ENBIS-13 Conference in Ankara, Turkey

Comparison of efficiencies of Kobetsu Kaizen and 6 Sigma

Approaches in Problem-Solving Process
Mustafa Konuk (mustafa.konuk@bellona.com.tr)
BELLONA Mobilya, Sytem and Job Development Department O.S.B. 14. Cd. No:14 38070, Melikgazi, Kayseri, Turkey.

1. Introduction and Objective

• In today's competitive environment, increasing business excellence day by day and gaining power for competition in all competitive branches of industry, particularly in industrial
organizations form the common goal of all activities (processes).
• Perspective on the problems of institutions, appears as the most important determining factor in providing motivation for detecting and solving the problems.
• Kobetsu kaizen, a Japanese approach, composed of a number of systematic study to eliminate the problem by employees' identifying the resources of problem within a specific plan and
developing and testing countermeasures according to the characteristics of these resources.
• 6 Sigma reveals experience as well as, cause-and-effect relationships moving through the process and data, detect the source of problem (variability) and develop the right solution.
• We applied these two approaches, in our enterprise in order to resolve problems that arise in the process of installation.
• We have brought recommendations in order to identify the strengths of each approach and weaknesses and get more effective results

3. 6 Sigma Approach
2. Kobetsu Kaizen Approach-General
• Contains elements of Japanese culture in approaching to problems.
•The Basis of Japanese problem-solving techniques are based.
• Great incomprehensible and frightening problem is divided into smaller
•Motorola was the first company that put 6 Sigma methodology in the business world
literature in 1980s.
pieces.

• The problem is solved in 8 or 10 steps containing PDCA (Plan-Do-Check-

•Problem is solved in 5 steps containing containing PDCA (Plan-Do-Check-Act) cycle.
Act) cycle.
•The process that led to the problem, by ensuring the validity of the measurement
• After the current situation and the project plan, the sources of problems are
system root causes of that led to the problem are revealed statistically with process
analysis. Interactions between the root causes are also determined.
determined and improvements are planned according to the result of Why-
Why' analysis performed by a team created by people in the problem
experiencing period in general.
•Root causes are presented as a result of experiences and experiments/evidences.
• In Why-why analysis the root causes that cause the problem can be
•How much improvement is needed can be demonstrated in accordance with the aim of
the project .
presented in the light of experience.

• Measures against the root causes are determined, are brought to life and the
•Improvements are carried out in a planned way. Solution is brought under control and
project indicator is monitored at least for 6 months.
goal of the project is confirmed. Solution is standardized.

4. Mounting Process 5. Problem solving with Kobetsu Kaizen Approach

• Glass-Mirror paste Job 1. Choice of Topic: Due to the 4 Current Situation Analysis (5W 1H)) 6. Analysis 7. Improvement
Glass/Mirror growing customer returns the (What?) Falling off event in Glass / Mirror doors Problem Cause-1 Cause-2 Cause-3 Cause-4 Decision Solutions A: Cleaning brush was performed on the band was.
B: The standard work on the line was confirmed every week.
prevention of the problem of glass C: Semi-acrylic foam was started to be used and adhesive strength
(Where?) At customer's home. was increased.
mirror shift is determined by Kaizen Glass
Glass
D: Weighting was performed need for waiting time was reduced.
(When?) More in summer months, less in winter months doesn’t keep OK
management. band
Component

(Which?) In the past, it was not so high. It's getting worse

2. Project team: a team of 5 gradually. Glass
Not Cleaned Staff forgot 8. Verification and Profit
NG
cleaning It is observed in the samples performed in the line that the
people from Quality and brush
A
(How?) Problem is increasing in the last 5 months. Body Dirty Body glass mirror shifts were blocked
Subsequently
Production departments. (Who?) There are differences in experiences of assembly Dirty
Waited for
part
NG
İade Adeti
workers. önce proje sonra
3. Target: It is primarily aimed Glass Miror NG
140
Band 120
to reduce monthly average 5. Project Plan falling aff Staff No Standard number Standard
B

Individual Value
100
Business
return form 110 pieces to 30 Actions to perform Oct. Nov. Dec. Jan. At NG
Verification 80

60
1
1
1
1
UCL=56,2
_
pieces. implement 40 X=40,8
LCL=25,3
Band type NG 20 1

Data analysis of No Semi acrylic

0

resisitance 08 08 08 08 09 09 09 09 10 10
foam C 20 20 20 20 20 20 20 20 20 20
past periods in adhesion High Curing transition Oc
ak
Ni
sa
n
m
mu
z
Ek
im
Oc
ak
Ni
sa
n
m
mu
z
Ek
im
Oc
ak
Ni
sa
n

Time Te Te
Both Sided NG A ylar
Customer opinons being late band

Panel Plate Band Analysis

Unsufficient
implement
Low wait
time
NG weighting D
9. Standardization
Actions performed was processed to standard forms. Business
development training courses are reinforced with visual
management and single point trainings.
Identification of
improvements 10. Dissemination
Performance was shared with those concerned and customers
Implentation of
improvements

