Post on 26-Mar-2015
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 1 2/13/90-PREZ-1rev. 5/7/91
ANALOG DEVICES AT A GLANCE
Headquartered in Norwood Massachusetts
Publicly Held (NYSE Symbol ADI)
$485 Million in Sales (FY1990)
48% of Sales Outside United States
5600 Employees Worldwide
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 2 5/8/91-PREZ-1a
ANALOG DEVICES AT A GLANCE
Products: ICs, assembled products, subsystems
Applications: precision measurement & control
Markets: data acquisition
Integrated supplier
(cont.)
50% industrial/instrumentation25% military/avionics13% computer12% other
designmanufacturing (8 locations)direct sales (100 locations)distribution
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 3 9/12/91-09121-1.doc
Period Accomplishments
Key Weaknesses Corrective Action
1984-1986 CEO awareness of TQM
TQM organizationTQM goal setting process
established Corporate TQM staff
1987-1990 established goals/metricsQIP teams10x improvement“it’s real”
top management involvementfocus on managementnot everyone involvedincomplete TQM infrastructure training slow progress -> plateau
seek outside help expert: Professor Shoji Shiba networking: CQM
1991-1992 business restructuringfinancial focusTQM planning
TQM activity on “back burner”
revitalize TQMimplement TQM infrastructure
1992-1994 Creating the New Analog >10x improvement
? ?
today Better understanding of the challenge
• TQM skills of senior managers• Alignment of TQM with business goals• Low quality of management systems• Total participation
• CQM course(s)
• Rigorous Hoshin deployment
• Baldrige self-assessment, benchmarking, process redesign• TQM diffusion
TQM at ANALOG DEVICES
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 4 3/24/91-03241-1
DIAGNOSIS
5
TQM INFRASTRUCTURE
TQMACTIVITY
OUTPUT
GOAL
SETTING
ORGANIZATION
SETTING
1
TRAINING &
EDUCATION
2
PROMOTION
3
PUSH PULL
MONITORING
6
INCENTIVES
REWARDS
7
DIFFUSION
4
ORIENTATION EMPOWERMENT SYNCHRONIZATION
7 REQUIRED COMPONENTS
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 5 9/12/91-09121-1
ADI CONTINUOUS IMPROVEMENT HISTORY
1986 1987 1988 1989 1990 1991 1992 1993 19940.01
0.03
0.1
0.3
1
3
10
30
100
year
defect level
12 month halflife
6 month halflife
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 6 10/28/86-QIP-7rev: 4/22/91-04080-2a
METRICS
IS
culture scorecard
projects
GOALS
SOLVINGPROBLEM ADI QIP GOALS
MARKET LEADERSHIP (RMS)
REVENUE GROWTH
PROFITABILITY
BE RATED #1 BY OUR CUSTOMERS
TOTAL VALUE DELIVERED
BUSINESS
OBJECTIVES:
DRIVERS:
IN
TIME TO M
ARKET
PROCESS PPM
MANUFACTURING CYCLE TIM
E
YIELDEXTERNAL LEVERS:INTERNAL L
EVERS:
PRODUCTS
DEFECT LEVELS
ON-TIME DELIVERY
LEADTIME
PRICE
RESPONSIVENESS
GENERALIZED CYCLE
TIME
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 7 10/29/91-10291-1rev 10/31/91
GETTING TO THE BENCHMARK
defe
ct
level
time
1
.1
.01
.003
.3
.03
today
us
best in class
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 8
Source: Kenzo Sasaoka, PresidentYokagowa-Hewlett-Packard 7/84
Process Quality Improvement (Dip Soldering Process)
Counter Action I
Counter Action IIICounter Action II
%
ppm
0.4%0.4
0.2
0.1
0.3
0
Failure R
ate
11 1 4 7 10 1 4 7 10
Masking MethodImprovement
Omit Hand ReworkProcess
40ppm
Dip Solder wasmade to oneworker job
Basic WorkingGuide Manual
Revision of ManufacturingEngineering Standards
PC Board DesignInstructions 3ppm
78 79 80 81 82 FY
10 1 4 7 10 1 4 7 10 1 4 7 10 month
40
20
10
0
30
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 9 12/12/85-ZD4
