weibull model

The University of New South Wales School of Mechanical and Manufacturing Engineering

Optimizing Life-Cycle Maintenance Cost of Complex Machinery Using Advanced Statistical Techniques and Simulation

By

Mustapha El Hayek

Under the Supervision of

Dr. Erik van Voorthuysenand

Prof Donald W. Kelly

A Thesis Submitted for the degree of Doctor of Philosophy March 2006

COPYRIGHT STATEMENT I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'

Signed ...........................

Date

...........................

AUTHENTICITY STATEMENT I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.

Signed ...........................

Date

...........................

ORIGINALITY STATEMENT

I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis.

I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.

Signed ...........................

Abstract

Maintenance is constantly challenged with increasing productivity by maximizing up-time and reliability while at the same time reducing expenditure and investment. In the last few years it has become evident through the development of maintenance concepts that maintenance is more than just a non-productive support function, it is a profit- generating function. In the past decades, hundreds of models that address maintenance strategy have been presented. The vast majority of those models rely purely on mathematical modeling to describe the maintenance function. Due to the complex nature of the maintenance function, and its complex interaction with other functions, it is almost impossible to accurately model maintenance using mathematical modeling without sacrificing accuracy and validity with unfeasible simplifications and assumptions.

Analysis presented as part of this thesis shows that stochastic simulation offers a viable alternative and a powerful technique for tackling maintenance problems. Stochastic simulation is a method of modeling a system or process (on a computer) based on random events generated by the software so that system performance can be evaluated without experimenting or interfering with the actual system. The methodology developed as part of this thesis addresses most of the shortcomings found in literature, specifically by allowing the modeling of most of the complexities of an advanced maintenance system, such as one that is employed in the airline industry. This technique also allows sensitivity analysis to be carried out resulting in an understanding of how critical variables may affect the maintenance and asset management decision-making process.

In many heavy industries (e.g. airline maintenance) where high utilization is essential for the success of the organization, subsystems are often of a rotable nature, i.e. they rotate among different systems throughout their life-cycle. This causes a system to be composed of a number of subsystems of different ages, and therefore different reliability

characteristics. This makes it difficult for analysts to estimate its reliability behavior, and therefore may result in a less-than-optimal maintenance plan.

Traditional reliability models are based on detailed statistical analysis of individual component failures. For complex machinery, especially involving many rotable parts, such analyses are difficult and time consuming. In this work, a model is proposed that combines the well-established Weibull method with discrete simulation to estimate the reliability of complex machinery with rotable subsystems or modules. Each module is characterized by an empirically derived failure distribution. The simulation model consists of a number of stages including operational up-time, maintenance down-time and a user-interface allowing decisions on maintenance and replacement strategies as well as inventory levels and logistics. This enables the optimization of a maintenance plan by comparing different maintenance and removal policies using the Cost per Unit Time (CPUT) measure as the decision variable. Five different removal strategies were tested. These include: On-failure replacements, block replacements, time-based replacements, condition-based replacements and a combination of time-based and condition-based strategies.

Initial analyses performed on aircraft gas-turbine data yielded an optimal combination of modules out of a pool of multiple spares, resulting in an increased machine up-time of 16%. In addition, it was shown that condition-based replacement is a cost-effective strategy; however, it was noted that the combination of time and condition-based strategy can produce slightly better results. Furthermore, a sensitivity analysis was performed to optimize decision variables (module soft-time), and to provide an insight to the level of accuracy with which it has to be estimated.

It is imperative as part of the overall reliability and life-cycle cost program to focus not only on reducing levels of unplanned (i.e. breakdown) maintenance through preventive and predictive maintenance tasks, but also optimizing inventory of spare parts management, sometimes called float hardware. It is well known that the unavailability of a spare part may result in loss of revenue, which is associated with an increase in system downtime. On the other hand increasing the number of spares will lead to an increase in capital investment

and holding cost. The results obtained from the simulation model were used in a discounted NPV (Net Present Value) analysis to determine the optimal number of spare engines.

The benefits of this methodology are that it is capable of providing reliability trends and forecasts in a short time frame and based on available data. In addition, it takes into account the rotable nature of many components by tracking the life and service history of individual parts and allowing the user to simulate different combinations of rotables, operating scenarios, and replacement strategies. It is also capable of optimizing stock and spares levels as well as other related key parameters like the average waiting time, unavailability cost, and the number of maintenance events that result in extensive durations due to the unavailability of spare parts. Importantly, as more data becomes available or as greater

accuracy is demanded, the model or database can be updated or expanded, thereby approaching the results obtainable by pure statistical reliability analysis.

Acknowledgements

I would like to praise Allah the almighty for providing me with the virtue of patience, knowledge, and determination to accomplish this project.

I would like to dedicate this work primarily to my father Ghazi, mother Sana, sisters and brother for their on-going psychological and mental support. I would like to express my deepest gratitude to my wife Marwa, and beautiful son Gazi, for this project would never have come to fruition without their care, endurance, love, and continuous support.

