The purpose of this paper is to present an empirical investigation of maintenance performance (MP) management practices from the sugar industry in India.
Empirical data for this study were collected through field visits, interviews and published reports. Statistical methods including correlation, multiple regression and cluster analysis are utilized to accomplish the objective of the study.
Explanation with multiple regression analysis showed that the sugar industry MP is significantly and positively related to maintenance approach (MA), continuous improvement (CI), financial approach and spare part management (SPM). Cluster analysis showed that sugar industries focusing on MA, CI and policy development and organization are having higher MP. The cluster analysis also pointed out that there is a substantial variation in MP due to the type of ownership (private and cooperative) while no variation has been observed due to installed capacity (low and high).
The generalization of the results obtained in this work for the sugar industry can be possible through a larger sample size.
The study contributes to the better understanding of maintenance measures in the sugar industry and provides insights on the role of maintenance managerial practices in enhancing the MP.
The findings provide empirical evidence that maintenance practices across the sugar industry are important to improve MP.
Gandhare, B., Akarte, M. and Patil, P. (2018), "Maintenance performance measurement – a case of the sugar industry", Journal of Quality in Maintenance Engineering, Vol. 24 No. 1, pp. 79-100. https://doi.org/10.1108/JQME-07-2016-0031Download as .RIS
Emerald Publishing Limited
Copyright © 2018, Emerald Publishing Limited
It is well known that effective maintenance policy is very much essential to upkeep the facilities and equipment for the better performance and profitability of an organization (Alsyouf, 2007). In manufacturing organizations, maintenance-related costs estimate 25 percent of the overall costs (Cross, 1988; Komonen, 2002). The primary aim of maintenance facilitates, intended to retain an item in, or restore it to, a state in which it can perform its required function (ISO 14224, 2004). Different approaches of maintenance performance (MP) measurement, including economical, technical, strategic, system auditing, composite formulation, partial productivity and maintainability have been reported in the literature (Tsang, 2002; Oke, 2005; Samat et al., 2011).
The nature of maintenance required varies across the industries. For example, in a fabrication unit (job shop production system), stoppage of one facility (e.g. welding machine or general purpose lathe machine) may not be as critical as that of an assembly line (mass production system) where the stoppage of one station can halt the complete line. In other words, when product varieties are larger and volume per product is small, a process layout is used where similar machines or facilities are grouped together. Hence, stoppage of one facility may not have a significant impact on production. In addition, the cost of the facility also dictates the weightage of maintenance. That is, the non-availability cost of a facility with very high initial equipment cost (e.g. CNC machine or highly automated equipment) will be more per unit time as compared to the less costly (or manual facilities).
In general, the process industry requires high capital investment and continuous availability of facilities that makes maintenance critical for the performance of an organization. The sugar industry is a seasonal process industry and availability of facilities during the sugar production period decides the performance of the organization. Hence, the maintenance of facilities in the sugar industry is important. Very few studies have been reported to understand the maintenance practices in sugar industries and its impact on the performance of an organization. In addition, differences in MP due to the factors such as type of ownership (private and cooperative) and installed capacity (low and high) has not been explored. The focus of this work is to address these issues.
The paper is organized as follows. The relevant literature review has been reported in Section 2, which focuses on identifying MP criteria and methodologies adopted by the researchers in understanding the MP such as survey or case study approach. The systematic methodology used in this research work is given in Section 3 and the results of regression, correlation and cluster analysis are presented in Section 4. Discussion and managerial implications are given in Section 5, while conclusion and future directions are given in Section 6.
2. Literature review
The review of 251 articles by Simões et al. (2011) from 1979 to 2009 clearly indicates that importance of MP measurement has increased over the last few years. Hence, a large body of literature, including the review articles are available that addresses the issues of maintenance management (Garg and Deshmukh, 2006), total productive maintenance (TPM) (Ahuja and Khamba, 2008), MP measurement (Parida and Kumar, 2006; Muchiri et al., 2010) and managing maintenance using performance measurement systems (Kutucuoglu et al., 2001). However, there is a paucity of literature on MP measurement in process industries, especially in the sugar industry. To identify the relevant literature, a combination of keywords (e.g. “maintenance performance”+“process industry,” “maintenance performance”+“sugar industry,” “maintenance”+“sugar,” etc.) has been searched on the portals of publishers such as Emerald, Elsevier, Springer, Taylor & Francis, Inderscience, Sage Publications and IEEE. The search resulted in only one article that discusses MP measurement by survey in the sugar industry (Mwanaongoro and Imbambi, 2014). However, there is a large body of literature in the domain of “MP,” including the review articles (Sherwin, 2000; Parida et al., 2015; Kumar et al., 2013; Ding and Kamaruddin, 2015). Hence, to identify the MP measurement criteria and methodologies adopted by the researchers, relevant literature has been reviewed and is summarized in Table I.
The analysis of literature clearly shows that the MP has been studied across various domains such as manufacturing, service and construction. Exploratory and case study with statistical tools like correlation, factor analysis and regression analysis has been widely used to establish the relation between maintenance practices and MP. Important maintenance practices used in evaluating the MP include maintenance approach (MA), policy development and organization structure (PDO), task planning and scheduling (TPS), financial aspects (FA), spare part management (SPM), Human Resource Management (HRM), and continuous improvement (CI). These are briefly discussed next.
In literature, MAs such as corrective, preventive, predictive, reliability centered maintenance (RCM) and TPM has been identified (Khazraei and Deuse, 2011). In corrective maintenance, the equipment is allowed to run until failure and then failed equipment is repaired or replaced. Preventive and predictive MA is used to avoid equipment breakdown. TPM seeks to improve equipment performance while continuing to avoid equipment failures. TPM and RCM are aggressive approaches that are committed for the long-term improvement of maintenance management in improving manufacturing competencies (Ahuja and Khamba, 2008). Swanson (2001) reported strong positive relationships between proactive and aggressive maintenance with MP. The availability of sugar plant machinery is very important as the failure of single equipment may stop the whole production. Hence, the type of MA (corrective or predictive) taken by THE sugar industry is crucial for its performance.
