The current issue and full text archive of this journal is available at

www.emeraldinsight.com/1355-2511.htm

JQME METHODOLOGY AND THEORY

16,2
Maintenance management in
Italian manufacturing firms
156
Matters of size and matters of strategy
Damiana Chinese
Università di Udine, Udine, Italy, and
Gianni Ghirardo
AISL – Associazione Italiana di Studio del Lavoro, Pordenone, Italy

Abstract
Purpose – This paper aims to give a picture of maintenance management in Italian manufacturing
firms supported by empirical evidence. The purpose is also to highlight how far maintenance
performance and strategies are influenced by context and which measures and goals are within reach
of small-sized firms.
Design/methodology/approach – Frameworks for describing maintenance management and
strategies derived from literature were reviewed and used to develop a questionnaire. A survey-based
empirical research involving a sample of 100 manufacturing firms was performed. Non-parametric
statistics are applied to highlight correlations and dependencies between contextual variables,
maintenance strategies and performance.
Findings – Maintenance performance hardly seems a matter of size, while many elements of strategy
certainly are. Some elements of strategy, in particular planning and control elements, seem to have
little impact on performance. By contrast, an enhanced use of preventive maintenance and, above all,
of condition-based maintenance is demonstrated to be a highly effective action for maintenance
improvement applicable to firms of all size.
Practical implications – The research may help managers to decide on maintenance strategic
variables by deducing from the experience of many different firms whether, and how, strategies affect
maintenance performance.
Originality/value – Besides giving a country’s portrait, the empirical research addresses the links
between strategies, context and performance, thereby understanding strategy in a broad sense and not
just in terms of maintenance policies and concepts. In particular, the maintenance practices of small
enterprises with fewer than 50 employees are studied, which has rarely been done in the literature.
Keywords Maintenance, Surveys, Italy
Paper type Research paper

1. Introduction
Although the tertiary sector is gaining growing importance in Italy as in most
western countries, the manufacturing sector remains a driving force in Italian
economy. The most recent national census (ISTAT, 2001) recorded more than 585,000
Journal of Quality in Maintenance manufacturing firms, employing about 5 million people, which is nearly 22 per cent of
Engineering
Vol. 16 No. 2, 2010
pp. 156-180 Cooperation by the Industrial Union of Pordenone and of Udine is gratefully acknowledged. The
q Emerald Group Publishing Limited
1355-2511
authors also wish to thank Mr Rudy Lunardelli and Mr Marco Bettiol for collaboration in data
DOI 10.1108/13552511011048904 collection, as well as all organizations, which took part in this survey.
the country’s labour force. Census data show that most firms belong to the metal Matters of size
working and machine manufacturing sectors (39 per cent), followed by typical “made and strategy
in Italy” sectors (textile and clothing, shoes and leader, wood and furniture
manufacturing industries, which involve 24 per cent of total firms). A distinctive
feature of the Italian manufacturing sector is the small average size of firms: about 94
per cent of manufacturing firms have less than 20 employees, 4 per cent of firms have
between 20 and 50 employees and just the remaining 2 per cent of total firms have 157
more than 50 employees and provide work for about 25 per cent of the labour force in
the manufacturing sector.
Given the important physical assets required by manufacturing industries and the
significant proportion of maintenance spending on total manufacturing costs reported
in literature (Tsang, 2002), industrial maintenance is likely to play an important role in
Italian economy, in terms of employees, value added and contribution to
competitiveness under the increasing international pressure. However, little is
known about the state of maintenance management in Italian organizations. To the
best of the authors’ knowledge, no official statistics exist, while some information can
be derived from a few empirical studies reported in national and to some extent in
international literature. A survey on maintenance management in small and medium
firms (Cattaneo, 2000) was carried out by AIMAN, the Italian Maintenance Society, in
the year 2000. In total, 174 organizations, having up to 200 employees, were involved,
mainly belonging to the mechanical and metal working sectors and to the chemical and
pharmaceutical sectors. The survey highlighted that an actual maintenance function
exists in about 20 per cent of the micro-firms (with less than 15 employees), in about 50
per cent of small firms (having between 16 and 50 employees) and in about 85 per cent
of medium sized firms. The survey was focused on maintenance costs, revealing that
they are around 2 per cent of turnover, with no significant difference depending on firm
size or sector, and on maintenance policies: it was found that a fire-fighting attitude
still prevails in many firms, with reactive maintenance accounting for about 40 per cent
of total maintenance activities, as reported also by Ferrari et al. (2002). Another survey
at a more local scale, involving 62 medium firms, was performed in 2002 by a regional
section of the AISL, the Italian Society for Work Studies (Ghirardo, 2004) and
confirmed similar shares of reactive maintenance and the same attitude towards
maintenance: for instance, it was found that only 20 per cent of the examined
organizations calculated and took into account inefficiency costs (costs of loss of asset
availability), caused, e.g. by stops due to reactive or delayed maintenance.
Nevertheless, the pursuit of excellence and a successful implementation of best
practices (e.g. TPM as defined by Nakajima (1988) in Italy is testified by single case
studies (Ferrari et al., 2002) or collections of case studies (Cigolini and Turco, 1997)
presented in international literature. Still, most surveyed cases concern either large
industries or some smaller manufacturing plants, which however belong to large trusts
or multinational groups.
Considering the structure of Italian manufacturing sector outlined previously,
where small sized firms prevail, it would be interesting to obtain a wider image of
maintenance management conditions in manufacturing firms. In particular, a
reasonable question is whether excellence is strictly limited to large sized firms, which
are a small part of the national production context. More generally, and beyond the
limits of a country’s portrait, it is useful to investigate how far maintenance
JQME performance and practices depend on firms’ size and to what extent they are rather
16,2 linked to strategic choices made by organizations. Therefore, we developed the
research presented in this paper with the objective of providing possible answers to the
highlighted questions on the basis of empirical evidence. With this aim, we structured
a questionnaire moving from frameworks and international surveys designed to
analyze maintenance status at regional or national level. Section 2 presents the
158 research methodology, including the background literature for our survey. The results
are discussed in section 3, while conclusions and implications for future research are
derived in section 4.

