Innovative Maintenance Management

Methods in Oil Refineries

M. Bevilacqua, F.E. Ciarapica, G. Giacchetta, C. Paciarotti

and B. Marchetti

This chapter describes the relevant steps for the design of a preventive maintenance
program in an oil refinery plant and its application. The method was developed
during a period of 3 years in one of the main Italian refinery.
The preventive maintenance program for the most critical equipments of the
reference plant is developed analyzing the used oil sector standards (ASA, API,
UNI, ISO, ASME). These standards have been used for several oil refinery
equipments such as pumps, tanks and processing furnaces.
The research has been focused on the turnaround process that has been analyzed
by two different approaches: a risk-based method and an innovative criticality
index. Both have been used as decision tools for clearly assessing the maintenance
tasks and equipment to include in the process. The relevant technical specifications
(operating conditions, process fluids, and safety system configuration) have been
gathered for each component. The data have then been used to assess the failure
mode effects and severity so that to calculate the criticality index. The outcome of
the analysis has highlighted important relations among variable, confirming, and
validating the results obtained through other reliability assessment techniques (see
Bevilacqua et al. 2000, 2012).
The results have highlighted a clear improvement in terms of resources usage
optimization, outage duration, production loss and total costs reduction, and
increase of the interval between two shutdowns.

M. Bevilacqua
F.E. Ciarapica
G. Giacchetta
C. Paciarotti

Dipartimento di Energetica, Università Politecnica della Marche,
via Brecce Bianche, Ancona, Italy
B. Marchetti (&)
Università degli Studi ECampus, via Isimbardi 10, Novedrate, Como, Italy
e-mail: [email protected]

© Springer-Verlag London 2016 197

H. Pham (ed.), Quality and Reliability Management and Its Applications,
Springer Series in Reliability Engineering, DOI 10.1007/978-1-4471-6778-5_7
198 M. Bevilacqua et al.

1 Introduction

In the contemporary highly challenging environment, a reliable production system

is a crucial factor for competitiveness, almost all the processing and manufacturing
sectors are required to maximize availability and efficiency of equipment, con-
trolling failure and deterioration, guarantee a safe and correct operation and mini-
mizing the costs (Zhaoyang et al. 2011). In case of oil and gas industry, the
challenge it is even harder due to the highly critical nature of its activities that
cannot afford unexpected failures (Telford et al. 2011).
For refineries maintenance, inspection functions are the backbone for safe and
reliable plant operations and play a pivotal role in efficiently achieving the desired
production target and profitability for the company. Maintenance functions in the
refinery include mechanical, electrical, instrumentation, and civil functions, which
are responsible for monitoring, repair, and maintenance of equipment in the
respective defined areas. Inspection is responsible for certification of quality
adherence on the repair and maintenance activities through monitoring of static
equipment functioning during plant operations. Fulfillment of statutory require-
ments and liaison with regulatory bodies, failure analysis, and remaining life
assessment of plant equipment to establish a repair replacement plan in advance for
reliable plant operations, metallurgy upgrading, etc., are also performed by
inspection. On the basis of the process condition, on-stream monitoring of all
critical equipment is decided by an individual group and monitored religiously.
Preventive maintenance, predictive maintenance, a structured repair system, and
full-fledged plant shutdown management must be reliable (Kosta and Keshav
2013).
Among others the maintenance of refineries raises the issue that, in the current
international context, average downtime in such industries can reach 10 % of the
production time, and refineries are often used only at 60 % of their capacity
(Dossier special Algérie/Special Issue for the Algerian Petroleum 2002). Thus,
important financial gains and safety improvements can be affected by optimizing
maintenance tasks. In this paper, we focus on the dynamic scheduling of these
maintenance tasks (Aissani et al. 2009).
Planned refinery turnarounds are major maintenance or overhaul activities and
represent the most critical and expensive maintenance task for industries.
A turnaround is a global maintenance operation that involves the partial or total
shutdown of the plant and involves either the critical equipment, that is not possible
to isolate during the normal functioning of the plant, and those that have shown
problems or need a periodical inspection. The frequency of this operations vary by
type of unit, and often requires 1–2 years of planning and preparation, and some-
times longer when major capital equipment changes are needed. A major unit
turnaround may last about 20–60 days and involves as many as 1500–2000 skilled
contractor workers brought on-site to perform a myriad of interrelated jobs that
require significant coordination and safety measures.
Innovative Maintenance Management Methods in Oil Refineries 199

In principle it could be possible to shut down only the portion of the plant
needing maintenance, but the work is normally too disruptive to continue operating.
Moreover, labor assets to perform the daily maintenance work in the operating
portion of the plant will be in short supply. Thus, the maintenance outage most
often involves a total plant shutdown. A turnaround in any major refinery unit can
affect production of finished products, such as gasoline or distillate. Safety is a
major concern when implementing refinery turnarounds. Refineries run with
materials at high temperatures and high pressures, and some of the materials
themselves are caustic or toxic and must be handled appropriately. Maintenance is
required to assure safe operations, and turnarounds themselves require extra safety
precautions.
The turnaround is a common process in the oil and gas industries so that many
best practices and case studies are available on literature as well as the related
regulations and standards; however, there are few research works dealing with the
development of innovative tools for managing and optimizing the correlated
operations.
The research wok presented in this chapter had the aim of filling this gap
proposing an innovative maintenance program applied to the turnaround manage-
ment and based on two different methodologies: the risk analysis and the appli-
cation of an innovative criticality index that has been developed by the authors with
the purpose of being an economic, flexible, simple, and complete decision tool to
apply for evaluating the criticalities of equipments and plants in order to optimize
the use of economical, human, and instrumental resources needed for the refinery
maintenance activities.

2 Relevant Literature

Several papers related to maintenance practices, methodologies, and tools have

been examined, as well as regulations and standards.
In the last decade, the Total Productive Maintenance (TPM) and reliability
centered maintenance (RCM) methodologies have been accepted as the most
promising strategies for improving maintenance performances (Ahuja and Khamba
2008, 2009); advantages and issues related to its application have been addressed in
many studies (Khanna 2008; Bhangu et al. 2011).
Innovative maintenance strategies and methods applied to a wide variety of fields
have been proposed by several authors in the past few years.
Savino et al. (2011) developed a modified FMECA methodology in which the
criticality evaluation is made considering both production performances and
users/workers safety. Tsakatikas et al. (2008), demonstrated the use of FMECA
together with a Decision Support System (DSS) for the establishment of spare parts
criticality with a focus on industrial maintenance needs. Sachdeva et al. (2008),
proposed a new maintenance decision strategy alternative to traditional FMEA
approach and based on Analytical Hierarchical Process (AHP) technique which
200 M. Bevilacqua et al.

