Process improvement for printingoperations through the DMAIC

Lean Six Sigma approachA case study from Northwest Ohio, USA

Nicholas Roth and Matthew FranchettiMechanical, Industrial and Manufacturing Engineering Department,

The University of Toledo, Toledo, Ohio, USA

Abstract

Purpose – This project aims to address the problems facing a small printing company during theprinting of sample boards. The company was unable to meet their projected yearly demand of 200,000boards. During the study, the team found that the company’s maximum output was 143,400 from twoprinting machines and thus faced the need to acquire new capital. The goal of this project is, therefore, tocreate a sustainable process that increased the client’s competitive advantage in the printing industry.

Design/methodology/approach – By applying Six Sigma and Lean principles, the team identifiedthe current situation that the printing company’s operations were in as well as determining themaximum possible output. The team identified methods to increase production output while findingthe optimum annual labor costs per unit for possible future situations.

Findings – Approximately 30 percent of the workers’ activities were found to be non-value added andthere are numerous machine delays that decreased productivity. The project also investigated theoptimal number of employees that are needed to staff additional machines, should the company wish toexpand its operations.

Originality/value – This project is unique in that it looked at a printing process with a fixed cycletime. This project is useful for any company that is restrained by cycle time and needs to find the mostcost-efficient way to expand their resources.

Keywords Process management, Six sigma, Kanban, Printing industry, United States of America

Paper type Research paper

1. IntroductionThe company being studied was a small printing and visual arts company based out ofToledo, Ohio with 120 employees and annual sales of USD$1.5 million. They have a longhistory of innovation in their industry and hope to continue this in the future. An issue thatthis company was facing was the rising costs associated with acquiring the capitalrequired to remain competitive in the printing business. They recently introduced alldigital equipment into their facility.

In the last few years, the company has entered the business of printing sample boardsfor shingles, wood samples, brick samples, and similar products. These sample boardshave a quality that could not have been achieved without current scanning and printingtechnology. The company is looking to create a market for these boards to reduce theircustomers’ need for actual samples. Printing these boards is currently cheaper and moreenvironmentally friendly than using actual samples for the boards. As with any newproduct that has the potential to save companies money, the market is growing rapidly

The current issue and full text archive of this journal is available at

www.emeraldinsight.com/2040-4166.htm

Processimprovement

119

International Journal of Lean SixSigma

Vol. 1 No. 2, 2010pp. 119-133

q Emerald Group Publishing Limited2040-4166

DOI 10.1108/20401461011049502

and the company currently does not have the capacity to handle the projected demand forthis year.

The problem with production was that the existing process for manufacturing sampleboards was not reaching its efficiency potential. The principles of Lean and Six Sigma areuseful to the manufacturing process in order to maximize the production output whilereducing the costs associated with defective parts (Henderson and Evans, 2000).Additional benefits include increased utilization of equipment and labor, improved safety,maximization of floor space utilization, and reduced inventory. These benefits will allowthe client to increase its competitive advantage by allowing them to reach their projectedproduction volume with minimal capital expenditure.

This project follows the five step methodology used in the Six Sigma process.Six Sigma ideology states that any project should be solved by using DMAIC, whichstands for define, measure, analyze, improve, and control (Pande et al., 2000). The definestep is outlined in Section 2 – about the case study. In Section 2, the problem is identifiedand specific goals are determined. The measure step is the step that focuses mainly ongathering raw data from the process. This is described in detail in Section 3.1 – measure.The third step, analyzing the data will be shown in Section 3.2 – analyze. This isa breakdown of what the gathered data means for the company. In Section 4 –improvement opportunities, the suggestions for the company are explained in moredetail. Improvement opportunities give possible ways to improve printing process andfulfill the fourth step in the Six Sigma process. Finally, the methods for sustaining thechanges (Step 5 – control) are discussed in this paper.

1.1 Project goalThe goal of this project is to reduce process waste and increase production capacity to meetprojected market demand for the new sample board product line. Six Sigma and Leanpractices will ensure that the changes will be made effectively and will be sustainable.

