Introduction
In order to gain business profit, a safety and health management system is indispensable to the enterprise. Hayes et al.1) indicated that the safe management prac-tice carried out by the supervisor was the most predictable index to the satisfaction of employee. Thompson et al.2) further explained the model of how the management affected the work safety by the safety culture and the man-agement’s support to the safety. But few of these researches focused on the topic of semi-conductor indus-try safety and health management. The semi-conductor industry is a very important key in the development of Taiwan’s science and technology industries, so this study focuses on the field of semi-conductor industry in Taiwan, hoping to find out the significant factors of the safety and health management in this area.
With the fast development of semi-conductor industry in Taiwan, Taiwan has already caught up with the
advanced countries in the manufacturing process ability of the semi-conductor industry. However, the related safety and health management technique was not pro-gressing simultaneously, which causing a lot of major accidents happened in the past few years. The purpose of this research is to construct a frame of semiconductor industry safety and health management system, and with questionnaires and on-the-spot interviews, to find out the main factors of safety and health management system in semi-conductor industry. Then an in-depth interview was executed to find out the safety and health management problems in the semiconductor industry. The investiga-tion domains are based on the most intensive and partic-ularly important area of semi-conductor industry in Taiwan —the Hsinchu Science-based Industrial Park. The main research hypotheses are as follows: First, “Whether the main factors of safety and health management system found in this research are the same with other manage-ment system manage-mentioned in the literature review (for
exam-A Study for Safety and Health Management
Problem of Semiconductor Industry in Taiwan
Chin-Jung CHAO
1*, Hui-Ming WANG
1, Wen-Yang FENG
2and Feng-Yi TSENG
31Department of Industrial Engineering and Management, National Chiao Tung University, 1001 Ta Hsueh
Road, Hsinchu, 300 Taiwan
2Department of Industrial Engineering, Chung Yuan Christian University, Chung Li, 320 Taiwan
3Department of Industrial Engineering and Management, National Tsing-Hua University, 101 Section 2,
Kuang-Fu Road, Hsinchu, 300 Taiwan
Received April 5, 2007 and accepted June 11, 2008
Abstract: The main purpose of this study is to discuss and explore the safety and health man-agement in semiconductor industry. The researcher practically investigates and interviews the input, process and output of the safety and health management of semiconductor industry by using the questionnaires and the interview method which is developed according to the frame-work of the OHSAS 18001. The result shows that there are six important factors for the safe-ty and health management in Taiwan semiconductor industry. 1. The company should make employee clearly understand the safety and health laws and standards. 2. The company should make the safety and health management policy known to the public. 3. The company should put emphasis on the pursuance of the safety and health management laws. 4. The company should prevent the accidents. 5. The safety and health message should be communicated sufficiently. 6. The company should consider safety and health norm completely.
health management in a semi-conductor industry?”
Literature review
After 2001, 12 inches of wafers factories pre-dominate the whole market. In order to avoid major accident which causes the operation to break off, the insurance fee to rise, and the orders to runs off, it is necessary to establish a comprehensive safety and health management system and carry out adequate actions for the manufacturing process, chemical toxicity, exhaust and gas usage etc. Along with the influence of global eco-environment becomes worsen gradually, the industry advanced countries have already treated product quality, environmental protection and industrial safety and health as the international competi-tion ability index. Therefore, the ISO (Internacompeti-tional Organization for Standardization), under the current ISO 9000 and ISO 14000 management structures, will com-bine quality, environmental protection and safety and health into a whole in the future. As for the internation-al industriinternation-al safety and heinternation-alth standards, the BS8800 and OHSAS 18001 (Occupational Health and Safety Assessment Series) are referenced frequently as the basic safety and health management structure. Therefore, by reviewing the existed occupational safety and health man-agement systems, a study is developed to examine the safety and health integrity of semi-conductor industry, to find out the potential safe problems, and to find the insuf-ficiency of the current management system. The results will provide the semi-conductor industry with a better safety and health management system.
Occupational accidents analysis in the
semi-conductor industries
During 1977–1997, there were 407 accidents happened in the global semi-conductor factories. Taiwan had 7 among them, and the estimated average total cost of two of the accidents was $710 million (Sherin)3). Hence, both safety and health management and production perfor-mance are related as closely as each breath is to the next. Ironically, in order to get work done faster, the safety and health rules are ignored, as a result the surprised accident occurred. The main causes of semi-conductor industry accident included fire (47%), fluid leak (22%), and explo-sion (5%). Most of the accident causes mainly lie in the latent hazard of semi-conductor manufacturing process and the production pressure that change with each pass-ing days continuously. Therefore, the tradeoff between production and safety and health is important issue for the enterprise, and this is the reason of why a better safe management system exists.