provides an aid to the maintenance managers/analysts to formulate an efficient and

effective priority ranking of the various components/failure modes based on a
number of maintenance issues.
Ghosh and Roy (2010) presented a Multiple-Criteria Decision-Making (MCDM)
methodology for selecting the optimal mix of maintenance approaches—Corrective
Maintenance (CM), Time-Based Preventive Maintenance (TBPM) and
Condition-Based Predictive Maintenance (CBPM)—for different equipment in a
typical process plant. According to Rosmaini et al. (2011), preventive maintenance
(PM) strategies have been as well addressed in many studies. One of the popular
strategies that is widely used is the preventive replacement (PR), which aims to
determine the optimum replacement time. The critical issue, however, is that most
studies assume the aging of a component to be time dependent. In reality, the failure
of a component is influenced by an external factor. They present the process of
revising or updating the PR time by considering external factors by using the
proportional hazard model (PHM). Zhou et al. (2010) proposed an opportunistic
preventive maintenance policy for multiunit series systems based on dynamic
programming, with integrating the online information of the intermediate buffers.
An optimal preventive maintenance practice is determined by maximizing the
short-term cumulative opportunistic maintenance cost savings for the whole system
which is a combination of the maintenance cost saving, the downtime cost saving,
the penalty cost for advancing the preventive maintenance action, and the penalty
cost for work in process.
Conn et al. (2010) analyzed the maintenance issues when dealing with multiple
heterogeneous systems, a mix of maintenance strategies (replacement and inspec-
tion), economic dependence among maintenance activity, etc. With this initial
determining factor, the design or reorganization of a maintenance plan must begin
with an exhaustive study of all the equipments and facilities, the purpose of which
is to obtain all the information necessary to justify and analyze the viability of each
maintenance task, thus to decide the best maintenance strategy for the plant. Such a
study must begin with a detailed inventory of equipments and facilities, including
their characteristics and functional interrelationships; records of past failures, if they
exist; the cost of acquisition and supply; the direct and indirect cost of maintenance,
if this information is available; needs and operational factors; the type of mainte-
nance to be carried out and any legal or contractual obligations concerning main-
tenance (such as periodic inspections subject to regulation, guarantee periods, etc.);
the legal or contractual obligations of the company, such as those related with legal
sanctions or penalties regarding the quantity or quality of the production; the means
(tools, auxiliary equipment, etc.) available for maintenance; human resources, and
the qualifications of the personnel available; the maintenance tasks that can or must
be contracted out and any other aspects that are relevant for the case in question,
Gomez et al. (2006).
Peres and Noyes (2003) present a methodology for the evaluation of mainte-
nance strategies by taking into consideration the effect of certain variables on the
dynamic of maintenance, on its structure and its context of evolution. This factual
approach considers the return to an operational state as a point of entry into the
Innovative Maintenance Management Methods in Oil Refineries 201

evaluation of a strategy. It is based on the treatment of data collected from the

history of the behavior of equipment on which the strategy to be evaluated can be
applied.
You-Tern et al. (2004) focused their work of the idea that the development of
maintenance strategies must take into account that resources are limited and,
therefore, maintenance will be imperfect.
Several papers on the evaluation of components reliability have also been taken
into account. Reliability is a fundamental aspect for proper maintenance execution,
current reliability evaluation methods are based on the availability of knowledge
about component states. However, component states are often uncertain or
unknown, especially during the early stages of the development of new systems. In
such cases it is important to understand how uncertainties will affect system reli-
ability assessment. Pham (2013) developed a new software reliability model
incorporating the uncertainty of system fault detection rate per unit of time subject
to the operating environment. Brissaud et al. (2011) presented a model of failure
rates as a function of time and influencing factors that allows to represent the
system life phases considering a large variety of qualitative or quantitative, precise
or approximate influencing factors. Zhang and Mostashari (2011) proposed a
method to assess the reliability of systems with continuous distribution of com-
ponent based on Monte Carlo simulation. They demonstrated that component
uncertainty has significant influence on the assessment of system reliability.
Garg et al. (2010) developed a reliability model for systems that undergoes
partial as well as direct total failure, for calculating both time dependent and steady
state availability under idealized as well as faulty Preventive Maintenance (PM).
Some of the relevant literature analyzed focused in maintenance strategies
applied to oil refineries.
Prabhakar and Jagathy Raj (2013) proposed a maintenance approach based on
the implementation of RCFA programs and reliability-centered maintenance,
alternative to traditional methods of reliability assurance, like preventive mainte-
nance, predictive maintenance, and condition-based maintenance that they consider
inadequate to face the extreme demands on reliability of the plant. The authors
developed an accelerated approach to RCM implementation, that, while ensuring
close conformance to the standard, does not require large amount of analysis, long
implementation, and a high number of skilled people as the traditional ones.
Aissani et al. (2009) presented a multiagent approach for the dynamic mainte-
nance task scheduling for a petroleum industry production system. Agents simul-
taneously ensure effective maintenance scheduling and the continuous improvement
of the solution quality by means of reinforcement learning, using the SARSA
algorithm. Reinforcement learning allows the agents to adapt, learning the best
behaviors for their various roles without reducing the performance or reactivity.
The results obtained in a petroleum refinery demonstrated the innovation of their
approach.
Christen et al. (2011), a competing risk model, namely a Random Sign model, is
considered to relate failure and maintenance times. They proposed a novel Bayesian
analysis of the model and applied it to actual data from a water pump in an oil
202 M. Bevilacqua et al.

refinery, developed an optimal maintenance policy under a formal decision theo-

retic approach, using a competing risk model, namely a Random Sign model, to
relate failure and maintenance times. They proposed a novel Bayesian analysis of
the model and validated it with data from a water pump in an oil refinery.
Laggoune et al. (2009) proposed a maintenance plan based on opportunistic
multigrouping replacement optimization for multicomponent systems applied to
continuous operating units such as oil refinery. Their approach is based on the
analysis of individual components, according to the well-known age-based model.
The optimization algorithm allows rearranging the optimal individual replacement
times in such a way that all component times become multiple of the smallest one to
allow for joint replacements. In this way, the times obtained by the multigrouping
approach do not give individual optimality conditions, but satisfies the optimal cost
regarding the whole system.
In this scenario, the optimization of major maintenance activities represents a
key factor for improving safety and economic aspects. Turnaround in particular, is
the most challenging within the maintenance tasks: industry surveys report that
between 35 and 52 % of maintenance budgets are expended in individual area or
whole plant shutdowns.
Kister and Hawkins (2006) stated that the majority of preventive and planned
maintenance work is performed while the manufacturing plant is in operation but
major maintenance works that cannot be performed while the plant is operating are
also periodically required.
An optimal maintenance approach is a key support to industrial production in the
contemporary process industry and many tools have been developed for improving
and optimizing this task.
As for the standard to consider, in the oil sector the current regulations to
elaborate a preventive maintenance program, (ASA, API, UNI, ISO, ASME) rep-
resents the reference for the key equipments of the plant: pumps, tanks, reservoirs,
and processing furnaces.