6. 6 Sigma 1- IDENTIFICATION 2- IDENTIFICATION-MEASUREMENT 3- MEASUREMENT-ANALYSIS 4- ANALYSIS - IMPROVEMENT 5- IMPROVEMENT – CONTROL

Project Description (What?): Improving the Glass-Mirror
SIPOC First Analysis for Fall Time. Main Effects .
adhesion strength
Term Effect Coef SE Coef T P
Main Effects Plot for FITS1
Constant 251,6 18,58 13,54 0,000 Data Means
Acquirements (How?): increasing the Customer A -216,9 -108,4 18,58 -5,84 0,000 Sıcaklık Bant

Satisfaction , preventing the risk of accidents at home, B

C
24,6
26,4
12,3
13,2
18,58
18,58
0,66
0,71
0,519
0,489
350

300

reduction of the costs of After-Sales Service D 16,6 8,3 18,58 0,45 0,662 250

200
E -32,6 -16,3 18,58 -0,88 0,395
Düşme Süresi

Method F 28,3 14,1 18,58 0,76 0,459 150

Targets (How much?): Reduce monthly return rate to 75%. G 134,6 67,3 18,58 3,62 0,003
-20
Sıvazlama
60 A
Bant Sayısı
E

Go down to the level of return to 27 returned from 110. Modeling of Glass Mirror Falling event with H 14,7 7,4 18,58 0,4 0,698 350

J -26,4 -13,2 18,58 -0,71 0,489 300

Plackett -Burman Experimental Design 13 Faktor, K 22,7 11,4 18,58 0,61 0,551 250

200

Time Planning: 28 Experiment, 1 reputation L

M
-75,6
-24,6
-37,8
-12,3
18,58
18,58
-2,03
-0,66
0,061
0,519 150

Phase Oct. Nov. Dec. Jan. Fab. Mar. Apr. May.

Place: Air Conditioning Testing Laboratory N 67,6 33,8 18,58 1,82 0,09 Var Yok 6 12

Project Start Detailed Road Map StdOrder RunOrder PtType Blocks Sıcaklık Yüzey Sıcaklık Farkı İstifleme Cam Ayna Bekleme Süresi Bant Takoz/Strafor Temizlik Sıvısı Ortam Tozu Sıvazlama Bant Sökme Bant Kullanım Oranı
Last Analysis for Fall Time.
İade Adeti
22 1 1 1 -20 Kirli 28 Alt Temiz 0 1 Yok Yok Temiz Yok Sökülmüş 12 önce proje sonra
Define 6 3 1 1 -20 Kirli 14 Alt Temiz 2 2 Var Yok Tozlu Var Sökülmemiş 12 Term Effect Coef SE Coef T P 140

120
Measerement 19 4 1 1 -20 Temiz 14 Alt Kirli 2 1 Yok Yok Temiz Var Sökülmemiş 12 Constant 251,6 16,42 15,32 0,000
Individual Value

100
18 5 1 1 -20 Temiz 28 Alt Temiz 2 2 Yok Yok Temiz Var Sökülmüş 6
A -216,9 -108,4 16,42 -6,6 0,000 80
Analysis 25 8 1 1 -20 Kirli 28 Alt Kirli 2 1 Var Var Temiz Yok Sökülmüş 12 1
1
1
1

5 10 1 1 -20 Kirli 14 Üst Temiz 0 2 Var Yok Temiz Yok Sökülmemiş 6 G 134,6 67,3 16,42 4,1 0,000 60 UCL=56,2
_
Improvement 24 12 1 1 -20 Temiz 14 Üst Kirli 0 1 Var Var Tozlu Var Sökülmüş 12 L -75,6 -37,8 16,42 -2,3 0,031 40 X=40,8
LCL=25,3
20 1
Control 28 16 1 1 -20 Kirli 14 Alt Kirli 0 1 Yok Var Tozlu Yok Sökülmemiş 6 N 67,6 33,8 16,42 2,06 0,051 0
20 19 1 1 -20 Kirli 28 Üst Kirli 0 2 Yok Var Tozlu Var Sökülmüş 12
08 08 08 08 09 09 09 09 10 10
20 20 20 20 20 20 20 20 20 20
3 21 1 1 -20 Temiz 28 Üst Temiz 2 1 Yok Var Temiz Yok Sökülmemiş 6 ak sa
n uz im ak sa
n uz im ak sa
n
Oc Ek Oc Ek Oc
Project Limits (where?): Glass-Mirror Drop Process 27 24 1 1 -20 Temiz 14 Alt Temiz 0 2 Yok Var Tozlu Var Sökülmüş 6 Analysis Result: Effective Factors in Ni
Te
m
m Ni
Te
m
m Ni