Reason4, 5)pointed out that organization factors would
influence environmental factors, and unsafe environment continued to bring up personnel unsafe action which pen-etrating the defense line of the organization, then the acci-dent just happened. Only through an effectively safety and health management, the accident can be prevented. Rahimi6) indicated the supervisor’s attitude is the impor-tant key to the success of safety and health management. Deming7) addressed that the safety and health manage-ment is related to the product quality, only when man-agement did not treat accident as a special condition, then the safety and health quality would get improved. Kennedy and Kirwan8) indicated there was currently no certain way to point out clearly what kind of safety and health management is the best, so an individual evalua-tion must be built-up to the safety and health manage-ment in the organization separately. Hee et al.9) also thought the first step to reduce the accident was to eval-uate the safety and health management. Generally used evaluation tool is checklists, which are based on the cur-rent standards or principles and extended to the require-ment for each company. Therefore, these checklists are good at checking machine, interface or operation proce-dures, but were not so useful when applied to the safety and health management aspect.
Hee et al.9)showed that the quantified reliability analy-sis of hardware equipment can reduce the injury degree of the accident, but it was not the most effective method for accident improvement. Nivolianitou and Papazoglou10) thought that the totally detailed safety and health man-agement failure data must be acquired before quantitative safety and health management evaluation. Kennedy and Kirwan8)indicated that in order to understand safety man-agement problems and acquire the useful factors of safe culture fully, it must adopt qualitative analysis method.
Some researchers put forward the subjective question-naires to measure the employee’s opinion about safety and health issues in the work place, to know indirectly the levels that safety and health management should pay attention to. Those questionnaires include the Occupational Hazards Survey (OHS)11); the Safety Climate Survey (SCS)12); The Physical Demands and Dangers (PDD)13); The Perceptions of Workplace Hazards Scale (PWHS)14); The Work Safety Scale (WSS)2) etc. But the above sur-veying methods have measuring problems. For example, the internal consistency coefficient of OHS is low (.28, .30, .61); SCS has excessive items (8); WSS encountered the representative problem in sampling, especially most of the samples are matching with the industrial accident investigation on request.
By the above analysis and comparison, in order to con-firm the effectiveness of the safety and health manage-ment, there must have a set of good procedure to exam-ine and evaluate safety and health management. This
study utilize quantified analysis to evaluate directly the failure of the safety and health management, and an induc-tive quality analytical method is used to determine the priority of the detailed and key results of the safety and health management system.
The development of investigating tool
The contents of safety and health management are very comprehensive, and by means of exterior audits only will not be able to find the core of the safety and health prob-lem. Therefore, the better way within a firm is to estab-lish a set of safety and health management plan basing on the interior condition actively and progressively. Good tools are prerequisite to the successful execution of a job. As a result, this research would like to raise a set of com-plete management system to understand and explore deeply the safety and health issues in the semiconductor industry.
In order to reach the above goal, the management sys-tem of OHSAS 18001 (BSI)15)was adopted. The OHSAS 18001 are developed mainly by referring to the structures of BS 8800 (BSI)16) and ISO 14000 with the spirits of closed-loop management and continuous improvement. The contents of OHSAS 18001 include organization structure, activities, job description, procedure, work pro-cessing, execution, correction, and the resources of main-taining the safety and health policy which concretely make public the meanings and procedures of how to put the management system into effect. Therefore, the inter-view questionnaire in this research is according to the management structure of OHSAS 18001. At the same time, Reason’s5)accident path model was also integrated to strengthen the completeness of the management sys-tem. The evaluation procedures are divided into four steps; decide the main system factors, develop a primary survey questionnaire, confirm the related factors, and make a conclusion.
Research method
Both the quantitative and qualitative analysis method are adopted in this research. First, by using a question-naire to collect the data toward safety and health man-agement from the employee, then factor analysis method is used to make sure the structure factors of the and health management. Next an analysis of variance (ANOVA), a least squares method, was utilized to test whether there were differences between the means of the dependent variable for each combination of levels of the experi-mental factors. Finally, a Tukey’s test was selected to do
imentwise error rate at the selected level α=0.05 (Montgomery)17).
Subject and questionnaire
The sampling subjects are the employee in the semi-conductor factories, and the questionnaire is answered by anonymity. The questionnaire is issued according to the company employee scale. 188 questionnaires were sent out and 168 completed questionnaires were returned, giv-ing a response rate of 89.36%. After excludgiv-ing those who did not fulfill the inclusion criteria, 162 subjects were included in the subsequent analysis.
