. Presently, the model is being updated on a monthly basis, but the facilities thatdesire to use the model want to be able to update it on a weekly or daily basis.Suggestions and buy in from individuals who have to collect the data are needed forsuccessful implementation.
There is very little data on idle time losses, and more is needed since it has a largeinfluenceon Production SystemYield. One reason that the facilitycannot be fullyscheduled is due to tooling shortages. Tooling shortages are not desirable if there is freecapacity. Tools are a capital investment,but giventhe potential monetary profits, excesstooling could make sense.
In the model Pareto charts are created from reported data. Pareto charts will assist inanswering questions about where to close gaps. Monthly run charts will assist in
43
-
7/27/2019 33960864
44/84
-
7/27/2019 33960864
45/84
Chapter 4: Production Planning Model
Planning allows an organization the opportunity to anticipate and prepare for customerorders. The ability to produce customer orders affects customer loyalty and goodwill, asshown in Figure 4-1. To ensure that no orders are delayed or missed there is a customersatisfaction tradeoff, as well as, a cost and feasibility tradeoff regarding capacity. Ordersmay have to be delayed if no material is available. Although adding machinery is costly, itmay provide needed extra capacity. Productivity improvements occur over time, and it isharder to predict when a capacity increase due to productivity improvements will beavailable. Other methods of increasing capacity include inventory leveling, support fromother facilities, and varying labor usage. To minimizeinconvenience to the customer,considering alternative plans will allow the facilities to be prepared when demand changes,or capacity is constrained, given existing machines, materials and production plan.
Ability to Prouce ustomer Perception of ustomer Loyalty andustomer Orders sponsiveness to Needs odwill(new and returnustomers)PrepareroductionlanstomeratisfactionradeofPrepare Production lan Percent Delayed/MissedotrdireAssess Present Cost/Feasibility Capacity TradeoffDemonstrated CapacityCapability Inventory Leveling, Material and LaborSupport from Other Facilities, Availability and UsageAdd Machinery,Productivity Improvements
Figure 4-1: The ability to produce customer orders on time affectscustomer loyalty and goodwill.
45
Compare DemonstratedCapacity to ForecastedDemand
-
7/27/2019 33960864
46/84
The purposes of the production planning model are to assist management in makingdecisions about the business unit on a regional level, and to assist management inestablishinga stronger communicationlink with all functions of the organization. Themajor components of the production planningmodel are aggregate sales, theoreticalcapacity, speed adjustments, capacity adjustments, demonstrated capacity,production/inventory plan and materials requirements. The production planning model is aspreadsheet tool that incorporates informationfrom the PSY model to establishdemonstrated capacity, indicatesmajor product loss categories and how these impactcapacity, provides abilityto analyzevarious demand scenarios,determines amount of goodproduct a region is capable of producing on an aggregate level and determines the amountof polymer needed for this production level. The production planning model linksforecasted demand with demonstrated capacity to determine a production plan andmaterials requirements. Raw material requirements are necessary well in advance due tomaterial shortages. Also, when demand exceeds capacity, advance planning assists inestablishingmethods to minimizeor avoid product shortfalls. Various managementdecisions can be considered with this model to understand how certain decisions influencedemonstrated capacity, the production plan, and materials requirements. Also, this tooladdresses how changes in the demand will affect the region's ability to meet customerorders. The model covers a two-year time frame, in monthly increments at a regionallevel. This region can be serviced by one or multiple plants. Generally, facilities do theirown planning. Regional planning on aggregate level provides more flexibility indetermininghow to best use assets to meet customer orders.
