International Journal of Food Science and Technology 2012, 47, 22432251

Invited review
Reliability, availability and maintainability analysis in food
production lines: a review
Panagiotis Tsarouhas*
Department of Logistics, Alexander Technological Educational Institute of Thessaloniki, Katerini, Pieria 60100, Greece
(Received 7 December 2011; Accepted in revised form 2 April 2012)

Summary

The scope of this study is to review reliability, availability and maintainability (RAM) analysis in the food
industry and aims to identify the critical points of the production systems that should be improved by the
operational performance and the maintenance eectiveness. RAM is an engineering tool that addresses
operations and safety issues of production lines and aims to identify areas within the system or process where
signicant improvement can be achieved. Food production lines consist of several machines supplied with a
common transfer mechanism and control system that have dierent failure modes. When a random failure
occurs, the failed machine stops and forces most of the line upstream of the failure to operate without
processing, whereas the material (raw, intermediate or end-product) of the line downstream may have to be
scrapped because of quality deterioration during the stoppage. The negative failure impact is the drop of line
reliability and production rate.

Keywords

Availability, food industries, maintainability, reliability management, strategic management.

Introduction

Reliability, availability and maintainability (RAM) management play an important role in the success of a
company. With these three measures, one can estimate the
eectiveness of the production system. The objectives of
RAM analysis are multifaceted and include operations
and safety issues and aim to identify areas within the
production system or process where signicant improvement can be achieved. Thus, the probability of failure,
equipment down time and availability of the system can
be computed. High costs motivate seeking engineering
solutions to reliability problems for reducing nancial
expenditures, enhancing reliability, satisfying customers
with on-time deliveries through increased equipment
availability and by reducing costs and problems arising
from products that fail easily (Barringer, 2000).
There exists a large volume of literature discussing
RAM analysis for various types of products and applications. Hajeeh & Chaudhuri (2000) have worked on
reliability and availability assessment of reverse osmosis,
which assessed the performance of reverse osmosis plants
in the Arabian Gulf region by analysing its failure
behaviour and down time patterns. de Castro & Cavalca
(2003) have presented an availability optimisation prob*Correspondent: Fax: +302351047861; e-mail: ptsarouh@logistics.
teithe.gr.

lem of an engineering system assembled in a series

conguration, which has the redundancy of units and
maintenance teams as optimisation parameters. McFadden (1990) has proposed techniques for developing the
database for RAM improvement programme for an
industrial plant or commercial building. Smith & Smith
(2002) have presented a method of calculating system
availability and reliability probability distributions using
permutations of inseparable system failure and restore
data sets. Biswas et al. (2003) have given a methodology
to calculate the availability of a periodically inspected
system, maintained under an imperfect repair policy.
Measuring the reliabilities of plant and equipment by
quantifying the annual cost of unreliability incurred by
the facility puts reliability into a business context.
Higher plant reliability reduces equipment failure costs.
Failure decreases production and limits gross prots
(Warburton et al., 1998). Barabady & Kumar (2008)
reported that system reliability, maintainability and
availability have assumed great signicance in recent
years because of competitive environments and overall
operating production costs; an unplanned failure can
result in signicantly higher repair costs than a planned
maintenance or repair. Increasing reliability by improving design, materials and production early on will lead
to fewer failures and may decrease maintenance costs
later on (Blischke & Murthy, 2003). Low levels of
reliability mean more accidental failures on the production

doi:10.1111/j.1365-2621.2012.03073.x
 2012 The Author. International Journal of Food Science and Technology  2012 Institute of Food Science and Technology