I would like to acknowledge the valuable and highly appreciated support of my supervisor Dr. Erik van Voorthuysen for his immense contribution throughout this project. I am endlessly grateful for his professional advice, technical expertise, and encouragements in tough moments. His friendship, persistence, and support will never be forgotten.

I would also like to express my sincere appreciation to Prof Don Kelly from the School of Mechanical and Manufacturing Engineering for his valuable contribution in securing the industrial partners, and Dr. R A Platfoot from Covaris Pty Ltd. for his technical support during the critical initial stages of this project.

I would like to thank Mr. Adrian Verkerk and Mr. Tom Bemstein from Qantas, and in particular Mr. John Napier from Qantas Defense Services for sharing their knowledge and expertise in regards to airline management and engine maintenance. I also would like to thank Mr. Gary Baunach from Simulation Modeling Systems, Newcastle for his valuable tips regarding the development of the simulation model. Finally, I would like to thank a bright 4th year student, Mr. Bourhan Chemaise for his valuable support in entering the raw data, and building the electronic database.

List of Publications

M El Hayek & E van Voorthuysen & D Kelly, Optimizing Life Cycle Maintenance Cost of Complex Machinery with Rotable Modules Using Simulation, Journal of Quality in Maintenance

Engineering.Vol 11 (4), 2005.

M El Hayek & E van Voorthuysen, Optimizing Inventory Levels Using Simulation, Submitted to the Journal of Quality in Maintenance Engineering, Dec 2005.

M El Hayek & E van Voorthuysen, A tool for Optimizing the Maintenance and Removal Strategies for Complex Machinery, Submitted to the Journal of Quality in Maintenance Engineering, Dec 2005.

M El Hayek & E van Voorthuysen, Optimizing Engine Up-time Using Simulation, International Conference of Maintenance Societies, Hobart Australia, 2005.

Table of Contents

Chapter One Introduction ------------------------------------------------------------------------------------- 11.1 1.21.2.1 1.2.2 1.2.3

Asset and Maintenance Management -------------------------------------------------- 1 Thesis Aims---------------------------------------------------------------------------------- 4Maintenance Efficiency -------------------------------------------------------------------------- 5 Reliability Forecast and Equipment Aging ---------------------------------------------------- 6 Inventory and Float Hardware------------------------------------------------------------------- 6

1.31.3.1

Approach------------------------------------------------------------------------------------- 7Why Simulation?---------------------------------------------------------------------------------- 7

1.4 1.5

Proposed Methodology-------------------------------------------------------------------- 9 Thesis Layout ------------------------------------------------------------------------------12

Chapter Two Theoretical Frame of Reference ------------------------------------------------------------152.12.1.1 2.1.2 2.1.3

Fundamental Knowledge ----------------------------------------------------------------16Maintenance Tasks and Activities -------------------------------------------------------------17 Maintenance Costs and Profits -----------------------------------------------------------------19 Impact on Business Performance --------------------------------------------------------------20

2.22.2.1 2.2.2 2.2.3 2.2.4 2.2.5 2.2.6

Reliability-Centered Maintenance (RCM) -------------------------------------------21Purpose of RCM ---------------------------------------------------------------------------------21 RCM Benefits ------------------------------------------------------------------------------------22 The RCM Method--------------------------------------------------------------------------------23 Views on RCM -----------------------------------------------------------------------------------24 RCM Implementation ---------------------------------------------------------------------------25 Other Strategies ----------------------------------------------------------------------------------28

2.32.3.1

Key Supporting Concepts ---------------------------------------------------------------29Imperfect Maintenance --------------------------------------------------------------------------30

I

2.3.2 2.3.3

Treatment models for Imperfect Maintenance -----------------------------------------------30 Maintenance strategies --------------------------------------------------------------------------32

2.42.4.1 2.4.2 2.4.3 2.4.4

The Weibull Distribution ----------------------------------------------------------------35Why Weibull? ------------------------------------------------------------------------------------35 Applications --------------------------------------------------------------------------------------36 MTBF or MFOP ---------------------------------------------------------------------------------39 Life Data Formats --------------------------------------------------------------------------------40

2.52.5.1 2.5.2

Maintenance Simulation -----------------------------------------------------------------42Simulating Maintenance Planning and Replacement Strategies ---------------------------43 Inventory Levels ---------------------------------------------------------------------------------45

2.6

Literature Shortages----------------------------------------------------------------------49

Chapter Three Preliminary Analysis--------------------------------------------------------------------------533.13.1.1

Background---------------------------------------------------------------------------------53Aero Engines -------------------------------------------------------------------------------------54

3.23.2.1 3.2.2 3.2.3

Engine Maintenance ----------------------------------------------------------------------56Engine Piece parts -------------------------------------------------------------------------------56 Engine Shop-Visits ------------------------------------------------------------------------------59 Life-Cycle Data ----------------------------------------------------------------------------------60

3.33.3.1 3.3.2

Engine Mean-Time-On-Wing (MTOW)----------------------------------------------61Data Collection -----------------------------------------------------------------------------------61 Engine Data Analysis----------------------------------------------------------------------------63

3.43.4.1 3.4.2 3.4.3

Criticality Analysis------------------------------------------------------------------------68Maintenance Cost --------------------------------------------------------------------------------69 Off-Wing Removal Cause ----------------------------------------------------------------------72 Module Criticality--------------------------------------------------------------------------------73