4.1 Components Of Production Planning ModelAn example of a production planning model is shown in Figure 4-2. The components ofthe model are explained below.4.1.1 Aggregate SalesAggregate sales is the forecasted sales for the region of interest. In section 1A of themodel forecasted sales for 1995 are shown. Generally, there is wide variation in the
46
-
7/27/2019 33960864
47/84
accuracy of the forecasted demand. To determinethe impact of a low sales forecast theforecasted 1995 sales were increased by 15 percent each month (and is stated as 1995-15%). Although it is not provided in the example, sales below 15 percent could also beevaluated. Forecasted sales for 1996 are shown. The value of having two years offorecasted sales informationensures that December 1995 production plan is based onability to meet customer orders in the future versus ability o meet end of the yearproduction goals. The monthly expected sales information is then discussed with allfunctions and will be used to establish a monthly production plan. Depending on themanufacturingstrategy, the amount of product produced in a given month may bedifferent than the amount of product sold.
4.1.2 Theoretical CapacityThe theoretical capacity is the monthly product output for 365 days of production, giventhe number of molders within the region and the molder design speed. Capacityis basedon molding since the process is designed for molding to be the rate limiting step. Insection 1B output for 1995 and 1996 are given in z units.* This example showstheoretical yearly output for 1995 and 1996 at 120 z. These yearly outputs could differ ifadditional molders were added over time or design speed increased as result of machinemodifications. Generally,the region will not be producing product at the theoreticalcapacity due to holidays, slower machine speeds, preventive maintenance and othercapacity losses.
z is picked as the unit of measure for the examples throughout the text, but the author does not specifyfurther detail in order to maintain facility confidentiality.
47
-
7/27/2019 33960864
48/84
-
7/27/2019 33960864
49/84
-
7/27/2019 33960864
50/84
-
7/27/2019 33960864
51/84
-
7/27/2019 33960864
52/84
4.1.3 Speed AdiustmentsSpeed adjustments are a revised output based on product losses due to speed. In section1C demonstrated UPC(units per time cycle) or actual machine speed is an aggregateproduct speed based on product mix and the speed at which each product is presentlymanufactured. The 1996 speed adjustments are shown in section 1J of the ProductionPlan. In the example, product not produced due to speed losses is given as a yearlypercentage of theoretical capacity(percent loss).
4.1.4 Capacity AdjustmentsCapacity adjustments are the losses from various production parameters that result in theregion not achieving revised output levels. The production system yield model providesdata to determine the monthly losses. The intention of the capacity adjustments is toreveal to management actionable items that generate capacity losses. By seeing how muchcapacity is lost by a certain parameter, management can make informed decisions on howto change certain parameters to increase capacity. In Section 1D of the model eightcapacity parameters are given. These parameters are: employee absence, holidays, majorR&M(repair and maintenance), PM(preventive maintenance), mold equip failures(moldingequipment failures), all other downtime, scrap and product losses due to tool shortages.The scrap parameter is comprised of the following subcomponents: scrap at mold, postmold, finishing, off line processing and other. Capacity adjustment parameters are thoseitems that have a significant impact on capacity, have the potential to be altered, andmanagement desires to know the present and future effects of these parameters.Depending on the regions' needs, the parameters that are detailed can be modified. Forcertain facilities, employee absence may not be a large capacity loss, and this may want tobe considered in with product losses due to all other downtime. Presently, product losesfrom a given parameter are a function of demonstrated unit per time cycle(or presentmachine operating speeds). Scrap is a function of revised output. As relationship betweenparameters is more fully understood, equations linking these could be put into the model.
52
-
7/27/2019 33960864
53/84
-
7/27/2019 33960864
54/84
4.1.7 Materials RequirementsMaterials requirements establishes how much raw material must be received to meetproduction requirements and scrap allowance. In Section 1G of the model only polymerraw material requirements for molding are presently established. Post-mold processingraw material requirements could be incorporated at a later date. An initial beginninginventory is entered and beginning and ending inventory of polymer is calculated forsubsequent months. An inventory target can be set monthly based on the amount of rawmaterial the facility intends to have availableat any giventime. By setting the inventorytarget and examining the difference to target number it can be determined if monthlyshipments need to be modified. Knowing the amount of usable units that will be producedand using the average weight per unit a calculation determines how much material will beneeded for production. Material wasted as scrap can be determined based on anticipatedscrap levels, given in the capacity adjustments section. Scrap material that will be recycledis not considered in the material requirements, even though there is a time lag on its use.Entering the intended receipts of material for the month allows the user to determine if theyearly material requirement will be met. If the ending monthly inventory is negative thenmore material receipts will be needed. To ensure that material ending inventory is zero orpositive, the production plan will need to be modified if increasing material receipts is nota feasible option.