YOKOGOWA HEWLETT PACKARD
60483624120
Failu
re R
ate
%
1
.1
.01
.001
.0001
months
10,000
1,000
100
10
1
PPM
50 % improvement each:3.6 months
Dip Soldering Defects
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 10 3/11/87-IMPDATA1
OBSERVED HALF-LIVESHALF-LIFE IMPROVEMENT
DESCRIPTION (months) CYCLES R2
operations sheet errors 0.6 4.2 0.834days late in delivery 0.8 7.6 0.774rejects due to bends and dents 1.3 1.7 0.590process sheet errors 1.4 2.1 0.535PCB photo imaging resist flake 1.9 3.3 0.748errors in purchase orders 2.3 1.5 0.531aluminum smears from IC test pads 2.4 5.1 0.717yield loss, die coat inspection 2.4 2.3 0.733scrap costs, die coat inspection 2.4 2.0 0.754defective stockings 2.7 2.2 0.843yield loss, PCB photo imaging 2.9 2.3 0.843typing errors in bank telegram dept. 2.9 2.0 0.754late orders to customers 3.0 2.7 0.838defects in PCB edge polishing 3.3 1.9 0.188insertion defect rate 3.3 3.4 0.738failure rate, dip soldering process 3.7 8.6 0.980down time of facilities 4.5 1.3 0.562COPQ, goggles manufacturer 4.7 1.9 0.942scrap and repair costs 5.0 1.6 0.918scrap and repair costs 5.0 0.8 0.746in-process defect rate 5.3 1.1 0.550late spare parts to customers 5.3 1.1 0.471defects due to pits, piston rings 5.5 3.5 0.968defects in vacuum molding 5.6 4.6 0.882vender defect level, capacitors 5.7 6.3 0.812
---------------------------------------------------------------------------------------------------------customer returns due to admin. error 6.3 3.8 0.941
WIP 6.3 1.1 0.979accounting miscodes 6.4 2.5 0.709manufacturing scrap 7.0 3.9 0.530vender defect level, transformers 7.2 5.0 0.842vender defect level, IC linears 7.4 4.9 0.906
WIP 7.5 2.1 0.759failure rate, line assembly 7.5 3.2 0.886manufacturing cycle time 7.6 2.7 0.741defects per unit 7.6 4.6 0.948
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 11 3/11/87-IMPDATA2
OBSERVED HALF-LIVES(cont.)
HALF-LIFE IMPROVEMENT
DESCRIPTION (months) CYCLES R2
rework rate 8.0 1.4 0.801off-spec rejects 8.8 5.1 0.513set up time 9.5 0.6 0.690
vender defect level, transistors 9.6 3.7 0.997defect levels, customers incoming QC 10.1 7.1 0.989defects 10.4 5.2 0.965
software documentation errors 10.5 1.2 0.173---------------------------------------------------------------------------------------------------------error rate, perpetual inventory 12.1 3.0 0.862customer returns due to product 12.4 2.9 0.974missing product features 12.5 2.9 0.947equipment downtime 13.1 2.1 0.940scrap costs 13.8 1.7 0.805
absenteeism due to accidents 14.8 4.0 0.956defects at turn-on 14.9 1.3 0.624manufacturing cycle time 16.9 2.5 0.937defects on arrival 16.9 2.0 0.848non-conformances 16.9 0.7 0.666
vender defect level, microprocessors 18.5 1.9 0.838post release redesign 19.0 2.5 0.842field failure rate 20.3 1.3 0.857accident rate 21.5 2.8 0.907defective lots received from vendors 21.6 1.7 0.976failure rate, PCB automatic test 23.7 0.5 0.182
---------------------------------------------------------------------------------------------------------first year warrantee costs 27.8 2.6 0.950computer program execution errors 29.9 0.4 0.364late deliveries to customers 30.4 0.8 0.994warranty fail rates 36.2 2.5 0.769
failure costs (internal + claims) 37.9 1.9 0.909product development cycle time 55.3 1.1 0.733
AVERAGE: 10.9 2.8 0.77
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 12 6/11/90-06110-1rev. 5/28/91
hi
med
low
himedlow
Org
aniz
ati
onal C
om
ple
xit
y