2. Research methodology
We began our research by reviewing literature about maintenance performance
measurement frameworks, as they are a tool to assess industrial achievements in
maintenance management, and about sector or countrywide surveys on maintenance
management. We found that general maintenance performance measurement
frameworks, which are designed to evaluate single firms and mostly rely on
quantitative indicators, have inherent limitations for sector wide surveys: in fact, it is
difficult for researchers to obtain uniform and reliable measures because of diverse –
and often lacking – recording and measurement procedures used in different firms (see
Pintelon and Van Puyvelde (1997) and Swanson (2003). Therefore, we decided to focus
on qualitative indicators and to start from frameworks and methodologies used in
country or sector wide maintenance surveys.
Several studies of this kind have been reported in literature in the last few years.
They could be grouped in two classes according to their objectives, that is surveys
aimed at giving descriptions and general pictures of maintenance management and
surveys aimed at inferring relationships between maintenance practices and other
business elements and outcomes. When the main objective is description, broad
frameworks and questionnaires encompassing numerous aspects of maintenance
management are proposed in order to give a most comprehensive picture of
maintenance in a specific region, e.g. Ikwan and Burney (1994), Jonsson (1997), Tse
(2002), Cholashuke et al. (2004) – or industrial context, e.g. Dowlatshahi (2008). For
instance, Jonsson (1997) introduces a framework of maintenance management
consisting of five components, i.e. goal and strategy, human aspects, support
mechanisms, tools and techniques and organization. A wider framework is proposed
by Cholashuke et al. (2004), who identify eight elements of maintenance management,
i.e. organization, maintenance approach, task planning, information management,
spare part management, human resource management, financial aspects and
continuous improvement. The widest and most recent theoretical framework for
describing maintenance state appears to be the one proposed by Pintelon et al. (2006),
which is grounded on an extended maintenance strategy concept including four
structural – i.e. long-term – and six infrastructural – i.e. operational – decision
elements. The structural decision elements are:
.
maintenance capacity;
. maintenance facilities;
.
maintenance technology; and
.
vertical integration,
while infrastructural decision elements include: Matters of size
.
maintenance organization; and strategy
.
policies and maintenance concepts;
.
planning and control;
.
human resources;
.
maintenance modifications; and 159
.
performance measurement and reward.
The same authors (Pinjala et al., 2006) used their framework as a basis for a survey on a
sample of Belgian manufacturing companies with more than 100 employees, with the
aim of inferring relationships between general business strategies and maintenance
strategies.
Actually, when the main research goal is to infer relationships, i.e. in the second
group of papers we found, just few maintenance variables are usually considered,
whose links with other variables are investigated. Interdependencies explored in
literature include, in particular:
.
the link between maintenance practices and production features of firms
(Swanson, 1997) or between maintenance strategies and general business
strategies adopted by firms (Pinjala et al., 2006) and the correlation between the
pursuit of best practices in maintenance management and in other areas of
operations management (e.g. Mitchell et al., 2002), particularly between the
introduction of TPM and the application of other best practices such as TQM and
JIT. In other words, authors try to find out which kind of firms adopt particular
maintenance practices or strategies; and
.
the correlation between maintenance concepts or polices and maintenance
performance i.e. which advantages are obtained by firms adopting various
maintenance approaches (McKone et al., 2001; Swanson, 2001).
Considering the objectives and scope of our study, i.e. to give a general picture but also
to understand which kinds of firms adopt best practices and achieve good
performances, we decided to start from the framework by Pintelon et al. (2006) for
structuring our survey, as it is the most recent and complete for its comprehensive
view of maintenance strategies and allows both a general view and the analysis of
correlations and relationships between maintenance strategic variables. Our effort was
to formulate questions suitable for a survey target group largely consisting of small
firms, and to cover most structural and infrastructural elements forming the adopted
framework. In particular, bearing in mind our target group, as to maintenance
performances we decided to ask for qualitative judgements about obtained results and
managers’ satisfaction, because the previously mentioned difficulties of successfully
asking for quantitative measures or maintenance PIs may be exacerbated in small firm
contexts.
With this in mind, we elaborated the questionnaire presented in Table I. As the
survey was the result of cooperation with local industrial unions, the questionnaire
structure and answer options were first tested through discussions with consultants in
safety and maintenance as well as through pilot questionnaires administered to
volunteer respondents.
 16,2

160

Table I.
JQME

Questionnaire structure
and descriptive statistics
Variable
Question area Question type Answer options Descriptive statistics

General features of the Activity sector of the Percentage of firms per class
firm organization
Categorical a. Mechanical and metal working a. 36 per cent
b. Wood working b. 23 per cent
c. Process industry (plastic, rubber, paper, c. 27 per cent
glass, ceramics, construction materials,
food industry)
d. Other (clothing, shoes, accessories and d. 14 per cent
decorations)
Number of employees Ordinal Percentage of firms per class
1. Small (up to 49 employees) 1. 31 per cent
2. Medium (50 to 249 employees) 2. 40 per cent
3. Large ($ 250 employees) 3. 29 per cent
Company’s yearly turnover Ordinal Percentage of firms per class
1. Small (less than 10 million e) 1. 32 per cent
2. Medium (from 10 to 50 million e) 2. 41 per cent
3. Large (more than 50 million e) 3. 27 per cent
Operations time Interval Number of shifts per day (1, 2 or 3 shifts) Mean SD
a. Working shifts per day Operations days per week (5, 6 or 7) Shifts: 1.77 0.98
b. Operations days per week Operations days 5.38 0.72
Structural Maintenance capacity: Ordinal 1. Less than 2 Percentage of firms per class
maintenance strategy number of internal
decisions maintenance operators
2. 3 to 5 1. 41 per cent
3. 6 to 15 2. 28 per cent
4. More than 15 3. 19 per cent
4. 12 per cent
Maintenance facilities: Ordinal 1. Less than 10,000e Percentage of firms per class
Value of maintenance spare 2. 11-50,000e 1. 31 per cent
parts inventory
3. 51-500,000e 2. 29 per cent
4. . 500,000e 3. 23 per cent
4. 17 per cent
(continued)
 Variable
Question area Question type Answer options Descriptive statistics

Maintenance technology: Categorical Yes-No Yes: 52 per cent of firms

Presence of condition-based No: 48 per cent of firms
or predictive maintenance
Maintenance technology: Categorical Yes-No Yes: 35 per cent of firms
Presence of a CMMS No: 65 per cent of firms
Vertical integration: Ordinal 1. Total in-house maintenance Percentage of firms per class
Type of outsourcing 2. Mostly work package contracts 1. 6 per cent
contracts
3. Mostly performance contracts 2. 78 per cent
4. Total outsourcing based on performance 3. 15 per cent
contracts
5. Total outsourcing with integrated 4. 0 per cent
facilities management contracts 5. 1 per cent
Infrastructural Maintenance organization: Categorical a. Centralized technical department 19 per cent Percentage of
maintenance strategy organization of maintenance firms per class
decisions function
b. Decentralized department depending on 54 per cent
production
c. Technical department integrated into 27 per cent
production
Maintenance organization: Ordinal 1. Skilled worker 1. 57 per cent Percentage of
firms per class
Hierarchical level of 2. Middle management 2. 38 per cent
maintenance function head
3. Senior management 3. 5 per cent
Policies and concepts: Interval All parameters are expressed as a Mean SD
percentage of total maintenance workload
Share of reactive maintenance RM þ PrM ¼ 100 per cent RM ¼ 55 per cent 25 per cent
(RM)
Share of preventive With CBM þ Cyclic Preventive ¼ PrM PrM ¼ 45 per cent 25 per cent
maintenance (PrM)
Share of condition-based CBM ¼ 10 per cent 11 per cent
maintenance (CBM)
(continued)
and strategy
Matters of size