2243

2244

RAM analysis in food production lines: a review P. Tsarouhas

system, leading to low availability. A machine has an

inherent reliability created at the design stage; thus,
machine reliability studies are necessary for equipment
design, modication and quality control during manufacturing, eld trials and life testing (Samanta, 2001).
Reliability, availability and maintainability is one of
several quality management techniques applied to
improving eciency and productivity on food production lines. It can be applied with other total quality
management (TQM) tools, that is, failure mode and
eect analysis, Pareto analysis, statistical process control, etc. However, there is a scarcity of the literature
regarding RAM analysis in food production lines. In
these lines, all the machines are connected in series.
When a failure occurs anywhere on the line, everything
upstream of the failure stops, creating a gap in production down the line. Moreover, material in some parts of
the stopped line may have to be scrapped because of
quality problems. Thus, the actual or eective processing rate of the line can be substantially less than the
nominal production rate (Liberopoulos & Tsarouhas,
2002). In another study, Tsarouhas & Liberopoulos
(2004) studied the problem of determining the frequency
of quality control inspections in a pizza production line
based on eld failure data analysis. Kumar et al. (1992)
developed an analytical model and some results from an
analytic study of reliability and availability of an actual
crystalliser system in sugar plants. The crystalliser
system consists of ve basic repairable subsystems in
series, where each subsystem is considered as being
good, reduced or failed. Tsarouhas (2005) analysed
some alternative solutions and proposed a maintenance
policy for increasing the eective processing rate of a
croissant production line without raising the cost
through reliability analysis. Entrup et al. (2005) developed mixed-integer linear programming models that
integrated shelf-life issues into production planning and
scheduling, which were based on an industrial case study
of a yogurt production line. Moreover, Tsarouhas
(2009) exhibited a methodology to classify the primary
failure modes in categories based on failure data of a
bread production line covering a period of 2 years.
Chakhlevitch et al. (2011) presented the application of
operational research methodology to the logistics of
food testing at a microbiology laboratory, using bottleneck analysis. The overall eect on the entire process is
to allow continuous testing throughout the day and an
increase in the capacity of the laboratory of over 20%,
with no additional sta or capital investment. In recent
research, Tsarouhas (2011) presented a comparison
study of performance evaluation between four pizza
production lines, and a statistical analysis including the
computation of descriptive statistics of eld failure and
repair data for the lines was performed.
This study reviews RAM analysis in the food industries
and aims to identify the critical points of the production

International Journal of Food Science and Technology 2012

systems that should improve the operational performance

and the maintenance eectiveness. This study is organised
into a presentation of RAM theory, its application to the
food industry (bread and bakery products, bottling and
canning and dairy products) and nally, conclusions.
Reliability, availability and maintainability theory

Reliability, availability and maintainability aims to

identify the critical points within the system or process
where improvement actions can be initiated. Thus, the
systems performance is aected by RAM, which
consists of three fundamental parameters, that is,
RAM. Figure 1 shows the role of these parameters in
improving the system eectiveness.
Reliability

Reliability is the probability that a machine or system will

perform a required function, under stated conditions, for
a stated period of time t. Thus, reliability is the probability
of nonfailure in a given period of time. If T is the
continuous random variable that represents the time
between failure (TBF) of the system with T 0, then the
reliability can be expressed as (Wang & Pham, 2006),
Rt PT  t

If F(t) is the unreliability of the system, then

Rt Ft 1

Thus, the unreliability is the probability that a failure

occurs before time t,
Ft 1  Rt PT<t

F(t) is also called the cumulative distribution function

of the failure distribution.
In reliability theory, the hazard or failure rate
function is indicated as
kt ft=Rt

where f(t) is the probability density function of the

failure distribution and is dened by
ft

dFt
dRt

dt
dt

The expected or mean value of f(t) is the mean time

between failure (MTBF) and is given by

MTBF

Z/
0

tftdt

Z/
Rtdt

The failure behaviour of various engineering components is called the failure rate function (Fig. 2). It is a

 2012 The Author

International Journal of Food Science and Technology  2012 Institute of Food Science and Technology

RAM analysis in food production lines: a review P. Tsarouhas

System effectiveness

Operational
readiness

Mission
availability

Design
adequacy

Reliability
characteristics

Maintainability
characteristics

Prolonging the life of items/

components/subsystems

Minimize the time to restore an item

back to service following a failure

Perform trade off analysis, Study

environmental and design factors

Estimate MTTR & Redesign for

increased accessibility to
maintenance spares, components &
consumables

Perform system modeling &

reliability analysis studies, such as
FMEA, FTA, common cause
failures

FMEA: Failure modes and effects analysis

FTA: Fault tree analysis
MTTR: Mean time to repair
Figure 1 Framework for improving system effectiveness (adapted from Sharma & Kumar, 2008). FMEA, failure modes and eects analysis; FTA,
fault tree analysis; MTTR, mean time to repair.

Failure rate

(t)

Burn-in
(I)

Useful life
(II)

Wear-out
(III)

Time t

Figure 2 The bathtub curve.

function of time and represents the failure rate of each

component (or system). Because of its shape, it is known
as the bathtub curve and is divided into three regions,
that is, I, II and III.
Region I is known as burn-in that has a decreasing
failure rate because failures are caused by manufacturing defects, poor workmanship and poor quality
control.