3.53.5.1 3.5.2 3.5.3

Critical Modules Data Analysis --------------------------------------------------------75Introduction ---------------------------------------------------------------------------------------75 Data Analysis -------------------------------------------------------------------------------------76 Discussion-----------------------------------------------------------------------------------------76

II

Chapter Four Maintenance and Replacement Strategies ------------------------------------------------794.1 4.24.2.1 4.2.2 4.2.3 4.2.4

Introduction --------------------------------------------------------------------------------80 Maintenance Strategy --------------------------------------------------------------------83Introduction ---------------------------------------------------------------------------------------83 Background ---------------------------------------------------------------------------------------85 Cost-Data Analysis ------------------------------------------------------------------------------87 Cost per Unit Time (CPUT) --------------------------------------------------------------------90

4.34.3.1 4.3.2 4.3.3 4.3.4 4.3.5 4.3.6 4.3.7 4.3.8 4.3.9

Replacement Strategy --------------------------------------------------------------------93Introduction ---------------------------------------------------------------------------------------93 Background ---------------------------------------------------------------------------------------94 Methodology--------------------------------------------------------------------------------------98 Data Analysis ----------------------------------------------------------------------------------- 106 Model Verification----------------------------------------------------------------------------- 108 Model Validation------------------------------------------------------------------------------- 109 Simulation Results ----------------------------------------------------------------------------- 116 Cost per Unit Time (CPUT) Analysis ------------------------------------------------------- 129 Soft-time Sensitivity Analysis---------------------------------------------------------------- 130 Optimal Replacement Strategy ----------------------------------------------------------- 134

4.3.10

4.4

Conclusion -------------------------------------------------------------------------------- 137

Chapter Five Reliability Forecast and Equipment Aging --------------------------------------------- 1405.1 5.25.2.1 5.2.2

Introduction ------------------------------------------------------------------------------ 140 Asset Life-Cycle ------------------------------------------------------------------------- 142Shop Visit Count (SVC) ---------------------------------------------------------------------- 142 Module Rotability------------------------------------------------------------------------------ 143

5.35.3.1 5.3.2

Estimating Engine Reliability--------------------------------------------------------- 145Data Analysis at Engine Level --------------------------------------------------------------- 146 Data Analysis at Module Level -------------------------------------------------------------- 149

5.45.4.1

The Simulation Model------------------------------------------------------------------ 158Model Overview ------------------------------------------------------------------------------- 158

III

5.4.1 Model Overview ------------------------------------------------------------------------------- 158 5.4.2 Simulation Results ----------------------------------------------------------------------------- 160 5.4.3 SVC Sensitivity Analysis --------------------------------------------------------------------- 163 5.4.4 Repair Versus Replace ------------------------------------------------------------------------ 165 5.4.5 Case study: Replacements Limited by Spares ---------------------------------------------- 168

5.5 Conclusion -------------------------------------------------------------------------------- 169

Chapter Six Inventory Theory and Maintenance Capacity --------------------------------------- 1716.1 Background------------------------------------------------------------------------------- 171 6.2 Introduction ------------------------------------------------------------------------------ 173 6.3 Methodology------------------------------------------------------------------------------ 1746.3.1 Data Analysis ----------------------------------------------------------------------------------- 174 6.3.2 Simulation Model ------------------------------------------------------------------------------ 175 6.3.3 Simulation Results ----------------------------------------------------------------------------- 179

6.4 Optimal Replacement Strategy------------------------------------------------------- 182 6.5 Case Study Multiple Lease Agreements------------------------------------------ 1846.5.1 No Lease Agreement -------------------------------------------------------------------------- 186 6.5.2 Lease Agreement A ---------------------------------------------------------------------------- 187 6.5.3 Lease Agreement B ---------------------------------------------------------------------------- 189 6.5.4 Optimal Agreement ---------------------------------------------------------------------------- 191

6.6 Conclusion -------------------------------------------------------------------------------- 192

Chapter Seven----------------------------------------------------------------------------- 194 Conclusion and Future Work ---------------------------------------------------------- 1947.1 Research Outcomes --------------------------------------------------------------------- 1967.1.1 Maintenance Tasks and Strategies ----------------------------------------------------------- 197 7.1.2 Replacement Efficiency ----------------------------------------------------------------------- 198 7.1.3 Module and Task Parameters ----------------------------------------------------------------- 198 7.1.4 Reliability and Equipment Aging ------------------------------------------------------------ 199 7.1.5 Allocation of Rotable Modules--------------------------------------------------------------- 199 7.1.6 Spares Strategy --------------------------------------------------------------------------------- 200 7.1.7 Repair Capacity -------------------------------------------------------------------------------- 200

IV

7.2.3

Real-Life Operation---------------------------------------------------------------------------- 202

7.37.3.1 7.3.2 7.3.3 7.3.4

Proposed Future Work----------------------------------------------------------------- 203Escalating PM Intervals for LLPs------------------------------------------------------------ 203 Simulating the Engine Maintenance Shop -------------------------------------------------- 204 Optimizing Engine Scrap Rate --------------------------------------------------------------- 205 Performance Condition-Monitoring --------------------------------------------------------- 205

