Journal of Research in Health Sciences

• Home
• About
• Log In
• Register
• Search
• Current
• Archives
• AuthorGuidelines

Home > Vol 8, No 2 (2008) > Azadeh

JRHS 2008; 8(2): 28-39

Copyright © Journal of Research in Health Sciences

An Integrated and Multivariate Model along with Designing Experiments Approach for Assessment of Micro- and Macro- Ergonomic Factors: The Case of a Gas Refinery

Azadeh A (PhD)a, Mohammad Fam I (PhD)b, Sadjadi M (Ms)c, Hamidi Y (PhD)d, Kianfar A (Ms)e

a Research Institute of Energy Management and Planning, Dept. of Industrial Engineering and Dept. of Engineering Optimization Research, Faculty of Engineering, Tehran University, Iran.

b Department  of Occupational Health and Faculty of Health, Hamadan University of Medical Science, Iran

c Department  of Industrial Engineering and Department of Engineering Optimization Research, Faculty of Engineering, Tehran University of, Iran.

d Department of Health Management, Faculty of Health, Hamadan University of Medical Science, Iran

e Department  of Occupational Health and Faculty of Health, Tehran University of Medical Science, Iran

*Corresponding Author: Dr Iraj Mohammadfam, E-mail: iraj_f@yahoo.com

Received: 2 July 2008; Accepted: 11 September 2008

Abstract

Background: The objectives of this paper are three folds. First, an integrated framework for designing and development of the integrated health, safety and environment (HSE) model is presented. Second, it is implemented and tested for a large gas refinery in Iran. Third, it is shown whether the total ergo­nomics model is superior to the conventional ergonomics approach. This study is among the first to examine total ergonomics components in a manufacturing system. 

Methods: This study was conducted in Sarkhoon & Qeshm Gas refinery- Iran in 2006. To achieve the above objectives, an integrated approach based on total ergonomics factors was developed. Second, it is applied to the refinery and the advantages of total ergonomics approach are discussed. Third, the im­pacts of total ergonomics factors on local factors are examined through non-parametric statistical analysis.

Results: It was shown that total ergonomics model is much more beneficial than conventional ap­proach. It should be noted that the traditional ergonomics methodology is not capable of locating the findings of total ergonomics model.

Conclusion: The distinguished aspect of this study is the employment of a total system approach based on integration of the conventional ergonomics factors with HSE factors.

Keywords: Health, Safety, Environment, Ergonomics, Gas refinery

Introduction

The main objective of this research was to present a comprehensive model in connec­tion with designing and analyzing the macro and micro factors of ergonomics. Since, there are a lot of factors in ergonomic design of work­place both in micro and macro parts, it seems inevitable to consider a model which in­clu­des all related factors.

Measurement methodology, analysis tech­niques, all factors and finally presentation of executive solutions are taken into considera­tion in designing this model. Because of us­ing the statistical methods and multi analy­ses as strong tools of result analysis, this model is different from the previous ones. Further­more questionnaire data, direct as­sessment of en­vironmental factors, compari­son of safety fac­tors in refineries, compari­son of ergo­nomic and macro ergonomic factors among work groups and shifts and anthropometry data are simultaneously used. Therefore, decision mak­ing and presenting effective solutions are eas­ier and more pre­cise.

Introduction of subdivisions and basic model

The ergonomic and macro ergonomic analy­sis model has a process approach to macro and micro ergonomic factors. The initial data is acquired by questionnaire, direct meas­ure­ments and data sources, existing as four dis­tinct inputs

Questionnaire subdivision

In this part, questions about quality of envi­ronmental indices (light, temperature, hu­midity, etc), musculoskeletal disorders and personal protective equipment are asked. These ques­tion­naires should be designed ac­cording to the type of activity in investigated organiza­tions. It is suggested to use Henrich and Nodric standard questionnaires, which are scientifi­cally valid and standard and also widely used in many ergonomic studies (1-3).

These questionnaires must be adapted to all industrial, production and service depart­ments; also, obtained data should be gath­ered in re­lated groups, to be used in later analysis. Fur­thermore, accuracy of answers can be evalu­ated by setting common ques­tions in ques­tion­naire. Figure 1 shows the first part of model's input.