2. About the case studyCurrently, there are two digital printing machines that are used to print the sample boards.These machines have the maximum capacity of about 20 boards per hour. Assuming threeshifts per day and five days per week, the most boards the company could produce peryear is about 75 percent of their projected yearly demand. These calculations do not allowfor any unexpected delays or take into account the scrap rates. In addition, the companytakes on other unrelated projects that utilize the printers on which the sample boards areprinted. Various studies have been performed to determine the true capacity. Thesestudies show that they will be unable to handle the forecasted demand.

In order to reach the maximum production capacity of the system, the major sources ofwaste need to be addressed. One of these issues is the quality of the incoming material usedto make the sample boards. These blank white boards are shipped into the facilities wherethey are stored in inventory. These boards are already pre-cut and are ready to be printedon when they arrive in the facility. While the cost associated with inferior raw material isnot a substantial issue (material with defects is simply not used), the disruptions caused bysuch quality issues are reducing the capacity of the process. The operator has to take timeto inspect the boards to verify there are no scratches, discolorations or warping. While he isperforming this inspection and moving the defective parts to a scrap pile, the printingmachine is standing idle.

IJLSS1,2

120

Another concern in the current process is the inconsistent nature of the printingmachines. Delay studies and feedback from the machine operator indicate a variety ofreasons the printer would be offline. Some of these issues include a blown fuse on thecomputer that controls the printers, inconsistencies with the print color and qualitybetween machines, downtime while the ink is being replaced, and general softwareissues that make it necessary to restart the computer. Although correcting the printerissues will greatly improve the overall productivity of the process, it was found that theonly preventable source of significant downtime comes from replacing the ink.

A third area of improvement would come from the work design of the process.Currently, there are many times when the operator is waiting for the printers to finish.Also, there are times when the printer is sitting idle and the operator is still performingsetups. One of the ways to adjust this is to design a specific set of instructions thatoptimizes the machine utilization without sacrificing operator effectiveness.

3. Application of the Six Sigma define, measure, analyze, improve, andcontrol methodologyThe methodology used in this project has been broken into two categories, the datacollection and the data analysis. In order to gain a better understanding of the printingprocess at the company, considerable data were collected using various data collectiontechniques such as delay studies, facility measurements, work sampling studies, andtime studies. The data were collected, analyzed, and used to create man machine charts,standardized work plans, and plant layouts.

3.1 Measure3.1.1 Delay study. A delay study is a study of the delays associated with the production ofa product. In this case, it is the delays associated with the printing of the sample boardsbeing produced by the company. Some of the delays that were initially predicted to takeplace at the company include: the time in between printings, the time necessary for data totransfer, the time lost due to machine breakdowns, and the time wasted because theoperator is working with another machine or performing another operation associatedwith a process.

Current delay studies have shown that there are a few common delays that can occur.The first and most common is the time taken to restart the machine in between printingcycles. Although this is usually a very short amount of time (1-3 minutes), the wasted timeadds up as the process continues over the course of the day. Fortunately, as long as theoperator is only operating only one machine, this issue can easily be resolved by creating astandard operating procedure (SOP). Otherwise, the use of different SOPs will becomenecessary to determine the number of operators necessary to operate a given number ofmachines efficiently.

Delays in the process, other than time in between print cycles, is either the result ofmachine break downs, maintenance, quality assurance, or issues involving the supply chain.Issues were encountered associated with a machine breakdown while trying to performdelay and process studies. At one point, a component of the machine had broken, and thecompany was not able to replace it for over a week. Once they had finally replaced it, they didnot have enough raw material in their inventory to continue working and were forced to waituntil the next day to receive more sample boards. Another issue that has been noticed at thecompany is the quality checking and covering of the printer before every shift. Neither of

Processimprovement

121

these time consuming actions add any value to the part, and are an example of waste. Byscaling back these two actions (assuming the quality could still be ensured), the companycould save up to an hour of wasted machine and employee time daily.

3.1.2 Process diagram. Two types of flow associated with the printing process ofsample boards have been identified. One is the flow of information and the other is thematerial flow. The combination of both gives the general flow of the printing process.Although several areas of the information flow can be addressed for optimization themain focus will be the material flow.