Region II is characterised as the useful life, during

which the component failure rate remains constant
because of random loads, human error, natural failures
and abuse. Region III is called wear-out that has an
increasing failure rate because failures are caused by
fatigue, ageing, corrosion and friction.
From the bathtub curve, it is obvious that the failure
rate of a component (or system) is dierent according to
the failure mode. Thus, in the burn-in region, the failure
rate decreases; in the useful life region, the failure rate is
constant, whereas in the wear-out region, the failure rate
increases. Therefore, each failure mode follows a dierent failure distribution and requires dierent mathematical models for its minimisation.
Availability

Availability is dened as the ability of an item (under

combined aspects of its reliability, maintainability and
maintenance support) to perform its required function
at a stated instant of time or over a stated period of time

 2012 The Author

International Journal of Food Science and Technology  2012 Institute of Food Science and Technology

International Journal of Food Science and Technology 2012

2245

RAM analysis in food production lines: a review P. Tsarouhas

(Rausand & Hyland, 2004). The availability of a

system can never be less than system reliability because
the availability is the probability that the component is
currently in a nonfailure state, even though it may have
previously failed and been restored to its operational
state (Ebeling, 1997). Consider a reparable system that is
in operation at time t = 0: when the system fails, a
repair action takes place to restore the system to an
operational state. The state of the system can be given
by a binary variable:

1 if the system is operating at time t
Xt
0 otherwise
Availability can be expressed mathematically in several ways:
1 Instantaneous or point availability, A(t): is the probability that the reparable system or equipment will be
available to operate at any random time t:
7

When the system is not repaired, then A(t) = R(t). The

unavailability At at time t of a reparable system is the
probability that the system is in nonoperating state at
time t, then
At 1  At PrXt 0

Am t2  t1

1
t2  t1

Atdt

t1

It is interesting that in some applications, the interval

availability from start-up time interval (0, T) is dened
as:
ZT

Am T

1
T

Atdt

kr t rt=1  Mt

In Fig. 3, typical diagrams for both parameters TBFs

and TTRs at line level are shown. It is observed that the
TBF initially decreases and then has a constant failure
Parametric Hazard Plot for TBF line
Weibull
Complete Data - ML Estimates

0.006
0.005
0.004
0.003
0.002
0.001
0.000
0

1000

MTBF
MTBF MTTR

3000

4000

5000

Parametric Repair Plot for TTR Line

Weibull
Complete data - ML Estimates

11

where MTBF is the mean time to failure, and the MTTR

is the mean time to repair a failure.

2.0

1.5

1.0

0.5

0.0
0

Maintainability

TTR Line (h)

Maintainability is the probability that a failed machine

or system will be restored to operational eectiveness
within a given period of time t when the repair action is

International Journal of Food Science and Technology 2012

2000

TBF line (h)

10

3 Steady-state or long-run availability, Ass: is the probability a system will be available to operate at any point of
time t, as

T!1

13

The MTTR is the average time it takes to restore a

system to operational status after it has failed to
function and can be calculated by:
Z/
14
MTTR 1  Mtdt

Ass lim At

12

The repair rate function is given by,

2 Mission, interval or average availability, Am(t2)t1): is the

proportion of time over the interval (t2)t1) when the
system or the equipment is available to operate:

Zt2

Mt PTr  t

Rate (failures per hour)

At PrXt 1

performed in accordance with the prescribed procedures. Maintainability is the probability of completing
the repair at a given time. If Tr is the continuous random
variable representing the time to repair (TTR) of the
system, having a probability density function of r(t),
then according to Ben-Daya et al. (2009), the maintainability is as follows:

Rate (repairs per hour)

2246

Figure 3 Typical diagrams for time between failures (TBFs) and time
to repairs (TTRs) at line level, using the maximum-likelihood (ML)
estimation method.