3 The Test Plant

The study was carried out in a pilot refinery located in central Italy that operates
directly in the supply of crude oil and semi-refined products destined for processing
holding one of the most modern and technologically advanced refineries in the
country. It is certified to ISO 14001 for environmental protection, OHSAS 18001
for safety and ISO 9002 for quality. Built in 1950 the refinery covers an area of over
700,000 m2, it has a refining capacity of 3,900,000 tons/annum, equivalent to
approximately 85,000 barrels per day, and a storage capacity of over 1,500,000 m3.
The refinery is also equipped with a dispatch facility by land which has a potential
capacity of over 12,000 tons/day. It is connected to the sea through a combination
of marine terminals which can accommodate tankers from 1,000 to 400,000 tons.
Innovative Maintenance Management Methods in Oil Refineries 203

The company owns also one of the first IGCC (integrated gasification combined
cycle) plants constructed in Europe.

3.1 The Production Cycle

The refinery offers a production system that ensures flexibility in operations on the
basis of the crude oil used and the best possible yield in distillates. The system
employs “Hydroskimming”, together with a conversion system “Thermal
Cracking/Visbreaking”: thus high-quality products can be made that meet envi-
ronmental specifications that are required by the regulations.
The processes of refining as the following sequence: Topping—Catalytic
Reforming—Isomerization—Vacuum—Visbreaking—Thermal Cracking. A syn-
thetic scheme of the process is presented in the following Fig. 1.

3.2 The Turnaround Process

The turnaround is a global maintenance operation that involves the partial or total
shutdown of the plant and consists in modifying and/or restoring its working
conditions through the intervention on its components with the aim of increasing

Fig. 1 Scheme of the refining process in the analyzed refinery

204 M. Bevilacqua et al.

the energetic efficiency, guarantee a homogeneous functioning and the integrity of

the security systems, limit the wear to increase the service life. It is a maintenance
process that involves either the critical equipment, that is not possible to isolate
during the normal functioning of the plant, and those that have shown problems or
need a periodical inspection. The turnaround management is very challenging since
all the interventions have to be executed in a very strict time span (4 weeks at
maximum), in which all the resources (human, materials, technical, economical)
have to be effectively organized and planned.
Maintenance activities during a planned turnaround might include:
• Routine inspections for corrosion, equipment integrity or wear, deposit forma-
tion, integrity of electrical and piping systems;
• Special inspections (often arising from anomalies in the prior operating period)
of major vessels or rotating equipment or pumps to investigate for abnormal
situations;
• Installation of replacement equipment for parts or entire pumps or instruments
that wear out;
• Replacement of catalysts or process materials that have been depleted during
operations.
Improvement activities could include:
• Installation of new, upgraded equipment or technology to improve the refinery
processing;
• Installation of new, major capital equipment or systems that may significantly
alter the refinery process and product output.
The main phases and activities of the turnaround management (TAM) process
are described in Table 1. The Kick-off meeting represents the first phase and
establishes:
• when the shutdown will start;
• for how long it will last;
• a rough cut estimate of the costs;
• the plants involved.
Phase 2 represents the Scope Challenge (S/CH), a tool for managing the shut-
down and optimizing the activities that need to be carried out as schematized in
Fig. 2. The main aim of this phase is to define the Scope of Work (SOW), a
document that contains the list of tasks and activities to carry out during the
turnaround and the resources needed; it reports also all the other aspects of the
maintenance actions identified such as safety, quality, duration, resources, material,
and equipment as defined by Duffuaa and Daya (2004).
According to Singh (2000), the following groups are primarily responsible to
identify potential turnaround work scope:
• Operations or production
• Inspection and mechanical integrity maintenance
Innovative Maintenance Management Methods in Oil Refineries 205

Table 1 Main phases of a turnaround process

Phase 1 definition and Kick-off meeting
approach ✓ Management strategy definition;
✓ Scope of Work definition;
✓ Team, organization, responsibilities, and roles approval;
✓ Determination of communication matrix, of contracts
strategies, of the Health, Safety, Environmental, and Quality
(HSEQ) general plan
Phase 2 scope definition Scope challenge
✓ Preliminary scope identification;
✓ Scope improvement;
✓ Definition of the operative system and needs;
✓ Development of initial plans and cost estimation;
✓ Definition of materials need and of items with long waiting
list;
✓ Strategies compilation;
✓ Risk analysis
Phase 3 ✓ Scope tune up and evaluation of improvement opportunities;
Scope improvement ✓ Development of works assignment;
✓ Programs and scheduling evaluation;
✓ Identification of the pre-shutdown works;
✓ Development of subcontracting plans;
✓ Tune up of HSEQ plans;
✓ Training needs development;
✓ Start of the logistic and materials planning;
✓ Compilation of the resources plan and integration with
contractors
Phase 4 budget & ✓ Fine regulation of timetable and record of communication
scheduling sections;
✓ Re-improvement of scope and methods;
✓ Critical path/Density analysis/Workload;
✓ Contingencies programming;
✓ Definition of procedures coordination;
✓ System variations approval;
✓ Exercitations HSE/quality;
✓ Details
Phase 5 Mobilization and ✓ Execution of all preparatory works;
execution ✓ Subcontractors integration;
✓ Collocation of material equipments;
✓ Assembling and scaffoldings access;
✓ Isolation removal;
✓ Production and preassembling;
✓ Cleaning and controls;
✓ Temporary installations
Phase 6 demobilization and ✓ Critics and lessons learned after TAM;
closedown ✓ Scheduling and budget control;
✓ Updating of documents with modifications;
✓ Shutdown reports;
✓ Verification of employed teams;
✓ Preparation of working list for next TAM
206 M. Bevilacqua et al.

Fig. 2 Scope challenge scheme

• Rotating and machinery

• Instrument, electrical and analyzer
• Process and technical
• Capital projects and engineering
• Turnaround planning
• Safety, health, and environment
• Outside parties, e.g., vendors, regulatory agencies, insurance, etc.
In order to achieve world-class results on plant shutdowns and turnarounds,
companies must strive to finalize the work scope 8–12 months before the
turnaround.
The poor exercise of work scope identification and validation done during the
preparation phase results in a lot of additional and emergent work creeping up
during the execution period. This can cause a lot of constraints on the available
resources resulting in schedule slippages and escalation of costs.
Phase 3 is used to tune up and optimize the schedule of the activities defined in
the SOW, to define the labor (internal and from subcontracts) and the training
needed, and the logistic and resources plan.
In phase 4 the detailed budget plan and activities scheduling is prepared as well
as a contingency program; moreover the coordination of the different procedures is
defined.
In phase 5, first the preparatory works and then the planned maintenance
activities are carried out.
Innovative Maintenance Management Methods in Oil Refineries 207

Fig. 3 Turnaround timeline

Phase 6 closes the process, summarizes the results obtained and starts the
preparation of the working list for the next turnaround.
The shutdowns are scheduled on a 10-year basis and this program is shared
within the refinery personnel through the Multiyear Shutdown Plan.
Although TAM can cause a production plant availability reduction of about
2–3 % on a yearly basis and its economical benefits due to reduction in equipment
failures can increase the firm profits due to the minor loss production up to 15 %.
As an example the last oil refinery TAM lasted 18 days, with a medium use of
more than 500 workers/day, a total of 100,000 worked hours and about 10 million
euro of investment. To reach all the planned objectives of the TAM it was necessary
the following timeline has been drawn as shown in Fig. 3.