The material flow describes the passage and transformation of raw material (blanksample boards) into finished product through the facility. The boards arrive stacked on apallet from the warehouse and are stored in the production area, close to the printer. Rawmaterial inspection takes place immediately before placing the board on the printerfollowed by surface cleaning. After printing takes place on the first surface, the boards areturned over; the other surface is cleaned, and then printed. The process is finalized withbar-code scanning of the boards, final inspection, packaging, and depositing of the packedmaterial in the shipping area (Figure 1).

Figure 1.Sample-board processdiagram

Order comes infrom IT

departmentLabel is printed File is sent to printer

Sent to next computerand converted to PDF file

Next 4 boardsare cleaned

Printing of first4 boards begins

First 4 boardsare cleaned

8 boardsplaced on

printer

1st board are done,removed and replaced

2nd 4 boards are donebeing printed, flipped

and cleaned

Printing on 2nd 4boards continues

Reverse side of 1st

boards are printed1st 4 boards are flipped

over and cleaned

Printing on 2nd 4boards begins

Printing on first fourboards is finished

Printing begins on 2nd

boards

2nd 4 boards are done,replaced and removed

Removed boards are boxedand sent to shipping

IJLSS1,2

122

3.1.3 Time study. Time studies have been performed on both first and second shiftoperators, and the data obtained have been used to identify and simulate different situations.The data collected are shown below. The data in the following table show the time inbetween Steps 1 and 2, Steps 2 and 3, etc. Table I is data collected from the first shift operator,Table I is data collected from the second shift operator, and Table I represents data collectedfrom Operators 1 and 2 and averaged together. All data are presented in minutes.

The time in between Steps 1 and 2, 2 and 3, 4 and 5, and 5 and 6 basically consist ofmachine run time only. For these steps, the operator has little effect on the speed of theprocess.

However, the time in between the Steps 3 and 4 and Steps 6 and 1 involved the operatorrestarting the machine. Therefore, there is more room for variation. Although littlevariation is seen in-between Steps 3 and 4, it can be observed by looking at the standarddeviations of the times between Steps 6 and 1. The larger deviation in the time betweenCycles 6 and 1 is due to the machine cycle ending and the operator sometimes packagingthe material instead of immediately restarting the machine for the next part, andsometimes immediately restarting the machine before packaging the boards.

3.1.4 Work sampling study. A work sampling study was performed on the operator toget a better understanding of the distribution of tasks that he performs. This was done byrecording what action he was performing once a minute, on the minute. The duration of thestudy was just over two hours, giving 121 data points. Results and analysis can be seen inTable II and Figure 2, respectively. These data were then analyzed and Pareto diagramswere created. One of the methods of analysis was to determine which activities were valueadded and which ones were non-value added. This was useful in finding improvementopportunities to reduce wasted time.

Analysis of the results showed that more than 30 percent of the operator’s time is spentperforming non-value added tasks. The two major activities that contributed to this werewalking to or from the printer and waiting for the printer to finish a batch. Most of the timehe spent walking was from the computer to the printer, but there was also a significantamount of time spent walking to pick up raw materials. In addition, it can be seen that theoperator spent 19 percent of his time cleaning the blank sample boards. Improvements inthis process would allow for more time to perform other tasks, such as running a secondmachine (Figures 2-4).

3.2 Analyze3.2.1 Layout evaluation and material storage and material handling. The main focus ofthis aspect of the study was the minimization of inventory and non-value added work

1 to 2 2 to 3 3 to 4 4 to 5 5 to 6 6 to 1 Process time

Operator 1AVG 4.95 5.66 0.93 5.09 5.59 2.29 24.18SD 0.17 0.08 0.28 0.05 0.07 1.66 2.07Operator 2AVG 5.29 5.36 1.39 5.68 5.27 3.37 26.08SD 0.37 0.48 0.45 0.81 0.46 2.99 3.43Operators 1 and 2AVG 5.12 5.51 1.16 5.38 5.43 2.83 25.13SD 0.27 0.28 0.36 0.43 0.26 2.32 2.75

Table I.Time study data

Processimprovement

123

such as transportation and storage. For the redesign of the layout, the appropriatenumber of machines and operators resulting from the time study analyses wereconsidered. Accurate measurements of the workspace were taken and a digital drawingof the current workspace was created. The measurements were then used to create floorplans that will allow three or four machines to operate in the current work space and willminimize the amount of necessary operator movement between machines, raw material,and the packaging area.