4.2 Facilitating Communication
The production planning model incorporates informationfrom various functions of theorganization. Some of these functions include: marketing/sales, manufacturing,design/research and development, purchasing and finance. Each function has a role withinthe organization. The results of the model provide managementa mechanism fordiscussion and information exchange on the important linkage each function has inassisting the organization in meeting its objectives. In order for a feasible production plan
54
-
7/27/2019 33960864
55/84
and materials requirements to be generated, forecast accuracy and a realistic demonstratedcapacity are critical.
The model allows each function to examine the impact of their information and othergroup information on the present and future organization's ability to meet customerorders. The marketing/sales group interacts with the customer and will provide forecasteddemand data for the model. Given historical sales data, as well as industry trends aconfidence level can be assigned to the forecast. Based on the forecast, manufacturingwill have to determine how many individuals to hire, and purchasing will decide onmaterials requirements. Manufacturing has data around their performance, and thisinformation will assist in determining a demonstrated capacity. If manufacturing does notdeliver the stated demonstrated capacity the implications could include unmet orders or ifthey exceed the expected demonstrated capacity, there could be excess capacity whichwas not utilized. The design/research and development group contributes technologyadvances that assist in increasing demonstrated capacity, as well as establishing newmarkets. These advances, which can be analyzed with the production planning model,could include a new design, or different materials. The purchasing group needs to knowhow much raw material to buy and when. Given raw material shortages, the earlier theyhave material needs the quicker they can ensure availability. Finance is involved becausethey can state how much does it cost to have a certain excess capacity position, as well aswhat technology advances make sense from a cost perspective. Also, if ways to increasedemonstrated capacity are examined, the finance group can suggest which options arefiscally feasible.
4.3 Base CaseA hypothetical example of a production planning model is shown in Figure 4-2. The 1996sales were calculated based on 2 percent growth for the industry. The theoretical capacityin 1995 and 1996 is the same, since this region is not expected to add any machinery. Thecapacity adjustment categories and product losses are listed in Table 4-1.
55
-
7/27/2019 33960864
56/84
-
7/27/2019 33960864
57/84
polymer received every month. The 1995 demand can be met given present capacity, butthere is very little excess capacity available during the year. If 1995 demand increases 15percent every month, starting in June all customer orders will not be met. Figure 4-3shows the cumulative demonstrated capacity and demand for 1995. When the cumulativedemonstrated capacity line is above the demand line all orders can be met, but if thecumulative demonstrated capacity line falls below the cumulative demand line a productshortfall will exist.
As shown in Table 4-1, the top three areas for opportunity to reduce product losses arespeed, total scrap, and tooling shortages. A comparison of similar products at differentfacilities may be done, to see if any speed improvements could be achieved bystandardizing product run speeds. As speed losses decrease, the losses that occur in othercategories will increase slightly since losses are presently a function of speed. More thanhalf of the scrap losses are due to other scrap losses that are not well defined. If theseother scrap losses were reduced this could have a significant impact - not only will morecapacity be available, but less material will need to be purchased. Although there arelosses due to tooling shortages, if sales can be generated in products where excess toolingis available, additional capacity would be available for use. Also, if the region anticipatesthat it will be short similar tooling in 1996, a cost benefit analysis should be done todetermine if additional tooling should be bought.
4.3.1 Comparison to 1994 DataUsing 1994 information the production planners for the facilities in the region state thatdemonstrated capacity is 83 z for the 1995 year. The difference between this number andthe production planning model demonstrated 1995 capacity of 84 z is small. Thisdifference could be attributed to differences in how losses from capacity adjustmentparameters are determined.
57
-
7/27/2019 33960864
58/84
-
7/27/2019 33960864
59/84
-
7/27/2019 33960864
60/84
region. Figure 4-4 shows the ability to meet demand at various demonstrated capacitylevels.