The questionnaire is divided into two parts, and part 1 contains 68 topics which covers: (1) safety and health pol-icy (2) risk identification, evaluation and control (3) safe-ty and health laws (4) safesafe-ty and health object and plan-ning (5) responsibility, knowledge, abilities, and traiplan-ning of safety and health (6) communication (7) operation con-trol and emergency response (8) performance evaluation (9) accident investigation and prevention (10) audit and reviewing. Part 2 is the basic information of the subject which includes age, gender, title, company employee number, work experience, educational background and work section etc.
Data analysis method
This research utilized reliability analysis first to find the internal consistency coefficient (Cronbach, α) of the questionnaire, and then a factor analysis was used to build up each factor immediately after. Next each factor was regarded as the dependent variable, and the safe health performance, company scale, company capital, gender, age, work title, work experience, work type and educa-tion level were served as independent variables. Finally, a statistics analysis was executed for each relation. Each factor score accounts by summing all scores of topics inside that factor, and an ANOVA is used to discuss the relationship of each factor with organization variation and personnel variation. Finally, a Tukey’s test was carried on to do the after-the-event comparison.
Results
The reliability and validation of the questionnaire In the beginning, a pilot study for the preliminary 68 questions of the questionnaire is used to test the similar-ity between question and question. After the reliabilsimilar-ity analysis, for those coefficients over 0.50 between ques-tions are deleted, and the correlation coefficient over 0.2
a factor analysis is carried on for all the variables. The KMO (Kaiser-Meyer-Olkin) value is 0.79 which means that the results of entire factor analysis are appropriate. This internal consistency index Cronbach coefficient is 0.89 (Cronbach of each factor is 0.81, 0.77, 0.72, 0.67, 0.77, and 0.62) which shows that these 32 questions mea-surement are rather consistent. Meanwhile, each factor reliability and validation is acceptable too. Finally, six factors are established and they explained 52.92% varia-tion totally. The first factor was named “The company should make employee clearly understand the safety and health laws and standards”, and it included 7 topics and explained the total variation of 23.13%. The second fac-tor was named “The company should make the safety and health management policy known to the public”, and it included 6 topics and explained the total variation of 10.72%. The third factor was named “The company should put emphasis on the pursuance of the safety and health management laws”, and it included 6 topics and explained the total variation of 6.43%. The fourth factor was named “The company should prevent the accidents”, and it included 4 topics and explained the total variation of 4.63%. The fifth factor was named “The safety and health message should be communicated sufficiently”, and it included 5 topics and explained the total variation of 4.21%. The six factor was named “The company should consider safety and health norm completely”, and it included 4 topics and explained the total variation of
3.81%. In order to understand further the effectiveness of the safety and health management factors established in this research, an interview with two supervisors in the semiconductor field is conducted to confirm the suitabil-ity of the results.
The relationship between organization factor and safety and health management variables
An ANOVA is used to test the relationships between six safety and health management factors and total safe-ty and health management with company scale, company capital, safety and health performance. The results are shown in Table 1. A few discoveries are shown as fol-lows:
1. The larger company scale. (1) The more employee agree that the importance of the total safety and health management. (2) The company should let the employ-ee understand clearly the safety and health laws and standards. (3) The safety and health information should be communicated fully.
2. The greater company capital, (1) The more employee agree that the importance of the safety and health management. (2) The company should let the employ-ee understand clearly the safety and health laws and standards. (3) The safety and health information should be communicated fully. (4) The company should make public announcement in drawing up the safety and health policy. (5) The company should
Industrial Health 2008, 46, 575–581
Table 1. The analysis of organizational factors and safety management variables
Category
Company scale Company capital Safety performance large medium small great moderate few good fair bad
F A C T O R S 1 X 26.04 23.40 21.61 25.29 21.42 23.24 25.34 22.51 22.20 sd 2.90 3.75 3.67 3.37 3.07 4.12 3.36 3.59 4.52 2 X 26.43 26.19 25.08 26.47 25.06 25.95 26.46 25.72 14.93 sd 2.48 2.51 3.26 2.45 2.64 3.12 2.46 2.64 3.90 3 X 23.68 23.17 22.42 23.64 22.03 23.24 23.68 22.71 22.47 sd 2.97 3.03 3.53 2.65 3.65 3.42 3.22 3.08 2.64 4 X 17.18 17.31 16.69 17.13 16.72 17.49 17.45 16.86 16.53 sd 1.81 2.00 1.85 1.71 2.36 1.83 1.83 2.02 1.55 5 X 19.54 2.89 16.58 19.42 16.69 18.10 19.24 17.82 17.20 sd 2.59 1.42 2.98 2.39 2.42 3.76 3.03 2.55 3.65 6 X 13.95 13.84 13.31 14.06 13.06 13.76 13.84 13.60 14.00 sd 1.55 1.42 1.75 1.49 1.47 1.59 1.58 1.54 1.56 : Significant difference at α=0.05 level.
carry out the safety and health laws without doubt. (6) The company should have the full communication on the safety and health message. (7) The company should have the integral consideration at safety and health standard.
