Keywords

1 Introduction

Industry workers performing manual operations are subjected to musculoskeletal disorders (MSD) [1, 2]. The high prevalence of MSD symptoms as a significant problem among Malaysia automotive workers [3,4,5]. The occupational health and safety management system (OHSMS) has been known to play an essential role in controlling the safety conditions of workplaces and health of the employees in the companies [6, 7]. However, occupational safety and health (OSH) practitioners appear to focus on checking the safety and health aspects rather than zooming in on getting to the human factors or ergonomics issues.

Most of the companies had no information with regards to the health, safety and ergonomics performance [8]. Thus, WMSDs consistently continues as one of the occupational safety and health (OSH) related problems due to OSH and ergonomic intervention have not been wholly implemented [9]. In particular, with regards to the OHSMS, the focus should not be only on the development of healthy and safe working conditions, but also equally to the workplace comfort and wellness, and employee’s well-being. Thus, a practical approach to predict workplace ergonomics risk exposure can be beneficial to occupational health and safety practitioners and production managers to prevent WMSD symptom among production workers. Therefore, this research is an effort to develop a knowledge-based ergonomics assessment system (KBEAS) that assists in evaluating the potential of WMSD risk factors at an automotive component manufacturer.

The proposed algorithm of the system is based on AHP method (T. L. Saaty 1980) for the rules and inference system. The AHP was used for multi-criteria decision-making process and provides reasonable support [10] and flexible approach to risk analysis [11]. Since decision maker can unbiasedly get numerical pair-wise comparisons by choosing proper numerical scale to quantify linguistic pair-wise comparisons [12]. The web-based system was set up as ergonomics assessment tool since it was fundamental for such a medium to be open efficiently and auspicious [13]. Web-based system application has upgraded the effect of a decision support system when the choice available is considered by an expansion in the quantity of other options to assess [14].

The final experiment results show that KBEAS could use to predict a particular risk factors level to obtain several appropriate measures proposed by OSH practitioners in engineering and administration controls. Consequently, a smart and practical tool can be developed to identify WMSD risk related to a variety of factors.

2 Knowledge Acquisition Process

The required information and knowledge to construct the knowledge base are obtained through knowledge acquisition process (KAP). KAP consists of a mixture of knowledge acquisition methods. The preliminary knowledge acquisition gains an overview of the workplace ergonomics problem. The WMSD symptom knowledge acquisition finds about the body pain complaint among production workers.

Initially, the potential risk factors are obtained from literature associated to back, neck, shoulder, and arm pain. The literatures contain a wealth of information about the possible causes of MSD [15,16,17,18]. Based upon the expertise, experience, and accessibility the group of subject matter experts (SME) consists of a managing director and general manager, as well as department managers from production, engineering, and safety health and environment are selected for tacit knowledge analysis. The recognized risk factors and common characteristics are acquired from the literature that represents the categories individual-related, organizational-related, physical-related and psychosocial-related risk factors.

To validate the risk factors related to WMSDs, the survey is considered useful in identifying the dominant risk factors. The questionnaire was distributed to practitioners at three automotive production plants. The consequences of the knowledge acquisition lead to the generation of various risk factors. The final risk factors thus acquired are summarised to formulate the AHP hierarchy, as shown in Fig. 1.

Fig. 1
figure 1

AHP hierarchy structure of ergonomics risks factors

Full size image

3 AHP Methodologies

The risk factors related to WMSD are diversity, various, and dynamic. Based on that, to achieve intelligence and dynamic in early detection and meantime facilitate employees to make an objective descriptive of subjective judgments, the current study applies AHP methodology. The pair-wise comparison is the relative importance of one criterion over another in meeting a specific goal [13, 19]. Given those, current study set up ergonomics assessment model of WMSD risk factors to do the quantitative analysis. The AHP-based method involves the following significant steps:

1st Step: Develop a hierarchy of factors based on KAP results (refer Fig. 1).