Technical Complexity
1 3 5
7 9 11
14 18 22
TARGET HALF-LIVESmonths
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 13 1/22/91-01221-1
PLAN
WHAT
IDENTIFICATION OF THEME
WHY CAUSAL ANALYSIS
WHO
WHEN
WHERE
HOW
SOLUTION PLANNING
EVALUATION OF RESULTS
STANDARDIZATION
DO
CHECK
ACTION
1
2
3
7
6
5
4AND IMPLEMENTATION
DATA COLLECTION
AND ANALYSIS
REFLECTION AND
NEXT PROBLEM
THE PDCA CYCLE and THE 7 STEPS
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 14 5/9/91-05091-1arev. 5/28/91
AUTOMOTIVE RELIABILITY IMPROVEMENT
Consumer Reports, April 1991
1980 1985 1990 19950
20
40
60
80
100
120
year
pro
ble
ms
per
100 c
ars
GM
Ford
Chrysler
Toyota
Nissan
Honda
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 15 5/9/91-05091-1rev. 5/28/91
AUTOMOTIVE RELIABILITY IMPROVEMENT
Consumer Reports, April 1991
pro
ble
ms
per
100 c
ars
year1980 1985 1990 1995
5
10
20
50
100
200
500
GM
Ford
Chrysler
Toyota
Nissan
Honda
80
76
79
205
73
94
halflife(mths)
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 16 QS-16B
ADI QIP GOALSIC OPERATIONS, ESTABLISHED PRODUCTS
METRIC 1987 HALF-LIFE 1992
On time delivery
Lead time
Manufacturing Cycle Time
Yield
Outgoing defect levels
Time to Market
EXTERNAL
INTERNAL
CORPORATE-WIDE COST MANAGEMENT
WHILE AGGRESSIVELY PURSUING
Process Defect Levels
85%
10 wks
500 PPM
9
9
9
>99.8%
<10 PPM<3 wks
15 wks
20%
9
9 >50%
5000 PPM 6 <10 PPM
4-5wks
36 mths 24 6 mths
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 17
ADI QIP GOALS
IC OPERATIONS, ESTABLISHED PRODUCTS
METRIC 1987 1990 1992
EXTERNALOn time delivery 85% 96% >99.8%Outgoing defect level 500 PPM 230 PPM <10 PPMLead time 10 weeks 5.4 weeks <3 weeks
% CRDs matched 31% 50% n/aexcess leadtime 3.9 weeks 2.8 weeks n/a
INTERNALManufacturing cycle time
15 weeks 8 weeks 4-5 weeks
Process defect level 5000 PPM 1100 PPM <10 PPMYield 20% 38% >50%Time to market 36 months ? 6 months
11/12/90-11120-3.doc
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 18 5/7/91-05071-8a
FY 1991 ADI CORPORATE SCORECARDACTUAL ACTUAL ACTUAL ACTUAL ACTUAL ACTUALBHMK BHMK BHMK BHMK BHMK
End FY 90 Q1 91 Q2 91 Q3 91 Q4 91 FY 91
ACTUAL
FY 90
ACTUAL ACTUAL ACTUAL ACTUAL ACTUAL
Q1 91 Q2 91 Q3 91 Q4 91 FY 91
FY 87 PLAN FY 87 PLAN FY 87 PLAN FY 87 PLAN FY 87 PLAN
FINANCIAL
QIP
MANUFACTURING METRICS: IC PRODUCTS
MANUFACTURING METRICS: ASSEMBLED PRODUCTS
NEW PRODUCTS
SALES
SALES GROWTH YTY
CONTRIBUTION MARGIN
ROA (CM)
ON TIME DELIVERY (to FCD)
% CRD's NOT MATCHED
EXCESS LEADTIME
EMPLOYEE TURNOVER
OUTGOING PPM
PROCESS PPM
CYCLE TIME
YIELD
OUTGOING PPM
PLUG-IN YIELD
CYCLE TIME
% COST OF SCRAP/REWORK
BOOKINGS POST-85 PROD
FORECAST 3rd YR BOOKINGSof new product releases
485.2
7.0
6.0
7.1
94.9
52.2
2.7
13.3
587
981
65.4
38.4
1503
90.9
23.0
8.1
165.3
140.0
27.7
5.7
7.0
95.0
49.9
16.7
4.3
647
772
34.0
386
48.5
10.6
5.9
56.7
133.1
21.4
4.6
5.7
89.9
54.5
3.0
11.0
541
816
35.6
849
48.8
8.6
4.1
54.5
144.0
23.7
8.9
11.1
95.8
47.5
16.7
4.1
508
657
33.6
312
48.6
10.3
5.5
64.7
146.0
21.1
20.2
12.7
96.4
45.7
16.6
4.2
411
546
37.9
296
48.8
10.2
5.2
70.5
150.0
8.3
12.5
16.0
97.0
44.2
16.7
4.0
329
450
40.0
282
48.9
10.1
4.9
77.3
580.0
19.5
9.4
11.8
97.0
44.0
16.7
4.0
328
453
39.7
273
48.9
10.1
4.9
269.2
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 19 5/22/90-05220-4rev. 5/7/91
GOALS
IS
culture scorecard
projectsMETRIC
SSOLVINGPROBLEM
PERFORMANCE MEASUREMENT
does not mean
measurement
improvement+
monitoring
If you don't monitor it,
it will get worse.
If you don't measure it,
it will not improve.