Activity Count Cumulative Value time NV time Percentage

Working at the computer 30 30 30 0 24.79Cleaning boards 23 53 23 0 19.01Walking 16 69 0 16 13.22Waiting 9 78 0 9 7.44Set down blanks 8 86 8 0 6.61Transporting blanks 6 92 0 6 4.96Flipping boards over 5 97 5 0 4.13Preparing next order 5 102 5 0 4.13Scanning barcodes 4 106 4 0 3.31Talking to co-workers 4 110 0 4 3.31Packaging 3 113 3 0 2.48Pick up completed parts 3 116 3 0 2.48Quality check of blanks 2 118 2 0 1.65Recording orders 1 119 1 0 0.83Break 1 120 0 1 0.83Reload spray bottle 1 121 0 1 0.83Totals 121 84 37

69.42% 30.58%

Notes: Work sample study; February 29, 2008 – First shift, morning

Table II.Work samplingstudy results

Figure 2.Work samplingsurvey results

35

30

25

20

15

10

5

0

142

120

100

80

60

40

20

0

Count

Cumulative

Wor

king

at t

he c

ompu

ter

Cle

anin

g bo

ards

Wal

king

Wai

ting

Set d

own

blan

ks

Tra

nspo

rtin

g bl

anks

Flip

ping

boa

rds

over

Prep

arin

g ne

xt o

rder

Scan

ning

bar

code

s

Talk

ing

to c

o-w

orke

rs

Pack

agin

g

Pick

up

com

plet

ed p

arts

Qua

lity

chec

k of

bla

nks

Rec

ordi

ng o

rder

s

Bre

ak

Rel

oad

spra

y bo

ttle

IJLSS1,2

124

3.2.2 Man machine chart. The creation of a man machine chart will be essential in theorganization of the manufacturing process. A man machine chart will highlight thetasks that need to be performed by the machine operator during the process, and the timethat those processes will take. That data will then be organized into a set of sequentialtasks that that outline the work necessary for the operator to perform in order toaccomplish all the necessary jobs in the optimal way.

By outlining the work necessary for the operator to perform, the work and the order inwhich it should be performed can be standardized. By “scheduling” the work that theoperator performs during the process, short breaks in the operator’s process can be found.In order to prevent this time from simply being lost, work can be assigned to the operatorto complete during his short breaks. In some cases, there are such large breaks in theprocess, the operator can run a second machine, or two operators can run three machines.

Figure 3.Value added

operator activities

35

30

25

20

15

10

5

0

9080

6070

4050

2010

30

0

Wor

king a

t the

compu

ter

Cleanin

g boa

rds

Set do

wn blan

ks

Flippin

g boa

rds o

ver

Prepari

ng ne

xt or

der

Scann

ing ba

rcode

s

Pick up

comple

ted pa

rts

Packa

ging

Qualit

y che

ck of

blan

ks

Recor

ding o

rders

Value added activites

Figure 4.Non-value added

operator activities

18

16

14

12

10

8

4

2

6

0

35

40

30

25

20

15

10

5

0

Walk

ing

Wait

ing

Transp

ortin

g blan

ks

Talking

to co

-wor

kers

Break

Reload

spray

bottl

e

Non-value added activities

Processimprovement

125

3.2.3 Standardized work. From the data collected, a standardized work plan can becreated to control how materials move through the process. Machining time, movement ofraw material, facility layout, and operator actions are all factored into the creation of thestandardized work plan. The machining time is an essential part of determining processtime. Because of the importance it plays in the process, other factors have been created andmodified around it. The facility layout has been designed to minimize the movementnecessary for the operator. By reducing the movement of the operator between themachines, raw materials, the computer, and the packaging area, the amount of timewasted in the transport of material will be minimized. Operator actions will also be takeninto account in determining a standardized work plan because the operator plays a crucialrole in how his or her workstation runs. The order in which he performs the necessarytasks in completing the process plays a crucial role in the efficiency of the process.