3. The better performance in the safety and health man-agement, (1) The more employee agree that the impor-tance of the safety and health management. (2) The company should have the full communication on the safety and health message. (3) The safety and health information should be communicated fully.
The relationships between the employee’s variables and safety and health management variables
The relationships between six safety and health man-agement factors and personnel variables are explored fur-ther by using ANOVA test. The results are shown in Table 2. From Table 2, it indicated that different work-ers do not view the effort of safety and health in the same way. The possible reason is that safety and health relat-ed workers are responsible for the safety and health work, so they are more understanding what the company have done in the safety and health affairs. Those most-expe-rienced employee are more understanding what the haz-ardous characteristics are and how serious when accident happened. And from the distribution of the sample char-acteristics, most of the personnel responsible for safety and health are male. Due to the average level of educa-tion for female employee is lower than that of male, female workers serve as the basic assistant or technician
work which is full of hazardous exposure. They are fac-ing hazardous operation and the disaster takes place most-ly on these areas. As a result, they expected that the com-pany can adopt the positive attitude to prevent accident. This also reflects the female employee may need to accept the related safety and health training and emergency response maneuver.
Discussion and Conclusion
From the factor analysis results, this research acquires six safety and health management factors in the semi-conductor industry. These six safety and health manage-ment factors are the mostly important factors that the semiconductor industry currently should put more efforts on these things. When compared these six safety factors with the key items of OSHAS 18001, there exist a slight dissimilarities. It is because that this factor analysis is aiming directly at the semiconductor industry, and as a result it can stand for the safety and health points of emphasis in the semiconductor industry.
The research inquires into the related safety and health management problems respectively in the semiconductor industry and the conclusions that it integrate are shown in Table 3. Although the semiconductor processes are dif-ferent from other manufacture industries in the types and quantities of material used, the findings in Table 3 shows that the six factors of safety and health management prob-lems in the semiconductor industry are almost same as problems in general manufacture industry. As for
Table 2. The ANOVA of the organizational variables and personnel variables of six safety management factors
Independent Variables
Work type Work Experience Gender Work rank
F A C T O R S 1 ** 2 * 3 4 * 5 6 *
Explanation Safety-related workers agree that the company should make the employee under-stand the safety and health laws and standards clearly (x=25.87, sd= 3.85, n=31), but the others do not think so (x=23.45, sd= 3.69, n=131)
The most- experienced employ-ee have higher scores in the factor that the safety policy of the company should be public (x=25.80, sd=2.75, n=92), the un-experienced employee are next (x=25.50, sd=2.67, n=34)
The female employee get a higher score than male employee at the factor that the accident should be pre-vented (female: x=17.59, sd=1.76, n=59), (male: x=16.86, sd=1.95, n=131)
The lower position got the higher scores than the medium position in the factor that the company should have integral consideration about the safety health laws and specifi-cations (lower position: x=14.14, sd=1.59, n=49; medium position: x=13.74, sd=1.49, n=92). But the higher position have lowest score in this factor (x=12.95, sd=1.50, n=21)
OHSAS 18001, its safety and health management is more universal to general industries and is provided to all the related business no matter how the scale and professions are. Therefore, its contents are common in the safety and health management structure. But, along with the differ-ent technique and managemdiffer-ent need of each industry, they can reference the basic safety and health management structure with individual characteristics included. As a result, the safety and health management knowledge can be readily applied to similar industries, and this is the value of OHSAS 18000 structure which is suitable to dif-ferent enterprise.
In order to make this research more attributed to the semiconductor industries, the recommendations about the future study are suggested as follows.
(1) Although this research investigates the overall study for the safety and health management factor of the semi-conductor industry only, the results show that the safety and health training plays an important role in the safety and health management system. Therefore, in the near future, the topics of safety and health training and emer-gency practices should be included in the safety and health management of semiconductor industries.
(2) Both the quantitative and qualitative analyses are adopted directly in this research, but there are topics (such
as the effects of organizational culture on safety and health management) which are not easy to express by quantified data. As a result, a qualitative method can be adopted to explore the related discussion.