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 20 4/8/90-04080-3rev 5/13/90
GOAL:IMPROVE CUSTOMER SERVICE
CUSTOMER SERVICE METRICS
ON TIME% late% early % on time
RESPONSIBILITYfactorywarehouse
creditcustomer
LATENESS/EARLINESSshipped late, how late?shipped early, how early?still late, how late?
months to ship late backlog
LEAD TIMEcustomer requested lead time% CRD's matchedexcess lead time
RESPONSIVENESStime to schedule an order
GOALS
ISculture scorecard
projectsMETRICS
SOLVINGPROBLEM
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 21 % Late 3Q91
Percent Of Lines Shipped LateQuarterly Data 1Q86-3Q91
ADS ADIPMIMDLDSPIPDCLDMEDADBV
100
10
1Half Life
(in months)15 211419 53920 17NA
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 22 c. 1991
ON TIME DELIVERY HALF LIFE - IN MONTHSAnalog Devices, Inc.
60
50
40
30
20
10
0
3Q88 4Q88 1Q89 2Q89 3Q89 4Q89 1Q90 2Q90 3Q90 4Q90 1Q91
21.57 22.77
18.3916.41
15.15 15.12 15.1516.95
21.18
43.50
60 +
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 23 % Late Sep 91
Percent of Scheduled Lines Shipped LateCustomer Service Performance (Oct 90 Through Sep 91)
Percent
Of
LInes
Late
Half Life(In Months)
100
10
1
0.1N/A N/A N/A N/A N/A N/A N/A6N/A
ADS CLD IPD DSP MDL PMI ADIBVADM MED
+
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 24 3/91-VRS Database
64 CUSTOMERS INVENDOR RATING DATABASE
ABFAGFAAirResearchAllen BradleyAllied SignalAmdahlAmetekAnalogicApolloAT&TBecton DickinsonBendix AvionicsBrown EngineeringCompugraphicCoulter ElectronicsDigital InstrumentEatonElectronics & SpaceEMCFinneganFlukeFord (Aerospace)GECGeneral DynamicsGeneral ElectricGouldHewlett-PackardHoneywellHughesInstronJET ElectronicsKodak
LittonLoralLucasM/A-COMMarquette ElectricMartin MariettaMasscompMcDonald DouglasMeasurexMicrocircuits SemiconductorParker Air & SpacePenastarPerkin ElmerPitney-BowesRaytheonRecognitionReliance ElectricRockwellSandersSiemensSikorskySpectra PhysicsTektronixTelecoTeledyneTeradyneTexas InstrumentsTrilliumUnited TechnologiesWalkins-Johnson Co.Waters AssociatesWestinghouse
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 25 c. 1991
Delivery PerformanceCustomer Measured% Late
100
50
30
20
10
5
3
2
1
HALF LIFE = 9*
QIP STARTED
Fe Ap Ju Au Oc De Fe Ap Ju Au Oc De Fe Ap Ju Au Oc De Fe Ap Ju Au Oc De Fe
Mr My Jl Se No Ja Mr My Jl Se No Ja Mr My Jl Se No Ja Mr My Jl Se No Ja Mr
1987 1988 1989 1990 1991
Actual Data Trend* Half Life is only calculated
over the period of time thatimprovement occurred Note: There is an average
of 22 companies per point
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 26 5/22/90-05220-1
HEWLETT-PACKARD VENDOR RATINGS
year ADI rank total suppliers category
linear IC suppliers
all IC suppliers
linear IC suppliers
all IC suppliers
16
8
15
12
8
5
5
1*
1986
1987
1988
1989
*tied with one other supplier
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 27 3/24/91-03241-3
DATAQUEST1990 SEMICONDUCTOR"SUPPLIER OF THE YEAR"
Awarded to "...manufacturers who exhibitextraordinary dedication to productquality and customer service."
Winners:
major supplier: Motorola
mid-size supplier: Analog Devices
niche supplier: Maxim
©1987-2000 Arthur M. Schneiderman All Rights Reserved.
Slide 28 3/25/91-03251-1rev 4/1/91
SOME LESSONS LEARNED(One Person's Opinion)
"97% is good enough" -> no planned improvement
PMI acquisition
Reorganization (centralization) Manufacturing, QA Consolidation
CLD+MDD+MED+IPD=IED
(ACE+LSP)+DSP=SPD ADBV+ADS+PMI=?
New business planning process Product/Market/Function Strategy Managers
Eventually, you run out of "slack" need for cross-functional problem solving need for management participation
If you don't monitor it, it will get worse!
Fundamental re-evaluation of our QIP efforts Center for Quality Management (CQM) Shoji Shiba
It's easy to be distracted (even by the right things)
Input / Output Data Base
TQM / CQM
anti-Hawthorn Effect
7 steps weakness -> standardization -> control