4. Improvement opportunities4.1 Production configurationOne of the main purposes of the project is to provide guidance for capital investment. Asdemand is expected to rise in the future, the company will need to invest in additionalequipment and labor in order to satisfy demand. With the current configuration, thecompany has an ideal output of 143,400 sample-boards per year based on the followingcalculations:

ðHr £ Sl £ S £ DÞ £ R £M ¼ Maximum Annual Output

½15 £ 7:5 £ 3 £ 250� £ 0:85 £ 2 ø 143; 400 pcs=year

where:

Hr ¼ maximum machine hourly production rate.

Sl ¼ shift length, in hours.

S ¼ number of shifts per day.

D ¼ number of days worked in a year.

R ¼ Reliability.

M ¼ number of machines used in the process.

These calculations represent an ideal maximum output based on equipment cycle timeand reliability. In the near future the company expects order levels to exceed 200,000sample-boards per year and thus will need to invest in additional equipment and labor.Table III presents different production configurations and their maximum annualoutput. In making the recommendations, the main emphasis was put on maximizingmachine utilization and labor utilization. Labor costs for the recommendedconfigurations were compared with the current costs in Table IV.

4.2 Plant layoutThe current production layout makes use of two printing machines. In the future, thecustomer will be required to invest in additional equipment according to the increase indemand. As new equipment will be brought in, the opportunity to positively affectproductivity through layout redesign will arise.

IJLSS1,2

126

Max

imu

man

nu

alou

tpu

tN

um

ber

ofm

ach

ines

Lab

orco

nfi

gu

rati

onT

ask

dis

trib

uti

onT

ime

nee

ded

per

cycl

e(s

econ

ds)

Lab

oru

tili

zati

ona

(%)

Mac

hin

eu

tili

zati

ona

(%)

Mac

hin

ere

liab

ilit

ya

(%)

75,4

001

1O

per

ator

All

task

sto

be

don

eb

yth

eop

erat

or1,

610

100

89.4

85

150,

800

22

Op

erat

ors

All

task

sto

be

don

eb

yth

eop

erat

ors

1,61

010

089

.485

160,

300

1O

per

ator

–1

hel

per

Op

erat

orb

Hel

per

:re

mai

nin

gta

sks

1,44

02,

200

100

100

9585

226,

300

33

Op

erat

ors

All

task

sto

be

don

eb

yth

eop

erat

ors

1,61

010

089

.485

253,

100

1O

per

ator

–3

hel

per

sO

per

ator

c

Hel

per

:re

mai

nin

gta

sks

900

1,31

062

.590

.910

085

301,

700

44

Op

erat

ors

All

task

sto

be

don

eb

yth

eop

erat

ors

1,61

010

089

.485

337,

500

1O

per

ator

–4

hel

per

sO

per

ator

c

Hel

per

:re

mai

nin

gta

sks

1,20

01,

310

83.3

90.9

100

85

320,

600

2O

per

ator

s–

2h

elp

ers

Op

erat

orb

Hel

per

:re

mai

nin

gta

sks

1,44

02,

200

100

100

9585

377,

100

55

Op

erat

ors

All

task

sto

be

don

eb

yth

eop

erat

ors

1,61

010

089

.485

421,

800

1O

per

ator

–5

hel

per

sO

per

ator

c

Hel

per

:re

mai

nin

gta

sks

1,50

01,

310

100 90

.910

085

Notes:

Tas

kti

mes

are

exp

ecte

dto

dec

reas

ed

ue

tow

ork

spec

iali

zati

onan

del

imin

atio

nof

non

-val

ue

add

edac

tiv

itie

s;ab

ased

onob

serv

edp

erfo

rman

ces;

bal

lin

spec

tion

s,ra

wm

ater

ial

insp

ecti

ons,

load

dat

a,st

art

mac

hin

ecy

cle,

pri

nt

lab

els;

c load

dat

a,p

rin

tla

bel

s(3

00se

con

ds/

mac

hin

e)