References ----------------------------------------------------------------------------------- 207 Appendices ---------------------------------------------------------------------------------- 233

V

Table of Figures

Figure 1.1.1 Asset Management versus Maintenance Management (van Voorthuysen 2005)------------- 2 Figure 1.1.2 Effect of effective maintenance on profitability -------------------------------------------------- 3 Figure 1.4.1 The sequence for the proposed methodology ----------------------------------------------------- 9 Figure 1.4.2 Model overview------------------------------------------------------------------------------------- 11 Figure 2.1.1 Types of maintenance activity (Hastings 2001) ------------------------------------------------ 17 Figure 2.1.2 Maintenance concepts development (Alsyouf 2004). ------------------------------------------ 18 Figure 2.2.1 - RCM analysis steps--------------------------------------------------------------------------------- 24 Figure 2.3.1 Scheduled maintenance intervals (Jayabalan and Chaudhuri 1992b) ---------------------- 31 Figure 2.4.1 Generic Failure Categories (Nelson, 1982 and 1985) ----------------------------------------- 41 Figure 2.4.2 Generic Failure Categories (Reliasoft, 2001)--------------------------------------------------- 42 Figure 2.5.1 Elements of the integrated simulation model (Duffuaa and Andijani 1999)---------------- 46 Figure 2.5.2 Spares-supply module (Duffuaa et al. 2001)---------------------------------------------------- 48 Figure 3.1.1 Rolls Royce engine modular breakdown (Rolls-Royce 2000) -------------------------------- 55 Figure 3.2.1 Effect of inefficient maintenance on life-cycle cost -------------------------------------------- 58 Figure 3.3.1 Data collection ------------------------------------------------------------------------------------- 62 Figure 3.3.2 Best-fit distribution test---------------------------------------------------------------------------- 64 Figure 3.3.3 The Weibull shape parameter at all phases in the bath-tub curve --------------------------- 66 Figure 3.3.4 - Weibull plot for engine SV------------------------------------------------------------------------- 67 Figure 3.4.1 Module Criticality---------------------------------------------------------------------------------- 68 Figure 3.4.2 Breakdown of maintenance cost (per SV) ------------------------------------------------------- 70 Figure 3.4.3 Modular breakdown of maintenance cost------------------------------------------------------- 72 Figure 3.4.4 Off-Wing removal cause frequencies for all engine modules--------------------------------- 74 Figure 3.4.5 Pareto analysis for module criticality factors -------------------------------------------------- 75 Figure 3.5.1 Weibull analysis results for critical modules --------------------------------------------------- 78 Figure 4.1.1 Proposed steps for maintenance and overhaul strategy--------------------------------------- 81 Figure 4.2.1 Illustration of the CPUT breakdown------------------------------------------------------------- 88 Figure 4.2.2 Engine and module CPUT ------------------------------------------------------------------------ 92 Figure 4.3.1 Replacement strategies breakdown -------------------------------------------------------------- 97 Figure 4.3.2 Process overview ---------------------------------------------------------------------------------- 100 Figure 4.3.3 Methodology overview---------------------------------------------------------------------------- 102 Figure 4.3.4 Simulation model overview ---------------------------------------------------------------------- 103

VI

Figure 4.3.5 Top: A top-level snapshot of the Arena model------------------------------------------------- 105 Figure 4.3.6 An exploded view of the Soft-time sub-model ---------------------------------------------- 105 Figure 4.3.7 An exploded view of the maintenance cost sub-model.------------------------------------ 106 Figure 4.3.8 An exploded view of the condition monitoring sub-model -------------------------------- 106 Figure 4.3.9 Full and partial refurbishment breakdown ---------------------------------------------------- 109 Figure 4.3.10 Variable display board used for model verification----------------------------------------- 110 Figure 4.3.11 Model validation---------------------------------------------------------------------------------- 110 Figure 4.3.12 - Failure data analysis results for the actual strategy ---------------------------------------- 114 Figure 4.3.13 - Cost data analysis results for the actual strategy -------------------------------------------- 114 Figure 4.3.14 Failure data analysis results for the block replacement strategy ------------------------- 119 Figure 4.3.15 Cost data analysis results for the block replacement strategy ----------------------------- 120 Figure 4.3.16 - Failure data analysis results for the on-failure replacement strategy--------------------- 122 Figure 4.3.17 Cost data analysis results for the on-failure replacement strategy------------------------ 122 Figure 4.3.18 - Failure data analysis results for the time-based replacement strategy ------------------- 124 Figure 4.3.19 - Cost analysis results for the time-based replacement strategy ----------------------------- 125 Figure 4.3.20 - Failure data analysis results for the condition-based replacement strategy ------------- 127 Figure 4.3.21 - Cost data analysis results for the condition-based replacement strategy----------------- 127 Figure 4.3.22 - Failure data analysis results for the combination of time and condition based replacement strategy --------------------------------------------------------------------------------------------------------------- 129 Figure 4.3.23 Cost data analysis results for the combination of time and condition based replacement strategy --------------------------------------------------------------------------------------------------------------- 129 Figure 4.3.24 Sensitivity analysis of module soft-time on flying hours ------------------------------------ 132 Figure 4.3.25 - Sensitivity analysis of module soft-time on shop-visit cost ------------------------------- 133 Figure 4.3.26 - Sensitivity analysis of module soft-time on CPUT ----------------------------------------- 134 Figure 4.3.27 - Failure data analysis results for the optimal strategy--------------------------------------- 137 Figure 4.3.28 - Cost data analysis results for the optimal strategy ------------------------------------------ 138 Figure 4.4.1 Replacement strategy comparisons ------------------------------------------------------------- 140 Figure 5. 1 Chapter layout -------------------------------------------------------------------------------------- 141 Figure 5.2.1 Shop Visit Count (SVC) effect on aging --------------------------------------------------------- 143 Figure 5.2.2 Engines composed of modules with different ages -------------------------------------------- 145 Figure 5.3.1 - An overview of the proposed methodology ----------------------------------------------------- 146 Figure 5.3.2 Engine SVC analysis------------------------------------------------------------------------------ 147 Figure 5.3.3 Effect of engine SVC on Weibull parameters -------------------------------------------------- 148 Figure 5.3.4 - Weibull Analysis for module M1 at all SVCs --------------------------------------------------- 150 Figure 5.3.5 - Weibull Analysis for module M2 at all SVCs --------------------------------------------------- 150 Figure 5.3.6 - Weibull Analysis for module M3 at all SVCs --------------------------------------------------- 151 Figure 5.3.7 - Weibull Analysis for module M4 at all SVCs --------------------------------------------------- 151