161

Table I.
 16,2

162

Table I.
JQME

Variable
Question area Question type Answer options Descriptive statistics

Policies and concepts: Categorical Yes-No Yes: 16 per cent

of firms
Presence of TPM No: 84 per cent
of firms
Planning and control: Categorical Yes, all – Some – No Yes, all: 35 per
cent of firms
Maintenance orders are Some: 46 per
written cent of firms
No: 19 per cent
of firms
Planning and control: Categorical Yes-No Yes: 11 per cent
of firms
Is a periodic maintenance No: 89 per cent
budgeting foreseen, at least of firms
on monthly basis?
Planning and control: Categorical Yes-No Yes: 39 per cent
of firms
Existence of a spare parts No: 61 per cent
stock book of firms
Planning and control: Categorical Yes-No Yes: 48 per cent
of firm
Do you account for No: 52 per cent
inefficiency costs? of firms
Planning and control:
If you have a CMMS, to what Ordinal Five-point Likert scale, from 1 ¼ Never to Mean SD
extent is it used for: 5 ¼ Always
a. Work order planning and a. 3.54 1.48
scheduling
b. Preventive maintenance b. 3.94 1.16
planning and scheduling
c. Spare parts inventory c. 3.26 1.36
management
d. Maintenance budgeting d. 2.14 1.31
e. Maintenance data filing? e. 4.29 0.99
(continued)
 Variable
Question area Question type Answer options Descriptive statistics

Human resources: Ordinal Percentage of firms per class:

Educational level of 1. Up to junior high school
maintenance head
2. At least high school diploma
Human resources: Ordinal Gross annual salary (in thousands e per Percentage of firms per class:
year)
Maintenance head 1. Less than 25 a. 24 per cent
remuneration
2. Between 25 and 50 b. 68 per cent
3. Between 50 and 75 c. 6 per cent
4. More than 75 d. 2 per cent
Human resources: Ordinal Number of years: Percentage of firms per class:
Maintenance head 1. Up to five years a. Position b. Company
professional experience
a. How long has he/she been 2. Five to ten years 1. 32 per cent 1. 22 per cent
in the position?
b. How long has he/she been 3. More than ten years 2. 19 per cent 2. 19 per cent
with the company?
3. 49 per cent 3. 59 per cent
Human resources: 1. Up to five years
Tasks of maintenance 2. Five to ten years
function:
How much emphasis is 3. More than ten years
placed on these activities as
responsibilities of your
plant’s maintenance function?
a. Monitoring the production Ordinal Five point Likert scale, from 1 ¼ Not Mean SD
equipment status important to 5 ¼ One of the most
important
b. Analyzing equipment a. 4.07 0.93
failure causes and effects
c. Restoring equipment to b. 3.88 1.06
operation
(continued)
and strategy
Matters of size

163

Table I.
 16,2

164

Table I.
JQME

Variable
Question area Question type Answer options Descriptive statistics

d. Planning preventive c. 4.19 0.90

maintenance work
e. Installing new equipment d. 3.61 1.11
f. Cooperating in the new e. 3.11 1.05
equipment purchasing
process
g. Helping design and f. 2.97 1.11
improve the production
process
g. 2.76 1.18
Maintenance Over the last year, how much Ordinal Five-point Likert scale, from Mean SD
performance has maintenance contributed 1 ¼ Negligible contribution to
to: 5 ¼ Decisive contribution
a. The improvement of a. 3.68 0.93
equipment availability
b. The reduction of b. 3.40 1.02
production costs
c. The improvement of c. 3.22 1.01
product quality
d. The improvement of d. 3.37 1.12
occupational safety?
Four major areas were investigated, that is general features of firms (manufacturing Matters of size
sector and company size, expressed in terms of number of employees and of yearly and strategy
turnover, and operations time), some structural decisions (maintenance capacity,
facilities, outsourcing type and maintenance technologies including the existence of
some computerized maintenance management system – CMMS in the following – and
of some predictive or condition based maintenance approach – CBM in the following)
and infrastructural decisions (maintenance organization, planning and control 165
including usage of CMMSs, adopted policies and features of human resources,
including also maintenance function tasks as identified by Swanson (2001)) – forming
maintenance strategy as defined by Pintelon et al. (2006) – and maintenance
performance, which was measured by Swanson (2001), with addition of a further
question about the contribution of maintenance to the improvement of occupational
safety.

2.1 Data collection

Our attention was focused on manufacturing firms with at least 20 employees.
Directories of the population of local manufacturing firms were obtained from the
Industrial Union of Pordenone and of Udine and the Chamber of Commerce of Treviso.
These three provinces are located in North Eastern Italy, which is a most industrialized
part of the country, having the highest density of privately owned firms and the
highest employment rate among Italian regions (ISTAT, 2008). The local scope of the
survey allowed to administrate the questionnaire through on-site visits by trained
personnel, rather than via e-mail or fax. Firms were contacted telephonically to arrange
appointments with maintenance heads, to who the questionnaires were addressed.
Thanks to this approach and to the support of associations, 100 usable questionnaires
were obtained out of an originally selected sample of 120 firms, drawn from the
population of about 1,400 firms with more than 20 employees derived from the
associations’ directories. The general features of the respondents’ sample are
summarized in the first block of Table I. As to organizations’ size, a fair composition of
the sample was obtained, which allows to investigate size related features, bearing
however in mind that the actual share of small firms in the considered population is
definitely larger (reaching about 70 per cent) and that of large firms is much smaller
(amounting to circa 2 per cent) than in the sample. As to manufacturing sectors, the
sample composition mirrors the population structure, where metalworking and
machine construction are the leading activities. Altogether, the considered population
is consistent with the general features of Italian manufacturing sectors as described in
section 1.

2.2 Data analysis

Statistical techniques were used to analyse collected data in order to obtain answers to
our main questions, i.e. whether there is significant diversity in maintenance strategies
and performance across different firm size classes or manufacturing sectors and how
far does obtained performance depend on structural and infrastructural maintenance
strategy decision elements.
The selection of statistical techniques mostly depended on the level of measurement
of gathered answers, which is shown in column entitled “variable type” in Table I. It
can be observed that our answer types fall within all measurement levels usually
JQME distinguished, that is categorical data types, ordinal data types and interval data types.
16,2 Table II shows the statistical methods that can be adopted and were generally used in
our analysis to investigate dependences between variables of different types.
In particular, for ordinal data types, nonparametric statistics and tests, such as the
Mann-Whitney U test, the Kruskal-Wallis H test and the Spearman’s rank correlation
coefficient, are appropriate, as they do not rely on assumptions about the probability
166 distribution of the sample populations and are based on rank statistics (for an overview
see Conover (1980) and Mendenhall and Sincich (2007)).
Actually, many questions of our survey are based on five point Likert type scales.
The way these data can be elaborated is controversial: it is generally agreed that Likert
scales fall within the order level of measurement because intervals between values
cannot be presumed equal (see Jamieson (2004)), but, while some authors strictly treat
related data as ordinal, employing the median as measure of central tendency and
nonparametric statistics for analysis (see, e.g. Jonsson (1997)), others deem it acceptable
to handle them as interval, using parametric statistics, especially if sample size is large
(see, e.g. Reiman and Oedewald (2004)). Although the size of our sample is also large,
our approach in elaborating Likert scale based data consists in using non parametric
statistics – which rely on comparing ranks or medians – nevertheless presenting
means and standard deviations as descriptive statistics rather than or beside rank
means or medians whenever this provides better information about behaviour within
categories, as done, for instance, by Ahmed et al. (2004).