Maintenance policy is the provision of maintenance and maintenance support based on the objectives and policies of an organization. Maintenance policy should be set to support the business and manufacturing strategy (Dunn, 1996; Kelly, 1997; Wilson, 1999). The performance of the maintenance operation is given by the organization structure deployed as it controls the allocation of capabilities in maintenance function and hierarchy of communication (Kelly, 1997). PDO plays an important role in improving the MP of an organization (Raouf, 1994; Jonsson, 1997). Hence, it is essential to understand the policies (such as written policy document, frequency of review and involvement of top management) deployed by the sugar industry.
It helps in scheduling and allocating resources (man, material, equipment, tools) for the work. Well-planned and scheduled maintenance work significantly meets the objective of servicing the organizational needs and lowers the maintenance cost. Proper planning can lead to better utilization of time, less unplanned work, less overtime, which reduces the total overall cost of maintenance. Levitt, 1997; Parida and Kumar, 2009) suggested planned, unplanned maintenance task as one of the MP indicators for process industries and its importance is also highlighted in the literature (Raouf, 1994; Cholasuke et al., 2004; Mckone et al., 2001). How far TPS is being practiced in the sugar industry is important to know as it has a direct impact on the overall efficiency of the sugar industry.
This criterion evaluates MP on the basis of costs as it helps to improve the margins. Cholasuke et al. (2004) used sub-criteria such as cost of maintenance, measured cost of loss production, the average cost of maintenance per plant to capture financial aspect. Swanson (2001) measured reduction in production cost as MP output and reported that reactive maintenance increases the production cost. Similarly, another indicator of FA is costs per ton as suggested by Parida and Kumar (2006). In the context of the sugar industry, it is important to know the various facets of FA (such as maintenance budget, budget control, production loss costs due to maintenance) that are used to control the overall costs.
It plays a vital role in achieving the operational efficiency of a capital-intensive firm. Availability of critical spares during the breakdown is crucial for operational continuity and hence the business success (Louit, 2007). On the other hand, SPM requires considerable investment. Hence, an effective SPM practice can help improve the organization performance. The indicators used to analyze SPM include material and tool control (Raouf, 1994) and waiting time of spares (Visser and Pretorius, 2003). In the case of sugar industry, this can be analyzed by understanding the process of recording, controlling and managing the spare parts.
Maintenance practices are planned, organized, controlled and executed by human resource department. Many authors agreed that HRM is one of the important factors for MP improvement (Raouf, 1994; Jonsson, 1997; Cholasuke et al., 2004). This criterion has also been used as the pillar of maintenance management (Gupta and Galloway, 2003) and factor to improve maintenance productivity (Visser and Pretorius, 2003). Similarly, another indicator for this criterion includes employee complaints (Parida and Kumar, 2006) and total number of people trained (Sun, 2004). HRM in the sugar industry can be analyzed through various aspects such as training, motivation and sufficiency of people to carry out the maintenance activities.
It mainly concerns with establishing the customer requirements (internal and external), meeting the requirements, measuring success and continue to check customers’ requirements to find areas in which improvement can be made (Chang, 2005). Tsang et al. (1999) reported that CI and staff development are the key elements for the maintenance. Maletic et al. (2012) presented a conceptual model for CI in the field of maintenance using “plan-do-study-act” (PDSA). Maletic et al. (2012) reported that CI significantly and positively contributes to the MP. MP in terms of CI can be improved by the commitment of the management and employees. CI of maintenance has been measured through the management committed to CI and the adoption of proactive maintenance (TPM, RCM). Hence, it will be interesting to know whether sugar industries are also using PDSA approach and separate teams for continuously improving the MP.
Accurate estimation of equipment utilization is very important in process industries being a capital-intensive industry. The identification and analysis of hidden losses are initiated from these estimates. Nakajima (1988) proposed the overall equipment effectiveness (OEE) metric for evaluating the progress of TPM). Based on the utilization estimated, managers can identify the causes of the time losses and attempt to reduce the losses. Before OEE, only availability was considered in equipment utilization (Ljungberg, 1998).
Jeong and Phillips (2001) claim that the original definition of OEE suggested by Nakajima (1988) is not appropriate for a capital-intensive industry, and suggested the definition of OEE based on loss classification scheme. Mean time between failures (MTBF) and mean time to repair (MTTR) are the two key performance measures in plant maintenance for achieving high maintainability. MTBF gives an average time elapsed from one failure to the next while MTTR states the time that it takes to repair after a failure. In the sugar industry, MP can be found out by considering the status of OEE and the maintainability.
The overall methodology for MP measurement is depicted in Figure 1. The details of various components of the methodology are discussed in the next section.
3.1 Sample and data collection
This research work focuses on one type of industry, that is, sugar industry. It has been reported that empirical studies that focus on a single industry are useful for identifying and measuring the firm’s critical resources (Hitt et al., 2001). The use of only sugar industry in analyzing the MP is more appropriate as the measure involved in the maintenance management across the sugar industry are likely to be more homogeneous. Hence, by focusing the data collection on one class of industry (sugar), one can reduce the range of extraneous variations that might influence the constructs of interest.
The data collection for this research work was carried out during 2012-2014, where 16 sugar industries from the western region of India were visited to understand the sugar manufacturing process and the MAs adopted by these industries. Each visit involved the personal interaction with four to six respondents (chief engineer, general manager, shift engineers, turbine/boiler engineer) responsible for maintenance activities. The purposeful sampling and quota sampling is used in this study, where data from 63 respondents from these 16 sugar industries were collected, of which 11 was discarded for incomplete information resulting in the usable sample size of 54. Demographic analysis of the data collected is shown in Figure 2(a)-(c). The data collection approach is discussed in the next section.
3.2 Measurement variables
The problem of measuring the MP in the sugar industry is an important area of management research. Perceptual measurement is commonly used for this purpose where judgments (perceptions) of respondents are recorded and analyzed statistically to arrive at the conclusion (Cholasuke et al., 2004; Swanson, 2001; Ali and Mohamad Nasbi Bin Wan Mohamad, 2009; Chinese and Ghirardo, 2010). This research has adapted the measuring instrument developed by Cholasuke et al. (2004) for MP measurement in the sugar industry.
In this research work, clarity and validity of MP have been carried out in two stages. In the first stage, a questionnaire has been developed and then discussed with five experts from the sugar industry and based on their feedback the questionnaire was modified. For example, a computerized maintenance management system (CMMS) and contracting maintenance criteria were deleted for being not significant in this study.
