Improvement of Contact Lens Field Complaint Rate Using the DMAIC Approach

Emil A. Pietri
Manufacturing Engineering
José A. Pesante, Ph.D.
Industrial Engineering Department
Polytechnic University of Puerto Rico

Abstract  Contact lenses are susceptible to a SITUATION SUMMARY

defect call “tearing”. This is common but seldom
• The complaints are spread over the entire
condition detected in the field by the customer. The
spectrum of “Proclear” TM products.
manufacturing process of a contact lens is a very
Occasionally, field complaints are mentioned
precise method of combining an injection molding
regarding similar failures of 2 or 3 lenses in the
processes with a chemically bonded monomer. The
same pack of 6 lenses (see Figure 1), for
liquid monomer is automatically placed between
assembly configuration reference of a lens.
two half’s of plastic injected molds then forming the
• The complaints have been particularly difficult
lens ,subsequent process of curing ,de-molding ,
to investigate due to lack of consistent
hydration and packaging would form the final
product for sale. The purpose of this study is to complaint documentation and few returned
samples of damaged lenses. The number of
demonstrate that the application of the 6 sigma
formal complaints has been historically lower
DMAIC approach to problem solving would
than what would be expected given the level of
provide a methodical and enriched conclusion to
the root cause determination of contact lens this problem as conveyed by our sales force.
“tearing” thus reducing the defect rate, • The investigation focused specifically on
subsequently customer complaints. potential causes for the incidence of “tearing”
(See Figure 2) in “Proclear” lenses. Using the
Key Terms  Define, Measure, Analyze,
6 Sigma approach to problem solving.
Control.

MOLD ASSEMBLY

MALE MOLD

FEMALE MOLD

LARGE FLASH RING

LENS SMALL FLASH RING

14

Figure 2
Figure 1 Contact Lens Tear
Contact Lens Assembly
 Key “Proclear” Material Findings Element Description Team Charter
Process: The process in which the All Proclear Omafilcon contact Lens manufacturing at Norfolk
opportunity exists. Plants.
• There is no evidence that product aging is a
Problem Describe the problem that High Noise-Complaints are being received from sales force
possible cause for the torn and split defects Description: needs to be solved, or the in that contact lenses are tearing- before use and in use.
opportunity to be
seen on “Proclear” products in the market. addressed.
Objective: What improvement is • Standardize with automation De-molding operation.
• The study confirmed that “Omafilcon targeted? • Material improvement with in monomer tint. Reduce
handling.
Proclear” material is significantly weaker than
4. Metrics: What are the Name of Baseline Goal Entitlement* Units of
the material used as control for the study Metric Measure
measurements that quantify
(Ocufilcon). Previous studies going back as far program progress and CPM’s 100 25 3.4 to 50 CPM’s
success?
as 2001 had also shown “Proclear” Material to *What is the best the
process is expected to
be weaker when compared with other CVI produce?
5. Business Results: What is the improvement in Cost Cost WIP/ Cash Labor Inc.
materials as well as competitor lenses. business performance? Reduction Avoidance Inventory Flow Savings Sales
Please list any other Reduction
• Tensile strength analysis indicates a difference improvements on a X
between the material formulation utilized for separate sheet as needed.
Program Scope: Which parts of our business Included Excluded
most “Proclear” products and the slightly processes will be Proclear Omafilcon lens
considered? Which Product.
different material formulation utilized for customer segments, Norfolk Facility
organizations, geographies,
“Proclear” Daily Disposable lenses, with the and timeframe?
Team Members: Names and roles of team Emil Pietri Project Leader
Daily Disposable material being stronger. members Brian Charlton
Dan Earnhardt - Engineering
• Tear test show that lens strength is significantly Joe Calcagno - Manufacturing
affected by induced splits to the material. The Klaus Hummel - Manufacturing
Tom Barrett - Engineering
force required to break a good strip of a lens
Benefit to External Who are the final External customer the user will experience a improvement in
versus a strip with a 2mm split is on average 7 Customers: customers, what are their lens tearing by the beginning of March 09.
most critical
times higher (average for materials tested). requirements/measurement
s, and what benefits do we
expect to deliver to them?
Key Project Dates
PHASE I - DEFINE Schedule: Give the key milestones
and dates. Project Start May 08
Define Complete June 08
The Define phase determines the initiation of a Measure July 08
Complete
project, customer requirements and key process Analyze September 08
Complete
outputs variables. To define this process a “team Improve April 09
charter” is created (See Figure 3 above). Complete
Control Complete May 09