All lines below the cumulative 1995-15% Increase in Demand line indicate a shortfall inmeeting orders. At the base case(Y machines) the region would be able to meet demanduntil May, and then a shortfall will be experienced. Even if customers are willing to waitfor their orders, by the end of the year there will be a shortfall of 7.6 z. At 1.08Ymachines demonstrated capacity, all orders will be met without delay until June. Ifcustomers are willingto wait for their orders, then by the end of the year there will be noshortfalls. To ensure no shortfalls at any time during the year, 1.13Y Machines would beneeded.
To add 13 percent more molding machines, there would be a major capital expenditure,since this would involve the need to buy some Post-Mold Machines and FinishingMachines so that molding would continue to be the bottleneck operation. Also, additionallabor and floor space to put the machines may be needed. In addition, more machineswould increase material costs. These costs would have to be weighed against the potentialrevenue lost. Will customers be willing to wait for their orders? The value of customergood will and future lost sales due to delays are difficult to measure. Table 4-2 shows thepotential percentage of revenue lost if customers are not willingto have their ordersdelayedversus if customers are willingto have their orders delayed. In the base case, evenif customers would accept delayed orders, there is a potential loss of revenue of ninepercent per year due to insufficient capacity. With eight percent additional machines(1.08Y machine Demonstrated Capacity), all orders will be met if customers acceptdelays. If customers willnot accept delays, there is a potential lost revenue ofapproximately2 percent per year.
60
-
7/27/2019 33960864
61/84
-
7/27/2019 33960864
62/84
Number of Potential Lost Potential Lost Shortfall MonthsMachines Revenue RevenuesWith No Delay With DelayAccepted Accepted(Percentage/Year) (Percentage/ Year)Y (Base Case) 10.12% 9.08% May -December1.08Y 2.27 0 July-October1.13Y 0 0 none
Table 4-2: With 1.13Y machines the region will be able to meet all orders without delayif demand increases 15 percent.
Cost of machinery,the value of meeting customer orders on time, and other alternativestomeeting customer orders are all factors that influence a decision to determine if machineryshould be added. If meeting customer orders on time is critical, then a certain amount ofexcess capacity at all times will be desired.
To assess the cost and feasibility of adding machines, a net present value(NPV) andinternal rate of return(IRR) calculation was performed.*" The company expects aminimumof a 20 percent rate of return on any investmentsthey proceed with. Fiveadditional molding machines and the corresponding Post Mold Processing and Finishingmachines would be a significant investment. It was assumed the machinery would beoperational immediately after investment and added to an existing facility in the region,where floor space was available. The corporate tax rate is 34 percent and a seven yearstraight line machinery depreciation schedule was used. If the existing product price,manufacturing cost, throughput rates, and 352 day yearly machine operation of all fivemachines are maintained, then a net present value analysis of this investment is positive ata 20 percent rate of return for four years of machineuse or beyond. The IRR of this
NPV is the present value of cash inflows less the present value of cash outflows, or the increase inwealth accruing to an investor when they undertake an investment. IRR is the discount rate at whichprojects net present value equals zero. Rate at which funds left in a project are compounding(c. f. rate ofreturn or yield obtainable on an asset)6 .
62
-
7/27/2019 33960864
63/84
-
7/27/2019 33960864
64/84
-.- 5 years IRR ......-+- 4 years IRR- 3 years IRR ...-
1%
Change n Price (percentage)Figure 4-5: The internal rate of return assuming existing machine throughput varies depending on the life of themachine.
EU7O
35%
, 30%C
0 20%o6-
05 10%
5%0%
-0.1 -0.05 0 0.05 0.1Change n Cost (percentage)Figure 4-6: The internal rate of return given a percentage change in product cost for a three, four and five yearoperational period.
64
45%
40%
35%
C) 30%0E 25%00.E 20%
15%10%
E 5%-S
o%-1- F
AnQ
-
7/27/2019 33960864
65/84
-
7/27/2019 33960864
66/84
3%Change in Throughput (percentage)Figure 4-7: An increase in throughput increases the internal rate of return.