(3) This research samples are based on the IC cate-gories in Science-based Industrial Park industry in Taiwan, and it is trying to construct an entire content of safety and health management in semiconductor industry. The future research could combine related industry (for example: computer and peripherals industry, communica-tion industry, light electricity industry, precise machine industries and bio-tech industry) to construct a set of safe-ty and health management system of all industries inside the area that applies to the industrial park.
References
1) Hayes BE, Perander J, Smecko T, Trask J (1998) Measuring perceptions of workplace safety: develop-ment and validation of the work safety scale. J Safety Res 29, 145–61.
2) Thompson RC, Hilton TF, Witt LA (1998) Where the safety rubber meets the shop Floor: a confirmatory model of management influence on workplace safety. J Safety Res 29, 15–24.
Industrial Health 2008, 46, 575–581
Table 3. The six factors of safety and health management problems in the semi-conductor industry
Factor Explanation
1. The company should make employee clearly understand the safety and health laws and stan-dards
When a company executed safety and health management, it should put efforts on how to make the basic employee understand what the company has done in the safety and health management. And it should hold the educational training and emergency responses actively.
2. The company should make the safety and health management policy known to the public
The supervisor should make a public explanation and commitment of the safety and health policies, and this will be a key factor to the success of the safety and health performance.
3. The company should put emphasis on the pursuance of the safety and health manage-ment laws
The performance of the safety and health management depends on the supervisor’s altitude towards active or passive; compiling to the standards or meeting the employee’s real need. Therefore, in addition to empha-size the training, carrying out the safety and health management actively and progressively is a good plan to reduce risk.
4. The company should prevent the accidents
The different altitude for accident prevention of personnel comes from the differences in the work content. That is, the basic personnel think the accident can be prevented actually, but not for the middle and high level personnel. It is very important for the higher manager to commit the promise about the safety and health management. Only such, the accident prevention can be carried out to all levels and the purpose of effective prevention accident can be reached.
5. The safety and health message should be communicated suffi-ciently
One of the reasons for better safety and health performance is that the related safety and health information can be communicated fully across the company. The makes the personnel inside company can clearly know related safety and health measurement and correction. The communication of safety and health information in semi-conductor industry counts for much. Both the internal message exchanges in the company and the external message exchanges among companies would reflect to the safety and health performance directly or indirectly.
6. The company should consider safety and health norm com-pletely
The understanding of the safety and health standard by the employer is different from the true meaning of the standard itself. Different personnel does not explain safety and health norm in the same way. Therefore, they need a specific safety and health standard to approach the special need of semi-conductor industry.
3) Sherin B (1998) Comprehensive risk management for IC Fabs., Solid State Tech 41, S11–S8.
4) Reason J (1992) Human Error. Cambridge University Press, Cambridge.
5) Reason J (1997) Managing the Risks of Organizational Accidents. APC, USA.
6) Rahimi M (1995) Merging strategic safety, health and environment into total quality management. Int J Ind Ergon 16, 83–94.
7) Deming WE (1982) Out of the Crisis. MIT Center for Advanced Engineering Study, Cambridge, MA. 8) Kennedy R, Kirwan B (1998) Development of a Hazard
and Operability-based method for identifying safety management vulnerabilities in high risk systems. Saf Sci
30, 249–74.
9) Hee DD, Pickrell BD, Bea RG, Roberts KH, Williamson RB (1999) Safety Management Assessment System (SMAS): a process for identifying and evaluating human and organization factors in marine system operations with field test results. Reliability Eng System Saf 65, 125–40.
10) Nivolianitou ZS, Papazoglou IA (1998) An auditing methodology for safety management of the Greek process industry. Reliability Eng System Saf 60,
185–97.
11) Guastello SJ, Guastello DD (1988) The Occupational Hazards Survey: second edition manual and case report. Guastello & Guastello, Milwaukee.
12) Zohar D (1980) Safety climate in industrial organiza-tion: theoretical and applied implication. J Appl Psychol
65, 96–101.
13) Sandman BA (1992) The measurement of job stress: Development of the Job Stress Index. In: Job satisfac-tion: how people feel about their jobs and how it affects their performance, Cranny CJ, Smith PC, Stone EF (Eds.), 240–54, Lexington Book, New York.
14) Smith CS, Kruger T, Silverman GHM, Hayes BE, Silverman M, Mattimore L (1992) A research method for assessing industrial accidents. Paper presented at the annual convention of the Human Factors Society. 15) BSI (1999) Occupational Health and Safety Assessment
Systems: guidelines for the implementation of OHSAS 18001.
16) BSI (1996) BS 8800 Guide to occupational health and safety management systems.
17) Montgomery DC (2005) Design and Analysis of Experiments, 6th Ed., 94, Wiley, New York.