Figure 1: Questionnaire subdivision of ergonomic and macro ergonomic factors' analysis

After acquiring data by parametric and non- parametric tests and plenty frequency dia­grams, the obtained answers will be evalu­ated. In this part, a lot of variant data will be obtained as self reporting data. Due to per­sonnel age, education and attitudes diver­gen­ces, unreal factors may be added to the model. The final objective of this subdivi­sion is to process and filtrate the obtained data. The ques­tionnaires can be distributed either to all of the organization's personnel or a group of personnel who are randomly chosen. Obvi­ously, applying the former will provide more data into the model and its re­sults will be more precise. Taking the latter, the sample must contain all jobs; therefore each job should have one representative. When several people do the same job, they should be randomly cho­sen. Although the second way is less precise, it is less expen­sive and faster.

Environmental indices subdivision

Another subdivision of the new model was environmental indices data. In this section data would be summarized with a special form and compared to the available standards. This sec­tion's output might be consistent or non con­sistent with international standards. Con­sid­ering environmental indices -in a way that per­sonal attitudes such as lighting, tempera­ture, humidity, etc are taken into account-is one of the significant point about this model (Figure 2). 

Figure2: Environmental subdivision of ergonomic and macro ergonomic factors' analysis model

For indices which bring up no consistency, it is expected to receive more complaint about the un­suitable environmental factors. But it is essential to say that unsuitable environ­men­tal indices have sometimes indirect in­flu­ences, for example insufficient light in work­place might cause accidents.

Safety subdivision

The third part of the model introduces the safety subdivision, which compare organiza­tions with other similar organizations (peers) nationally or internationally (Fig. 3). A Multi analysis is also used in this section. Since in the previous sections ergonomic and macro ergonomic standards were taken into con­sid­eration and analyzed, in this section the or­ganization is assessed in comparison with the other organizations. Due to work complex­ity, astronomical costs, increasing of final price and lack of required technology, reaching the mandatory standard level is impossible (4). Therefore real criteria can be for the indices, which are in disconformities with standard, organization assessment touchstone.

Figure3: Safety subdivision of ergonomic and macro ergonomic factors' analysis model

This approach is applied to the cases, which were ignored in previous studies or do not have a statistical basis. Using of Multi Analyses (PCA & Numerical Taxonomy) evaluates the organization's position among its peers and implicitly emphasizes on the improvement of factors, which merely have deviation with due attention to the world's reality (5). If we directly want to gather data and facts from peers ,a great deal of money and time will be spent and it almost brings the economic jus­tification of ergonomic' improvement struc­ture under question. Therefore, in this sec­tion we have to make use of available data even though it is very limited. One of the most im­portant domains, in which it is possible to compare the organization with other peers, is safety. Since in the most of organization such as refineries, power plants, and military sites, etc the first error is considered as last one, there is no margin of safety for safety issues; therefore it is inevitable to compare organization to the less risky ones.

Ergonomic and macro ergonomic factors' subdivision

The 4th subdivision of the model is to design the matrix of ergonomic and macro ergo­nomic indices between work shifts on one hand and work groups on the other hand (Fig. 4). The work groups include jobs, which are in each work shifts. Contrary to the 3rd subdivision, that compares organization to the peers, this section will compare the internal parts of or­ganization according to the designed indices. One of the advantages of this subdivision is determining the priority of improvement proc­ess and execution of ergonomic for organi­za­tion's parts or individuals, which are in unac­ceptable condition. (PCA & Numerical Tax­onomy) is also used to rank work groups Multi Analyses.

Figure 4: Ergonomic and macro ergonomic subdivision of model

In this main structure, outputs of A, B, C, D in addition to anthropometry's data is present to­gether and provide executive solutions to im­prove the organization's work condition (Figure 5). This executive recipe has two im­prove­ment aspects:

1. engineering improvement
2. managerial improvement

Figure 5: Comprehensive model of ergonomic and macro ergonomic factors' analysis

Engineering im­prove­ment includes, redesigning or replacement of tools, equipment, work station, packings, etc (6). These improvements can be very ef­fec­tive, because they can omit or decrease the principal reasons concerning effective factors. The best time for applying engineering im­provement is at the time of installations, proc­esses designing or new design methods.