5%
Machine Utilization (Percentage)Figure 4-8: A decrease in machinery utlization lowers internalrate of return.
66
35%
4)
Cl.-00.CI-
I.-0
Cc.
25%-
20%
15%
10%
.. ..... . .... .. .- . - .. .-. - -.-....... .- -
-- 5 years IRR .--- 4 years IRR^- 3 years IRR
0%0%
30I
25%H
20%
15%
10%
091
-5%
-10
-15%N
30%
A
_
5.% -
. _ .
-
7/27/2019 33960864
67/84
-
7/27/2019 33960864
68/84
-
7/27/2019 33960864
69/84
-
7/27/2019 33960864
70/84
model. Data on improved defect levels and effects on tooling were not available, thereforethese numbers were unchanged. The new design production planning model can be foundin the appendix. Speed losses were reduced by two percent for the year 1995, resulting ina two percent increase in demonstrated capacity. All orders were met in 1995 and a 0.1 zexcess demonstrated capacity was maintained, or slightly greater than one percent ofdemonstrated capacity. In 1996, a 0.2 z excess demonstrated capacity was able to bemaintained. If demand increased 15 percent in 1995 all orders would not be met after themonth of June, but implementing the new product does allow the facility to produce 1.6 zmore product than the base case.
4.6 Limitations of Production Planning Model
The production planning model is an effective tool at revealing product losses, providingproduction levels and materials requirements, but a framework for analysis of the modelwill need to be customized by the company. As a company, business philosophy andstrategy will determine how certain scenarios the model provides are analyzed. Whatpriority is material conservation and meeting customer orders on time? These decisionswill influence the recommended material safety stocks and demonstrated excess capacitythat is desired when using the model. A make to order strategy may require longer leadtimes in peak season. Starting the year with a given amount of excess capacity may becostlier than none, but will allow more flexibility when order load varies unexpectedly.
The assumptions used to create the model will affect the production information. Once aregional production plan is established, and tradeoffs of various management decisionsare examined, the next step is implementing a regional plan at the facility level. Althoughthe facilities were involved in providing information for the production planning model,disaggregating the demand and production plan can be challenging depending on theassumptions. One facility may be required to make certain product due to in countrymanufacturing requirements or customer preference. In the production planning model,
70
-
7/27/2019 33960864
71/84
-
7/27/2019 33960864
72/84
-
7/27/2019 33960864
73/84
-
7/27/2019 33960864
74/84
-
7/27/2019 33960864
75/84
-
7/27/2019 33960864
76/84
management in establishing methods to meet as many customer orders as possible. Thesemethods may include a combination of buying machinery and productivity improvements.The real gain that the tools provide is that since they are customized to the facilities andregional markets they serve, as the issues change or more accurate data is available, themodels can be quickly adapted and updated if need be to measure other parameters andexact results and information in appropriate areas.
5.2 Recommendations For Future Work
Continuous improvement efforts to increase quality and productivity and productionplanning efforts to decrease delayed or missed orders are dependent on management'stimely access to and analysis of accurate data and their ability to implement methods thatassist the organization in profitably meeting orders. The following recommendationswould enhance implementation and usage of the production system yield model, theproduction planning model, quality measurement and control efforts and other strategictools:
* Accelerating and broadening the quality effort will result in increases in productivity inseveral areas within the organizations. A study on quality benefits would be helpful inspecifying the tangible as well as intangible benefits. Tracking and monitoring must bedone to produce improvements. Management should consider increasing quality to apoint where facilities can share resources.
* Forecasting affects the ability for manufacturing to plan and effectively execute theplan. It has been stated that customers really do not know their demand. Identifyingtools and techniques that can be implemented will allow this organization to betterunderstand the customers' requirements. In addition, historical trends and cost benefitanalysis, will assist in establishing guidelines which indicate the desired levels ofregional and facility demonstrated capacity and product inventory.