4 Materials and Methods

Bahrami and Price (2000) demonstrated that different methods have been formu-
lated for identifying the critical equipments of a process, some centerd exclusively
on the effect of failure on the service and others are based on the involved risk, such
as the HAZOPs (Hazard and Operability Studies) method, Casal et al. (1999), or
safety equipment (Cepin 2002; Hokstad et al. 1995). In other cases, the aim is to
classify the maintenance activities to be carried out rather than to classify the
equipment itself, e.g., the FMEA (Failure Mode and Effects Analysis) method
(Bevilacqua et al. 2000). Risk-Based Inspection (RBI) is proposed as a sound
methodology for identifying and assessing the risks of operating plant equipment
and is widely used by many upstream and downstream oil and gas, petrochemical
companies worldwide to various degrees (Arunraj and Maiti 2007). The RBI
implemented in the analyzed refinery is the Shell RBI method (S-RCM Training
Guide 2000). It uses equipment history and the likely consequences of equipment
failure to determine Inspection regimes focused on actual risks, so as to prevent
unsafe incidents from occurring, and is based on API 581 base resource document
which involved representatives from a number of major oil and petrochemical
208 M. Bevilacqua et al.

companies, and a comprehensive statistical analysis of petrochemical facilities over

a number of years (API Recommended Practice 580–581 2002a, b).
In this paper we propose a methodology that, using the synergies provided by
the simultaneous adoption of risk-based analysis and maintenance management
methods, enables a preventive maintenance program to be made with a view to the
production of servicing plans that ensure greater reliability at the lowest possible
cost.
The method applied in the refinery for managing the TAM (Shell method) has
allowed to reorganize the whole process highlighting measures to achieve better
results in terms of performance and flexibility. The validation of the method
focused on two main aspects: the achievement of the defined objectives and its
economic convenience.
The process plants of the analyzed refinery include more than 10,000 equip-
ments. After the collection period of 18 months, the technical specifications (op-
erating conditions, process fluids, and safety system configuration) have been
collected for each component and, in a following step, the failure mode effects and
severity have been analyzed so that to calculate the criticality and reliability
indexes.
The development and the implementation of both approach (risk-based method
and Criticality Index), and the application to a specific stage in the maintenance
activities of a medium-sized refinery was carried out by a panel of experts.
The panel was made up of 10 participants, and included 2 academics, whose
research studies are mainly focused on risk analysis and maintenance management,
3 technical operators and 3 managerial operators involved in the maintenance
processes, 2 operators from an external service company (ESC) called into manage
the maintenance activities on the basis of a global service contract.
The following steps were involved in the criticality analysis of the analyzed
plant:
• identify and list all the equipment and components liable to maintenance;
• define the relevant data, the preventive maintenance plans, and the recording of
any faults;
• attribute the factors value to the risk items considered;
• extract only the objects that can be effectively monitored using Computer
Maintenance Management System (CMMS);
• determine the criticality of the single item and select the candidates for
monitoring.
The most critical items have been classified in four severity classes according to
their possible effects:
• First class catastrophic; complete failure of the equipment or component, with
very high risk for workers and environment.
• Second class critical; the failure mode limits significantly equipment
performances.
• Third class marginal; the failure mode causes a degradation of the performances.
Innovative Maintenance Management Methods in Oil Refineries 209

• Fourth class small; the failure mode causes small problems to the users, but it is
not possible to notice an important deterioration.
The equipment selected for predictive maintenance was then given ad input in a
machine-monitoring matrix in order to attribute the related monitoring activities
(Ai) and their frequency for each machine Mj.

5 Risk-Based Methodology

5.1 Scope of Work (SOW) Development

The elaboration of the SOW list has to be based on a rational evaluation of the
costs–risks–benefits that, starting from a structured analysis of each alternative, then
defining the TAM duration, the costs and the availability of plants relative to the
inclusion of the different items in the turnaround.
A strategic issue to develop an effective SOW is the project team: in the pre-
sented case, it was composed by the shutdown manager and leader, the operation
lines manager, the technology, inspections, planned services, costs and program-
ming responsible, as well as the engineers team and the members of the senior
management team. The methodology for assessing the best alternative is composed
by three main processes:
• Simple process;
• Extended process;
• Shell Global process.

5.1.1 The Simple Process: Application of the RAM Matrix

for the Risks Management

In this type of process the risk is defined as:

RISK ¼ ðPÞ x ðEÞ x ðCÞ

Being P = probability of the thread occurrence; E = level of exposure to the

thread; C = thread consequences. The quantification of the probability is quite
complex and if reliable data are not available it is necessary to rely on keywords; an
example for the probability assessment is presented in Table 2.
Keywords are used also to quantify the exposition, the consequences of the
health and safety, the loss of image for the company, the environmental impact, the
economical consequences (the relative tables are not reported here).
210 M. Bevilacqua et al.

Table 2 Keywords to quantify the probability parameter (P)

Probability
Keyword Probability Description Example
Very rare—probability almost 0.001 A Fire from pump with severe
null damages
Unlikely but possible 0.01 B Explosion of gas cloud
Rare but possible 0.1 C Security valve blocked
Quite probable—isolated 0.5 D Fire with moderated
possibilities economical damages
Possible occurrence— 1 E Unexpected pump stop
possibility of repetition

In order to manage the risk, the R&M (reliability and maintenance) management
team of the refinery applied the decision matrix presented in Table 3. This matrix is
used to choose the best alternative in a set characterized by a calculated level of
risk.
If the risk level assigned with the RAM fells within the critical range, the R&M
team performs a Root Cause Analysis (RCA), of the failure for assessing causes and
avoid repetitions of the event. For each improvement choice, it is necessary to
recalculate probability and consequences to ascertain the acceptability of the new
level of risk.