Table III.Labor and machine

utilization for differentproduction

configurations

Processimprovement

127

Three machines production layout. If three machines are required to satisfy demandthan the layout in Figure 6 is proposed. This layout was chosen based on the workconfiguration of three operators and zero helpers and the space available. Threeoperators results in the cheapest price per unit for three machines. Fortunately, thestandardized work design will not change for the three operators. The operators wouldhave to offset the starts to the printing to have the computer free. This proposed layoutrequires no modifications of the production space, assures proper safety distances exist,and the flow of materials is kept to a minimum.

Four machines production layout. If four machines are required to satisfy demand,the layout in Figure 7 is proposed. This layout requires some modifications tothe production area. This layout also takes into consideration safety and efficiency.The optimal way to run this process with four machines is with two operators and twohelpers (Figure 5-7).

4.3 Man-machine chartFrom the data, it was found that the average operator only works for about 13-14minutes out of the 21-22 minutes that it takes for the printer to complete the process. Thismeans that if the operator was operating two machines at once, he would be busy for26-28 minutes out of the 22 minutes that the machine runs, and the machine would beforced to sit idle for six to eight minutes. In fact, there is just enough time for the operatorto complete all of the processes associated with the machine, but there is no machine timeleft over for the operator to transfer the data from the computer to the machine and themachine sits idle.

Number of machines Labor configuration CommentsAnnual labor cost

($)aLabor cost per unit

($/unit)b

1 1 Operator 135,000 1.7902 2 Operators 270,000 1.790

1 Operator – 1 helper Best solution 243,000 1.5163 3 Operators Best solution 405,000 1.790

2 Operators – 1 helper –c

1 Operator – 2 helpers –c

1 Operator – 3 helpers 459,000 1.8144 4 Operators 540,000 1.790

3 Operators – 1 helper –c

2 Operators – 2 helpers Best solution 486,000 1.4401 Operator – 3 helpers –c

1 Operator – 4 helpers 567,000 1.7695 5 Operators 675,000 1.790

4 Operators – 1 helper –c

3 Operators – 2 helpers –c

2 Operators – 3 helpers –c

1 Operator – 4 helpers –c

1 Operator – 5 helpers Best solution 675,000 1.600

Notes: aBased on an annual salary of $45,000 for an operator and $36,000 for a helper; three shifts perday; bfound by dividing the annual labor cost by the maximum annual number of units (from Table III);ctasks allocation and work equity makes this configuration not feasible

Table IV.Labor cost for differentproductionconfigurations

IJLSS1,2

128

Assuming that there is sufficient demand for the sample boards, it may be necessary torun more than two machines. The company’s current capacity was calculated based onthe assumption that two machines run full time. The company should fall shortof producing their projected forecast of 200,000 boards annually. Additionally, this isassuming that both machines can be dedicated to producing only sample boards, whichis not the case. If the company’s demand does meet its projected forecast, it will have to

Figure 5.Current layout

configuration

Section 1Section 2

Figure 6.Proposed layout – three

machines

Section 1

Sect

ion

2Sect

ion

3

Processimprovement

129

invest in a third machine in order to dedicate production on two machines entirely tosample boards.

While running two machines, the management will have to decide whether time is theirmost important constraint of if money will be their critical constraint. If they decide toconserve money, and use one operator, they will be sacrificing machine time because themachines will be forced to sit idle while the operator performs other functions. If theydecide to operate both machines with two operators, the operators will sit idle while themachine prints. So, the number of operators on each machine will be up to the discretion ofthe management at the company.

If the company decides to run three machines, it is possible to run all three machineswith only two operators. Two operators can successfully operate three machines howeverthere will be machine downtime and little operator downtime. Three machines runningcontinuously are not necessary to maintain production of only 200,000 sample boardsannually. However, if there is a surge in demand, or if the demand grows, it is possible torun three machines successfully with only two operators.