76
-
7/27/2019 33960864
77/84
-
7/27/2019 33960864
78/84
-
7/27/2019 33960864
79/84
-
7/27/2019 33960864
80/84
-
7/27/2019 33960864
81/84
L o oo W)CR 9 8 2 8 o . . . .~l~> Lf to Ici-I c 1 "l Y1 1i 0 Lf 1 00 dz I I-
vr- OR-8048 0 8 8 8 8 2~~~~~~~~~~~~~ ~ ~ 2~~dlrlldld 'Id ,I, cr:ili CN c)0 ~~~~~~~~~o al I I I I I I 1~1I I I 1~1 I I I --{n,- l CC! ~~8 8O~8 8 18 aisi~s Y18c 'o c~, ccdl~~1~6Id 'Id 0 d dluiN9
'~ < >1IIIII-u. aD- >-s{ w OQ ts 0 1; ( 8 Ui CRlu ls 01118~8P,'-d - ,-o cm cluc~ c
co-,,ro :P-co l Cs 10 C,r c0 io8118~1 0D 1 >" I >" - II7 I~jlI Il 1 R 1 8 &1 P1o' , - oim C{dl~l~dl I I I I I'ldl I I I'ld 0 ~0, "i"~srr,7i~;~, co8 C' 0 v 1~'0V M08088Nf: VOa~ o uilcddl -wo In Cs U4 ci
12 U,8 CN N lm woo 6 cs0iodl 0
(A m~~~~m a a x a- cI0L) u) -0 X:uiCli:6.9 0 . -M2 '41 IUL J =) m 0r 9, 0~~~~~~~~a -5 ~~ilS u, o L u(A -O Ct m
IE Ei o 32i~8~8r U~;i~~~C 0 CL 'S. V41: ~ 9 la~a (~z z 0 0 z lo I0
Qcmt-P
E'8'CLcE
._4!0ig._vQ:Sc.9E
-
r'SEtm.S
81
-
7/27/2019 33960864
82/84
( r- I- 0) W() m%*L A2~~qcc~ a, nt0L'UT) ()'(C) ' ' ,i. i ci N .- :'idcricrcd d 00 6 0 C4W
0 o o i,C4 (70 0 CO 00 aLe. i~. q F5 R aR ,. ~ ,.~ ca c58N , " ,T~~~~) co I-- CD.,..,C ! O co cci 6a 6 6 .... : ~ -M-o ~ I0.~~ ~ 0~~~0 t-(0~~~~~(04 0(e O 01. .5: o - -dU5 dd dN) c); N
~~~~~E~ c C,.. ?D,- : ~o (,, . .d~~~~~~~ o o aio a I ',4 ,"r- co ig6)u,, .. .C"ICD ; s$-(6 04- LA LACqDW0 C4C', E~~~ ~a,_o o ''dd' c ~1. o cl c . -_. LI, co c 0 0 oIncN c eLo'cr-- cN P.
,.~~~~~~~~~~~~~~~~~~~.3 I , 0 0 P:a 60 ej~.- (N ,I~lod~dh, ( coC - - -.iN 0 O 0 I00 ((D DL L~~~~~~~6 o o (0 cv)' 00(W 0)o 'o. . dr d2 a 0 0 0 F.-Lr-c0 0 910 _ r Ui CLi02 0 0 0 oo 'i 0 0U0 load Vdid I1I- l~dI I.-C'CO '0 ~0 0 0 00 flv(4 0(0 '0~ 9Ir-(0 ~~~~0*r-~~~~ 00 m ov-0 )'LA.. ....o t:.. L0 W
EZE ~ P','e, -a. LL -c C? 86u 6 090 0 y0 0 00~~~CQ)$3 0~~~~~~~~~~~~oV ~ iU) ) ECo* 0 c Li~ ~~. E Do20 a0 co 0 CDa,I~~~~~~~~~~~a_ a i a a =CM ';; ~~@1F U. IV r C1
-
7/27/2019 33960864
83/84
-
7/27/2019 33960864
84/84