5.2 Extended Process: J-Factor Application

In addition to the conditions above defined, the team has to evaluate the most
profitable alternative in terms of maximum risk reduction efficacy.
The Alternative 0 is the reference alternative and it is used to compare all the
others alternatives. The index for ranking the different solution is called Justification
Factor (J- Factor) and is given by:

OR  NR
J  Factor ¼
CAA

where the OR = original risk, represents the initial risk calculated for the thread
(Alternative 0); NR = new risk for implementing a different solution (Alternative 1);
CAA = Cost of the alternative solution. The J-Factor gives an indication about the
amount of risk reduction for each euro invested, the higher the J-Factor, the better is
the alternative.
Table 3 Decision matrix (RAM) applied in the analyzed refinery

Decision matrix
Index Consequences Probability
A B C D E
Severity Keyword Health and Economical Environment Image Very rare: Unlikely Rare Quite Possible
safety probability but but probable occurrence
almost null possible possible Isolated Possibility
possibilities of repetition
1 Minor Limited consequences Negligible Negligible Negligible 1 2 3 4 10
discomfort: damage effects impact
medication/accident <10,000 within
1–3 days company
confines
2 Moderate Poor health 3–10 days Minor Short- term Limited 2 4 6 16 30
damage effects impact
>10,000 surrounding
<100,000 areas
3 Severe Occupational disease Localized Short- term Local 3 6 18 24 60
Reversible in 10– damage effects noted territory
Innovative Maintenance Management Methods in Oil Refineries

30 days >100,000 outside

<1M
4 Very severe Permanent damage to Important Transient Regional 4 16 36 64 80
health: > 30 days damage reversible level impact
Accident involving >1M damage
several people < 10 M
5 Catastrophic Lethal exposure: fatal Extended Permanent National 10 30 60 80 100
accident damage environmental level impact
> 10 M damage
211
212 M. Bevilacqua et al.

5.3 Shell Global Process

This method consists of five main steps:

1. Risk definition by using criticality matrix and confidence rating;
2. Generation of a list of alternatives for each item falling in the high criticality
area;
Alternative 0 = no action
Alternative 1 = selection of action 1
Alternative 2 = selection of action 2
Alternative n = selection of action n
3. Calculation of risk reduction (J-Factor) for each alternative selected;
4. Definition of the MUST (mandatory actions) and WANTS (suggested actions;
desirable conditions associated with convenience indexes);
5. Classifying and choosing the best alternative.

5.4 Decision-Making Method Applied for an Optimal

Definition of TAM Equipment List

A key factor for the execution of a TAM capable to ensure proper maintenance
actions, to respect the defined timeline, the estimated costs and resources usage, is
the definition of the equipments and components that have to be included in the
process. A common mistake in the management of this kind of actions is the
inclusion of items that could be treated during plant working conditions with
ordinary maintenance. To limit their number to the minimum possible means to
increase the possibility of being compliant with the TAM schedule and avoiding
unnecessary loss of production.
Following the Shell global process it has been possible to classify items that
have necessarily to be included in TAM (i.e., equipment that cannot be isolated
while the plant is in normal operation), and items that can be subjected to other
maintenance alternatives.
This method allowed also to evaluate the failure risk and the benefits deriving
from any preventive measures (the product of the cost of the measures multiplied by
the new probability of the failure’s occurrence) and thus compare the failure risks.
In Fig. 4 a scheme of the decision-making method applied is presented.
Other than identifying the phenomena of critical failure to analyze, the risk
matrix determines the priorities, i.e., the deadline for starting analysis of the event
and maintenance plans.
The decision-making process carried out using the risk matrix require the
involvement of an experts team able to evaluate every aspects of the different
operations examined. In this study, they defined the likelihood of failure (LOF) and
the consequence of failure (COF) of each item.
Innovative Maintenance Management Methods in Oil Refineries 213

Fig. 4 Decision-making method

Table 4 Criticality matrix

Risk of not doing the work Criticality classes

(likelihood of failure)
H L H VH U U
M L M H VH U
L N L M H VH
N N N L M H
Consequences N L M H VH
Economical <10 k 10– 0.1–1 M$ 1–10 M$ >10 M$
$ 100 k
$
Safety No Slight Major Single Multiple
cons. cons. cons. fatality fatality
Environment No Slight Localized Extensive Very
effect effect effect effect severe
effect

LOF is divided in 4 classes: N = negligible, L = low, M = medium, H = high;

COF are evaluated taking into account economical, safety, and environmental
aspects and divided in 6 classes: N = negligible, L = low, M = medium, H = high,
VH = very high, U = unacceptable as expressed in Table 4.
214 M. Bevilacqua et al.

Fig. 5 Criticality matrix combined with confidence rating

Entering in the criticality matrix (LOF x COF) it possible to associate a risk level
to each item.
It is then necessary to combine the criticality matrix with the “confidence rating”
that represents the reliability of the likelihood of failure assessment by the expert
team as expressed in Fig. 5.
Items falling in area III or IV are excluded from the TAM list; for items falling in
the area I or II the experts propose a list of maintenance alternative calculating for
each of them the J-Factor. If the alternative with the highest J-F does not need the
shutdown of the equipment, this can be eliminated from the list otherwise it is
included.
The application of this method has allowed the company to reduce the amount of
turnaround equipment by 24 % with respect to the previous TAM, with a conse-
quent reduction of costs of about 18 % and of resources usage of about 20 %. The
item excluded were subjected to current maintenance due to their low associated
risk.

6 Criticality Index-Based Method

This main aim in developing this innovative index was to create an economic,
flexible, simple, and complete decision tool to apply for evaluating the criticalities
of equipments and plants in order to optimize the use of economical, human, and
instrumental resources needed for the maintenance of the refinery as schematized in
Fig. 6. The objective was not only to use it for TAM optimization but as an
instrument to also manage current maintenance actions. It represents an alternative
to the risk-based method; nevertheless it could be also possible to apply them in
combination to better exploit their strongest characteristics and verify the results of
the analysis.
The questions to which this tool has to respond are: where, when, and how to
intervene.
To reach this goal it is necessary to know the initial condition of each equipment
and update them yearly. This allows to assess the critical equipment for the
maintenance of safety and environmental standard and of production performances
analyzed in the refinery, to define the causes of such criticality (failures, costs,
Innovative Maintenance Management Methods in Oil Refineries 215

Fig. 6 CI characteristics and

output management

accidents, etc.) and the preventive and mitigating actions necessary to restore an
acceptable level of criticality.
The CI differs from the common risk analysis index and is obtained by adding,
with a defined weight, data intrinsic to each equipment related to:
• temperature;
• pressure;
• fluid;
• complexity;
• presence of spare equipment;
• Effects of failure on the plant;
• Failures (from historical data available on the plant);
• Environmental inconvenient;
• Working inconvenient;
• Accidents;
• Results from inspections;
• Improvement actions;
In order to calculate the index, 15 operating factors have been collected for each
item.