The key elements of the Charter are as follows: Figure 3

• Process: the process in which the opportunity
exist.
• Problem description: describes the problem to
solved.
• Objective: what improvement is targeted.
• Metrics: quantifies the project success.
• Business results: improvement in business
performance.
• Program scope: what part of the business is
considered.
• Team Members: name and roles of the team.
• Benefit to external customers: who are the final
customers or who will benefit from the project.
• Schedule: key milestones and dates.
Project Charter
PHASE II - MEASURE
The measure phase concentrates in four mayor
aspects, understanding the process by developing a
comprehensive process map (See Figure 4), process
input variables and process output variables are
evaluated using a cause and effect matrix, this tool
decides the most important inputs sorted and
included in the failure mode and effect analysis
table (See Figure 5).
Preliminary conclusions and examined results
from the C&E and the FMEA guided the team to
start the data mining process at the de molding
process, cure process, and hydration process.
 After the lens is made it passes thru a cure resources adequately and finalizing plan
oven process and then is de-molded by semi activities.
automatic, automatic or manual machine (The GT • Compare manufacturing processes of sphere
100’s de- molders) these processes introduced product vs. a-sphere product that could explain
extreme variation form machine to machine, shift to differences in outgoing quality results.
shift and perhaps product to product. • Investigate other areas of the manufacturing
The Key Outcomes from the Measure Phase process that might result in subtle surface
were geared to be focused in different areas such defects being induced.
as: • Verify detection capability of current vision
• Brain Storm Section Using the 5 M’s to aided inspection system technology to detect
determine areas where to focus assigning surface splits on dry/wet lenses.

A
C D E F
Raw materials B Mixing of
Incoming
specification inspection Storage of monomers Storage of
Transport materials
and suppliers materials monomers
Mix and filter
HEMA
Storage conditions PC HEMA
Polypropylene resin Delivery and receipt In coming inspection monomers EGDMA
Monomers polypropylene AIBN

G H I J K L
Set up Mold Male &
molding Female Molds Convey Molds Mold Fill Mold with Load into
machine Cast Mold with cooling Monomer & Oven
Qualified Tools Assemble Male &
Female Molds
Polypropylene supply
Tool qualification Temperature & Pressure Molds cool and shrink Storage of monomer Load Robot
Barrel temperatures Timing Hamilton Pump Dispenser Chain Index
Preventive maintenance Cooling Assembly Rate Timing
Cooling water to mold plates Wetting & Capping
Mold fit

M N O P Q R

Mold Audit Decide De-Mold &

Pass Store Transfer to De-Mold
Curing Inspection Disposition Stack
molds
of Lot

Time Separate Male & Female

Temperature 5 Lenses / 2 hr / cavity Finished Molds Crimp Male & Remove Lens
Power, CT, Surface Defects Fail - scrap lot held until De-Ionize & Stack in hydration
required trays

U X
S T V W Dose Blister Tubs,
De-Stack & 100%
Transfer to Hydration Hydrate Transfer to Inspection Insert Inspected
Transfer to Visual
lenses Lenses & Place Foil
Inspection Trays Inspection

Branson or tanks Spray, Flip,

temperature and time Stack, Wrap Pick'n'place operation

Y Z BB CC
Heat-Seal AA DD
Check Sterization of
Blister Tubs Transfer to Sterilizer QC Audit
Seal Decide
lenses in of AQL Pass
and foil Width Disposition
blister tubs Quantity of Lot
Time, Temperature
Pressure, Positioning

Fail - scrap
or re-work

EE FF GG HH
Label, II
QC Audit Decide
Store Pass Transfer to
Box & of labeling Disposition
Package records Distribution centers
of Lot

Figure 4
Process Map
 The team selected what key process input analyzing data from the RPM’s higher than 90.
variables were more important in respect to the The result of the FMEA analysis is summarized in
customer requirements in this case the Key (Figure 6 below) the “Pareto Chart”.
processes outputs variables KPOV’s and input them
into the FMEA (See Figure 5). This is the First
Part of the FMEA-selection to start working and

S O D R
Potential Failure Potential Failure
Process Step Key Process Input E Potential Causes C Current Controls E P
Mode Effects
V C T N
What is the

d e te c t c a u s e o r
What is the Key Process Input? In what ways does What is the impact on What causes the Key What are the existing

H o w S e v e re is
th e e ffe c t to th e

H o w w e ll c a n y o u
H o w o fte n d o e s
c u s o tm e r?

cause or FM
process step the Key Input go the Key Output Input to go wrong? controls and procedures
wrong? Variables (Customer (inspection and test) that
Requirements) or prevent eith the cause or
internal requirements? the Failure Mode? Should
include an SOP number.