Managerial improvement contains changing working operations or organizational work­ing methods (7). The improvement may not state the relevant reasons or other problems.  Managerial improvement usually needs con­tinuous management and employees' feed­back with the intention of having confidence on policies and new operations' efficiency.

Whit respect to the fact that this model con­tains all ergonomic factors, no repetitive sug­gestion-which has later modified the ergo­no­mic indices in the other way, will be given. Besides, by performing each suggestion, it defines all effects of executive recipes in other domains. Therefore the speed and cost of ergonomic interference performance will be decreased.

Having presented and performed executive solutions in work circumstances, the indices must be again evaluated after a specific pe­riod and this will continually go on. In this process, the internal organization's parts al­ways improve independently in relation to each other; the organization's total procedure in comparison with the organization patterns will also be studied to get the organization's development relative rate.

Methods

This study was conducted in Sarkhoon & Qeshm Gas refinery- Iran in 2006.  The in­te­grated model was implemented in Sark­hoon & Qeshm Gas refinery. Sarkhoon & Qeshm Gas refinery was established to pro­vide a part of country needs of energy, and con­sump­tion of natural gas as an environ­mental friendly energy, which is suitable for indus­trial and civil consumption in Hor­muzgan and Kerman provinces. This refin­ery began to work in 1978 with producing refined natu­ral gas and gas liquids in Sork­hoon and Ga­varzin, which are located in Bandarabbas and Qeshm. This refinery has been taken into consideration as a case study for exe­cuting the integrated model. Due to a great deal of data in different parts, only a few as­pects of subject have been considered.

The control room of Gheshm and Sorkhoon refinery has different panels, which are re­lated to the special parts and all of the parts are coordinated by controllers and display­ers. Generally, a refinery consists of the fol­lowing units:

Separation

Sweetening

De moisten (adjustment of the dew point)

In order to apply the last subdivision's con­ceptual model in organizations, a system must be designed, which via model's subdi­visions accesses and processes the intro­duced indi­ces and factors by evaluation techniques. This system can be designed with two different ap­proaches (8):

Traditional systems

These systems have been used since many years ago and one or several units of organi­zation perform their task collaborate (9).

Suppose that organization X which has sys­tem analysis, ergonomic, safety and health or­ganization's unit, want to apply the intro­du­ced model. According to this approach, the matrix of different organization's subdi­vi­sions involves on the basis of organization nature and their rolls in diverse model's sub­divisions (Table 1).

At first, system analysis division gathers the required data. To do this, questionnaires are copied according to number of each sub­divi­sion questioned personnel and will be dis­trib­uted among them by system analysis ex­perts.

Table 1: Organization matrix

Then, the completed questionnaires are col­lected, the responses of questionnaires as quantitative data are entered to a relative software by the system analysis division, and figures and diagrams are made by them to be used in later analysis. On this point the sys­tem analysis division can show a great deal of present organization's problems by de­signing the proper hypothesis tests and plenty frequency diagrams. After this proc­ess, the output of questionnaires subdivision is prepared to be entered into the other or­ganization's units: ergonomic, safety and health, and each of these units in conjunction whit system analysis division in a parallel manner accomplish the related subdivisions according to the above matrix.

The ergonomic subdivision performs the process of D subdivision with the aid of system analysis division and detects the ad­vantages and disadvantages of organization in ergonomic and macro ergonomic issues. Evaluating the work safety factors and an­thropometric data are also done by safety and health experts. Finally, all data is en­tered into the main model, and a procedure will be planned according to each division's weakness and strength. After change exer­tion, aforesaid indices and factors are again evaluated and improvement quantity is studied and required changes are possibly exerted by systems, safety and health divi­sion's experts or executive managers in the case of macro ergonomic factors. Also, the safety and health subdivisions accomplish B&C subdivisions. The indices of environ­mental factors' subdivision can be measured by the safety and health experts of the or­ganization. For the purpose of easier deci­sion-making, data which are mostly: light, sound, thermal stress and environmental pollutants, are compared to the national and international values and weakness and strength points are mentioned. In this subdi­vision "the analysis of organization's present condition by PCA & Numerical Taxonomy" is only allocated to the system analysis or­ganization's unit and is done by this unit. Fi­nally with the subdivisions' fulfillment, the related results enter into the main model and executive solutions are detected with the co­operation of organization's units' common com­mittee: systems analysis, ergonomic, safety and health and are communicated to the re­lated unit's manager in order to accom­plish. It is essential to mention that ,after gathering other required data and presenta­tion of ex­ecutive solutions in respect of or­ganization's dynamic and changeable nature ,system re­evaluation & copying , delivering and sum­marization of questionnaires is in­evitable. Due to expensive and long evalua­tion period and complexity of administrative affairs, it is very expensive, and these are counted as its disadvantages. Therefore, it is essential to use modern systems and new technologies. The disadvantages of tradi­tional systems are:

High costs of execution

Long periods of factors evaluation

Exerting personal taste

Imprecise data and achieved results

Unreliability to data and gained results

The modern technology based systems

As it was mentioned in previous section, the application of the discussed model in organi­zations with traditional methods imposes problems and costs (10). With appearance of modern technologies in present epoch such as information technologies and information systems, application of discussed general model can be executed and used with less costs and problems (11). In this section, the manner of model application in organization is given by using of information technolo­gies and information systems. On the basis of web technology, firstly information sys­tem of questionnaire delivering and collect­ing web based is designed and set at the re­lated organization. On this basis, the person­nel of different organization's unit are con­nected to the system by username and pass­word and the relative information system. Then, the gathered responses are saved in the information system basis. The responses are changed to quantitative data ,then A subdivi­sion is performed in system analysis organi­zation's unit by connecting information sys­tem to the statistical and mathematical soft­ware's such as Microsoft Excel, SPSS, SAS, etc and  decision support sys­tems and sta­tistical-mathematical analysis. This process continues by connecting infor­mation system to the steps of B, C and D model subdivi­sions. Finally in order to com­pile executive solutions by data sources, or­ganization tools and solutions are selected from tools and solutions, and are electrically communicated to the managers and organi­zation units' per­sonnel by web based infor­mation system. The electrical performance of model proc­esses is very helpful to pursue the gained re­sults and determine the proper decisions and actions, and these are the electrical perform­ance advantages of model:

Fast and precise performance of subdivi­sions' and total models' different processes Low price of model performance in organi­zation

Ability of making and producing different in­formation by complex and intelligent analyses

Results

As it was mentioned, the questionnaires are the most common way for collecting data and providing information for the model. In the first subdivision, the systematic solution of data analysis was studied. We can analyze data by the introduced indices and tools. We study all of the indices, which were in the questionnaire, by frequency diagrams. Then characteristics between shifts and work groups are shown by parametric and non­parametric statistic techniques. Finally, positive and ne­gative points are explained. 

As in the introduction of Gheshm and Sork­hoon refinery was mentioned, Sarkhoon & Qeshm gas refinery has 4 work shifts. Every shift has 8 h and shifts change periodi­cally. Each group works continuously 3 weeks and rests 1 week. In Sarkhoon Gas refinery, each group filled the questionnaires separately; therefore, we can compare the shifts. In this part, we will exam, if there is a meaningful relation between work shifts and personnel's complaints about tiredness and work.

In Sarkhoon refinery, 60 people of the ex­ploi­tation unit have been asked. Gheshm-Gavar­zin refinery is physically and on the point of work shift different from Sarkhoon refinery. Gheshm-Gavarzin refinery has two 12 h work shifts and 4 persons work on average in each shift in the exploitation unit. Because the few number of the personnel, the work shifts aren’t separated. The real reason of this prob­lem is the few number of statistic population, which effects analysis results of the compari­son between shifts. Each group works 2 weeks and rests for 2 weeks.

According to the personnel's answers, 67.7% of personnel are satisfied with the program of work shift. Nowadays, designing of the work shifts is one of the most important er­go­nomic discussions. Because the type of work shift has a great effect on the natural circle of the body such as circle of Cir­cadian, Infra, ultra dine. In the case of wrong design, it ca­uses mentally and physically damaging effects.

As mentioned, the groups, which were intro­duced as work shifts, has been investigated in a special section. So, their answers cannot completely show the effect of shifts. In the following part, the tables of chi square test are shown separately according to the ques­tionnaires (Table 2). Every question can show, if there is any relation between shifts and answers. Then, we can finally conclude by number of questions and cases, which show relation or lack of relation.