VII

Figure 5.3.8 Module M1 Weibull Parameters behaviors for each SVC------------------------------------ 153 Figure 5.3.9 Module M2 Weibull Parameters behaviors for each SVC------------------------------------ 154 Figure 5.3.10 - Module M3 Weibull Parameters behaviors for each SVC----------------------------------- 155 Figure 5.3.11 - Module M4 Weibull Parameters behaviors for each SVC----------------------------------- 156 Figure 5.3.12 - Weibull Parameters stored in an Access Database ------------------------------------------ 157 Figure 5.4.1 - Overview of the simulation model --------------------------------------------------------------- 158 Figure 5.4.2 A snapshot of the Arena simulation model ----------------------------------------------------- 160 Figure 5.4.3 - Engine MTOW of SVC (3, 2, 2, 1) --------------------------------------------------------------- 161 Figure 5.4.4 - Off-Wing removal causes of SVC combination (3, 2, 2, 1) ----------------------------------- 162 Figure 5.4.5 - Sensitivity analysis results ------------------------------------------------------------------------ 164 Figure 6.1.1 Balancing capital and unavailability costs ----------------------------------------------------- 174 Figure 6.3.1 - A snapshot of the Arena model------------------------------------------------------------------- 178 Figure 6.3.2 - Simulation model overview ----------------------------------------------------------------------- 179 Figure 6.3.3 - No of spares versus fill rate based on simulated data ---------------------------------------- 180 Figure 6.3.4 - Fill rate versus repair capacity at engine spares equal two --------------------------------- 183 Figure 6.4.1 - Availability for the optimal and real life strategy at different levels of stock -------------- 184 Figure 6.5.1 Case study 1 PV Analysis ------------------------------------------------------------------------ 188 Figure 6.5.2 Case study 2 PV analysis ------------------------------------------------------------------------ 190 Figure 6.5.3 Case study 3 PV analysis------------------------------------------------------------------------- 191 Figure 6.5.4 Optimal PVs for all cases ------------------------------------------------------------------------ 192 Figure 7.3.1 An overview of the PM escalation method ----------------------------------------------------- 204