 2012 The Author

International Journal of Food Science and Technology  2012 Institute of Food Science and Technology

RAM analysis in food production lines: a review P. Tsarouhas

rate. On the other hand, the TTR presents an increased

repair rate, meaning that the maintenance sta receive
more expertise with time and manage to minimise the
TTR.
According to the international standard regarding
maintenance, dependability and quality of service (CEN
EN 13306, 2001), maintenance is dened as a combination of actions intended to retain an item in (or restore it
to) a state in which it can perform the function that is
required for the item to provide a given service. The
maintenance strategies that are usually applied in the
food industry are as follows: (i) corrective maintenance
that is a recovery action performed after failure occurs
on the system to restore it to its operational status. It is
unscheduled and has high application cost. (ii) Preventive maintenance, which is performed periodically to
reduce the probability of failure of the system. It is
scheduled with low application cost, and (iii) predictive
maintenance, which is based on the past trends to
predict failures. It makes use of factors such as
continuous monitoring, analysing, detection, diagnosing
and responding actions. This maintenance focuses on
eliminating the root causes of failure before it leads to a
major breakdown.
Application of reliability, availability and
maintainability in the food industry

Bread and bakery products

In the manufacture of bread and bakery products, it is

observed that when an unexpected failure occurs, the
failed equipment stops and forces most of the line
upstream of the failure to operate without processing
material and causing a gap in production downstream of
the failure. Moreover, if the failure is suciently long, it
may cause an additional gap in production upstream of
the interruption, because the material (raw, intermediate
or end-product) upstream of the interruption may have
to be scrapped because of quality deterioration during
the stoppage. The time corresponding to the total gap in
production is referred to as the time of lost production
(TLP). Thus, the actual production rate of the line can
be substantially less than its nominal production rate
(Liberopoulos & Tsarouhas, 2005). The same has been
reported in other studies for strudel and bread production lines (Tsarouhas et al., 2009a; Tsarouhas & Arvanitoyannis, 2010a). Therefore, the inconvenience of
failures on the actual production rate of the production
lines puts pressure on bread and bakery products
manufacturers to assess and improve the reliability of
their lines.
Liberopoulos & Tsarouhas (2005) presented a statistical analysis of failure data of an automated pizza
production line, covering a period of 4 years. The
analysis included the computation of the most

important descriptive statistics of the failure data, the

identication of the most important failures, the computation of the parameters of the theoretical distributions that best t the failure data and the investigation
of the existence of autocorrelations and cross-correlations in the failure data. It was pointed out that (i) the
availability of the line is 95.45%, while the eciency is
90.43% because the equipment failures cause an additional production gap in the line, and the material
downstream of the failure will have to be scrapped
because of quality deterioration during stoppages. (ii)
The TBF and the TTR at line level are Weibull
distributed. (iii) The failure rate decreases, because
proactive (preventive and predictive) maintenance improves the operating conditions at dierent parts in the
line, and (iv) there is a marginally statistically signicant, positive correlation between TBFn,n + 1 and
TTRn + 1 and between TTRn and TBFn,n + 1, at the
two-tailed 0.05 signicance level. This suggests that the
longer the time between two failures, the more problems
accumulate, and therefore, the longer the time it takes to
x the latter failure.
In a recent study, Tsarouhas (2011) performed a
comparison study of performance evaluation between
four pizza production lines (L1, L2, L3 and L4). Statistical analysis, including the computation of descriptive
statistics of eld failure and repair data for the production
lines, was carried out. The failure and repair data covered
a 1-year period. The best t of the failure and repair data
between the common theoretical distributions was found,
and its parameters were presented. It was concluded that:
(i) all the pizza production lines followed the lognormal
distribution for both TTF and TTR. (ii) A failure may
occur in the rst shift of operation of the line, after which
the probability for a failure to occur in the next shift
diminished with time, and (iii) if the repair had not been
completed in the rst 45 min and goes on for a rather long
time, then this indicated serious problems on the line.
Tsarouhas & Arvanitoyannis (2010a) displayed a
statistical analysis of the bread production line of the
failure and repair data at machine and line levels. The
experiment covers a period of 25 months. The best t of
the failure data between the common theoretical distributions was found, and its parameters were computed.
The reliability and hazard rate modes for all machines
and the entire production line were calculated as well. It
was pointed out that (i) the availability of the bread
production line is 90.74% and went down to 86.76%,
because the equipment failures caused an additional
production gap in the line. (ii) The best t for both
failure and repair data was the Weibull distribution. (iii)
53.5% of all failures were revealed at the bread machine,
cooling tower machine and volumetric-divider machine
and (iv) the machines of the bread production line that
displayed increasing hazard rate functions were identied.