In the second stage, content validity, criteria validity and reliability has been carried out. Content validity is a judgment, by experts, of the extent to which a summated scale truly measures the concept that it intended to measure based on the content of the item (Flyaan et al., 1990). In this research work, initially the contents are taken from the relevant literature on MP, which was then subjected to the expert’s opinion to ensure that issues of concern are correctly addressed and also to ensure the clarity and validity of the questions either by reframing or modifying the questions (Table II). For example, “Allowed maintenance cost/budget required” has been modified to “In my organization, there is sufficient budget for maintenance.”
Criteria (predictive) validity investigates the empirical relationship between the scores on a test instrument (predictor) and an objective outcome (the criterion) (Flyaan et al., 1990) which is measured through validity coefficient. In this research work, the criteria validity of measurement instrument has been determined by examining the multiple correlation coefficient (R) computed from the model. Obtaining high value is an indication that measurement instrument has a criterion-related validity. Reliability is concerned with consistency, accuracy and predictability of the scale. It refers to the extent by which a measurement process is free from random errors. Reliability analysis is a correlation based procedure and estimated using the reliability coefficient, Cronbach α, ranging between the values 0.00 and 1.00. The minimum generally acceptable α value is 0.70.
The questionnaire is developed in two parts. The first part contains general information of the organization and employees, such as names, address, age, working experience, ownership, business information of an organization, etc. The second part contains 22 questions seeking MP information. The perception of the respondents was obtained in a 1-5 Likert scale, where 1 – strongly disagree and 5 – strongly agree. Statistical Package for Social Sciences Version 21 has been used to analyze the data using correlation, regression, cluster and discriminant analysis. These are briefly discussed below.
4. Results and discussion
4.1 Reliability analysis
The Cronbach’s α reliability analysis provides the most efficient measure of internal consistency for the items considered in the questionnaire. Tavakol and Dennick (2011) pointed that Cronbach’s α should be reported in a critical way with adequate understanding and interpretation, and accepted value in the range of 0.70-0.95. The Cronbach’s α value was found to be 0.761 indicating the reliability of the survey instruments (Table III).
4.2 Correlation analysis
The Pearson correlation test has been conducted to understand the relationship between maintenance measurement factors. Higher value (greater than 0.7) of correlation coefficient is considered to imply a strong correlation, while the range of 0.3-0.7 is considered as a moderate relationship (Hussey and Hussey, 1997). Table IV, correlation matrix shows that MA has the highest correlation (0.510) with MP followed by CI (0.389) and FA (0.372). The correlation analysis also shows the high positive correlation among some independent variables. For example, the correlation coefficient between CI and MA is 0.309, CI and PDO is 0.362 and CI and FA is 0.425. Correlation analysis cannot explain whether CI influences MA or MA influences CI or the possibility of any third variable which may be influencing both the variable resulting in a higher correlation. To overcome this limitation regression, analysis has been carried out and the results are presented next.
4.3 Regression analysis
Regression analysis has been used to obtain the relationship between a dependent variable (MP) and independent (predictor) variables (MA, TPS, PDO, FA, SPM, HRM and CI). The procedure discussed by Hair et al. (2006) has been followed in developing the multiple regression model. To establish the best regression model we compared the R2 (coefficient of determination) and the standard error of estimate (represents the standard deviation of actual dependent values around the regression line).
For each of the first four variables (MA, CI, FA and SPM) when added one-by-one to the regression model, the value of R2 has increased from 0.260 to 0.425 and the standard error of estimate has reduced from 0.829 to 0.754 indicating the improvement in the regression model in explaining the maintenance performance (Panels A and B of Table V). However, when other dependent variables (HRM, TPS and PDO) are added to the regression model there is no significant improvement in R2 value and reduction in standard error of estimate. Similarly, the value of R2 (coefficient of determination=0.425) indicates 42.5 percent of the variation in the dependent variable (MP) is explained by these four independent variables (MA, CI, FA and SPM) indicating significant and positive relation with MP (Samson and Terziovski, 1999). The adjusted R2 (37.7 percent) indicates that these four independent variables have a reasonably high degree of validity (predictive or external) when taken together. The overall model for estimating the MP is given by (Table VI):
The value of multiple R (sample correlation coefficient) is 0.652, which indicates a strong positive correlation between the regression variables (Panel A of Table V). The overall regression model fit is determined by ANOVA in terms of the F ratio and t-value that determine the significance for each individual independent variable in the regression model. The model is significant with the F-value and t-values as 8.857 and 3.986, 1.562, 1.376, 2.625, respectively for the four independent variable MA, CI, FA and SPM (Panel A of Table V and Table VI). The standard error of the regression coefficient is an estimation of how much the regression coefficient will vary between samples of the same size taken from the same population. A smaller standard error implies more reliable prediction and therefore smaller confidence intervals. The standard error of MA is 0.073, denoting that the 95 percent confidence interval for MA would be 0.290±(1.96×0.073), ranging from a low of 0.146 to a high of 0.433 and the same is to be calculated for CI , FA and SPM variable.
The regression unstandardized coefficient (b) and the standardized coefficient (β) reflect the change in the dependent measure for each unit change in the independent variable. The comparison between regression coefficients allows for the relative importance of each variable’s important in the regression model. The value 0.290 is the regression coefficient for the independent variable MA and 0.152 for CI, 0.092 for FA and 0.138 for SPM. The standardized regression coefficient (β) value for these variable is MA (0.486), CI (0.193), FA (0.170) and for SPM (0.300) (Table VI). In this case, MA is having the highest (0.400) standardized coefficient (β) than SPM (0.300), CI (0.193) and FA (0.170). This indicates that MA is having positive and highest contribution in predicting the MP for the sugar industry followed by SPM (0.300), CI (0.193) and FA (0.170). The finding shows that MA (β=0.400, t=3.986), SPM (β=0.30, t=2.625), CI (β=0. 193, t=1.562) and FA (β=0.171, t=1.376) helps improve the MP in the sugar industry.
Variance inflation factor (VIF) is a measure of multicollinearity (other independent variable substantially shared variance), which is calculated as the inverse of the tolerance value. Multicollinearity can have substantive effects not only on the predictive ability of the regression model, but also in the estimation of the regression coefficients and their statistical significance tests. High degree of multicollinearity is reflected in lower tolerance value and higher VIF values. In this work, the value of VIF factor is less than 5 for all the variables which indicates that analysis is free from multi-colinearity.