VIII

Table of Tables

Table 2.2.1 - RCM potential benefits (Backlund 2003) --------------------------------------------------------- 23 Table 3.2.1- Input review report data ----------------------------------------------------------------------------- 60 Table 3.3.1 Observed data deficiencies ------------------------------------------------------------------------- 63 Table 3.4.1 Module maintenance cost breakdown results (per SV) ----------------------------------------- 69 Table 3.4.2 Best fit distributions for modules maintenance costs ------------------------------------------- 71 Table 3.5.1 Module MTBF --------------------------------------------------------------------------------------- 78 Table 4.1.1 Purpose of the proposed steps for maintenance and overhaul strategy ---------------------- 82 Table 4.2.1 Cost data analysis results -------------------------------------------------------------------------- 88 Table 4.2.2 Estimated values for CPM and CCM ---------------------------------------------------------------- 89 Table 4.3.1 - Simulation Condition for time-based replacement strategy ----------------------------------- 103 Table 4.3.2 - Simulation Conditions for the condition-based replacement strategy------------------------ 103 Table 4.3.3 Shop-visit cost statistical parameters ------------------------------------------------------------ 106 Table 4.3.4 - Shop-visit cost 95% interval ----------------------------------------------------------------------- 107 Table 4.3.5 Actual strategy -------------------------------------------------------------------------------------- 111 Table 4.3.6 Simulation Settings for the actual strategy ------------------------------------------------------ 111 Table 4.3.7 - Simulation results for the actual replacement strategy ---------------------------------------- 112 Table 4.3.8 Standard result table ------------------------------------------------------------------------------- 116 Table 4.3.9 Simulation results for the block replacement strategy ----------------------------------------- 117 Table 4.3.10 - Simulation results for the on-failure replacement strategy ---------------------------------- 120 Table 4.3.11 Actual values for module soft-time-------------------------------------------------------------- 122 Table 4.3.12 - Simulation results for the time-based replacement strategy --------------------------------- 123 Table 4.3.13 - Simulation results for the condition-based replacement strategy --------------------------- 125 Table 4.3.14 - Simulation results for the combination of time and condition-based replacement strategy127 Table 4.3.15 - CPUT values for different replacement strategies -------------------------------------------- 129 Table 4.3.16 Optimal strategy ---------------------------------------------------------------------------------- 135 Table 4.3.17 Simulation settings for the optimal strategy --------------------------------------------------- 135 Table 4.3.18 - Simulation results for the optimal replacement strategy ------------------------------------- 136 Table 5.4.1 Off-wing removal frequencies--------------------------------------------------------------------- 163 Table 5.4.2 - SVC of spares---------------------------------------------------------------------------------------- 168 Table 5.4.3 Simulation results ---------------------------------------------------------------------------------- 169 Table 6.3.1 - Average delay times -------------------------------------------------------------------------------- 181

IX

Table 6.5.1 - No Lease Agreement-------------------------------------------------------------------------------- 186 Table 6.5.2 - Lease Agreement A --------------------------------------------------------------------------------- 188 Table 6.5.3 - Lease Agreement B --------------------------------------------------------------------------------- 189

X

List of Acronyms

APK AUD CBM CCMS CFR CI CM CPFH CPUT CSLSV CSMO ECM EGT ESN F FMECA FOD HP HPC HPT HSLSV HSMO HSN

Available Passenger Kilometers Australian Dollar Condition-Based Maintenance Computerized Maintenance Management System Constant Failure Rate Confidence Interval Corrective Maintenance Cost per Flying Hour Cost per Unit Time Cycles Since Last Shop Visit Cycles Since Major Overhaul Engine Condition Monitoring Exhaust Gas Temperature Engine Serial Number Failure Failure Modes, Effects, and Consequence Analysis Foreign Object Damage High Pressure High Pressure Compressor High Pressure Turbine Hours Since Last Shop-Visit Hours Since Major Overhaul Hours Since New

XI

IFR IPC IPT KPI LCC LCD LLP LPC LPT MFOP MFOPS MSG MSN MTBF MTOW MTTF NPV OEM PM PV PR RAAF RCM RRX S SRM SV SVC

Increasing Failure Rate Intermediate Pressure Compressor Intermediate Pressure Turbine' Key Performance Indicator Life-Cycle Cost Life-Cycle Data Life-Limited Parts Low Pressure Compressor Low Pressure Turbine Maintenance Free Operating Period Maintenance Free Operating Period Survivability Maintenance Steering Group Module Serial Number Mean Time Between Failures Mean Time On-Wing Mean Time to Failure Net Present Value Original Engine Manufacturer Preventive Maintenance Present Value Passenger Revenue Royal Australian Air Force Reliability-Centered Maintenance Rank Regression on the x-axis Suspension Standard Ranking Method Shop Visit Shop Visit Count

XII

TAT TGT TSHM USAF WO

Turn Around Time Turbine Gas Temperature Time Since Heavy Maintenance United States Air Force Work Order

XIII

Glossary

Asset A physical item which has value over a period exceeding one year (Hastings 2001).

Asset Management A set of business processes, disciplines, and professional practices. It is an integrated, holistic, performance focused, whole life- costed, data-based, people inclusive, and risk managed spectrum of modern methods (Woodhouse 2001).

Capital Velocity A measure of the efficiency of the use of capital investment and a key indication of sustainable profitability in the future (van Voorthuysen 2005). Condition Monitoring The continuous or periodic measurement and interpretation of data to indicate the degraded condition of an item to determine the need for maintenance (BS 3811: 1993). Condition-based Maintenance Maintenance carried out according to the need as indicated by condition monitoring (BS 3811:1993).

Corrective Maintenance Maintenance carried out after fault recognition and intended to put an item into a state in which it can performed a required function (PrEN 13306). Criticality Analysis Quantitative analysis of events and faults and the ranking of these in order of seriousness of their consequence (BS 3811:1993).

XIV

Effectiveness The accomplishment of the right thing on time and within the quality requirements specified (Sink and Tuttle 1998).

Efficiency A measure of how economically the firms resources are utilized when providing a given level of requirements (Sink and Tuttle 1998).

Failure Mode One of the possible states of faulty item for a given function (prEN 13306). Hard-Life The age of the component, at or by which the component has to be replaced (Crocker and Kumar2000).

Life-Cycle Cost The total costs estimated to be incurred in the design, development, production, operation, maintenance, support, and final disposition of a major system over its anticipated useful life span (Barringer and Weber 1996).