2nd Step: Construct judgement matrix and make pair-wise comparisons: Practitioners are asked to evaluate the relative importance of risk factors within and among the leading ergonomics factors. A matrix of element evaluation, denoted as A, will be formed using the comparisons. Each entry aij of the matrix, in the position (i, j) is obtained comparing the row element Ai with the column element Aj: Means of pair-wise comparison administered the survey questionnaire. A numerical scale is then assigned to each pair of alternatives (Ai, Aj) by the experts (refer Table 1).

Table 1 Pair-wise comparison scale used with AHP adapted from [3]
Full size table

3rd Step: Synthesize judgments: The process is to calculate a vector of local weights or priorities of each risk factor in term of its contribution to the overall goal. It includes the following steps.

  1. Step 3.1

    Once the overall expert judgments are created and calculated using the geometric mean [refer Eq. (1)],

    $$ {\text{Geometric}}\,{\text{mean}}, {\text{GM}}_{\text{i}} = \sum\limits_{i = 1}^{n} {n\sqrt {a_{ij} } } = \sqrt[n]{{a_{ij\,1} x\,a_{ij2} x \ldots x\,a_{ijn} }} $$
    (1)

    where n = number of participants

  2. Step 3.2

    The next step is to calculate a vector of local weights or priorities of elements [refer Eq. (3)]. The principal eigenvector w of the matrix can be calculated using the Eq. (2).

    $$ {\text{Eigenvector = w}}_{\text{i}} \frac{{GM_{i} }}{{\sum\nolimits_{i = 1}^{n} {GM_{i} } }} $$
    (2)
    $$ {\text{Eigenvalues}}\,{\text{ = local}}\,{\text{weight = }}\frac{{\sum\nolimits_{\text{i}}^{n} {w_{\text{i}} } }}{n} $$
    (3)

    where n = number of elements

  3. Step 3.3

    The synthesized weight or global priority vector for elements can be presented as in the Eq. (4):

    Global priority vector, Wi;

    $$ {\text{W}}_{\text{i}} = \left( {{\text{Criteria}}\, {\text{local}}\, {\text{weight}}} \right)\left( {{\text{Sub - criteria}}\, {\text{local}}\, {\text{weight}}} \right) $$
    (4)

4th Step: Upon having the local priority vector determined, it is then necessary to evaluate the consistency of the pairwise comparison matrix. The consistency index and consistency ratio can be seen in the Eqs. (5) and (7).

$$ {\text{Consistency}}\,{\text{ index }}\left( {\text{CI}} \right) \, = \frac{{\lambda_{ \hbox{max} } - n}}{n - 1} $$
(5)
$$ \lambda_{ \hbox{max} } = \sum\limits_{i = 1}^{n} {\left[ {\left( {\sum\limits_{i = 1}^{n} {GM_{i} } } \right)\left( {w_{j} } \right)} \right]} $$
(6)

where, ʎmax = maximum eigenvalue and n is the number of elements

$$ {\text{Consistency}}\,{\text{ ratio }}\left( {\text{CR}} \right) \, = \frac{{{\text{Consistency index }}\left( {\text{CI}} \right)}}{{{\text{Random index }}\left( {\text{RI}} \right)}} \le 0.10 $$
(7)

If the value of consistency ratio (CR) is smaller or equal to 0.1 the inconsistency is acceptable. If the CR is greater than 0.1 then the subjective judgment need to be revised.

4 KBEAS Development

In the present convenience working environment, it is fundamental to propose and set up a reasonable system which will consolidate the power of web-based systems, AHP analysis, and empirical data and therefore advance the efficiency of prioritizing processes. The KBEAS was designed as a web-based system since it was planned to accessible anytime and anyplace inside the organization.

4.1 Structure of System

The basic structure of the web-based system includes the user interface, a knowledge base, and an AHP inference engine. The proposed system is implemented by using XAMPP and MySQL server is used as a database. XAMPP is a free and open source cross-platform web server solution stake package. XAMPP consists mainly of the Apache HTTP Server, Maria DB database, and interpreters for scripts written in the PHP and Perl programming languages. The structure of the developed web-based system for ergonomics assessment purpose is given in Fig. 2.