Table 2: Chi square test for ergonomic questionnaire

It is obvious, that the atmosphere frequently changes out of the control room at different work shift and this matter gives meaning to the above question. The most important changes are changes of relative moisture and radia­tion temperature. This matter inconven­iences the per­sonnel because of its quick and acute effects contrary to other long lasting ef­fects.

According to the chi square test, it was con­cluded that shifts cannot make block. They have been chosen from a uniform popula­tion. Chi square test only considers the inde­pend­ence of shifts and it doesn’t point the rela­tion between the averages. Kruskal-Wal­lis test stud­ies the equality of population sev­eral aver­ages hypothesis. Kruskal-Wallis test is a non- pa­rametric synonym for one-factor analysis of variance. The test uses the zero hy­potheses of same distributive populations against the hy­pothesis, which expresses that some popula­tions produce bigger observa­tions. The method has been designed in a man­ner that averages' difference is sensitive to the test. It is better to take the Kruskal-Wal­lis test into con­sid­era­tion as the test of equality of different popu­la­tions' averages. The results of test can prove that the shifts don’t make block (Table 3).

Table 3: Kruskal-Wallis test for ergonomic questionnaire

If we accept the hypothesis that our data is normal, we can test the above question on the basis of one-factor analysis of variance. The results of the test have been presented for 2 questions of the ergonomic question­naire (Table 4).

Table 4: one-factor ANOVA for ergonomic questionnaire

Analysis of environmental indices

Here, assessment of noise is evaluated as one of the environmental indices. In the control room of Gheshm and Sorkhoon gas refinery and Gourzin's installations, noise measure­ment by level meter of model and Quest calibrator is done to provide a dis­trict map and finally compare to the per­mis­sible boundaries of noise (Fig. 6). The noise map of the refinery's control room is as fol­low:

Safety limit (SPL<65dBA) with S code

Precaution limit (65<SPL<85dBL) with C code

Danger limit (SPL>85dBL) with D code

Figure 6: The noise map of refinery control room

As observed, according to the national stan­dards of Iran about permissible boundary of noise for 8 h, noise is not harmful on the point of injuries caused by noise. Neverthe­less, according to the filled questionnaires, 43.3% of persons inside the control room and 81.7% of persons out of the control room know the noise harmful.

Analysis of the safety indices

As in the 3rd subdivision of the comprehen­sive model of ergonomic and macro ergo­nomic factors analysis was explained, the aim of this part of model is to compare Sarkhoon & Qeshm Gas re­finery to other gas refineries in regard to the introduced factors (Table 5). The most important Sarkhoon & Qeshm Gas refinery's indices, which are at data bank, are related to the safety indices. Because of the cost and time limitations, we content with the data of this part. Although in this man­ner, we compare only one special aspect of refineries to each other, other aspects can be gradually studied with the information ex­pansion.

Table 5: Some of the safety indices of gas refinery

Indices, which are accessible, are:

Number of events

Wasted days according to the above events

Fire accident

This information is gathered in 1380 and 1381.

Refineries, which are chosen as the compari­son pattern, are:

Asalooyeh gas refinery

Fadjr gas refinery

Bidboland gas refinery

Khangiran gas refinery

In addition to the comparison between re­fineries, gas refining companies as a set of gas national company have been compared to three sub-set companies in regard to indi­ces (Table 6). The below comparative table shows indices and frequency rate of each index. Like the previous case, theses indices are related to 1380 and 1381.

Sarkhoon & Qeshm refinery had gained the second place among four other refineries dur­ing 1380 and 1381 by both methods of  PCA and Numerical Taxonomy, which shows the high place of this refinery in the vie of na­tional safety indices.

Table 6: ranking the refineries on the basis of safety indices

Discussion

The comparison of internal companies of gas national company shows a great difference by PCA method and Numerical Taxonomy (12). The PCA method is generally more ef­ficient than other methods and it is taken into consideration as the basis for this section. Treatment company and gas works stand re­spectively in the last place and the last place but one. These results are expected due to the hazardous work type of these companies in comparison with central administration, warehouses and pipeline affairs. However, on account of these results, it is necessary to consider and lead the gas national company's investment to promote safety indices grade in refineries all over the country, and in re­gard to refineries' ranking results, it is vitally significant to lay stress on Khangiran, Fadjr and Bideboland refineries.