 2012 The Author

International Journal of Food Science and Technology  2012 Institute of Food Science and Technology

International Journal of Food Science and Technology 2012

2247

2248

RAM analysis in food production lines: a review P. Tsarouhas

In another study, Tsarouhas et al. (2009a) developed a

reliability and maintainability analysis of a strudel production line at machine, workstation and entire line level.
Descriptive statistics of the failure and repair data were
carried out, and the best tness index parameters were
determined. Data collection from production line and
their analysis were analysed over a period of 16 months.
Furthermore, the reliability and hazard rate modes for all
workstations and the entire production line were calculated. It was pointed out that (i) the availability of the
strudel production line was 93.76% and dropped to
89.68%, because the equipment failures caused an additional production gap in the line; some or all of the inprocess material downstream of the failure had to be
scrapped because of quality deterioration during stoppages. (ii)The Weibull distribution provided the best t for
the strudel production line to describe time between
successive failures and repair times. (iii) The failure rate
decreased and the repair rate increased, because of the
maintenance policy applied on the line while the maintenance personnel improved their expertise with time, and
(iv) the reliability and hazard rate models both at workstations and entire line level were determined.
In Table 1, the characteristics of some bread and
bakery production lines are shown, and the following
conclusions can be deduced: (i) the availability is
between 90.74% and 96.95%, whereas the eciency is
86.76% (bread line) and 93.71% (pizza L2). (ii) The
maximum failure rate is observed in pizza (L3) production line with 0.7888 failures per shift. (iii) The maximum value of mean TBF is in pizza (L2) line with
1110 min, and (iv) the minimum value of mean TTR is
in pizza (L3) line with 33 min, whereas the maximum
value of mean TLP is observed in bread line with
94 min.
Bottle and canned products

Tsarouhas et al. (2009b) conducted a reliability and

maintainability analysis for the juice bottling industry

by applying statistical techniques in eld failure data.

Reliability, maintainability, failure rate and repair rate
models of the production line for all workstations and
the entire production line were developed. Data collection from the line and their analysis were collected over
a long time of 45 months. It was pointed out that (i) the
availability of the juice production line was 85.66% and
the eciency amounts to 82.10% because of equipment
failures. (ii) The production line presented a failure
every 12.5 h or the equivalent two failures per day
approximately, while the MTTR is approximately 2 h.
(iii) The TTF followed the Weibull distribution, while
the TTR followed the lognormal distribution, and (iv)
the machines with the lowest availability of the juice
bottling industry were identied.
Reliability analysis for packaging during beer production over a period of 9 months was carried out
(Tsarouhas & Arvanitoyannis, 2010b). The most important failure modes were identied, and the descriptive
statistics at failure and machine level were calculated.
Several theoretical distributions were applied, and the
best t for failure data was identied. The reliability and
hazard rate models of the failure data were determined
to provide an estimate of the current operation management and improve the line eciency. The main
research ndings can be summarised as follows: (i) the
availability of the beer lling capping machine was
94.80%, (ii) the failures because of mechanical and
pneumatic causes amounted to 57.1% of all the failures
of the machine, (iii) the TBF was very short indicating a
high probability of failure, which suggested that the
current maintenance policy should be revised, and (iv)
the TBF followed the normal distribution, whereas the
TTR followed the logistic distribution.
In a recent study, Tsarouhas & Arvanitoyannis (2011)
developed the reliability and maintainability analysis of
an automated peach production line. Descriptive statistics of all the failures at machine and line level were
shown, and the most critical machines under failures
according to several criteria were determined. The best

Table 1 Characteristics of the bread and bakery production lines

Production lines

Total shiftsa

Number of failures

Failure rate

Mean TBFb

Mean TTR

Mean TLP

Availability

Efficiency

Pizza
Pizza L1
Pizza L2
Pizza L3
Pizza L4
Bread
Strudel

4473
696
696
696
696
2301
1224

1773
420
285
549
500
1512
711

0.3964
0.6034
0.4095
0.7888
0.7184
0.6581
0.5809

716.88
802.284
1110.318
668.634
703.44
636.708
651.774

34.2600
35.3929
34.8947
33.4153
33.8900
60.7110
36.2588

72.0976
76.5763
74.5825
70.9490
67.3040
94.1501
62.5570

0.9545
0.9577
0.9695
0.9524
0.9540
0.9074
0.9376

0.9043
0.9129
0.9371
0.9041
0.9127
0.8676
0.8968

Adapted from Liberopoulos & Tsarouhas, 2005; Tsarouhas et al., 2009a; Tsarouhas & Arvanitoyannis, 2010a; Tsarouhas, 2011.
TBF, time between failure; TLP, time of lost production.
a
A shift corresponds to 8 h of operation.
b
All the times are expressed in minutes.