Fig. 2
figure 2

Web-based of knowledge-based ergonomics assessment system structure

Full size image

Component of the system. Figure 3 demonstrates the system components and their corresponding functions. The system has four major components. First one is a database component in which all factors information are stored. Users would retrieval be able to the risk factors record by time and embed new record.

Fig. 3
figure 3

Components of the KBEAS and their corresponding functions

Full size image

The second component is data input part. Concerning user’s determination comparing data is recovered from the database. The third component is a data processing part which allows the server to use AHP method to calculate each criterion and sub-criteria weights. The consistency test for data input also performed. This component can retrieve the corresponding numerical values regarding the submitted specific condition about those criteria and sub-criteria mentioned previously. The fourth component is data output part. This component is to prioritize for criteria and sub-criteria of risk factors. Also, rank all the criteria and sub-criteria within the group and all groups of factor.

Section of a web-based system. The system is divided into two sections. One of them is administrator side and the other one is user side. Administrator or user can access the system wherever he/she needs. The flow diagram of the system is illustrated in Fig. 4. Under administrator section, administrator of the system makes some definitions on the purpose of use. These definitions are as follows: remove a record, remove a user, modify risk factors, modify questions, and adding a legend.

Fig. 4
figure 4

Flow diagram of the system

Full size image

Company’s Safety Health Officer (SHO) User: Safety Health Officer User has primary responsibility for ergonomics risk report for the production plant. Safety Health Officer can login both administrator and user. Other Users: system’s other defined users, registered user who is employee representative from the related department and had authority over the ergonomics assessment.

4.2 System Application

To validate the proposed KBEAS, an application in a local automotive component manufacturer has been done. The evaluation group was made up of fifteen practitioners from several departments under production plant, whose expertise was different. All practitioners that participated had at least 5 years of working experience in the company. The practitioners were selected according to their job tasks, roles and influences on OHSM system practices.

The system works as follows, after registering to the system (refer Fig. 5), the user can log into the system for data input purpose.

Fig. 5
figure 5

User interface: registration page of the web-based system

Full size image

Then, once the user logs into the system the comparison module page of the system appears. Figure 6 displays one out of seven comparison modules. Each question defined under different potential risks is answered and scale is selected. After the comparison module is completed, user click button “Next questions” and system allows the user to pass on to the next page. Afterward, the user clicks the button “submit and finish” (see Fig. 7), the end of questions and data input process.

Fig. 6
figure 6

User interface: comparison module for job task risk factors

Full size image
Fig. 7
figure 7

User interface: comparison module for psychosocial risk factors

Full size image

The registration page (refer Fig. 5) will appear for next data input. Once completing the comparison modules, the system calculates the risk factors weight. After the prioritization process, the system provides the user with intermediate results consisting of the priority factors and the consistency of the pairwise comparison module.

4.3 Results

In this study, the experience with the system and ergonomics risk factors data obtained from the company resulted in a priority vector as illustrated in Table 2.

Table 2 Priority results for ergonomics risk factors related to WMSD
Full size table

Based on the results acquired in the AHP analysis, the following statements can, therefore, be made:

  1. i.

    Organizational risk factors are more critical than physical risk factors.

  2. ii.

    Physical risk factors are more critical than psychosocial risk factors.

  3. iii.

    Psychosocial risk factors are more critical than individual risk factors.

  4. iv.

    Individual risk factors are less significant than organizational risk factors.

The weights for each of the factors in the hierarchy were calculated in line with their perceived contribution to the risky situation to the workers in automotive component production plant (refer Table 3).

Table 3 Ranking of risk factors associated with WMSD
Full size table

The overall prioritization revealed that ‘exposure to physical demands’ had a priority vector of 26.33%, and it was perceived as the item with the most critical factor. It is compulsory to execute a consistency validation for each hierarchy and the model as a whole for every comparison module form. In this regard, Table 4 shows consistency ratio (CR) value consistency check. Since the CR for all comparison modules is less than 0.1, the hierarchy has therefore passed consistency validation.

Table 4 Consistency validation
Full size table