4.4 Cluster analysis
Cluster analysis has been used to understand the difference in maintenance practices across the surveyed sugar industry and its impact on MP due to the type of ownership and the installed capacity. This is briefly discussed next.
Cluster analysis is a statistical technique that groups objects according to the characteristics that they possess, such that clusters exhibit high internal homogeneity and high external heterogeneity (Hair et al., 2006). A k-mean cluster method has been used to find cluster centers while the ward method with Euclidian distance is used to group the sugar industries. Determining the number of clusters can be somewhat subjective (Everitt, 1993); however, as suggested by Hair et al. (2006), we ascertain the number of clusters using changes in the agglomeration coefficient. We have named the three groups resulting from the cluster analysis as high, medium and low maintenance performer. The classification result shows that 98.1 percent of original cases and 96.3 percent of cross-valid grouped cases are correctly predicted. High, medium and low maintenance performer cluster represents 31.48 (17 cases), 29.64 (16 cases) and 38.88 (21 cases) percent of the sampled population, respectively and the difference in mean values of MP attributes is shown in Figure 3.
However, the medium and high performance clusters are almost similar in case of FA and SPM. This may be due to the difference in the ownership (private and cooperative) and better financial practices. To indicate which variable contribute most in the cluster solution, analysis of variance (ANOVA) test has been carried out. Table VII shows the result indicating PDO (47.734) as the highest contributing factor, followed by SPM (42.559), HRM (16.586) and MA (10.82).
Linking MP with practices: One-way ANOVA has been performed on the data to analyze the contribution of practices in the MP, especially in the MP1 and MP2. The analysis shows that for MP1 (higher OEE MP), sugar industries are focusing on CI (F=5.350, sign=0.003), MA (F=7.436, sign=0.000), and PDO (F=2.788, sign=0.050). Similarly, for MP2 (maintenance satisfaction), sugar industries are focusing on FA (F=1.001, sign=0.399), SPM (F=1.911, sign=0.140) and HRM (F=2.253, sign=0.094); whereas one or more of the clusters are significantly different from another by the activities specified by the factors such as MA (F=10.827 sig.=0.000), PDO (F=47.73 sig.=0.000), SPM (F=42.559 sig.=0.000) and HRM (F=16.586 sig.=0.000). That is, the largest mean square is represented by PDO, followed by SPM, HRM and MA). It should be noted that analysis is only for descriptive purpose because the cluster have been chosen to maximize the differences among cases in different clusters (Table VIII). Discriminant analysis is used to find out which independent variables are relatively better at discriminating between groups and also to validate the accuracy of classification done by cluster analysis (Figure 4). A discriminant analysis (Table VII) of group mean is carried out to test the difference in the group mean. Wilks’ λ is the ratio of the within-groups sum of squares to the total sum of squares. Wilks’ JQME605053 ranges from 0 to 1.0. Smaller values indicate strong group differences. Table VII shows the factors: PDO (0.348), SPM (0.375), HRM (0.606) and MA (0.702) are contributing to differentiate the sugar industries.
Similarly, other two types of cluster analysis have been performed based on the ownership of sugar industry (private and cooperative) and capacity of the sugar plant (low and high) (Figures 5 and 6). The main objective is to understand whether there is any significant difference in the maintenance practices in these groups which in turn will result in the difference in the MP of the plant. There is a significant difference between MPs in private and cooperative sugar industry based on independent sample t-test. The variables t-values are significant (<0.05) for MA (t−0.001<0.05), CI (t−0.012<0.05), PDO (t−0.000<0.05), FA (t−0.014<0.05) and SPM (t−0.007<0.05). However, there is no significant difference between high and low capacity sugar industries’ MP. Only one variable, FA (t−0.036<0.05), is near to the range of significant value of 0.05.
The primary purpose of this research is to verify whether the maintenance practices adopted in the manufacturing industries, especially the discrete, are applicable to the sugar industry being a seasonal process industry. This research supports the works of Cholasuke et al. (2004) and Swanson (2001) for the MA and CI practice that significantly affects the MP in the sugar industry. Commonly used MA in sugar industries is the overall maintenance during off-season periods while periodic maintenance (time based), corrective maintenance and condition-based maintenance are performed during the season period.
In this work, three clusters have been identified based on the MP and they are termed as high, medium and low performers. In the high group the top management is concerned for MP and realizes the benefits from it. This can be clearly observed in this group as it has performed better in PDO, HRM, CI and SPM (Table VIII). This group is significantly different from a low maintenance performer in CI, MA and PDO practices, and significantly different from a medium maintenance performer in SPM and HRM practices.
In the medium group, the top management is moderately involved; however, there are some maintenance and financial approaches that enables better maintenance performance compared to the low group. This group ranked significantly higher for FA and MA with a mean of 16.75 and 12.00, respectively (Table VIII). This cluster also had only moderate levels of involvement in CI, PDO, and HRM practices, however, it is significantly different from a low maintenance performer in CI, MA and PDO practices. In low performing group top management is least concerned for MP and the group does not employ the expected maintenance practices, however the MP may be the output due to the responsibility of the functional heads and the group that has focused on SPM and TPS.
Based on the discriminant analysis, Wilk’s λ (Table VII), maintenance practices differ across sugar industries and they are differentiated based on PDO, SPM, HRM and MA. Further, to understand the maintenance practices influencing the MP within each cluster regression analysis has been carried out. Within each cluster, correlation analysis has been performed to identify the highly correlated variable to be added into the regression model. The statistic of regression model clearly shows different variables contributing into different clusters (Table IX). For example, in the low performance cluster, the main contributing variables are MA and CI and other variables, namely, PDO, TPS and HRM are not entered in the regression model. In the medium group, the MP is increased due to moderate involvement of top management and systematic FA. However, the MP is further improved in the high group due to HRM and TPS practices.
This work has reported the empirical analysis of measuring the MP of the sugar industry by evaluating the six important maintenance practices, namely, MA, PDO, TPS, FA, SPM, HRM and CI. Statistical tools such as correlation, multiple regression and cluster analysis and been used to obtain the insights.