X
15
CIm ¼ fim
pi
i¼1

where:
CIm criticality index for the m item
fim value of factor i for the component m
pi weight of the factor i
216 M. Bevilacqua et al.

Table 5 Factors fi and relative weight

FACTOR fi Pi (weight of each Positive or negative
factor i) effects
(A1) corrosion sensitivity 4 +
(B1) economical sustainability 3 +
(C1) sensitivity to get dirty or scraping 1 +
(D1) process criticality (RBI) 1 +
(E1) safety and environmental impact 6 +
(F1) warning of technical committee 1 +
(G1) sensitivity to transitory phases 3 +
(H1) equipment complexity 2 +
(I1) impact on production efficiency 2 +
(RCM)
(L1) impact on maintenance plan (work 2 +
orders)
(M1) sensitivity to serious failurea 4 +
(N1) sensitivity to slight failureb 1 +
(O1) possibility of spare −3 −
(P1) item under preventive–predictive −4 −
maintenance
(Q1) item under improving maintenance −6 −
a
Shutdown total shutdown of operations in a plant due to any type of anomaly
b
Slowdown reduced working capacity of the plant, less than 75 % of the expected value

According to the panel of expert the most important factors to consider are listed
in Table 5. The panel also indentified the weight of each factor, on a 6-point scale
from low importance to high importance, and the possible contribution to CI
(positive or negative).
Figure 7 represents the scheme used for the CI development.
Each factor fi was then divided into subclasses and the relative factor value was
defined as described in Table 6.
Four classes of criticality were defined for each item: Iper critical (CI ≥ 50);
Critical (40 ≤ CI < 50); Subcritical (35 ≤ CI < 40); Noncritical (CI < 35).
The experimental phase was carried out for a period of 2 years for evaluating the
results of the maintenance actions implemented. In the following Tables 7 and 8 the
CI calculated for the first and the second years is reported.
In the following Table 9 the comparison of the index obtained in the first and
second years shows that one equipment (White) maintained the same level of
criticality; 6 items increased their CI passing from the subcritical class to the iper
critical (Gray I); 6 items decreased their CI going from iper critical to subcritical
class due to the maintenance program implemented (Gray III); finally two items
remained in the same class even if with a lower CI due to the maintenance actions
carried out (Gray II).
Innovative Maintenance Management Methods in Oil Refineries 217

Fig. 7 Scheme representing the CI development steps

The maintenance activities implemented on the items that in the first year fell in
the iper critical or critical class produced a decrease of their criticality level. In the
second year of the study were identified:
• 12 iper critical equipments;
• 29 critical equipments;
• 27 subcriticals.
In order to define which item include in the turnaround other simulations have
been carried out; in particular the weight of factors fi linked to a particular speci-
fication established by the panel of experts have been modified. In the first simu-
lation it was decided to modify the weight of the cost related parameters (B1, G1,
H1, M1, N1, defined in Table 6). The weight value has been increased by a factor of
two. In the second simulation a different weight has been given to parameters
related to the refinery production process (D1, I1, E1 of Table 6). In this case it was
incremented by a factor of 4. In the last simulation, only the weight of the factor E1,
safety and environment, has been halved.
The simulation demonstrated that the most critical items remained the same and
only the CI value changed but without modifying the criticality level. The example
of the first simulation is shown in Table 10.
218 M. Bevilacqua et al.

Table 6 fi factor classes

FACTOR fi Number Factor Item description
of classes value
(A1) corrosion sensitivity 3 1 Item with low thickness
3 Slight corrosion
6 Drilling
(B1) economical sustainability 3 1 10,000 € < maintenance cost of last
18 months < 100,000 €
3 100,000 € < maintenance cost of last
18 months < 1,000,000 €
6 maintenance cost of last
18 months > 1,000,000 €
(C1) sensitivity to get dirty or 6 1 1–2 warning in the last 18 months
scraping 2 3–4 warning in the last 18 months
3 5–6 warning in the last 18 months
4 7–8 warning in the last 18 months
5 9–10 warning in the last 18 months
6 >10 warning in the last 18 months
(D1) process criticality (RBI) 4 1,5 Very high
3 High
4,5 Medium
6 Low
(E1) safety and environmental 6 1 Minor
impact 2 Moderate
3 Medium
4 Severe
5 Very severe
6 Catastrophic
(F1) warning of technical 6 1 1–20 warning in the last 18 months
Committee 2 21–40 warning in the last 18 months
3 41–60 warning in the last 18 months
4 61–80 warning in the last 18 months
5 81–100 warning in the last 18 months
6 >100 warning in the last 18 months
(G1) sensitivity to transitory 6 1 1–4 stopping events in the last
phases 18 months
2 5–8 stopping events in the last
18 months
3 9–12 stopping events in the last
18 months
4 13–16 stopping events in the last
18 months
5 17–20 stopping events in the last
18 months
(continued)
Innovative Maintenance Management Methods in Oil Refineries 219

Table 6 (continued)
FACTOR fi Number Factor Item description
of classes value
6 >20 stopping events in the last
18 months
(H1) equipment complexity 3 1 Equipment value <= 10,000 €
3 10,000 < Equipment value <= 100,000
€
6 Equipment value > 100,000 €
(I1) impact on production 4 1, 5 Very high
efficiency (RCM) 3 High
4, 5 Medium
6 Low
(L1) impact on maintenance 6 1 1–6 work orders in the last 18 months
plan (work orders) 2 7–12 work orders in the last 18 months
3 13–18 work orders in the last
18 months
4 19–24 work orders in the last
18 months
5 25–30 work orders in the last
18 months
6 >30 work orders in the last 18 months
(M1) sensitivity to serious 6 1 1–25 events in the last 18 months
failure 2 25–50 events in the last 18 months
(N1) sensitivity to slight 3 51–100 events in the last 18 months
failure
4 101–150 events in the last 18 months
5 151–200 events in the last 18 months
6 >200 events in the last 18 months
(O1) possibility of spare 3 1 No automatic standby
3 1 spare
6 2 spare
(P1) item under preventive– 2 0 No preventive–predictive maintenance
predictive maintenance in the last 18 months
−6 Yes preventive–predictive
maintenance in the last 18 months
(Q1) item under improving 3 0 No improving maintenance in the last
maintenance 18 months
−3 1 improving maintenance in the last
18 months
−6 2 or more improving maintenance in
the last 18 months
 220