International Journal of Food Science and Technology 2012

 2012 The Author

International Journal of Food Science and Technology  2012 Institute of Food Science and Technology

RAM analysis in food production lines: a review P. Tsarouhas

t of the failure data between the common theoretical

distributions was found, and the respective parameters
were identied. Both the reliability and hazard rate
models for the entire production line were calculated,
providing useful tools to assess the current situation and
to predict reliability for upgrading the operation management of the peach production line. It was pointed out
that (i) the availability of the peach production line was
90.30%, (ii) the failure times followed the logistic
distribution, whereas the repair times comply with the
Weibull distribution, (iii) the TTR decreased, thereby
implying that the maintenance sta expertise increased
with time and (iv) the most important, machines with
failures that need immediate actions of the peach
production line were identied.
Table 2 shows the characteristics for some bottle and
can production lines, and the following observations can
be made: (i) the availability of the juice bottling line is
85.66%, whereas for the beer packaging is 94.80%. (ii)
The failure rate of the beer packaging is 0.1026 failures
per shift and 0.6641 failures per shift for the peach
production line, and (iii) the maximum MTBF is
4432 min for the beer packaging, whereas the minimum
MTTR is 70 min for the peach line.
Dairy products

In dairy lling and packing lines, because of line

imbalance, an extra time is required for the upstream
lling station, which has low production rate to work in
advance to build work-in-process. Thus, the downstream packing station can operate at higher productivity during a shift to achieve the desired production
volume. In food manufacturing, production lines aim to
reach a steady state in a short time. In dairy production
lines, products are so perishable that they cannot be
safely left in buers at the end of the shift, leaving such

systems to start the next shift with empty buers and the
consequent loss of production until a steady state is
reached. Wang et al. (2010) presented a study to
determine the buer capacity in dairy lling and packing
lines through transient analysis and investigated the
system production rate and work-in-process during
transients, the inventory build-up time, shift operating
time, nal clean up time and high production rate
period. Moreover, sensitivity analysis for larger buer
capacity, higher lling station throughput and initial
inventory build-up was carried out. Doganis & Sarimveis (2007) proposed a mixed-integer linear programming model for optimal production scheduling in a
single yogurt production line. The model was applied to
a yogurt production line of a major dairy company
where the availability of the machines constituted a
signicant restriction to production. The model took
into account all the standard constraints encountered in
production scheduling (material balances, inventory
limitations, machinery capacity, labour shifts and manpower restrictions). Furthermore, it considered special
features that characterise yogurt production, which are
limitations in production sequencing mainly because of
dierent fat contents and avours of various products
and sequence-dependent set-up times and costs.
Tsarouhas et al. (2009c) investigated the RAM of a
cheese production line over a period of 17 months. The
reliability and hazard rate modes at the entire production line were calculated as well. The models are
anticipated to be a useful tool to assess the current
conditions and to predict the reliability for upgrading
the maintenance strategies of the production line. It was
found out that (i) the availability of the cheese production line was 91.20% and went down to 87.03%. (ii)
Both the TBF and TTR were lognormal distributed. (iii)
The dominant four failure modes comprised 62.2% of
all the failures of the cheese production line. (iv) The

Table 2 Characteristics of the bottle and can production lines

Production lines

Total shifts

Number of failures

Failure rate

Mean TBF

Mean TTR

Availability

Efficiency

Juice bottling
Beer packaging
Peach

2283
750
393

1261
77
261

0.5523
0.1026
0.6641

748
4432
650

125
244
70

0.8566
0.9480
0.903

0.8210

Adapted from Tsarouhas et al., 2009b; Tsarouhas & Arvanitoyannis, 2010b, 2011.
TBF, time between failure.

Table 3 Characteristics of cheese (feta) production line

Production lines

Total shifts

Number of failures

Failure rate

Mean TBF

Mean TTR

Availability

Efficiency

Cheese (feta)

456

292

0.6403

750

66

0.9120

0.8703

Adapted from Tsarouhas et al., 2009c.

TBF, time between failure.