3. Results and discussion

Descriptive statistics reported in Table I provide a general picture of average
maintenance strategies in the examined sample of firms. With regard to structural
decisions, we can observe that internal maintenance capacity is quite small (about 70
per cent of firms have up to five maintenance operators and about 60 per cent of firms
have a spare part inventory value of less than 50,000e). In most cases, the internal
capacity is complemented by external support: vertical integration in maintenance
apparently characterizes only 6 per cent of firms, who declare to be completely
self-sufficient. The majority of firms primarily use the most basic form of maintenance
contract, i.e. work package contracts. As noted by Tsang (2002), this mode of
contracting is task oriented and this hinders firms from reaping all the benefits of

First variable type Second variable type Statistical technique

Categorical Categorical Contingency tables and Chi-Square tests

Fisher’s exact test
Categorical Ordinal Categorical variable with two levels: Mann-Whitney test
Categorical variable with more than two levels: Kruskal-
Wallis test and multiple comparisons among pairs
Categorical Interval Student’s t-test or one-way ANOVA
Ordinal Ordinal Spearman’s r-rank correlation coefficient
Table II.
Ordinal Interval Spearman’s r-rank correlation coefficient
Adopted statistical
techniques Interval Interval Pearson’s r-product moment correlation coefficient
maintenance outsourcing, because occasional service suppliers are bound to minimize Matters of size
their investments in staff development, equipment and new technologies. The more and strategy
advantageous but more complex performance-contracting mode is, however, selected
only by 15 per cent of firms as main option.
As to technologies, CBM is widely spread, involving 52 per cent of firms, whereas
the diffusion of CMMSs is really limited, especially considering that the proportion (35
per cent of firms) is comparable with those found in other countries more than ten 167
years ago – see Ikwan and Burney (1994), Jonsson (1997), Swanson (1997).
Looking at corresponding infrastructural decisions, we can also note that the
prevalent usage of CMMSs is for data recording (mean Likert score 4.29) and
preventive maintenance planning (mean Likert score 3.94), while more complex
activities involving elaboration of data are seldom performed (e.g. maintenance
budgeting, mean Likert score 2.14).
As to organization, a decentralized maintenance department depending on
production functions represents the prevailing structure (54 per cent of firms).
In 57 per cent of firms, the head of this department (or of the maintenance function
anyway) is a skilled worker and only in remaining 43 per cent of firms does he belong
to middle or senior management. This seems consistent with the limited dimensions of
internal maintenance capacity, but also with more general human resource aspects
disclosed in this research: 22 per cent of maintenance heads have only a junior high
school degree and 24 per cent have a gross yearly salary of less than 25,000e; thus, it
looks as if the “status” of maintenance managers is quite low in almost one firm out of
four. Experience is apparently a determinant of the career of maintenance heads, as 49
per cent of them have been in charge for more than ten years and 59 per cent have been
with their companies for more than ten years.
As to maintenance planning and control, formalization is limited: maintenance orders
are all written only in 35 per cent of firms, a spare parts stock book exists in just 39 per
cent of firms and only a minority of companies (11 per cent) have monthly budgeting.
Compared with previous Italian studies (Ghirardo, 2004), a positive sign is the growing
awareness of inefficiency costs, which are – however roughly – taken into account in 48
per cent of firms. Finally, concerning maintenance policies and concepts, we found a
limited diffusion of TPM (present in 16 per cent of the organizations) and a reactive
maintenance proportion of about 55 per cent, which is well above recommended values of
30-40 per cent (see, for example, Jonsson (1997)). The result is similar to those reported in
older surveys concerning Italy (Ferrari et al. (2002) or Ghirardo (2004)) and other
countries (Jonsson (1997)). Although widespread, CBM has a limited weight among
normal maintenance policies (about 10 per cent of total maintenance).
All these remarks refer to the average profile of examined sample of firms, which is
however composed by a balanced mix of small, medium and large ones belonging to
different sectors. Having obtained a general picture, our aim was to highlight and
understand context specific differences, i.e. to recognize whether performance and
strategies vary with size, operation time or manufacturing sector.
In order to interpret context specific differences correctly, we preliminarily tested
our sample for interdependencies between contextual variables. For completeness, size
was separately evaluated both in terms of employees and turnover, although we
proved that these two variables are strongly correlated and measure the same
dimension. Performing the x 2 test on the size-sector contingency tables, we verified
JQME that classification of firms according to size and sector are independent, i.e. the way
16,2 firms are distributed among size classes does not vary significantly with
manufacturing sector. On the other hand, patterns of operation time appear to be
significantly correlated to size and hardly correlated to sector. Nonparametric tests –
preferred because working shifts per day and operations day per week were far from
normally distributed – showed that medium and large firms operate on more shifts
168 than small firms and that the variables firm size and number of shifts grow together.
Also the number of operating days per week is positively correlated with size, although
more feebly than the number of shifts (see Table III).
In interpreting our results, we should thus bear in mind that larger firms often adopt
a more extensive operation time, which makes the overcoming of fire-fighting
approaches more impellent and the introduction of advanced fault prevention
technologies more rewarding.
Finally, regarding sector dependence, the average weekly operation time in process
industries (mean x
¼ 104, standard deviation s ¼ 53.9) is higher than in other sectors
(
x ¼ 74.7 s ¼ 42.1 for metal working, x
¼ 56.7 s ¼ 37.7 for wood working, x
¼ 74.8
s ¼ 43.8 for other activities), but from a statistical viewpoint the difference is not
significant.
Having ascertained these general features, we then investigated the relationships:
.
between contextual variables (sector, size, operation time) and performance (in
terms of equipment availability, production cost reduction, product quality
improvement, improved occupational safety);
.
between contextual variables and strategic decisions; and
.
between strategic decisions and performance
with the aim of supporting or rejecting hypotheses of mutual dependence.

3.1 Contextual variables and performance

Our first step was to investigate how far performance is affected by operation context.
As observed by Cua et al. (2001) usually “large organizations are more centralized and
formalized than small organizations and have more resources to deploy for the
implementation of manufacturing programs”. We could hypothesise that such
deployment of resources will cause larger plants to have better performance than
smaller plants.
Similarly, as mentioned previously, a more extended operation time with higher
inefficiency costs will make investments in more advanced maintenance management
technologies and practices more attractive. Therefore, we also want to test the
hypotheses that better performance will be achieved in plants with longer operation
time and in the process industry, rather than in other sectors, because in that sector
inefficiency costs are generally higher.
The dependence of performance on sector was tested with Kruskal-Wallis ANOVA.
The hypothesis of better performance in process industries was not confirmed. In spite
of some variations of average scores in distinct sectors, with process industry reporting
better cost reduction performance, the only difference resulting statistically significant
concerns safety related performance, which is perceived to be substantially higher in
the metal working and machine manufacturing sectors than in process industry and in
other sectors (Likert score mean x
safety ¼ 3.88 in metal and machine manufacturing
 Cost
Number No. of Weekly Weekly reduction Quality
Standard of Yearly working operation operation Availability Likert improvement
Mean deviation employees Turnover shifts time days Likert score score Likert score