De-molding Jaw design damage lense design none 800

8 10 10
de-mold - crimpers
De-molding Fit too tight damage lense mold shrinkage none 720
8 9 10
Mold fits
De-molding Fit too loose damage lense mold shrinkage none 720
8 9 10
Mold fits
De-molding Operator handling / damage lense Human variability none 720
pressure 8 9 10
handling in blister transfer - swabs
De-molding Variation of damage lense Incorrect temperatures none 720
temperature 8 9 10
Heating of molds
De-molding Different processes damaged lenses Equipment and human SOP for machine set up 720
8 variability 10 9
GT 100 vs semi-automatic vs manual process
De-molding curved / flat / uneven damage lense deformation of mold in none 720
/ diameter too small 8 pressing 9 10
de-molding press punch profile design
Figure 5
FMEA (Extract)

P a r e to C h a r t o f K e y P r o c e s s In p u t F r o m T h e F M E A
12000
100
10000
80
8000
Percent
RPN

60
6000
4000 40

2000 20

0 0
K e y P ro c e s s I n p u t
fi t
s rs
ab
s n ss ds or
s g te e) e) s) of
)
ss
) y
o ld or
s
re
)
pe si g ce ol at lin ra m us ep ne tr a at tu
ol
d
ri m sw de ro m ar nd ng (ti ho rc ck
ck n m
ar ra
M -c r - f il e a lp of ep - ha c ki o ns a re (f o t
( la
t hi c t io a le ep pe
e u g s i i n n e m s
ol
d
ns
f
pr
o
an t in -
ol
d p it
s/
w t io ta e sp fe - em
ea o ld nd ec ac dg in in 0 (t
-m t ra nc
h m -m co it i
on sp rf (e to 10 s
de er vs H -m de e su s ck T on
st pu ic de ur nd in
t er i n g s tu G d it i
li s t c o in e n
b es a (c g m ic k co
in pr om ge li n ra st re
in
g g ut ra nd pa cu
n dl ld in i- a s to ha n s
o m le
ha -m se ol
d
de vs m
0
10
T
G
RP N 28161584 1224 720 720 720 688 640 448 400 400 224 160 144 120 112 96 64
P e rc e n t 25 14 11 6 6 6 6 6 4 4 4 2 1 1 1 1 1 1
C um % 25 39 50 56 63 69 75 81 85 88 92 94 95 97 98 99 99 100

Figure 6
Pareto Chart from the FMEA
 PHASE III - ANALYSIS There was no evidence of differences in
detection capabilities at quality audits visual
The group started the process of data mining
inspection and sampling. The following graphs
and analysis where conclusion were drawn; some of
suggest that there is difference between shifts in
techniques utilized were Box plots, analysis of
picking and placing lenses very manual operation
variance, and comparison by shift and by de-
damage to lens could happen here.
molding equipment (See Figures 7, 8).
Key observations were:
Several Key findings:
• Handling and picking lenses for inspection is
• Thinner Lens Edge suffers greater vulnerability
causing more “tearing” in different shifts (See
to “lens tear”.
Figure 9).
• A significant interaction exists between
• Detection inspection audit is not different (See
Norfolk Cast Mold Line 2 a-sphere Lens (70-
Figure 10).
micron Lens edge thickness). (See Figure 7).
• Tear Rates and De-Mold GT100 #2. This
0.005
interaction is not exhibited to this degree on
other Lines or Lens Types (See Figure 8). P-Value = 0.000
0.004
The difference IS

A vg Tear Rate found, all Lots

statistically
All Product Tear Rate by Lens Type: Oct1,2007-Jun19, 2008 0.003 significant
0.080
0.075
0.070 0.002
0.065
0.060
0.055
0.001
FLIP Te a r Ra t e

0.050
0.045
0.040
0.035 0.000
# Lots ==> 737 674 721 653
0.030
0.025
A B C D
0.020 FLIP Shift
0.015 P-Value = 0.000
0.010 Figure 9
0.005
Edge Thkns==> 80 80 80 70 100 80 Plot difference in manual pick and place inspection
0.000
A4BL BLST EMFB PCAB PCT1 PMFB
PCCM Lens Type Each Auditor is assumed to have audited the same number of Lots; 245 each
0.005
Figure 7
Box Plot Difference Lens Type vs. Edge Thickness
0.004
P-Value = 0.415
A vg Tear Rate found, all Lots