The study validates the use of maintenance practices by the sugar industry similar to that of the manufacturing industry. However, the regression model shows that the MP in the sugar industry is mainly given by MA, CI, financial approach and SPM. Further, the cluster and discriminant analysis showed that the contribution of these maintenance practices varies across the three groups, namely, low, medium and high maintenance performers. The regression analysis within each cluster made it clear that in the low maintenance performer cluster, the main contributing variables are MA and CI while in the medium group, it is FA. In the high maintenance performer group, the MP is further improved due to HRM and TPS practices. The cluster analysis also reveals that the MP in privately owned and cooperative sugar industries are significantly different, however, for low and high capacity type sugar industries, no significant difference in MP has been observed.
The research contributes in identifying and testing the important maintenance practices used by the sugar industry and shows that MA, CI, financial approach and SPM are important for the improved MP in the sugar industry.
Review of maintenance management performance measurement
|S. No.||Author (year)||Type of research||Respondent/data collection||Research tool/technique||Maintenance practices||Remark/application|
|1||Mwanaongoro and Imbambi (2014)||Survey||51, respondents supervisors, managers||Questionnaire data collection, Likert scale, weighted average method||Independent variable – maintenance strategies – break down, preventive, proactive, and predictive maintenance
Dependent variable – organizational policies – maintenance leadership, human resource management, inventory management, asset management, technology adoption
|Relation between plant maintenance strategies and performance of sugar industry of Kenya|
|2||Carnero (2014)||Case study||Benchmarking in big buildings||AHP, fuzzy AHP, utility function criteria||8 criteria – maintenance organization, maintenance cost, quality environment and safety standards, out sourcing maintenance, control, maintenance computerization, maintenance training, maintenance manager and 50 sub-criteria||Competitive benchmarking and general benchmarking|
|3||Horenbeek and Pintelon (2014)||Case study||4 companies||ANP methodology||Maintenance cost, technical aspects, plant design life, support, people and environment||Assist maintenance manager ANP methodology in selection of relevant MPI|
|4||Gustafson et al. (2013)||Case study||4 years’ data||Statistics – mean SD||Production KPI, production cost per ton, ton per month. maintenance KPI, MTTF, MTTR, availability, maintenance cost, equipment downtime||Analysis – manual vs semi-automatic load hull dump (LHD)|
|5||Srinivasan et al. (2012)||Survey||23 SME industries in India||Questionnaire data collection, and analysis||Maintenance practices adopted by industries, creating awareness about the maintenance, guidance for implementation of the advanced maintenance, feedback from resource persons||Benchmarking of maintenance practices in Indian SMEs|
|6||Maletic et al. (2012)||Survey||53 respondent organizations||Questioner design, statistic – exploratory analysis, correlation analysis, principal component analysis (PCA), regression analysis||Continuous improvement||Relationship between continuous improvements and maintenance performance|
|7||Simões et al. (2011)||Review||na||na||Effective utilization of maintenance resources; total maintenance and information systems support; Measurement, measures, and human factor management||Reviewed 251 papers published up to 2010. Application area wise, journal wise, country, etc.|
|8||Oganga (2013)||Case study||One industry. Out of 48 respondents, 25 filled 1-5 scale||Multiple regression analysis, one-way ANOVA||Safety and health incidents, unscheduled plant breakdowns, shut downs, overtime cost, work backlogs, spillage||Relation between maintenance performance and improvements in maintenance|
|9||Muchiri et al. (2010)||Survey||Unit analysis of one company. One respondent. maintenance manager, maintenance engineering, senior professional with 41 respondents||Questioner design, statistic – 1-5 scale, simple random sampling||Maintenance objective – selected 21 KPIs||Related with choosing KPI for maintenance strategies and maintenance performance. Used logistic regression analysis. Belgium and European industry analysis|
|10||Chinese and Ghirardo (2010)||Survey||100 manufacturing industries||Non-parametric tests. 1-5 scale||Structural decision – capacity, facility, technology –CMB, CMMS, vertical integration –out sourcing infrastructural decision – maintenance organization, policies and concept, planning and control, human resource||Maintenance strategies and maintenance performance in industry having less than 50 employees. Italian Industry Survey|
|11||Shah Ali (2009)||Case study and survey||62 building maintenance managers||Correlation analysis||Maintenance building cost (existing building conditions, building cost, availability of funding, safety and health requirement, clients request)||Maintenance building cost depends upon the important factors in decision making|
|12||Ali and Mohamad Nasbi Bin Wan Mohamad (2009)||Exploratory case study||Hospital director, departmental heads, consultants, contractor’s customer service manager||Interview, field visit||Leadership, policy plans, training orientation, monitoring and supervision, service||Continuous involvement of all organizations is important to provide guidance and direction to the maintenance management. Hospital industry|
|13||Marqueza and Gupta (2006)||Review||na||na||Strategic Level (Break down maintenance, Schedule Maintenance, Routine maintenance, CBM) Tactical Level (Planning and Scheduling, Work order execution policy, long-term planning and maintenance assessment), operational level (work order complication, historical records, management information trends)||Modified three pillar of maintenance management with three levels of maintenance management|
|14||Amoedo and Modarres (2006)||Case study||Expert judgment||AHP||People, logistic, training, reactive and proactive||Concept of basic pillar of maintenance from balanced score card (BSC) and goal tree analysis is used to select maintenance performance indicator. automotive industry|
|15||Parida and Chattopadhyay (2007)||Theory/case study||na||Interview, field visits||Measuring value created by maintenance, justifying investment, revising resources, health, safety and environment issues, focus on knowledge management, organizational structural change, adopting new trends in maintenance||Developed multi-criteria hierarchical frame work for maintenance performance measurement|