Maintenance The combination of all technical and administrative actions intended to retain an item in or restore it to a state in which it can perform its required function (BS 3811: 1993).

Maintenance Function The engineering decisions and associated actions necessary and sufficient for optimization of specified capability (MESA). Maintenance Management All activities of the management that determine the maintenance strategy, objectives, and responsibilities and implement them by means such as maintenance planning, maintenance

XV

control, and supervision, improvement of methods in the organization including economic aspects (BS 3811: 1993).

Opportunistic Maintenance The opportunity that may be taken during the following shut down to carry out preventive maintenance on other maintenance-significant components which have not yet failed (Savic etal. 1995).

Ownership Costs All costs associated with the acquisition, use, and maintenance (Ellram and Siferd 1993).

Performance The level to which a goal is attained (Dwight 1999). Preventive Maintenance Maintenance carried out at predetermined intervals or according to prescribed criteria and intended to reduce the probability of failure or the degradation of the function of an item(prEN 13306).

Productivity The relationship between what comes out of an organization system divided by what comes into an organization system (Sumanth 1998).

Profitability The best overall indicator of the company performance; it measures the outcome of all management decisions about sales and purchase prices, levels of investment and production, and innovation, as well as reflecting the underlying efficiency with which the inputs are converted to outputs (Rantanen 1995).

RCM A process used to determine the maintenance requirements of any physical asset in its operating context (Moubray 1997). XVI

Soft-life The age of the component after which it will be rejected the next time the engine or one of its modules, containing it, is recovered (Crocker and Kumar 2000).

XVII

Chapter One Introduction

1.1 Asset and Maintenance Management

Asset Management is a broad concept that is aimed at managing return on investment. Since many organizations have significant capital invested in fixed assets, it is imperative that an effective asset management plan aimed at minimizing investment required to achieve business goals is in place. Depending on the nature of the asset, Asset Management will consist of a set of critical business, operating, and technical processes that will support the achievement of business goals (van Voorthuysen 2005). Asset Management includes all processes and activities including identifying, acquiring, supporting, maintaining, and disposing or renewing assets. The objectives of asset management are (van Voorthuysen 2005):

1. Minimize investment, 2. Minimize ownership costs, 3. Maximize commercial return, 4. Optimize strategic value, 5. Manage risk

Maintenance management is an important subset of Asset Management. Even though the two concepts have been used interchangeably, the role of maintenance, which is aimed at achieving integrity and availability, is distinct from that of Asset Management, which is of a higher level: Return on Investment. The relationship between Asset Management and


Page 2

Maintenance Management is shown in Figure 1.1.1. This thesis will focus on managing the maintenance side of assets.

Figure 1.1.1 Asset Management versus Maintenance Management (van Voorthuysen 2005)

Traditionally, maintenance has been regarded as a cost-center and thought of as a necessary evil. Consequently, management did not treat it as a core function, but rather a non-productive support function. This is due to the fact that the effect of maintenance on improving system reliability and availability is often unclear. When the availability of the system is reduced due to breakdowns, it is not difficult to prove that it is due to ineffective maintenance. On the other hand, if breakdowns were prevented, it is often more difficult to prove that it is caused by effective maintenance. It is easy to calculate yearly maintenance expenditure, but it is not easy to estimate the benefit or return in that maintenance investment or even how it can be measured (Alsyouf 2004).

Recently, and with the increase in the level of competition and severity of the financial climate, the role of maintenance as a revenue gathering function is more important than ever, and consequently has been more emphasized. The growing importance of maintenance has generated an increasing interest in the development as well as the


Page 3

implementation of optimal maintenance strategies, so improving system reliability can be achieved at minimum cost. Maintenance managers have been highlighting the fact that more effective maintenance plans and practices can generate more profit.

Effective performance measures are vital to ensure successful implementation of maintenance strategies. This is due to the fact that measurements are the link between strategy and action (Neely et al. 1995; Schalkwyk 1998). Therefore, integrating maintenance into the core function of the company is vital especially in capital-intensive industries (Alsyouf 2004). There is a need for a holistic performance measurement system that is capable of evaluating the contribution of the maintenance function on the strategic objectives of an organization (Ahlmann 1984; Tsang 1998; Tsang et al. 1999; Muthu et al. 2000).

Increased Profitability

Increased Revenue

Satisfied Customers

Satisfied Society

Higher Availability

Fewer defects, rejects, and complaints

Lower risk of adverse impact

Improved Reliability

Improved Maintainability

Better Quality Service

Health, Safety and Environment

Effective Maintenance Program

Figure 1.1.2 Effect of effective maintenance on profitability


Page 4

In todays industry, the vision of maintenance is:

For equipment to be available to meet the required business and operational plans, To deliver the best sustainable economic return on assets at a level of risk that is both understood and acceptable.

To meet its requirements, the maintenance strategy and plan needs to be effective and efficient. In addition, the maintenance plan should comply with guidelines in regards to quality (ISO9001), safety (AS3931 and AS3460), and the environment (ISO14001). It is also vital that the maintenance process is subject to continuous improvement.