All Product Tear Rate by DeMold: Oct1,2007-Jun19,2008

0.09 0.003 The difference is
0.08 NOT statistically
0.07 0.002
significant.
0.06
FLIP Te a r Ra t e

0.05
0.001
0.04

0.03
0.000
0.02
P-Value = 0.000 HR JM RBC
0.01
A-Shift Auditor Initials
0.00
A B C D C B C D A B C D A B C D A B C D A B C D A B C D A B C D
A A A A B C C C G1 G1 G1 G1 G2 G2 G2 G 2 G3 G 3 G3 G 3 G4 G4 G4 G4 G5 G5 G5 G5 G 6 G6 G 6 G6 Figure 10
DeMold Equipment/Shift (last letter = shift) Plot no Difference in Auditors
Figure 8
Box Plot Difference in De-molding Equipment
 Further studies revealed that cast mold line 2 in Asphere GT100 #4 Tear Trend vs DeMold Shift: 10/1/07-6/23/08
a combination with de mold equipment GT 100 4 Data are exponentially smoothed by factor of 0.99
line had a significant different in tear rate (See 0.11

0.105
Figure 11). There was also observed the GT 100 0.1

line 4 was higher in tear rates by shift specially 0.095

30% Cas t Mo ld
shift # 4 (See Figure 12 & 13). 0.09 Line 2 GT4A GT4B

F L I P T ea r R a te
0.085 GT4C GT4D
Key observations: 0.08

• Cast mold line 2 had a high “tear” rate when 0.075

compared with de molder # 4 so focus on 0.07

11% Cas t Mo ld
0.065
reducing variation in cast mold vs. de-molder 4 0.06
Line 2

is imperative. 0.055

• De-mold line GT100 # 4 in combination with 0.05

6% Cas t Mo ld
0.045 Line 2 No Line 2 pro duct
shift D. Shift D appears to be manipulating de- 9/25 10/5 10/15 10/ 25 11/ 4 11/ 14 11/24 12/ 4 12/14 12/ 24 1/3 1/ 13 1/23 2/ 2 2/ 12 2/22 3/3 3/ 13 3/23 4/ 2 4/ 12 4/ 22 5/ 2 5/ 12 5/ 22 6/ 1 6/11 6/ 21

mold equipment and set ups. Mold Date

Figure 13
0.26 Box Plot Difference De-molding Equipment vs. Shift
0.24
0.22
0.20
P-Value = 0.000 Thinner Lens Edge suffers greater vulnerability
0.18 to “tear” mechanisms. A significant interaction
FLIP Tear Rate

0.16
0.14
exists between Norfolk Cast Mold Line 2 a-sphere
0.12 Lens (70-micron Lens edge thickness) Tear Rates
0.10
0.08
and De-Mold GT100 #4.
0.06
0.04
0.02
PHASE VI - IMPROVEMENT
0.00
2 4 9 12
In Phase VI several improvements were
Cast Mold Line examined and implemented to reduce the incidence
of In-process lens tears, chip edges. After
Figure 11
determining that the critical to quality causes to be
Box Plot Difference Line 2 & De-molding Equipment
lens material strength, handling post cast molding
& poor inspection detection the team projects were
0.15
0.14
configured concurrently.
0.13 • The system called the “Low Volume
0.12
0.11 Automation” to reduce lens handling by
0.10 P-Value = 0.000
operators and aided with automated visual
FLIP Tear Rate

0.09
0.08 inspection system - cameras.
0.07
0.06
• The introduction of a new monomer initiator
0.05 “Perkadox” strengthen the lens to avoid
0.04
0.03 potential tearing when handling by patient.
0.02
0.01
• The introduction of an air gauge system aided
0.00 in the identification of “mold fit” the mold fit is
G4A G4B G4C G4D
GT4_Shift
an important part of the cast molding process
as it determines the gap between the female
Figure 12
mold and the male mold when together, when
Box Plot Difference De-molding Equipment vs. Shift
fit is not monitored properly and corrected at
 set up the mold too loose may cause excessive It is important to note that the mold fit is
monomer material thus permitting that at the de important to the lens cast molding manufacturing
mold process lenses can be ruptured as it pulls. stage, with this measurement process the technician
can measure exactly what is the circumference of
The Air Gauge System: The nearly-vertical
the mold and make minute adjustments to the
portion of the critical diameter region in the Male
injection molding process this will avoid potential
and Female Molds is that region where seal-off
problems.(See Figure 14).
occurs, to set the trapped Monomer volume and to
form the Lens edge. This region is typically about
80 microns beyond a “transition edge.”
(See Figure 13 & 14).