|16||Cholasuke et al. (2004)||Survey||150 organizations, 21 respondents, production manager/manager SMEs||Questionnaire design, radar chart, statistics – t-test, correlation||Maintenance approach, human resource management, policy and development, spare part management, financial aspects, continuous improvement computer management system, task planning and scheduling||Success factor of maintenance. Relationship between maintenance strategies and continuous improvement. UK’s manufacturing industry|
|17||Marquez et al. (2003),||Review||na||na||IT pillar; maintenance engineering methods pillar; organizational (or behavioral) pillar||Identified three pillars of maintenance management|
|18||Tsang (2002)||Theory-review||No||No||Value-based measure, balanced score card, and system audits||Four dimensions of maintenance management|
|19||Swanson (2001)||Survey||Maintenance managers (MM) – 125, Production managers (PM) – 162||Questioner design, statistic – correlation, regression, factor analysis, correlation||Aggressive maintenance, proactive maintenance, reactive maintenance||Relationship between maintenance strategies and maintenance performance|
|20||McKone et al. (2001)||Survey||115, respondents, three industries from four countries||Structural equation modeling (SEM)||TPM variable considered three maintenance practices namely preventive maintenance, inventory
|Positive and Indirect relationship between TPM and manufacturing performance (MP). electronics industry|
|21||Jonsson (1997)||Strategy, human aspects, support mechanisms, tools and techniques, and organization||Emphasized on improvement of the maintenance management components and suggested for formulating clear maintenance strategies that are linked to manufacturing and corporate strategies. Swedish manufacturing organizations|
Maintenance practices identification
|S. No||Criteria||Items||Interview support||References|
|1||Policy development and organization||Maintenance policy and production strategy||Yes||Dunn (1996), Kelly (1997), Wilson (1999), Ismail Mostafa (2004), Shafeek, 2014), Cholasuke et al. (2004)|
|2||Maintenance approach||Preventive maintenance||Yes||Ismail Mostafa (2004), Shafeek (2014), Márquez et al. (2009), Cholasuke et al. (2004), Mobley (1990), Jonsson (1997)|
|Time-based preventive maintenance|
|3||Task planning and scheduling||Task scheduling/completion||Yes||Levitt (1997), Shafeek (2014), Cholasuke et al. (2004)|
|4||Information management and CMMS||A computerized maintenance management system used||No||Cholasuke et al. (2004), Dunn (1996), Kelly (1997), Wilson (1999)|
|Have an effective key performance measures|
|Better utilization of CMMs|
|5||Spare part management||Spare recording||Yes||Ismail Mostafa (2004), Shafeek (2014), Cholasuke et al. (2004), Louit (2007), Visser and Pretorius (2003), Raouf (1994)|
|Inventory of spare|
|6||Human resources management||Training||Yes||Nakajima (1988); De Groote (1995); Kelly (1997), Jonsson (1997), Shafeek (2014), Rezaie et al. (2009), Cholasuke et al. (2004)|
|7||Contracting out maintenance||High benefits of contract out maintenance||No||Márquez et al., (2009), Cholasuke et al. (2004)|
|8||Financial aspects||Low cost with effective Maintenance||Yes||Moubray (1994), Kelly (1997), Cholasuke et al. (2004), Swanson (2001), Shafeek (2014)|
|Production loss cost|
|Finance optimum maintenance cost|
|9||Continuous improvement||Continuous improvement||Yes||Cholasuke et al. (2004) Maletic et al. (2012), Shafeek (2014)|
|Proactive approach (RCM TPM)|
Reliability statistics of Cronbach’s α
|Cronbach’s α||Cronbach’s α based on standardized items||Number of items|
|Dependent variable MP||0.510**||0.216||0.252||0.372**||0.130||0.203||0.389**|
Notes: MA, maintenance approach; TPS, task planning and scheduling; PDO, policy development and organization; FA, financial approach; SPM, spare part management; HRM, human resource management; CI, continuous improvement. *,**Correlations are significant at 0.05 and 0.01 levels (two-tailed), respectively
Multiple regression analysis
|Panel A. Multiple regression analysis|
|Dependent variable||Maintenance performance|
|Analysis of variance (ANOVAa)||df||Sum of square||Mean square|
|Panel B. Model summary of stepwise multiple regression model|
|Overall model fit||R2 Change Statistics|
|Step||R||R2||Adjusted R2||SE of the estimate||R2 change||F-value of R2 change||df1||df2||Significance of R2 change|
|Step 1||Maintenance approach (MA)|
|Step 2||Maintenance approach (MA), continuous improvement (CI)|
|Step 3||Maintenance approach (MA), continuous improvement (CI), financial approach (FA)|
|Step 4||Maintenance approach (MA), continuous improvement (CI), financial approach (FA), spare part management (SPM)|
Notes: Panel A: aDependent variable: MP. Predictors: (constant), FA, MA, CI
Regression analysis coefficients
|Unstandardized coefficients||Standardized coefficients||95.0 percent confidence interval for B||Collinearity statistics|
|Model||B||SE||β||T||Sig.||Lower bound||Upper bound||Tolerance||VIF|
|Variable entered into the regression model|
|Variable not entered into the regression model|
Note: Dependent variable: MP
ANOVA for cluster analysis
|Mean Square||df||Mean square||df||Wilks’ λ||F||Sig.|
|Low cases: 21
|Medium cases: 16
|High cases: 17
|Cluster mean||7.19 (2, 3)||8.00 (1)||8.11 (1)||F – 3.74|
|Cluster mean||9.95 (2, 3)||12.00 (1)||11.23 (1)||F – 10.82|
|Task planning and scheduling|
|Cluster mean||12.65||12.93||13.41||F – 0.76|
|Policy development and organization|
|Cluster mean||7.23 (2, 3)||11.43 (1)||12.17 (1)||F – 47.73|
|Cluster mean||15.71||16.75||16.70||F – 2.20|
|Spare part management|
|Cluster mean||13.38 (2)||10.00 (1, 3)||13.70 (2)||F – 42.55|
|Human resource management|
|Cluster mean||12.85 (3)||13.37 (3)||16.00 (1, 2)||F – 16.58|
Note: Numbers shown in parentheses indicate that the group was significantly different at the p<0.05 level according to the Student t-test and F-statistics and associated significant p-values are derived from one-way ANOVA
Regression statistics for low, medium and high clusters
|Cluster 1||Cluster 2||Cluster 3|
|Regression statistics||Low performance||Medium performance||High performance|
|SS – Regression||15.231||2.994||10.158|
|SS – Residual||13.911||2.756||8.783|
|SS – Total||29.143||5.750||18.941|
|Constant – intercept||− 0.919||0.734||−2.430|
|Constant – MA||0.436||0.148||0.155|
|Constant – CI||0.215||0.111||NA|
|Constant – PDO||NA||0.31||0.116|
|Constant – TPS||NA||NA||0.085|
|Constant – FA||0.084||0.246||NA|
|Constant – SPM||0.170||0.100||0.157|
|Constant – HRM||na||na||0.317|
|No. of respondents||21||16||17|
|No. of industries||4||5||6|
Ahuja, I.P.S. and Khamba, J.S. (2008), “Total productive maintenance: literature review and directions”, International Journal of Quality & Reliability Management, Vol. 25 No. 7, pp. 709-756.