 2012 The Author

International Journal of Food Science and Technology  2012 Institute of Food Science and Technology

International Journal of Food Science and Technology 2012

2249

2250

RAM analysis in food production lines: a review P. Tsarouhas

TBF was higher during the rst 50 min of production,

but decreased thereafter, indicating that the probability
of failure was high initially but then diminished. (v)
Similarly, the TTR was higher during the rst 40 min of
production, then declined, indicating that if a repair
process has not been completed in the rst 40 min and
goes on for a rather long time, then this indicates serious
problems on the line, that is, inadequate skill of
maintenance sta, no spare parts are available on the
warehouse, insucient management, etc.
The following conclusions can be derived for cheese
(feta) production line based on Table 3: (i) the failure
rate of the line is 0.6403 failures per shift, and (ii) the
MTBF is 750 min, whereas the MTTR is 66 min. The
distributions for both TBF and TTR of the production
lines under study are summarised in Table 4.
Conclusions

Reliability, availability and maintainability analysis in

the food industry is a tool that can increase the
performance and the production rate of the lines and
evaluate the impact of failures and proposed changes to
current operations management. Machines with high
risk of failure can be identied and so extend the useful
life of the production system. In addition, the reliability
and hazard rate models could prove to be useful to
assess the current conditions and to predict the reliability for upgrading the maintenance policy of the food
production line. Thus, a companys decisions can be
improved by quantifying losses and benets.
Moreover, the RAM will contribute to improving the
operations management of food production lines by
providing real and up-to-date data concerning the actual
state of the plant. Therefore, it should be viewed as an
ongoing process that will need to be continuously
improved and developed.
For future research, it is essential to investigate the
failures in more food production lines in several sectors
Table 4 Distributions for both TBF and TTR of the production lines
under study
Production lines

TBF

TTR

Pizza (2005)
Pizza L1, L2, L3 and L4 (2011)
Bread
Strudel
Juice bottling
Beer packaging
Peach
Cheese (feta)

Weibull
Lognormal
Weibull
Weibull
Weibull
Normal
Logistic
Lognormal

Weibull
Lognormal
Weibull
Weibull
Lognormal
Logistic
Weibull
Lognormal

Adapted from Liberopoulos & Tsarouhas, 2005; Tsarouhas et al.,

2009a,b,c; Tsarouhas & Arvanitoyannis, 2010a,b, 2011; Tsarouhas, 2011;.
TBF, time between failure.

International Journal of Food Science and Technology 2012

and to apply RAM analysis based on maintenance

strategy, maintenance sta (i.e. technician level, training, number of technicians), spare parts warehouse
policy, etc. Then, the researchers should then re-estimate
the reliability of these production lines and measure the
eciency and the productivity within the frame of TQM
principles.
References
Barabady, J. & Kumar, U. (2008). Reliability analysis of mining
equipment: a case study of crushing plan at Jajarm Bauxite mine in
Iran. Reliability Engineering & System Safety, 93, 647653.
Barringer, P.E. (2000). Reliability Engineering Principles. Humble, TX
77347, USA: Barringer & Associates, [email protected].
Ben-Daya, M., Duua, O.S., Raouf, A., Knezevic, J. & Ait-Kadi, D.
(2009). Handbook of Maintenance Management and Engineering.
NY, USA: Springer-Verlag London Limited.
Biswas, A., Sarkar, J. & Sarkar, S. (2003). Availability of a periodically
inspected system, maintained under an imperfect-repair policy.
IEEE Transactions on Reliability, 52, 311318.
Blischke, W.R. & Murthy, D.N.P. (2003). Case Studies in Reliability
and Maintenance. USA: John Wiley & Sons.
de Castro, H.F. & Cavalca, K.L. (2003). Availability optimization with
genetic algorithm. International Journal of Quality & Reliability
Management, 20, 847863.
CEN EN 13306 (2001). Maintenance Terminology. Brussels: European
Standard (European Committee for Standardization).
Chakhlevitch, K., Glass, A.C. & Sadd, A.P. (2011). Alleviating
bottlenecks in a microbiology laboratory. Journal of Food Engineering, 103, 377387.
Doganis, P. & Sarimveis, H. (2007). Optimal scheduling in a yogurt
production line based on mixed integer linear programming. Journal
of Food Engineering, 80, 445453.
Ebeling, C.E. (1997). An Introduction to Reliability and Maintainability
Engineering. New York, NY: McGraw Hill.
Entrup, M.L., Gunther, H.-O., van Beek, P., Grunow, M. & Seiler, T.
(2005). Mixed-integer linear programming approaches to shelf-lifeintegrated planning and scheduling in yogurt production. International Journal of Production Research, 43, 50715100.
Hajeeh, M. & Chaudhuri, D. (2000). Reliability and availability
assessment of reverse osmosis. Desalination, 130, 185192.
Kumar, D., Singh, J. & Pandey, P.C. (1992). Availability of the
crystallization system in the sugar industry under common-cause
failure. IEEE Transactions on Reliability, 41, 8591.
Liberopoulos, G. & Tsarouhas, P. (2002). Systems analysis speeds up
Chipitas food processing line. Interfaces, 32, 6276.
Liberopoulos, G. & Tsarouhas, P. (2005). Reliability analysis of an
automated pizza processing line. Journal of Food Engineering, 69,
7996.
McFadden, R.H. (1990). Developing a database for a reliability
availability and maintainability improvement program for an
industrial plant or commercial building. IEEE Transactions on
Industry Applications, 26, 735740.
Rausand, M. & Hyland, A. (2004). System Reliability Theory:
Models, Statistical Methods, and Applications, 2nd edn. NJ, USA:
John Wiley & Sons Inc.
Samanta, B. (2001). Reliability analysis of shovel machines used in an
open cast coal mine. Mineral Resources Engineering, 10, 219231.
Sharma, R.K. & Kumar, S. (2008). Performance modeling in critical
engineering systems using RAM analysis. Reliability Engineering &
System Safety, 93, 891897.
Smith, J.B. & Smith, W.B. (2002). Probabilistic assessment of
availability from system performance data. In: Proceedings of
Reliability and Maintainability Symposium (edited by IEEE). Pp.
569576. New York, NY, USA: IEEE.

