Taguchi methods are powerful quality engineering methods developed by the Japanese engineer and statistician Genichi Taguchi to improve the quality of manufactured goods. The methods were first introduced in Japan, where they have been used extensively to improve the quality of products and processes within the organizations over the past 60 years.
In the 1980s, they were introduced to the US and quickly adopted by some of the most successful American corporations, including Ford Motor Company, AT&;T Bell Laboratories, General Motors and Xerox. In the late 1980s, shortly after their debut in the US, the Taguchi method of experimental design was introduced to Europe and the UK.
However, in contrast to US and Japan, research has indicated that the use of TMs(Taguchi Method) is far from pervasive within European Industries. Although many case studies are published in the existing literature on TMs , most academics are of opinion that TMs are still largely underutilized within the Southeast Asian manufacturing organizations.
However, there appears to be either minimal or no empirical evidence to verify this opinion, although this is not the case in Japan , the US and other European countries. To these authors' knowledge , this research study is the first attempt to understand the true impact of TMs on the Southeast Asian manufacturing organizations.
Back to basics
Dr. Genchi Taguchi describes TMs as: "the evaluation and improvement of the robustness of products, tolerance specifications, the design of engineering management processes and the evaluation of the economic loss caused by the functional variation of products". Given the importance of design of experiments within Taguchi's overall engineering strategy, he went to great lengths to develop his own standardised approach to the design of experiments in an attempt to make it easier to apply and more user friendly.
It has been speculated that as much as 80% of the Japanese quality gains which occurred after the second world war can be attributed directly to TMs. A recent survey by Fujita and Matsuo in 2005 found that from a sample of 118 Japanese manufacturing organizations, approximately 40% of those companies were applying the methodology, clearly showing that the techniques is still in common use today.
Research carried out by Jugulum and Dichter into the US manufacturing companies was primarily focused on assessing the impact of TMs through a survey questionnaire in 2001. Generating a modest 270 responses to their questionnaire, Jugulum and Dichter found that about 30% of the companies were applying the Taguchi approach to experimental design. Moreover, they found TMs most commonly used to optimise process performance, as opposed to improving a product design.
Survey conducted within Spain, Germany and Ireland have shown that TMs were being used by less than 10% of the responding companies. In 2003, Gremyr et al. discovered that the application of TMs was among the highest in Europe, but still low in comparison to Japan and US.
There are several case studies within the existing literature which document successful applications within the UK manufacturing industry. Another study carried out by Araujo et al. in 1996 showed that the application of TMs is more prevalent within the aerospace and automotive sectors.
The survey results
As part of the survey research strategy, it was decided that a survey strategy would be adopted. The strength of the survey is that it allows a large amount of data to be collected from a large population in an economical way.
Adopting the same strategy as used by Jugulum and Dichter in 2001, the survey was conducted through a two stage questionnaire process. A short primary questionnaire was designed and distributed to around 750 manufacturing organizations from the Philippines, Singapore , Malaysia,Vietnam , Hongkong and Indonesia. Many of the questions used for the primary questionnaire were web-based one created and administered by http://www.surveymonkey.com/
Total response received from the primary survey were 105. A detailed secondary questionnaire was then designed and distributed to about 20 companies. The secondary questionnaire was only sent to those companies who had indicated that they were applying TMs and were deemed to have sufficient knowledge and experience to complete the questionnaire. A total of 15 questionnaires were returned, 14 of which contained valid and complete information. This corresponds to a very good response rate of 70%.
The primary questionnaire had been designed to collect general background information about respondents. Company size was divided into four categories;less than 50 employees;50-249 employees; 250-500 employees and over 50 employees. Most of the responses came from companies with over 50 employees.
Figure 3: Distribution of respondents across the sectors
Type of manufacturing industry Number of responses
Chemical and Pharmaceuticals 20
Aerospace and defense 16
Food and beverage 12
Electronic component manufacturers 10
Plastics and rubber products 8
Fabricated metal product manufacturing 6
Automotive manufacturing 8
Printing and packaging 5
Electrical equipment and consumer appliances 5
Other manufacturing sectors 17
Figure 3 presents responses recieved in relation to all sectors that participated in the study.
Nearly 40% of the respondents represented the manufacturing function.Moreover, the majority of the respondents were from upper or middle management, where it is assumed they have a good overall understanding of the business and its operations. The results of the study also revealed that over 80% of the respondents conduct industrial experiements, 15% of which conduct industrial experiments frequently.