Ali, M. and Mohamad Nasbi Bin Wan Mohamad, W. (2009), “Audit assessment of the facilities maintenance management in a public hospital in Malaysia”, Journal of Facilities Management, Vol. 7 No. 2, pp. 142-158.
Alsyouf, I. (2007), “The role of maintenance in improving companies’ productivity and profitability”, International Journal of Production Economics, Vol. 105 No. 1, pp. 70-78.
Amoedo, M.M. and Modarres, M. (2006), “A structured methodology for identifying performance metrics and monitoring maintenance effectiveness”, RAMS'06, IEEE Annual Reliability and Maintainability Symposium, pp. 239-246.
Carnero, M.C. (2014), “A decision support system for maintenance benchmarking in big buildings”, European Journal of Industrial Engineering, Vol. 8 No. 3, pp. 388-420.
Chang, H.H. (2005), “The influence of continuous improvement and performance factors in total quality organization”, Total Quality Management & Business Excellence, Vol. 16 No. 3, pp. 413-437.
Chinese, D. and Ghirardo, G. (2010), “Maintenance management in Italian manufacturing firms: matters of size and matters of strategy”, Journal of Quality in Maintenance Engineering, Vol. 16 No. 2, pp. 156-180.
Cholasuke, C., Bhardwa, R. and Antony, J. (2004), “The status of maintenance management in UK manufacturing organisations: results from a pilot survey”, Journal of Quality in Maintenance Engineering, Vol. 10 No. 1, pp. 5-15.
Cross, M. (1988), “Raising the value of maintenance in the corporate environment”, Management Research News, Vol. 11 No. 3, pp. 8-11.
De Groote, P. (1995), “Maintenance performance analysis: a practical approach”, Journal of Quality in Maintenance Engineering, Vol. 1 No. 2, pp. 4-24.
Ding, S. and Kamaruddin, S. (2015), “Maintenance policy optimization – literature review and directions”, The International Journal of Advanced Manufacturing Technology, Vol. 76 No. 5, pp. 1263-1283.
Dunn, S. (1996), “A framework for achieving best practice in maintenance”, paper presented to the West Australian Maintenance Conference, October 30, available at: www.plant-maintenance.com/change.html
Everitt, B.S. (1993), Cluster Analysis, 3rd ed., Edward Arnold, London.
Flyaan, B.B., Sakakibara, S., Schroeder, R.G., Bates, K.A. and Flynn, E.J. (1990), “Empirical research methods in operations management”, Journal of Operations Management, Vol. 9 No. 2, pp. 250-284.
Garg, A. and Deshmukh, S.G. (2006), “Maintenance management: literature review and directions”, Journal of Quality in Maintenance Engineering, Vol. 12 No. 3, pp. 205-238.
Gupta, M. and Galloway, K. (2003), “Activity-based costing/management and its implications for operations management”, Technovation, Vol. 23 No. 2, pp. 131-138.
Gustafson, A., Schunnesson, H., Galar, D. and Kumar, U. (2013), “Production and maintenance performance analysis: manual versus semi-automatic LHDs”, Journal of Quality in Maintenance Engineering, Vol. 19 No. 1, pp. 74-88.
Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E. and Tatham, R.L. (2006), Multivariate Data Analysis, Pearson Prentice Hall, Upper Saddle River, NJ.
Hitt, M.A., Bierman, L., Shimizu, T. and Kochhar, R. (2001), “Direct and moderating effects of human capital on strategy and performance in professional service firms: a resource-based perspective”, Academy of Management Journal, Vol. 44 No. 1, pp. 13-28.
Horenbeek, A.V. and Pintelon, L. (2014), “Development of a maintenance performance measurement framework – using the analytic network process (ANP) for maintenance performance indicator selection”, Omega, Vol. 42 No. 1, pp. 33-46.
Hussey & Hussey (1997), Business Research – A Practical Guide for Undergraduate and Postgraduate Students, Macmillan Press Ltd, London, p. 229.
Ismail Mostafa, S. (2004), “Implementation of proactive maintenance in the Egyptian glass company”, Journal of Quality in Maintenance Engineering, Vol. 10 No. 2, pp. 107-122.
ISO 14224 (2004), Petroleum and Natural Gas Industries – Collection and Exchange of Reliability and Maintenance Data for Equipment, 2nd ed., International Organization for Standardization, Geneva.
Jeong, K.Y. and Phillips, D.T. (2001), “Operational efficiency and effectiveness measurement”, International Journal of Operations & Production Management, Vol. 21 No. 11, pp. 1404-1416.
Jonsson, P. (1997), “The status of maintenance management in Swedish manufacturing firms”, Journal of Quality in Maintenance Engineering, Vol. 3 No. 4, pp. 233-258.
Kelly, A. (1997), Maintenance Organization and Systems, Butterworth-Heinemann, Oxford.
Khazraei, K. and Deuse, J. (2011), “A strategic standpoint on maintenance taxonomy”, Journal of Facilities Management, Vol. 9 No. 2, pp. 96-113.
Komonen, K. (2002), “A cost model of industrial maintenance for profitability analysis and benchmarking”, International Journal of Production Economics, Vol. 79 No. 1, pp. 15-31.
Kumar, U., Galar, D., Parida, A., Stenström, C. and Berges, L. (2013), “Maintenance performance metrics: a state‐of‐the‐art review”, Journal of Quality in Maintenance Engineering, Vol. 19 No. 3, pp. 233-277.
Kutucuoglu, K.Y., Hamali, J., Irani, Z. and Sharp, J.M. (2001), “A framework for managing maintenance using performance measurement systems”, International Journal of Operations & Production Management, Vol. 21 Nos 1/2, pp. 173-195.
Levitt, J. (1997), The Handbook of Maintenance Management, Industrial Press Inc, New York, NY.
Ljungberg, O. (1998), “Measurement of overall equipment effectiveness as a basis for TPM activities”, International Journal of Operations & Production Management, Vol. 18 No. 5, pp. 495-507.
Louit, D. (2007), “Optimization of critical spares parts inventories: a reliability perspective”, PhD thesis, University of Toronto.