 2012 The Author

International Journal of Food Science and Technology  2012 Institute of Food Science and Technology

RAM analysis in food production lines: a review P. Tsarouhas

Tsarouhas, P. (2005). Analysing event data from a repairable

automated systems: a case study. Information Technology Journal,
4, 228232.
Tsarouhas, P. (2009). Classication and calculation of primary failure
modes in bread production line. Reliability Engineering & System
Safety, 24, 551557.
Tsarouhas, P. (2011). A comparative study of performance evaluation
based on eld failure data for food production lines. Journal of
Quality in Maintenance Engineering, 17, 2639.
Tsarouhas, P. & Arvanitoyannis, I. (2010a). Reliability and maintainability analysis of bread production line. Critical Review in Food
Science and Nutrition, 50, 327343.
Tsarouhas, P. & Arvanitoyannis, I. (2010b). Assessment of operation
management for beer packaging line based on eld failure data: a
case study. Journal of Food Engineering, 98, 5159.
Tsarouhas, P. & Arvanitoyannis, I. (2011). Quantitative analysis for
peach production line management. Journal of Food Engineering,
105, 2835.
Tsarouhas, P. & Liberopoulos, G. (2004). Determining quality
inspection frequency in an automated production line based on
eld failure data analysis. Operational Research: An International
Journal, 4, 305315.

Tsarouhas, P., Varzakas, T. & Arvanitoyannis, I. (2009a). Reliability

and maintainability analysis of strudel production line with experimental data; a case study. Journal of Food Engineering, 91, 250259.
Tsarouhas, P., Arvanitoyannis, I. & Ampatzis, Z.D. (2009b). A case
study of investigating reliability and maintainability in a Greek juice
bottling Medium Size Enterprise (MSE). Journal of Food Engineering, 95, 479488.
Tsarouhas, P., Arvanitoyannis, I. & Varzakas, T. (2009c). Reliability
and maintainability analysis of cheese (Feta) production line in a
Greek medium-size company: a case study. Journal of Food
Engineering, 94, 233240.
Wang, H. & Pham, H. (2006). Reliability and Optimal Maintenance.
NJ, USA: Springer-Verlag London Limited.
Wang, J., Hu, Y. & Li, J. (2010). Transient analysis to design buer
capacity in dairy lling and packing production lines. Journal of
Food Engineering, 98, 112.
Warburton, D., Strutt, J.E. & Allsop, K. (1998). Reliability-prediction procedures for mechanical components at design stage.
Proceedings of the Institution of Mechanical Engineers, 212 (Part
E), 213224.

 2012 The Author

International Journal of Food Science and Technology  2012 Institute of Food Science and Technology

International Journal of Food Science and Technology 2012

2251

Copyright of International Journal of Food Science & Technology is the property of Wiley-Blackwell and its
content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.