Figure 4: Reasons for performing or conducting industrial experiments
Reasons Percentage of respondents
To solve detected problems 75%
To optimise process performance and 70%
product characteristics
To learn about process or product behavior 50%
Product or process innovation 40%
To confirm initial assumptions 30%
To develop a rubost design unsensitive
variation 25%
Figure 4 presents the most common reasons for conducting industrial experiments in the Southeat Asian manufacturing organizations. The top three reasons given for industrial experiements were:
- To solve detected problems
- To optimize processes or products
- To learn about and explore the behavior of processes
The respondents were also asked to indicate the strategies of experimentation that they use when performing experiments. It was found that over 50% rely on the one factor at a time (OFAT) approach to experimentation for process optimization problems. This confirms the results of the study carried out by Antony in 2002, and is also consistent with the findings of Tanco et al in 2008 for companies in Spain.
The studies carried out by Tanco et al. and Jugulum and Dichter indicated that 50% of Spanish organizations were aware of TMs, compared to 75% of US organizations. The analysis of data indicated that there was no correlation between the use of TMs and company size , which is consistent with the findings of Tanco et al.
The results also showed that the use of TMs is higher within ISO 9001 certified companies which supports the findings of Ozgur et al. in 2002.
The respondents were asked about the common reasons for the application of Taguchi method of experiemental design. The top three common reasons for the application of Taguchi method of experimental design were:
- To improve or to optimise those existing processes
- To solve detected problems
- To learn and to understand about process or product behavior.
The secondary questionnaire was used to obtain a more detailed understanding of the application of TMs with Southeast Asian manufacturing organizations.
The number of companies involved in the detailed questionnaire was 14. It was observed that most companies carry out less than five Taguchi projects per year, with only two companies claiming conduct between six and 10 project each year.
It was interestting to compare these results with Jugulum and Dichter's 2001 study. Their findings showed that US manufacturing companies are performing more industrial designed experiments than their Southeast Asian counterparts. It was also found that most applications occur within manufacturing , followed by the applications in product development.
The author also discovered that TMs are most commonly introduced to US organizations as part of corporate quality programme or product development process, as opposed to a stand-alone technique. However, in the SEA manufacturing organizations it was observed that TMs are more commonly promoted as stand-alone tool.
Barriers to implementation of TMs
The respondents were asked to list the typical barriers that they have experienced when applying TMs. The following bariers were subsequently identified from the survey:
- Lack of a clear methodology when applying TM
- Insufficient financial resources
- Negative image of statistics within the workforce
- Insufficient human resources
- Lack of management commitment
- Organizational resistance
- Poor or inadequate training on TMs
- Lack of knowledge of engineers on the methodology
- Engineers generally have insufficient knowledge of statistics
- TMs are thought to be just to complex to apply.
- The poor standard of satistical consultancy that exist.
- Previous negative experiences in applying the methodology
- Lack of time
- Poor leadership
It was found that the lack of clear and systematic methodology to apply TMs was one of the most common barriers across the participating companies.
Interestingly, Tanco et al, in 2008 also identified this as one of the biggest barriers to the design of experiments within Spanish organizations, despite there being numerous methodologies proposed for TMs within the literature.
The findings of our study also revealed that the most respondents perceive there to be a negative image about the use of statistics within their workforce. Furthermore, over 80% of the respondents believe that engineers generally have insufficient knowledge of statistics to apply the TMs.
Another common barier to the application of TMs was "poor or inadequate training" and "insufficient time" which also appear to be key barriers within many US organization.
Conclusion
This research has shown that only a limited number of companies within the SEA manufacturing sector are currently realizing the benefits of this powerful methodology.
Although some companies are now applying TMs, most are failing to realize its full potential due to various barriers in the application and deployment of TMs across the business.
The true value of TMs lies in the application of TMs for designing rubostness into the design stage of products and processes. This study attempts to show the status of the application of TM within the SEA manufacturing companies.
The next stage of this research is to then carry out a semi-structured interviews with selected companies to obtain a greater insights into the many applications of TM within the SEA manufacturing sector.
A comparison on the application of TM in the SEA service sector will also be a worthy research topic over the forthcoming years.
Walang komento:
Mag-post ng isang Komento