In the 4th subdivision of the comprehensive model, the ergonomic and macro- ergonomic matrix was introduced. This subdivision ranks work shifts and some of equal indices. As it was explained in the questionnaire analysis, work shifts haven’t usually mean­ingful dif­ference in the light of statistic, and its reason was explained in detail. But this time even though few differences were studied with a view to PCA and Numerical Taxonomy. Be­low, work shifts and work groups are stated as the under study sample (DMU), so evalu­ating indices are. Pains and some of macro- ergonomic indices are stud­ied as samples.

For the pain's matrix scoring is on the basis of following points, which are considered for each answer. Meanwhile, Gourzin group score is multiply by 15.12 on account of dif­ference in the statistic population in order to standardize.

Following ranking is obtained for work shifts in connection with variable indices, on the basis of the main component analysis and Numerical Taxonomy.

As it is shown, morning and afternoon work shifts stand respectively in the lowest places among shifts concerning pains indices, and this is a significant factor for designing and compiling work shifts program of Govrzin and Sorkhoon gas refinery. It is essential to state that Gourzin has the best rank in these lines.

For macro ergonomic indices ,Govrzin work shift stands again in the first place, due to the smallness of  the refinery and the simplicity of organization's structure (although ,it is one part of Sorkhoon refinery) simple rela­tion between managers and operators ,this work shift has the highest point. Though in the case of the most critical work shifts ,two work shifts are different in the light of voca­tion satisfaction ,it is clear that one of the daily work shifts (morning or afternoon) suf­fers from vocation dissatisfaction ,and be­cause of managers  presence in these work shifts in the refinery, it is vague.

This problem should be realized and studied, and after exerting the executive solutions, these indices should be again evaluated. The distinguished aspect of this study is the em­ployment of a total system approach based on integration of the conventional ergonom­ics factors with HSE factors.

Acknowledgements

The authors would like to thank all personals of the Sarkhoon & Qeshm Gas refinery who participated in survey and kindly responded to the questionnaire. Our appreciation also goes to the executives of the Sarkhoon & Qeshm Gas refinery for the support in fa­cilitating the data collection process.

The authors declare that they have no conflict of interests.

References

1. Bongers PM, Kompier MJ, Hildebrandt VH. Psychosocial factors at work and musculoskeletal disease. Scandinavian Journal of Work, Envi¬ronment and Health. 1993; 19(3): 297-312.
2. Smedley J, Egger P, Cooper C, Coggon D. Manual handling activities and risk of low back pain in nurses. Occupa¬tional and Environmental Medicine. 1995; 52 (1): 160-63.
3. Fadier E. Safety design: towards a new philosophy. Safety Science. 2006; 44(4): 55-73.
4. Hendrick HW. Ergonomics in organ¬iza¬tional design and manage¬ment. Ergo¬nomics. 1991; 34(6): 743-56.
5. Winsor MP.The English Debate on Tax¬onomy and Phylogeny, 1937-1940. History and Philosophy of the Life Sci¬ences. 1995; 17(2): 227-52.
6. Kolluru R, Bartell S, Stricoff R. Risk Assessment and Management Hand¬book. McGraw-Hill Inc, Canada, 1996: pp. 102-43.
7. Hopkins A. What are we to make of safe behavior programs? Safety Sci¬ence. 2006; 44(3): 583-97.
8. George JM, Jones GR. Organizational behavior. 3rd. Prentice-Hall, New Jer¬sey, 2000: pp. 80-90.
9. Laurence D. Safety rules and regula¬tions on mine sites – the problem and a solution. Journal of Safety Research. 2005; 36(1): 39–50.
10. Hasle P, Limborg HJ. A review of the literature on preventive occupational health and safety activities in small enterprises. Industrial Health. 2006; 44 (3): 6-12.
11. Hernberg S. Introduction to occupa¬tional epidemiology. Lewis Publishers, Chelsea ,1992: pp. 42-76.
12. Azadeh A. An integrated DEA PCA numerical taxonomy approach for en¬ergy efficiency assessment and con¬sump¬tion optimization in energy inten¬sive manufacturing sectors. Energy Policy. 2007; 35(7): 3792-806.