4.4 Inspection processCurrently, the sample boards are inspected twice during the printing operations. Theraw materials are checked visually before they are placed on the printing machine.These white boards are inspected not only for surface finish flaws, such as scratches,chips and flakes, but also for the coloring of the board. If there are any colorimperfections, the printed color of the sample image will not show up accurately, as thecomputer is balanced to print on pure white boards. Another imperfection thatthe operator is examining the board for is any warping or bending of the sample board.If the board does not sit flat on the printing table, the suction used to hold the board inplace will not hold it effectively. If the boards have edges that are bent up, there could be

Figure 7.Proposed layout – fourmachines

Section 1

Section 2

Section 3

Sectio

n 4

IJLSS1,2

130

inadequate clearance between the boards and the printer head, causing discoloration,smudging, and potential damage to the printer.

The other inspection process comes at the end of the operation. Because the companyis competing mainly on innovation and quality of material, the sample boards must fullymimic the appearance of the sample. Any problems with the printer head will cause inkblotches, so routine maintenance needs to be performed on the printer. In addition, at thebeginning of every shift, the operator will run a quality check to make sure that the colorsare balanced correctly.

Unfortunately, because quality is such an important aspect of this process and it isextremely difficult and expensive to create a gauge that can measure the surface qualityof this type of product, the human eye is really the only viable option for maintaining thequality of the product in a cost effective manner. Addressing the issues associated withthe warping of the raw material (i.e. causes, suppliers, storage, etc.) is beyond the scope ofthis project and unable to be addressed at this time.

4.5 Implementation of E-kanban systemsAn important determinant of the success of “push” production scheduling is the qualityof the demand forecast which provides the “push”. Kanban, by contrast, is part of a pullsystem that determines the supply, or production, according to the actual demand of thecustomers. In contexts where supply time is lengthy and demand is difficult to forecast,the best one can do is to respond quickly to observed demand. This is exactly whata kanban system can help: it is used as a demand signal which immediately propagatesthrough the supply chain. This can be used to ensure that intermediate stocks held in thesupply chain are better managed and are usually smaller. Depending on the number ofmachines used in the process, different reorder points will exist for printing supplies.

E-kanban will make use of self adhesive labels printed with a barcode representingthe stocked item. Labels will be attached to the received supplies before storage in theproduction area. After the materials have been taken from the storage area the kanbanlabel will be removed by the operator and handed to the stock-keeper which will scan itinto the computer data-base. E-kanban has the advantages of being simpler and morereliable; the stock-keep will not have to enter the values manually but scan them with abarcode scanner. The short amount of time necessary for this operation will also ensurepromptness in which stocks are handled. Table V gives levels at which the stock shouldbe held at to optimize the E-kanban system.

2 3 4Number of machines Point Quantity Point Quantity Point Quantity

Alcohol (gallons) 1 2 1 2 1 2Ink: C, M, Y (each) 2 2 2 2 2 2Ink: black 2 2 2 2 2 2Rags (lb) 3 10 4 10 5 10UV light bulbs 2 2 2 2 2 2Label stickers (boxes) 1 3 1 3 1 3

Table V.Printing supplies

reorder point

Processimprovement

131

5. ControlIn order to ensure that the proposed methods of improvement are sustained, thecompany will need to implement a set of control systems. These control methods willalso aid the company in being more responsive to process variations in the future andwill be better equipped to handle unexpected deviations. There are several tools thatwould be of use to the company given their circumstances.

One useful tool would be a SOP for the printer operators. As shown in this paper, theoperators do not necessarily use the same methods to print the samples boards, whichcreated costly delays in the printing process. A SOP consisting of very clear operationswill be useful for reducing or eliminating the variations in operator procedure.

Another tool that is useful for controlling the system is a check sheet. Every time thereis an unplanned delay in the process, the operator should mark the delay with the timeand date. This can be useful in planning for the delays. For example, the operationsmanager at the facility might notice a trend that the ink jet is more likely to misfiretowards the end of the life of the ink cartridge. This would be useful in anticipatingprinter misfires.