Number of employees * 58.24 29.62

Yearly turnover * 58.17 29.6 0.792
No. of working shifts 1.77 0.91 0.474 0.471
Weekly operation time (hours) 78.5 47.5 0.437 * * 0.447 * * 0.963 * *
Weekly operation days 5.38 0.72 0.193 0.234 * * 0.480 * * 0.667 * *
Availability Likert score 3.68 0.93 20.061 20.173 0.033 2 0.001 20.078
Cost reduction Likert score 3.40 1.02 20.111 2 0.132 0.210 * * 0.181 0.036 0.439 * *
Quality improvement Likert score 3.22 1.01 0.098 0.074 0.194 0.188 0.138 0.099 0.303 * *
Safety improvement Likert score 3.37 1.12 0.024 0.016 0.006 2 0.026 20.067 0.244 * * 0.102 0.540 * *
Notes: For ordinal variables (marked with *) rank-based descriptive statistics are presented. Correlations marked with * * (values above 0.20) are
significant at p , 0.05. Correlations in italics (absolute values above 0.17) are significant at p , 0.10
and strategy

Spearman’s correlations

variables and
between contextual

performance variables
Matters of size

Table III.
169
JQME against x
safety ¼3 in the process industry and x
safety ¼2.71 in other activities, with
16,2 p , 0:05). This outcome may be related to sector specific culture in metalworking and
machine manufacturing, which could be the object of future investigations.
On the contrary, the hypothesis of better performance in plants with longer operation
time is partially supported, as can be deduced from the correlation analysis presented in
Table III. Firms having a higher number of working shifts assign higher scores to the
170 contribution of maintenance to cost reduction (r ¼ 0:21, significant at p , 0:05) and a
weaker positive correlation (r ¼ 0:19, significant at p , 0:10) can also be found between
number of working shifts and product quality performance, while no effect can be
observed on availability and safety scores. The effect of the number of working days is,
conversely, not relevant and, as a result, even the correlation between weekly operation
time and cost and quality performance scores is significant only at p , 0; 10.
Finally, as can be observed in Table III, performance scores are generally unrelated
to size. In the only case where a weakly significant correlation (p , 0; 10) can be
detected, that is between yearly turnover and contribution to availability improvement,
the correlation coefficient is negative, i.e. the smaller the firm, the better the perceived
performance. This single result is of limited significance and would not lead us to
formulate contradictory hypotheses, i.e. that smaller mean better, nevertheless we can
argue that our original hypothesis of better performance in larger plants is far from
being supported by empirical data.

3.2 Contextual variables and strategic decisions

As can be deduced from results summarized in Table IV, sector depending differences
in maintenance strategy decisions are minimal. Significant ones concern organization
(with a prevalence of dependence on production and transfer of maintenance tasks to
production in the metal working sector, of centralized maintenance departments in
process industry and other industries and of a combination of both structures in the
wood working sector) and the diffusion of TPM, which is concentrated in the metal
working and machine manufacturing industries (it involves 31 per cent of the firms of
this sector but only 7-8 per cent of firms of other sectors).
On the other hand, Table IV shows that size related contextual variables are
correlated with most structural elements of maintenance strategies examined in this
research. Clearly, the measures of maintenance capacity and facilities show a
statistically significant increase with both firm size and firm operation time. Larger
firms are also more capable of adopting more advanced forms of contracting out
maintenance (performance contracting is the main option in 31 per cent of large firms
against 10 per cent of medium firms and 6 per cent of small ones).
The presence of CMMSs is directly connected with firm size: CMMSs exist in 29 per
cent of small firms, in 37 per cent of medium ones and in 41 per cent of large ones
(Mann Whitney test significant at p , 0; 05 with size measured in terms of number of
employees). Interestingly, there is however no statistically significant difference in the
corresponding infrastructural decision variable, i.e. in the way CMMSs are used across
small, medium and large companies (see Table V).
The diffusion of condition based maintenance is not statistically dependent on firm
sizes (although it is adopted by 39 per cent of small firms, 60 per cent of medium ones
and 55 per cent of large ones) but it grows significantly with operation time (CBM is
present in about 39 per cent of firms operating on a single shift basis, but in more than Matters of size
65 per cent of firms operating on a two or three shift basis). and strategy
Size influences significantly almost every infrastructural variable of maintenance
strategy, as can be deduced from statistical results summarized in Table V. As to
maintenance organization, dependence on production and transfer of maintenance
tasks to production operators is the most common choice in small organizations (77 per
cent of small firms, against 47 per cent of medium and 36 per cent of large ones), while 171
a centralized technical department is more common in large enterprises (44 per cent)
and in medium enterprises (24 per cent) than in small ones (16 per cent), and a
combination of centralized function integrated into production is more common in
medium firms (29 per cent) than in small (7 per cent) or large ones (20 per cent).
As can be expected, also the hierarchical level of maintenance head gets
significantly higher with firm size (head belongs to middle or senior management only
in 16 per cent of small firms against 52 per cent of medium ones and 59 per cent of large
ones) and with operation time (head belongs to middle or senior management in 33 per
cent of firms operating on a single shift, 44 per cent of firms operating on two shifts and
59 per cent of firms operating continuously). The latter result is consistent with the fact
that, with growing operation time, organization of maintenance as a centralized
independent department is more common (this form is selected by about 15 per cent of
firms operating on a single shift, 21 per cent of firms operating on two shifts and 56 per
cent of firms operating on three shifts). Increasing firm size goes with a generally
higher status of maintenance management, i.e. with higher education (93 per cent of
maintenance heads of large firms hold at least an high school diploma, against 74 per
cent in medium firms and 71 per cent in small ones) and experience (93 per cent of
maintenance heads in large firms have been at least five years in the position, against
56 per cent in medium firms and 58 per cent in small ones). Low retribution is
associated with small firm size (an annual salary of less than 25,000e is earned by 39
per cent of maintenance heads in small firms, 18 per cent in medium ones and 11 per
cent in large ones). Furthermore, we found that the larger the firms, the higher the
Likert scores attributed to the importance of various department tasks, with the
exception of restoring equipment, which is almost equally rated as most important
irrespective of firm size. This goes with the general empowerment of maintenance
heads in large firms. On the contrary, although an extended operation time is expected
to imply larger responsibilities of maintenance head, correlations between this
contextual parameter and an empowerment of human resources are limited to higher
education and experience in the company (but not necessarily in the position) and with
more importance placed on monitoring and preventive maintenance planning as tasks
of maintenance functions.
Differences in planning and control tools are coherent with higher formalization in
larger organizations (Cua et al., 2001): main differences concern monthly budgeting
(present in 24 per cent of large firms, 10 per cent of medium ones and in none of the
small ones), spare parts stock books (present in 62 per cent of large firms, 38 per cent of
medium ones and just 17 per cent of small ones) and inefficiency cost accounting
(accounted for in 72 per cent of large firms, 45 per cent of medium ones and 29 per cent
of small ones). The first two aspects also display a similar behaviour with reference to
operation time (e.g. monthly budgeting applies to 33 per cent of firms operating on
three shifts, 6 per cent of firms operating on two shifts, 0 per cent of single shift firms)
 16,2