Aylar
15 25 1 1 -20 Temiz 28 Üst Kirli 2 1 Var Yok Tozlu Yok Sökülmemiş 6
Project Team: 11 26 1 1 -20 Temiz 28 Üst Temiz 0 2 Var Var Temiz Var Sökülmemiş 12 Glass / Mirror falling and their Impact
4 27 1 1 -20 Kirli 14 Üst Kirli 2 2 Var Yok Tozlu Yok Sökülmüş 6
Levels
* Project Leader: Quality Management Chief (Green Belt) 12
14
2
6
1 1 60 Kirli
1 1 60 Temiz 14
28 Alt Temiz
Alt Temiz
2
2
1
2
Var
Var
Yok
Var
Tozlu
Tozlu
Var Sökülmüş
Yok Sökülmemiş
12
12 It is observed that temperature, band type
* Project Champion: Mustafa Tasci (Quality Management Data Collection Plan 21 7 1 1 60 Kirli 14 Alt Temiz 0 1 Var Var Temiz Yok Sökülmüş 6 Pie Chart of Source C ategory
A and patting after sticking are effective on
16 9 1 1 60 Kirli 28 Üst Kirli 2 2 Var Var Temiz Var Sökülmüş 6 G
L

Manager) 7
9
11
13
1 1 60 Temiz 28
1 1 60 Temiz 28
Alt
Alt
Kirli
Kirli
0
0
2
1
Yok
Var
Yok
Yok
Tozlu
Tozlu
Yok Sökülmüş
Var Sökülmemiş
6
6
Diğer
24,7% N
Diğer adhesion strength.
* Project Team: Production and Quality Management Staff 17
26
14
15
1 1 60 Temiz 14
1 1 60 Kirli 14
Üst Temiz
Üst Kirli
0
0
1
1
Var
Yok
Yok
Yok
Temiz
Temiz
Var Sökülmüş
Var Sökülmemiş
6
12 A
46,9%
N
10 17 1 1 60 Temiz 14 Üst Kirli 2 2 Yok Yok Temiz Yok Sökülmüş 12 4,6%

VOC (Voice Of The Customer), 2 18 1 1 60 Temiz 14 Üst Temiz 2 1 Yok Var Tozlu Yok Sökülmüş 12
L
13 20 1 1 60 Kirli 28 Üst Temiz 0 2 Yok Yok Tozlu Yok Sökülmemiş 12 5,7%

Stakeholder Management, 8 22 1 1 60 Temiz 28 Alt Kirli 0 2 Var Var Temiz Yok Sökülmemiş 12
G
23 23 1 1 60 Kirli 14 Alt Kirli 2 2 Yok Var Temiz Var Sökülmemiş 6 18,1%

Communication Plan, 1 28 1 1 60 Kirli 28 Üst Temiz 2 1 Yok Var Tozlu Var Sökülmemiş 6
S&S Matrisi

7. Results 8. General Assessment

In Kobetsu Kaizen study ;
Companies are facing a lot of problems in their daily activities and are implementing different
➢5W approaches in solving these problems. Another important feature in today's business world in which
➢1H made ​by the project team set out specifications for the problem. problems' become visible is as important as eliminating the problems, is deriving lessons form mistakes.
➢As a result of The Why-Why analysis root causes that lead to the fall of the glass-mirror have been learning lessons) these lessons provide information to prevent repeating the same mistakes or to
determined and 4 improvement areas. have been identified. determine a new problem rapidly.
➢Cleaning brush application is deemed appropriate in order to prevent the part's getting dirty thus protecting ❖In Kobetsu kaizen methodology workers detected significant problem resources, and they implement
the adhesion strength. actions to reduce the problem with the solution suggestions.
➢Identification and improvement of standard work is determined as a solution for worker's performing the
adhession job proper to band number and order of business. ❖However, affect of significant variation source such as temperature was determined as a result of
➢ It is determined that the type and structure of double-sided band have to be changed.. why-why analiysis. Because this factor is effective when products are being carried to customers
outside of the process, during logistics (the temperature inside the containers).
In 6 Sigma study;
❖Kobetsu Kaizen methodology requires effective team work in order to reveal the root causes of the
➢In identification stage Stakeholder Management, Communication Plan, S & S matrix studies were performed problem correctly. If assessments done during the determination of the current situation (5N1K) are
. based oh (5W1H) numerical analysis, it may be easier to get root causes more effectively.
➢A detailed process diagram was prepared in order to present how Glass-mirror falling process takes place.
➢13 process parameters (source of variability) affect the adhesion strength was determined in a detailed ❖6 Sigmamay may get the root of the problem with experimental designs and besides explain the
process diagram . Mirror-Glass Falling mathematically.
➢An experiment design that contains 13 was fulfilled .
❖It is necessary to sahre steps of 6 Sigma project and the results obtained during the project with the
➢Number experiment to be done with full factorial experiment design will be 213 = 8192. Under the current employees and the employees should make understand all the developments clearly.
circumstances, it is not possible to perform this test.
➢It was decided to use Plackett Burman experimental design method that can be used in identifying as little ❖6 Sigma gives the more positive results on improving thoughts and behaviors of operational
experimental point as possible and effective factors. With this design, only the main effects can be detected employees than Kaizen Kobetsu more delayed.
and important factors are set out. It may also provide predictive information about the interactions in the
second degree.
➢Temperature was considered as a factor In the study performed with 28 experiments and those experiments
were performed at air-conditioning rooms of Boytaş.
➢As a result of analysis temperature, band type, patting process and the number of bands used were
observed to be effective.
➢Improvements were made in this direction.