6. ConclusionThe overall goal of this project was to make the sample board printing process moreefficient in order to increase the company’s competitive advantage. It has been shown thatthe proper application and implementation of Lean and Six Sigma techniques can be usedto create a better process that is more cost effective and can meet the demands of thecustomers (Park, 2002). After defining the problem, it was clear which performancemeasures needed to be studied. From this data, improvement opportunities were analyzedand action plans were created to help implement those plans. Finally, a standardizedmethod was developed in order to ensure that the process would be sustainable. If thecompany is able to successfully carry out the proposed suggestions, they will be able tofind the optimal number of employees to minimize the labor costs per unit produced. Inaddition, the proposed improvement methods will decrease the number of productiondefects and improve the overall quality of the finished goods.

References

Henderson, K.M. and Evans, J.R. (2000), “Successful implementation of Six Sigma:benchmarking: general electric company”, Benchmarking: An International Journal,Vol. 7 No. 4, pp. 260-81.

Pande, P.S., Neuman, R.P. and Cavanagh, R.R. (2000), The Six Sigma Way: How GE, Motorola,and Other Top Companies are Honing Their Performance, McGraw-Hill, New York, NY.

Park, S.H. (2002), “Six Sigma for productivity improvement: Korean business corporations”,Productivity Journal, Vol. 43 No. 2, pp. 173-83.

Further reading

Breyfogle, F.W., Cupello, J.M. and Meadows, B. (2000), Managing Six Sigma: A Practical Guideto Understanding, Assessing, and Implementing the Strategy That Yields Bottom-LineSuccess, Wiley-Interscience, New York, NY.

Graham, R.C. (1993), Data Analysis for the Chemical Sciences: A Guide to Statistical Techniques,VCH Publishers, New York, NY.

IJLSS1,2

132

Kwak, Y.H. and Frank, T.A. (2006), “Benefits, obstacles, and future of Six Sigma approach”,Technovation, Vol. 26, pp. 708-15.

McAdam, R. and Lafferty, B. (2004), “A multilevel case study critique of Six Sigma: statisticalcontrol or strategic change?”, International Journal of Operations & ProductionManagement, Vol. 24 No. 5, pp. 530-49.

Tennant, G. (2001), Six Sigma: SPC and TQM in Manufacturing and Services, Gower, Vermont.

Yang, K. (2005), “Design for Six Sigma and value creation”, International Journal of Six Sigmaand Competitive Advantage, Vol. 1 No. 4, pp. 355-68.

About the authorsNicholas Roth is a Project Manager for the Environmentally Conscious Design and ManufacturingLaboratory at The University of Toledo. He earned his BS in Industrial Engineering in 2007 andMS in Industrial Engineering in 2010. Mr Roth has also worked as an engineer for HamiltonSundstrand.

Matthew Franchetti is an Assistant Professor of Mechanical, Industrial and ManufacturingEngineering and the Director of Undergraduate Studies of the Mechanical and IndustrialEngineering Programs at The University of Toledo. He also is the Director of the EnvironmentallyConscious Design and Manufacturing Laboratory and Principle Investigator of the BusinessWaste Reduction Assistance Program, a joint effort with the Lucas County Solid WasteManagement District. Additionally, Dr Franchetti is working as the Principal Investigator withthe Health Science Campus at The University of Toledo to improve the business processes for renalimplant patients by applying six-sigma concepts. Dr Franchetti received his PhD in 2003 and MBAin 2000 from The University of Toledo. He has worked as an Industrial Engineer and TechnicalManager for the US Postal Service in Washington, DC, Pittsburgh, PA, and Columbus, OH and hasconducted research at Daimler Chrysler, General Motors, and Ford before joining the MIMEDepartment in the Fall of 2007. Dr Franchetti is Certified Six-Sigma Black Belt from the AmericanSociety of Quality and has consulting and research experience with over 25 companies across thecountry. Matthew Franchetti is the corresponding author and can be contacted at: matthew.franchetti@utoledo.edu

Processimprovement

133

To purchase reprints of this article please e-mail: reprints@emeraldinsight.comOr visit our web site for further details: www.emeraldinsight.com/reprints