172
JQME

Table IV.

performance
elements and
decisions, contextual
Links between structural
Organization size Manufacturing sector Operation time
Correlation Correlation Correlation
or or or Maintenance performance
dependence Statistical dependence Statistical dependence Statistical Statistical
Structural decisions is verified? test used is verified? test used is verified? test used Availability Costs Quality Safety test used

Capacity Internal Yes * r No K-W Yes * r No Yes * Yes * Yes * r

operators
Facilities Value of Yes * r No K-W Yes * r No No Yes * No r
spare parts
store
Technology Presence of No M-W No x2 Yes * M-W No Yes * Yes * Yes * M-W
CBM
Presence of Yes * * M-W Yes * x2 No M-W No No No No M-W
a CMMS
Vertical Outsourcing Yes * r No K-W No r No No No No r
integration type
Notes: *Verified at p , 0.05; * *verified at p , 0.05 for the number of employees only; p ¼ Spearman’s rank based correlation; M-W ¼ Mann-Whitney
U-test; K-W ¼ Kruskal-Wallis H-test
 Organization size Manufacturing sector Operation time
Correlation Correlation Correlation
or or or Maintenance performance
dependence Statistical dependence Statistical dependence Statistical Statistical
Infrastructural decisions is verified? test used is verified? test used is verified? test used Availability Costs Quality Safety test used

Maintenance Organization Organization Yes * K-W Yes * x2 Yes * K-W No No No No K-W

organization of function of function
Hierarchical Hierarchical Yes * * r No K-W Yes * r No No No No r
level of head level of head
Policies and Share of PM Share of PM No r No K-W No r Yes * Yes * Yes * Yes * r
concepts
Share of Share of CBM Yes * * r No K-W Yes * r No Yes * Yes * Yes * r
CBM
2
TPM TPM No M-W Yes * x No M-W No No Yes * Yes * M-W
Planning Written Written No K-W No x2 No K-W No No No No K-W
and control orders orders
Monthly Monthly Yes * M-W No x2 Yes * M-W No No No No M-W
budgeting budgeting
Spare parts Spare parts Yes * M-W No x2 Yes * M-W No No Yes * No M-W
stock book stock book
2
Inefficiency Inefficiency Yes * M-W No x No M-W No No No No M-W
costs costs
accounting accounting
Use of Work order No r No K-W No r No No No No r
CMMS for planning
PM planning No r Yes * K-W No r No No No Yes * r
Spare part No r No K-W Yesb r No Yes * No No r
management
Budgeting No r No K-W Yesc r No Yes * No No r
Data filing No r No K-W No r No No No No r
(continued)
and strategy

performance
infrastructural decisions,
contextual elements and
Links between
Table V.
Matters of size

173
 16,2

174
JQME

Table V.
Organization size Manufacturing sector Operation time
Correlation Correlation Correlation
or or or Maintenance performance
dependence Statistical dependence Statistical dependence Statistical Statistical
Infrastructural decisions is verified? test used is verified? test used is verified? test used Availability Costs Quality Safety test used

Human Maintenance Education Yes * r No K-W Yes * r No No No Yes * r

resources head
Remuneration Yes * r No K-W No r No No No No r
Years in Yes * * * r No K-W No r No Yes * No No r
position ,-
Years in Yes * * * r No K-W Yes * r No No Yes * No r
company
Department’s Monitoring Yes * r No K-W Yesa r No No Yes * No r
tasks
Cause Yes * r No K-W No r No No Yes * Yes * r
analysis
Restoring No r Yes * K-W No r Yes * No No Yes * r
PM planning Yes * r No K-W Yesa r Yes * Yes * No No r
Installation Yes * r No K-W No r No No No Yes * r
Purchasing Yes * r No K-W No r No No Yes * Yes * r
Process Yes * r No K-W No r No No Yes * No r
design
Notes: * p , 0.05; * * * p , 0.05 with NE only; * * p , 0.10 with NE, p , 0.05 with YT; ap , 0.05 with number of shifts only; bp , 0.05 with operation time, at p , 0.10 with
number of shifts; cp , 0.05 with number of shifts, p , 0.10 with operation time; r ¼ Spearman’s rank based correlation; M-W ¼ Mann-Whitney U-test; K-W ¼ Kruskal-
Wallis H-test; ? ¼ Negative correlation coefficient
and this fact goes with an increased use of CMMS for monthly budgeting and spare Matters of size
parts management when operation time grows. and strategy
Finally, the weight of maintenance policies and concepts show different behaviours
across size classes. While the diffusion of CBM is size-independent, its proportion on
total maintenance is positively correlated with firm size (on average it equals 5,5 per
cent in small firms, 12 per cent in medium and 11 per cent in large ones). On the other
hand, the share of preventive maintenance as a whole does not change with size and, 175
although the average diffusion of TPM concepts grows with size, involving 10 per cent
of small firms, 15 per cent of medium and 24 per cent of large ones, these variations are
not statistically significant. Observed relationships of maintenance policies and
concepts with operation time are similar, i.e. positive correlation of CBM share, no
change in preventive maintenance share and diffusion of TPM increasing but not in a
statistically significant way.