This fit-region is short (20-250 microns) and is

elliptical in shape an average “Mold-Fit” (Female
Mold fit too loose
Dia. – Male Dia.) is the same, regardless of Mold- May come apart at the filling
station. Mold fit too loose
to-Mold relative orientation, but “At-Gate” and May come apart at the filling
station.

Off-Gate” fits vary considerably, depending on

relative orientation.
Precise measure of both the At-Gate and Off- Mold fit too tight
Male too deep
Gate diameters, to the “micron-level,” is required in Lenticular radius is cut short. Will also cause
fluctuating lens center thickness. ACCEPTABLE
order to control these fits.
Figure 14
Cross Section of Lens with Irregular Mold Fits

The Air Gauge explanation: The mold is

placed on top of the air gage. The critical fit
diameter is positioned exactly at the nozzle exit.
MONOMER Nitrogen is supplied through the gage head.
RETENTION RINGS
Added to help GT100
throughput
Because of the small gap between the mold and the
DETAIL A
gage head, the escape of nitrogen is restricted. The
SLEEVE LAND AREA
POINT AT WHICH THE
INSERTION DEPTH IS SET,
AT THE FILLING STATION
CRITICAL
FIT restriction causes back pressure and this back
DIAMETER

pressure is measured in volts. The voltage

Figure 13 measurement is used to calculate a precise mold
Mold Fit Regions diameter. (See Figure 15).

LENS EDGE
FIT INTERFACE
Also called:
- Vertical land area
- Land length

LENS
FLAT

GAP

SLEEVE
LAND AREA DETAIL B

DETAIL A NITROGEN
FIT AREA
DETAIL B
18
Figure 14
Mold Fit Regions-lens Edge Figure 15
 Cross section of Air Gauge a pre requisite for non extract process, meaning that
The use of the Air gauge provides the the lens will not have to undergo the process of
technician an accurate measurement, this helps the external tint addition and lens hydration, the
early detection of fit too loose or too tight which in formulation tint will be part of the monomer
turn would potentially cause variation in the preparation and will flow with cast molding
process, this variation of fit will yield product process, hydration will occur in blister sealed with
susceptible to tearing or chip edges. As evidenced saline water.
the higher the mold loosens the more propensity of
the product to be tearing at de-molding stations, the
Lens ultimate strength at break
tighter to mold fit the thinner the edge, more
susceptible to tearing as it weakens the area of edge Product using VB6 formulation
Analysis of Means Plot
contact. (See Figure 16). With 95% Decision Limits
(X 10000)
12
UDL=75089.31
10
CL=47464.92
Lines 1, 2, 4, 10, 11, 12 8
LDL=19840.54

Mean
Bartlett's Test
1 -75to-45 6

33 Lots (0.9%) Test Statistic 29.74 4

P-Value 0.000
2
2 -45to-35 57 Lots Levene's Test
A v g M o ld - F it R a n g e ( m ic r o n s )