Table 7 Example of CI calculated for some selected equipment in the first year
Item C8901 C8903 R80011 E8105 R80012 E8001A FV81004 E8002B R8602 D8102
CI year I 83 71 62 59 51 49 49 48 52 47
Sensitivity to corrosion 3 6 3 3 3 3 6 3
Economical sustainability
Sensitivity to dirt and abrasion 1
Process criticality
Safety and environmental impact 6 5 2
Operating work load 3 1 2 1 1 1 1 1 1 1
Sensitivity to transient state 5 5 5 5 5 5 5 5 5 5
Equipment complexity 6 6 6 6 6 6 3 6 6 6
Impact on plant productivity 4, 5 4, 5 1, 5 4, 5 1, 5
Impact on maintenance 1 2 1 3 1 1 1
programming
Severe sensitivity to failure 1 1 1 1 1 1 1 1 1
Low sensitivity to failure 1 1 1 1 1 1
Availability of spare parts
Subject to preventive maintenance
Subject to improvement
maintenance
Ranking 1 2 3 4 5 6 7 8 9 10
M. Bevilacqua et al.
Table 8 Example of CI calculated for some selected equipment in the second year
Item C8901 P8004A R8602 E8920 TK8101 F3751 C8903 D8902B F1401 R80011
CI year II 97 77 70 61 61 58 58 56 55,5 52
Sensitivity to corrosion 3 1
Economical sustainability 3 1 1 1 1 3 3 3
Sensitivity to dirt and abrasion 3 1
Process criticality 4,5
Safety and environmental impact 6 5 6 3 3 3
Operating work load 3 1 1 1 1 1 1 2
Sensitivity to transient state 6 6 6 6 6 4 6 6 2 6
Equipment complexity 6 3 6 6 6 6 6 6 6 6
Impact on plant productivity 4,5 6 6 4,5 4,5 6
Impact on maintenance programming 1 2 1 2 4 1 3 1
Severe sensitivity to failure 1 1 1 1 1 1 1 1
Innovative Maintenance Management Methods in Oil Refineries

Low sensitivity to failure 1 1 1 1 1 2 1

Availability of spare parts 1
Subject to preventive maintenance
Subject to improvement maintenance
Ranking 1 2 3 4 5 6 7 8 9 10
221
222 M. Bevilacqua et al.

Table 9 Comparison between CI of the first and second years of the study

Item IdC year I Ranking year I IdC year II Ranking year II Color
C8901 83 1° 97 1° White
P8004A 36 Sub critical 77 2° Gray I
R8602 52 9° 70 3°
E8920 39 Sub critical 61 4°
TK8101 28 Sub critical 61 5°
F3751 38 Sub critical 58 6°
C8903 71 2° 58 7° Gray II
D8902B 37 Sub critical 56 8° Gray I
F1401 39,5 Sub critical 55,5 9°
R80011 62 3° 52 10° Gray II
E8105 59 7° <40 Subcritical Gray III
R80012 51 10° <40 Subcritical
E8001A 49 12° <40 Subcritical
FV81004 49 13° <40 Subcritical
E8002B 48 14° <40 Subcritical
D8102 47 15° <40 Subcritical

Table 10 CI variation after II year standard weight II year increased weight of B1, G1,
increasing the weight of the H1, M1, N1 factors
cost related factors
Item CI Rank Item NEW Delta rank
CI
C8901 97 1 C8901 131 0
P8004A 77 2 P8004A 101 0
R8602 70 3 R8602 96 0
E8920 61 4 TK8101 91 From 5th to 4th
TK8101 61 5 D8902B 90 From 8th to 5th
F3751 58 6 E8920 87 From 4th to 6th
C8903 58 7 R80011 86 From 10th to
7th
D8902B 56 8 F3751 84 From 6th to 8th
F1401 55.5 9 F1401 83.5 0
R80011 52 10 C8903 82 From 7th to
10th
Innovative Maintenance Management Methods in Oil Refineries 223

Applying the CI method to the turnaround consists ofperforming the maximum

number of tasks, allowed by the established TAM duration, starting from the items
with highest CI value. However, all the maintenance operations related to items in
the iper critical and critical class should be performed.
The results of the study demonstrate the effectiveness of such method that can be
used as planning tools by the company management, together with the maintenance
and reliability team, for selecting the items that need major maintenance and
scheduling the turnaround activities. In this way, obtaining an optimized use of
human and economical resources, as well as the highest level of reliability and
safety.
Simulating the application of this decision tool on the same TAM, it was pos-
sible to estimate a reduction of the equipment of about 25 %, with a consequent
costs decrease of 11 % and of resources usage of about 7 %.

7 Conclusions

Maintenance engineering represents the technical and organizational answer to

promote the continuous improvement and the excellence of the performances,
together with the costs reduction. The application of the maintenance engineering in
the TAM allows to reduce the number of equipments that need to be included in the
shutdown, decreasing the total time of the outage and the production loss, and
increasing the interval between two shutdowns.
Two methods have been tested for defining the items to include in the pro-
grammed turnaround and discard those being potentially managed with current
maintenance.
The decision-making one has been used as a tool to quantify the risk and define
the priority actions in terms of ‘‘action/event risk—associated risks—possible
improvements—resources used—result.’’ This approach allowed to identify the
really critical events and the items to include in TAM in terms of safety, envi-
ronment, plant availability, quality of the product and maintenance costs. The
saving generated by the risk-based method, in terms of costs, has been evaluated in
the order of 1.8 million € and in terms of resources usage led to a reduction of about
20,000 working hours. Moreover it also brought to the development of a common
technical language facilitating the communication between all the people interact-
ing in the plants.
The criticality index, developed and demonstrated for the first time in the ana-
lyzed refinery, was applied in the simulation of the same turnaround management.
The results showed that also with this method it was possible to improve the
performance of the process in terms of equipment reduction, operations, and costs.
A different result was obtained for the resources; more personnel were indeed
involved in the management, due to the great amount of data needed to calculate the
index and the effort made to create and update the database.
224 M. Bevilacqua et al.

Despite this drawback the criticality index tool provides a complete overview of
the current equipment state of maintenance; the major efforts requested are paid off
by the minor complexity in the management of the reliability and maintenance
programs that for a refinery constitute one of the main issues.

Acknowledgments The authors would like to acknowledge the reviewers for their constructive
and helpful comments and suggestions that helped in improving the paper value.