McKone, K.E., Schroeder, R.G. and Cua, K.O. (2001), “The impact of total productive maintenance practices on manufacturing performance”, Journal of Operations Management, Vol. 19 No. 1, pp. 39-58.
Maletic, D., Maletic, M. and Gomišek, B. (2012), “The relationship between continuous improvement and maintenance performance”, Journal of Quality in Maintenance Engineering, Vol. 18 No. 1, pp. 30-41.
Marqueza, A.C. and Gupta, J.N. (2006), “Contemporary maintenance management: process, framework and supporting pillars”, Omega, Vol. 34 No. 3, pp. 313-326.
Marquez, A.C., Gupta, J.N. and Heguedas, A.S. (2003), “Maintenance policies for a production system with constrained production rate and buffer capacity”, International Journal of Production Research, Vol. 41 No. 9, pp. 1909-1926.
Márquez, A.C., Léon, P., Fernándes, J.F., Márquez, C.P. and Campos, M.L. (2009), “The maintenance management framework: a practical view to maintenance management”, Journal of Quality in Maintenance Engineering, Vol. 15 No. 2, pp. 167-178.
Mobley, R.K. (1990), An Introduction to Predictive Maintenance, Van Nostrand Reinhold, New York, NY.
Moubray, J. (1994), Reliability-Centered Maintenance, 4th ed., Butterworth-Heinemann, Oxford.
Muchiri, P.N., Pintelon, L., Martin, H. and De Meyer, A.M. (2010), “Empirical analysis of maintenance performance measurement in Belgian industries”, International Journal of Production Research, Vol. 48 No. 20, pp. 5905-5924.
Mwanaongoro, S. and Imbambi, R.M. (2014), “Assessment Of relationship between plant and equipment maintenance strategies and factory performance of the Kenya sugar firms”, Asian Journal of Basic and Applied Sciences, Vol. 1 No. 2, pp. 18-28.
Nakajima, S. (1988), Introduction to TPM: Total Productive Maintenance (translation), Productivity Press, Inc., Cambridge, p. 129.
Oganga, M.R. (2013), “Performance measurement practices and maintenance improvement among Coca-Cola bottling plants in Kenya”, MBA dissertation, School of Business, University of Nairobi, Nairobi.
Oke, S.A. (2005), “An analytical model for the optimisation of maintenance profitability”, International Journal of Productivity and Performance Management, Vol. 54 No. 2, pp. 113-136.
Parida, A. and Kumar, U. (2006), “Maintenance performance measurement (MPM): issues and challenges”, Journal of Quality in Maintenance Engineering, Vol. 12 No. 3, pp. 239-251.
Parida, A. and Kumar, U. (2009), “Maintenance productivity and performance measurement”, Handbook of Maintenance Management and Engineering, Springer, London, pp. 17-41.
Parida, A. and Chattopadhyay, G. (2007), “Development of a multi-criteria hierarchical framework for maintenance performance measurement (MPM)”, Journal of Quality in Maintenance Engineering, Vol. 13 No. 3, pp. 241-258.
Parida, A., Kumar, U., Galar, D. and Stenström, C. (2015), “Performance measurement and management for maintenance: a literature review”, Journal of Quality in Maintenance Engineering, Vol. 21 No. 1, pp. 2-33.
Raouf, A. (1994), “Improving capital productivity through maintenance”, International Journal of Operations & Production Management, Vol. 14 No. 7, pp. 44-52.
Rezaie, K., Dehghanbaghi, M. and Ebrahimipour, V. (2009), “Performance evaluation of manufacturing systems based on dependability management indicators – case study: chemical industry”, The International Journal of Advanced Manufacturing Technology, Vol. 43 Nos 5/6, pp. 608-619.
Samat, H.A., Kamaruddin, S. and Azid, I.A. (2011), “Maintenance performance measurement: a review”, Pertanika Journal of Science & Technology, Vol. 19 No. 2, pp. 199-211.
Samson, D. and Terziovski, M. (1999), “The relationship between total quality management practices and operational performance”, Journal of operations Management, Vol. 17 No. 4, pp. 393-409.
Shafeek, H. (2014), “Continuous improvement of maintenance process for the cement industry – a case study”, Journal of Quality in Maintenance Engineering, Vol. 20 No. 4, pp. 333-376.
Shah Ali, A. (2009), “Cost decision making in building maintenance practice in Malaysia”, Journal of Facilities Management, Vol. 7 No. 4, pp. 298-306.
Sherwin, D. (2000), “A review of overall models for maintenance management”, Journal of Quality in Maintenance Engineering, Vol. 6 No. 3, pp. 138-164.
Simões, J.M., Gomes, C.F. and Yasin, M.M. (2011), “A literature review of maintenance performance measurement: a conceptual framework and directions for future research”, Journal of Quality in Maintenance Engineering, Vol. 17 No. 2, pp. 116-137.
Srinivasan, K., Muthu, S., Sugumaran, C., Sathiyavanan, R. and Kumar, R.S. (2012), “Benchmarking study of maintenance performance monitoring practices in SMEs”, European Journal of Scientific Research, Vol. 79 No. 3, pp. 437-448.
Sun, S. (2004), “Assessing joint maintenance shops in the Taiwanese Army using data envelopment analysis”, Journal of Operations Management, Vol. 22 No. 3, pp. 233-245.
Swanson, L. (2001), “Linking maintenance strategies to performance”, International Journal of Production Economics, Vol. 70 No. 3, pp. 237-244.
Tavakol, M. and Dennick, R. (2011), “Making sense of Cronbach’s alpha”, International Journal of Medical Education, Vol. 2, pp. 53-55, available at: www.ijme.net/archive/2/cronbachs-alpha.pdf
Tsang, A.H.C. (2002), “Strategic dimensions of maintenance management”, Journal of Quality in Maintenance Engineering, Vol. 8 No. 1, pp. 7-39.
Tsang, A.H.C., Jardine, A.K.S. and Kolodny, H. (1999), “Measuring maintenance performance: a holistic approach”, International Journal of Operations & Production Management, Vol. 19 No. 7, pp. 691-715.
Visser, J.K. and Pretorius, M.W. (2003), “The development of a performance measurement system for maintenance”, SA Journal for Industrial Engineering, Vol. 14 No. 1, pp. 83-98.
Wilson, A. (1999), “Asset management and maintenance strategy”, Maintenance and Asset Management, Vol. 14 No. 1, pp. 3-10.