0
Test Statistic 1.52

PCMT Scottsville

PCM Scottsville

PCTXR Scottsville

PCSPH Hamble
PCM Norfolk

PCMEP Norfolk

PC1D WC

PC1D PR
PCT Norfolk

3 -35to-25
4 -25to-15
5 -15to-05
6 -05to+05
7 +05to+15
8 +15to+33
0.050
Figure 16
Tear & Edge Defect Rate Variance by Mold Fit
Measurement
Key Highlight: The air gauge has the
capability to identify precisely de gap between
female and male molds fit, as observed in the data
graph (Figure 16).
The new developed monomer formulation:
The implementation of a new “Perkadox” Initiator
appears to have more strength (See Figure 17), all
“Proclear” product required FDA submissions to
the 510K as supplements, approval granted in les
than 180 days, sites were FDA inspected after the
submissions. The formulation already used in other
products has more stability when stressed, it is also
PCXC Norfolk
P-Value 0.154
215 Lots
Product type – related to formulation
584 Lots
1173 Lots Figure 17
Brake test results new “perkadox” formulation
1061 Lots The Low Volume Automation System: The
LVA system provides a long range of capabilities
551 Lots
that enhances several issues with the lens
100 Lots performance pertaining to tears and split edges. The
system is design to automatically handle lenses
0.075 0.100 0.125 0.150
after the molding where we believe most defects
occur as discovered in the FMEA de-molding stage.
The system provides two automatic camera
inspections that will replace the use of human eye;
this will be done to 100% of lenses as if done in
regular wet lens inspection by an individual.
In addition, with the use of the new monomer
“perkadox” lenses will now be hydrated in the
blister it does not have to pass thru the tedious
process of bath hydration as exhibited in the
process map, this will also eliminate lens handling
and picking the lens manually. (This is called the
extract process). The system was fully implemented
on September 2008 with exceptional results
improving internal yields pertaining “lens tearing”
and improving-reducing confirmed customer
complaints in the field. Size (mm)
Condition
Key component of the system: 17.5 Description
Actual 10X
X
• (AIS) Automatic inspection system and lens
Any foreign matter that cannot be
handling system. (See Figures 19 & 20). Embedded
0.34 6 3.4 removed which is embedded in
Particles
the lens matrix.
Any particle attached to the lens
Attached to
surface that cannot be removed
surface
by cleaning. Reject if exceeds
Particles 0.11 2 1.14
size or more than three.
Excess lens material protruding
Flash
from the edge.
Enclosed void in the matrix of the
lens. Exception: None in OZ.
Bubbles 0.11 2 1.14
Reject if exceeds size or more
than three.

Figure 19 Split from edge of lens extending

AIS Rotary Table inwards Exception: for ProClear
lenses no edge splits are allowed.
Edge Splits 0.29 5 2.86
Reject if exceeds size or more
than three, except for daily
disposable non ProClear lenses.

The lens circumference is

incomplete, due to 1 or more
Edge Chip
indentations that exceed the
(depth or 0.17 3 1.71
image of the edge inwards (i.e. V
width)
or U shaped chips). Reject if
exceeds size or more than three.

Figure 20 Lens has an irregularly shaped

Rotary Table AIS Vision System Rough Edge edge which is not smooth.
1.14 20 11.43
(length) Irregularity does not exceed the
image of the edge inwards.
PHASE V - CONTROL
Sharp dark line which appears
In this phase two mayor standard operating Surface Split Reject All more obvious when lens is
procedure were implemented, the generation of stretched
visual standards (see Figure 21) for objective Figure 21
classification of tears and other defectives and the Acceptance standard

development of a new procedure to trigger Trigger for Corrective Actions: The

corrective actions at in-process and customer
complaints. These two controls were generated as
result of an analysis of the complaints and the
feedback from the field representatives.
The success and control was going to
monitored on a monthly basis by the means of
complaints per million by product platform.
purpose was to define the requirements to detect
and identify product or process data trends,
qualitatively or quantitatively, the following
diagram shows the process as to when to rise a
CAPA (see Figure 22).
 Figure 22
CAPA generation Process

Results of the project: The following • Part II of the FMEA-RPN’s was reduced in
indicators provided congruent data that the the critical key input variables from the
project was successfully implemented: process (See Figure 23).

R
Actions S O D R
P Resp. Actions Taken E C E P

N Recommended V C T N

800 Standarize Tom B Develop an gauge that can control "air gauge"
design low
detection, 8 10 1 80
damaged lenses
are detected
720 Study of mold fit Dan E Develop an gauge that can control "air gauge"
ve lens tear reate 8 9 1 72

720 Study of mold fit Dan E Develop an gauge that can control "air gauge"
ve lens tear reate 8 9 1 72

720 Study tear rates Klaus

manual process Develop an Automated system for Handlig & 8 9 2 144
vs auto process inspecting lens
720 Check on temp Tom B
and make sure
8 9 2 144
they work all the Develop an Automated system for Handlig &
time. inspecting lens

Figure 23
Part II FMEA (extract)

Conclusion: The optimization project to
reduce the amount of complaints prove to be a
success, with improvements the market is
expanding to Asia and growing because of the
recent excellent quality of the “Proclear” product
noticed in the field.
The “6 Sigma-DMAIC” process methods
gave the team the know how to determine the
Key process input variables that made an effect
in the process for “Tears, Chipped Edge or
Split”[1].
The goal was to achieve less than 3.4
complaints per million for tears –split edges by
February 2009 as stated in the charter.
REFERENCES
[1] Breyfogle F,W, “Implementing Six Sigma” second
edition, 2003, published by John Wiley & Sons, Inc.
New Jersey.
.