3.3 Strategic decisions and performance

Last columns of Tables IV and V summarize the relations between structural and
infrastructural elements of maintenance strategies and maintenance performance, as
qualitatively measured in our survey. Notes on tables report statistical significance
data; it should be stressed that, apart from one highlighted case, all significant
correlation coefficients obtained are positive, i.e. the higher the intensity or quality of
maintenance practice, the higher the performance.
We first noted that, for a substantial set of measures, no significant correlation with
maintenance performance could be observed.
In particular, we could find little or no relations between maintenance performance
and:
.
outsourcing type and presence of CMMS, among structural decision elements;
and
.
organizational features as well as most planning and control measures, among
infrastructure decision elements.
As to the outsourcing type, it is possible that advantages brought about by a more
extensive performance contracting cannot be highlighted statistically because the
number of firms who apply this evolved outsourcing mode is really limited.
Concerning CMMS presence, it is helpful to underline that the way CMMS is used
actually has a significant impact on performance. In particular, a more frequent usage
for PM planning is associated with better safety performance, while a more intense
usage for spare parts management and maintenance budgeting is significantly linked
to a stronger contribution to lower production costs. Thus, the point is apparently in
using a CMMS rather than in having a CMMS. Many firms acquire systems which they
do not seem to use and benefit from: this is coherent with other observations in
literature, e.g. as argued by Carnero and Novés (2006) “a high percentage of CMMSs are
inoperative, little-used or eliminated after brief periods of time because they fail to
provide the promised benefits (. . .) or certain modules are not used due to lack of
personnel training or the opposite attitude to the change experienced by the human
resources”. According to our survey CMMSs are mainly used for data filing: storing
maintenance data in CMMS is practical and helpful to prove compliance with
occupational safety regulations, especially as far as preventive maintenance
JQME obligations are concerned (which explains the link with safety performance), but little
16,2 use or elaboration of stored data seems to be done. This is totally consistent with
similar remarks in literature, e.g. according to Garg and Deshmukh (2006) “CMMS
appears to be used less often as a device for analysis and coordination. The uses so far
appear to be only a storehouse for equipment information, as a preventive maintenance
tool and a maintenance work planning tool”. From our results, three levels of use are
176 roughly distinguished: the basic storehouse use – with no impact on performance – a
more developed PM use – with some impact on safety performance – and the most
advanced stage of budgeting use, with investments in CMMSs starting to pay off also
from a cost reduction perspective.
Organizational features and planning and control measures hardly have significant
impact on performance, except for some sporadic correlation (between spare parts
recording and product quality improvement, which is consistent with the correlation
between spare parts store value and product quality and with a positive correlation –
expected and observed – between spare parts store value and spare parts recording).
Such infrastructural decision elements, which are substantially size dependent, mostly
seem to be necessary conditions to keep control and preserve maintenance performance
levels when organizations become larger and more complex.
On the other hand, there is a subset of variables, which significantly influence
performance scores, impacting on several indicators at the same time. The main ones
are, in order of decreasing importance:
.
the proportion of preventive maintenance to total maintenance;
. the presence of CBM and the share of CBM on total maintenance;
.
maintenance capacity, i.e. the number of internal operators;
.
the presence of TPM; and
.
human resource aspects, in particular maintenance department tasks.
Although “the solution for maintenance is not always just to increase planned
preventive maintenance (because) preventive maintenance programmes are not always
kept up to date, nor do they always work in harmony with production schedules”
( Jonsson, 1997), our empirical results confirm that, as repeatedly stressed in literature
(Swanson, 2001), more preventive maintenance is associated with better performance
as to all examined performance areas. What is more interesting, enhancing preventive
maintenance is a cross-sector and cross-size measure, which is within easy reach of
small firms and enables them to achieve optimum maintenance performance scores.
Among preventive maintenance approaches, condition based maintenance
demonstrates to be extremely effective, being positively correlated with better cost,
quality and safety performance. Also this practice appears easily available to small
and medium firms, who could improve their performance by further extending the use
of predictive maintenance approaches to proportions comparable with those of large
firms.
The introduction of TPM also proves advantageous, however mostly in terms of
quality, and safety, as a statistically significant impact on costs is not perceived. This
finding is corroborated by similar results of other empirical studies, e.g. by McKone
et al. (2001), who hypothesise that “TPM allows for effective use of the budgeted
maintenance expenses and is able to improve inventory turns, quality, and delivery
while maintaining stable production costs”. Also for this maintenance concept Matters of size
applicability and effectiveness do not depend on firm size. and strategy
As to maintenance capacity, the relevant impact on performance may be explained
by the fact that when more staff is employed, maintenance represents a larger portion
of total costs and of total personnel concerned by occupational health and safety
measures, and so changes in performances have more substantial impacts. The
observed positive correlation between capacity and product quality improvement is 177
less immediate, but somehow coherent with the finding that also larger spare parts
inventories are linked with higher quality performance and with the consideration by
Pinjala et al. (2006) that “quality environments require high speed of maintenance
response to prevent defective products”, which could be attained through
overcapacities both in terms of internal personnel and of spare parts inventories.
Finally, human resource aspects usually impact on single performances and are
perhaps more representative of personal and organizational visions of maintenance
and its objectives, rather than direct determinants of its results. Maintenance tasks, in
particular, are quite representative of the organization’s vision of maintenance: those
who see restoring malfunctioning equipment and planning preventive maintenance as
main duties are coherently more satisfied with equipment availability performance,
while firms focusing on quality as an underlying competitive priority see practices
with a longer term impact, such as continuous monitoring and root cause analysis, as
most important and are, as we derive also from Swanson (2001) and Pinjala et al. (2006),
those who most actively pursue proactive and aggressive practices of process
improvement.

4. Conclusions and future research

Our survey investigated the state of maintenance management in Italian firms
considering some structural and infrastructural decision elements of manufacturing
strategy as defined by Pintelon et al. (2006). The general picture evidences some
criticalities, such as too much fire-fighting and limited preventive approaches, and,
particularly in small firms, low status of maintenance management as to internal
capacity, retribution and education of persons in charge and inadequate diffusion and
use of planning and control tools, especially CMMSs. Most of these critical issues were
nevertheless pointed out by research on manufacturing firms in several western
countries (see, e.g. Cooke (2003) or Carnero and Novés (2006)). On the other hand, also
some strong points were brought to light, including the long experience of most
maintenance heads, growing awareness of inefficiency costs and the increasing
diffusion of condition based maintenance and of TPM across all industries,
independently of size. Mean judgements on maintenance performance are generally
positive (see Table I): although they are generally more optimistic than similar
measurements presented in literature (see Swanson, 2001), the size of our sample (100
firms) and the numerous and consistent correlations between maintenance best
practices and performances highlighted in this study support our trust that these
scores describe correctly, if qualitatively, the obtained maintenance results.
As to performance, in particular, an interesting finding is that no direct correlation
between firms’ size and maintenance performance could be observed: good results are
equivalently reported by smaller and larger firms and therefore seem not “a matter of
size”. We should anyway bear in mind that this finding also depends on the way
JQME maintenance performance could be measured here, i.e. only with qualitative scores.
16,2 Furthermore, the analysis of links between other contextual variables, specifically
operation time, and performance leads us to argue that, if there is any positive effect of
size on maintenance performance, then it seems to be rather of indirect type:
performance correlates positively with operation time and operation time is positively
correlated with firm size.
178 A common cause, which could be examined in future investigations, could be the
level of automation and of advanced manufacturing technologies.
We can, in fact, hypothesise that organizations investing in automation and
production technologies are more likely:
.
to extend their operation time, in order to make investments in automation and
productivity more rewarding;
.
to risk having higher inefficiency costs, both because hourly productivity is
higher and because total operation time is longer; and
.
to contrast this risk by pursuing more actively good maintenance practices (see
Swanson, 2003) and thus to achieve better maintenance performance.

Such firms could be more easily large ones, because they could reap economies of scale
and above all possess larger capitals to invest in AMT: given the results of this survey,
this is however not necessarily true and should be the subject of further studies. In
particular, since Pinjala et al. (2006) highlighted that quality competitors make a more
extensive use of AMT and of good maintenance practices, it would be interesting to
investigate differences in business strategies across organization size classes, to
disclose, for instance, whether larger firms have more quality oriented strategies.
From a practical viewpoint, we could observe that maintenance visions and
strategies influence maintenance results significantly, but the most effective and direct
determinants of maintenance performance seem to be those “matters of strategy”
directly linked with maintenance policies and with TPM concepts. In particular, a
practical implication for maintenance managers we could derive from this study is the
confirmation that CBM substantially contributes to improve performance: with
predictive maintenance technologies becoming more widespread and cheaper, this
practice can be easily applied even by small firms, leading them to optimize their
maintenance results.
Finally, the analysis of our empirical data from the examined area confirms the
general indication that the proportion of preventive maintenance, including CBM, should
be extended: however, this could be done effectively only if opportune maintenance
engineering instruments are adopted to assure that preventive maintenance programs
harmonize well with production schedules and with the actual state of manufacturing
equipment. This can be achieved through a more extensive and above all more conscious
use of CMMSs by adequately trained human resources: maintenance personnel could
thus be empowered, waste of resources could be avoided and a synergic positive effect
could add up to the extension of proactive maintenance practices.