References

Ahuja, I. P. S., & Khamba, J. S. (2008). Total productive maintenance implementation in a

manufacturing organization. International Journal of Productivity and Quality Management, 3
(3), 360–381.
Ahuja, I. P. S., & Khamba, J. S. (2009). Investigation of manufacturing performance achievements
through strategic total productive maintenance initiatives. International Journal of Productivity
and Quality Management, 4(2), 129–152.
Aissani, N., Beldjilali, B., & Trentesaux, D. (2009). Dynamic scheduling of maintenance tasks in
the petroleum industry: A reinforcement approach. Engineering Applications of Artificial
Intelligence, 22, 1089–1103.
American Petroleum Institute. (2002a). API Recommended practice 580 (1st Edn.). Risk-based
Inspection.
American Petroleum Institute. (2002b). API Publication 581 (1st Edn.). Risk-based Inspection.
Base Resource Document.
Arunraj, N. S., & Maiti, J. (2007). Risk-based maintenance—techniques and applications. Journal
of Hazardous Materials, 142(3), 653–661.
Bahrami, G K., & Price, J. (2000). Mantenimiento basado en el riesgo. Gestion de activos
industriales, ano III, numero 9, enero/2000. Alcion. Madrid.
Bevilacqua, M., Braglia, M., & Gabbrielli, R. (2000). Monte Carlo simulation approach for a
modified FMECA in a power plant. Quality and Reliability Engineering International, 16,
313–324. doi:10.1002/1099-1638(200007/08).
Bevilacqua, M., Ciarapica, F., Giacchetta, G., & Marchetti, B. (2012). Development of an
innovative criticality index for turnaround management in an oil refinery. International Journal
of Productivity and Quality Management, 9(4), 519–544.
Bhangu, N. S., Singh, R., & Pahuja, G. L. (2011). Reliability centered maintenance in a thermal
power plant: a case study. International Journal of Productivity and Quality Management,
7(2), 209–228.
Brissaud, F., Lanternier, B., & Charpentier, D. (2011). Modeling failure rates according to time
and influencing factors. International Journal of Reliability and Safety, 5(2), 95–109.
Casal, J., Montiel, H., Planas, E., & Vılchez, J. (1999). Analisis del riesgo en instalaciones
industriales. Barcelona: Ediciones UPC.
Cepin, M. (2002). Optimization of safety equipment outages improves safety. Reliability
Engineering System Safety, 80, 77–71.
Christen, A. J., Ruggeri, F., & Villa, E. (2011). Utility based maintenance analysis using a
Random Sign censoring model. Reliability Engineering and System Safety, 96, 425–431.
Conn, A. R., Deleris, L. A., Hosking, J. R. M., & Thorstensen, T. A. (2010). A simulation model
for improving the maintenance of high cost systems, with application to an offshore oil
installation. Quality and Reliability Engineering International, 26, 733–748. doi:10.1002/qre.
1136.
Dossier Special Algerie. (2002). Special Issue on the Petroleum Industry in Algery (2nd part).
Petrole et techniques, no. 440.
Innovative Maintenance Management Methods in Oil Refineries 225

de Leon, Gomez, Hijes, F., & Cartagena, J. J. R. (2006). Maintenance strategy based on a
multicriterion classification of equipments. Reliability Engineering and System Safety, 91,
444–451.
Duffuaa, S. O., & Daya, B. (2004). Turnaround maintenance in petrochemical industry: practices and
suggested improvements. Journal of Quality and Maintenance Engineering, 10(3), 184–190.
doi:10.1108/13552510410553235.
Garg, S., Singh, J., & Singh, D. V. (2010). Availability and maintenance scheduling of a repairable
block-board manufacturing system. International Journal of Reliability and Safety, 4(1),
104–118.
Ghosh, D., & Roy, S. (2010). A decision-making framework for process plant maintenance.
European Journal of Industrial Engineering, 4(1), 78–98.
Hokstad, P., Flotten, P., Holmstrom, S., McKenna, F., & Onshus, T. (1995). A reliability model
for optimisation of test schemes for fire and gas detectors. Reliability Engineering System
Safety, 47, 15–25.
Khanna, V. K. (2008). Total productive maintenance experience: a case study. International
Journal of Productivity and Quality Management, 3(1), 12–32.
Kister, T. C., & Hawkins, B. (2006). Maintenance planning and scheduling: Streamline your
organization for a lean environment. Elsevier Inc. ISBN: 978-0-7506-7832-2.
Kosta, A. L., & Keshav, K. (2013). Refinery Inspection and Maintenance, Petroleum refining and
natural gas processing, ASTM manual series MNL 58. West Conshohocken, PA: ASTM
International.
Laggoune, R., Chateauneuf, A., & Aissani, D. (2009). Opportunistic policy for optimal preventive
maintenance of a multi-component system in continuous operating units. Computers &
Chemical Engineering, 33, 1499–1510.
Pérès, F., & Noyes, D. (2003). Evaluation of a maintenance strategy by the analysis of the rate of
repair. Quality and Reliability Engineering International, 19, 129–148. doi:10.1002/qre.515.
Pham, H. (2013). A software reliability model with Vtub-shaped fault-detection rate subject to
operating environment, Proceedings of the 19th International Conference on Reliability and
Quality in Design. Honolulu. (August 5–7).
Prabhakar, P., Jagathy Raj, V. P. (2013). A new model for reliability centered maintenance in
petroleum refineries, International Journal Of Scientific & Technology Research, 2(5), 56–64.
Rosmaini, A., Shahrul, K., Ishak, A. A., & Indra, P. A. (2011). Preventive replacement schedule: a
case study at a processing industry. International Journal of Industrial and Systems
Engineering, 8(3), 386–406.
Sachdeva, A., Kumar, D., & Kumar, P. (2008). A methodology to determine maintenance
criticality using AHP. International Journal of Productivity and Quality Management, 3(4),
396–412.
Savino, M. M., Brun, A., & Riccio, C. (2011). Integrated system for maintenance and safety
management through FMECA principles and fuzzy inference engine. European J. of Industrial
Engineering, 5(2), 132–169.
S-RCM Training Guide. (2000). Shell-reliability centered maintenance. Shell Global Solution
International.
Singh, A. B. (2000). World-Class Turnaround Management. Houston, USA: Everest Press.
Telford, S., Mazhar, M. I., & Howard, I. (2011). Condition based maintenance (CBM) in the oil
and gas industry: An overview of methods and techniques. Proceedings of the 2011
International Conference on Industrial Engineering and Operations Management. Kuala
Lumpur. (January 22–24).
Tsakatikas, D., Diplaris, S., & Sfantsikopoulos, M. (2008). Spare parts criticality for unplanned
maintenance of industrial systems. European Journal of Industrial Engineering, 2(1), 94–107.
You-Tern, T., Kuo-Shong, W., & Lin-Chang, T. (2004). A study of availability centered
preventive maintenance for multi-component systems. Reliability Engineering & System
Safety, 84(3), 261–70.
226 M. Bevilacqua et al.

Zhang, C., & Mostashari, A. (2011). Influence of component uncertainty on reliability assessment
of systems with continuous states. International Journal of Industrial and Systems
Engineering, 7(4), 542–552.
Zhaoyang, T., Jianfeng, L., Zongzhi, W., Jianhu, Z., & Weifeng, H. (2011). An evaluation of
maintenance strategy using risk based inspection. Safety Science, 49, 852–860.
Zhou, X., Lu, Z., & Xi, L. (2010). A dynamic opportunistic preventive maintenance policy for
multi-unit series systems with intermediate buffers. International Journal of Industrial and
Systems Engineering, 6(3), 276–288.