References
Ahmed, S., Hassan, M.H. and Taha, Z. (2004), “State of implementation of TPM in SMIs: a survey
study in Malaysia”, Journal of Quality in Maintenance Engineering, Vol. 10 No. 2,
pp. 93-106.
Carnero, M.C. and Novés, J.L. (2006), “Selection of computerised maintenance management Matters of size
system by means of multi-criteria methods”, Production Planning and Control, Vol. 17
No. 4, pp. 335-54. and strategy
Cattaneo, M. (2000), Maintenance in SMEs in Italy, AIMAN, Milan, available at: 06/memorie_
seminario_ancona_genn_2006/mantenere_per_competere_gennaio_2006_cattaneo.pdf
(in Italian) (accessed 11 August 2008).
Cholashuke, C., Bhardwa, R. and Antony, J. (2004), “The status of maintenance management in 179
UK manufacturing organizations: results from a pilot survey”, Journal of Quality in
Maintenance Engineering, Vol. 10 No. 1, pp. 5-15.
Cigolini, R. and Turco, F. (1997), “Total productive maintenance practices: a survey in Italy”,
Journal of Quality in Maintenance Engineering, Vol. 3 No. 4, pp. 259-72.
Conover, W.J. (1980), Practical Nonparametric Statistics, John Wiley & Sons, New York, NY.
Cooke, F.L. (2003), “Plant maintenance strategy: evidence from four British manufacturing
firms”, Journal of Quality in Maintenance Engineering, Vol. 9 No. 3, pp. 239-49.
Cua, K.O., McKone, K.E. and Schroeder, R.G. (2001), “Relationships between implementations of
TQM, JIT and TPM and manufacturing performance”, Journal of Operations
Management, Vol. 19 No. 6, pp. 675-94.
Dowlatshahi, S. (2008), “The role of maintenance management in the maquiladora industry:
an empirical analysis”, International Journal of Production Economics, in press.
Ferrari, E., Pareschi, A., Persona, A. and Regattieri, A. (2002), “TPM: situation and procedure for
a soft introduction in Italian factories”, The TQM Magazine, Vol. 14 No. 6, pp. 350-8.
Garg, A. and Deshmukh, S.G. (2006), “Maintenance management: literature review and
directions”, Journal of Quality in Maintenance Engineering, Vol. 12 No. 3, pp. 205-38.
Ghirardo, G. (2004), “Maintenance in Pordenone province”, Manutenzione Tecnica e
Management, February, pp. 41-4, available at: www.manutenzione-online.com (in
Italian) (accessed 11 August 2008).
Ikwan, M.A.H. and Burney, F.A. (1994), “Maintenance in Saudi industry”, International Journal
of Operations & Production Management, Vol. 14 No. 7, pp. 70-80.
ISTAT (2001), “88 General census of industry and service sectors, data warehouse”, available in
Italian at: http://dwcis.istat.it/cis/index.htm (accessed 11 August 2008).
ISTAT (2008), Hundred Statistics for the Country (in Italian), ISTAT, Rome, available at: www.
istat.it/dati/catalogo/20080507_01/testointegrale20080507.pdf (accessed 11 August 2008).
Jamieson, S. (2004), “Likert scales: how to (ab)use them”, Medical Education, Vol. 38 No. 12,
pp. 1212-18.
Jonsson, P. (1997), “The status of maintenance management in Swedish manufacturing firms”,
Journal of Quality in Maintenance Engineering, Vol. 3 No. 4, pp. 233-58.
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.
Mendenhall, W. and Sincich, T. (2007), Statistics for Engineering and the Sciences, Pearson
Prentice-Hall, Upper Saddle River, NJ.
Mitchell, E., Robson, A. and Prabhu, V.B. (2002), “The impact of maintenance practices on
operational and business performance”, Managerial Auditing Journal, Vol. 17 No. 5,
pp. 234-40.
Nakajima, S. (1988), TPM – An Introduction to Total Productive Maintenance, Cambridge
Productivity Press, New York, NY.
JQME Pinjala, S.K., Pintelon, L. and Vereecke, A. (2006), “An empirical investigation on the relationship
between business and maintenance strategies”, International Journal of Production
16,2 Economics, Vol. 104 No. 1, pp. 214-29.
Pintelon, L. and Van Puyvelde, F. (1997), “Maintenance performance reporting systems: some
experiences”, Journal of Quality in Maintenance Engineering, Vol. 3 No. 1, pp. 4-15.
Pintelon, L., Pinjala, S.K. and Vereecke, A. (2006), “Evaluating the effectiveness of maintenance
180 strategies”, Journal of Quality in Maintenance Engineering, Vol. 12 No. 1, pp. 7-20.
Reiman, T. and Oedewald, P. (2004), “Measuring maintenance culture and maintenance core task
with CULTURE-questionnaire: a case study in the power industry”, Safety Science, Vol. 42
No. 2004, pp. 859-89.
Swanson, L. (1997), “An empirical study of the relationship between production technology and
maintenance management”, International Journal of Production Economics, Vol. 53
No. 1997, pp. 191-207.
Swanson, L. (2001), “Linking maintenance strategies to performance”, International Journal of
Production Economics, Vol. 70 No. 3, pp. 237-44.
Swanson, L. (2003), “An information-processing model of maintenance management”,
International Journal of Production Economics, Vol. 83, pp. 45-64.
Tsang, A.H.C. (2002), “Strategic dimensions of maintenance management”, Journal of Quality in
Maintenance Engineering, Vol. 8 No. 1, pp. 7-39.
Tse, P.W. (2002), “Maintenance practices in Hong Kong and the use of the intelligent scheduler”,
Journal of Quality in Maintenance Engineering, Vol. 8 No. 4, pp. 369-80.

About the authors

Damiana Chinese is assistant professor in Industrial Plant Engineering at the Department of
Energy Technology, University of Udine, Italy. She holds a Master’s degree in Industrial
Engineering from the University of Udine and a PhD degree in Energy Technology from the
same institution. Her current research interests include the design and operational optimization
of energy conversion plants and of manufacturing facilities as well as reliability and
maintenance optimization for industrial equipment and plants. Damiana Chinese is the
corresponding author and can be contacted at: damiana.chinese@uniud.it
Gianni Ghirardo holds a Master’s degree in Mechanical Engineering from the University of
Padua. Having enjoyed a long and distinguished career as a plant and safety manager in several
large firms based in North-Eastern Italy and within multi-national groups, he is now also
chairman of the regional section of AISL – the Italian Society for Work Studies – based in
Pordenone, Italy and committed to fostering maintenance and operations culture among
industrial associations and managers.

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com

Or visit our web site for further details: www.emeraldinsight.com/reprints