Figure 1.1.2 illustrates the effect of a cost-effective maintenance plan on the long-term profitability of an organization. The improved reliability characteristics of the operating assets would not only reduce the direct maintenance cost, but when combined with improved maintainability may improve system availability. This is crucial for the success of an organization especially in industries where high utilization is required. In addition, an effective maintenance program will enhance the service quality, and will also comply with international standards regarding health, safety, and the environment.

1.2 Thesis AimsAn overall maintenance strategy has two major components:

The establishment of the initial baseline maintenance plan or program, A feedback loop and ongoing assessment of the performance and reliability of the equipment allowing the effectiveness of the above maintenance plan or program to be improved over time. This is sometimes called a Reliability Program.

Many organizations are under substantial pressure to streamline the cost of maintenance and continuously improve reliability and performance, frequently under increasingly


Page 5

demanding operational requirements. The model developed in this thesis supports these aims through continuous improvement with system reliability and Cost per Unit Time (CPUT). The work presented in this thesis shows that great importance needs to be placed on estimating the reliability characteristics of machinery with rotable modules, establishing the basic maintenance tasks and elements, and optimizing the inventory levels jointly with the maintenance tasks at the most appropriate hierarchy in the system. The level of the hierarchy will depend on a number of factors including system, subsystem, and component criticality, as well as the availability of reliability and performance data. The model has been designed specifically to incorporate existing available data, which in some instances may be sparse. As more data becomes available, the model can be updated accordingly. In this way, the model can be used for designing maintenance strategies and constructing maintenance plans or programs, estimating the effect of the performed maintenance on aging, and optimizing the number of spare parts as a function of operating costs.

The thesis is therefore aimed at addressing the reliability program of complex machinery by optimizing life-cycle maintenance costs. The model developed applies to all complex machinery in different industries; but has focused on the airline industry since this project was sponsored by an airline. The main characteristic of the model presented in this thesis is its ability to effectively model the different aspects of the life- cycle of complex machinery with rotable modules.

1.2.1 Maintenance EfficiencyIt is essential for a cost-effective maintenance plan to balance improvements in system availability with expenditure on maintenance tasks and operations. This is particularly important in industries where high utilization at minimum cost is vital for the success of the organization. It has been mentioned that the outcomes of maintenance are often complex and hard to quantify. One of the main objectives set by this thesis is to quantify the outputs of a maintenance function by estimating the CPUT for different maintenance and replacement strategies including time as well as condition-based replacements, and even a combination of both using opportunistic replacements.


Page 6

1.2.2 Reliability Forecast and Equipment AgingAnother key objective set by this thesis is developing a methodology that aims at increasing system productivity by maximizing up-time and reliability while at the same time reducing maintenance expenditure and investments. To meet this demand, it is required to provide maintenance managers with reliability trends and forecasts for systems and subsystems in a short time frame, and based on the available data. Those forecasts can be used not only to estimate the reliability behavior of complex machinery with modules of rotable nature, but are also capable of quantifying the effectiveness of a maintenance function in relation to system aging. In addition, it is necessary to use those forecasts as inputs to decision regarding the life-cycle of the system (e.g. repair versus replace).

1.2.3 Inventory and Float HardwareEven though organizations have focused on reducing levels of unplanned (i.e. breakdown) maintenance through preventive and predictive maintenance tasks, aligned with critical failure modes and mechanisms, it is imperative, as part of an overall reliability and lifecycle costing program, that we also consider the optimal inventory of spare parts, sometimes called float hardware. Another aim of this thesis is to estimate the optimal number of spares for a fleet of complex machinery with rotable modules by addressing some of the limitations of models previously proposed in literature. It is required to estimate the number of spares in conjunction with a chosen maintenance strategy. It is important to determine the relationship between the fill rate (the probability of having a spare engine on-hand when required) and the number of spares used, and also to estimate the delay times in the case of unavailability of a spare engine. In addition, it is required to highlight the possibility of further reductions in machine life-cycle costs by estimating the effect of repair capacity on the fill rate.


Page 7

1.3 ApproachMaintenance managers are often interested in studying a system through measuring its performance and improving its operation. Their primary aim is to focus on understanding how the entire system works. Often analysts find that the process of defining how the system works, which must be done before constructing models, provides an insight on what changes need to be made. This is mainly because most often, there is more than one individual responsible for an entire process (Kelton 2002).

1.3.1 Why Simulation?Stochastic simulation refers to methods for studying a wide variety of models of real world systems by numerical evaluation using software designed to initiate the systems operations and characteristics over time (Kelton 2002). Practically, it is the process of representing a real or proposed system on a computer, based on well-defined experiments so system performance can be evaluated, and to gain a better understanding of the behavior of the system for a given set of conditions (Law and Kelton 1993; Kelton 2002). Advantages of this method are that it is often cheap compared to physical experimentation, it is safe, and it allows experimentation beyond practical limitations, which is often not possible using traditional methods. In addition, the model may contain as much detail or complexity as necessary in order to represent the system faithfully. Other methods often require simplifying assumptions about the system, bringing the model validity into question (Kelton 2002).

Even though the importance of maintenance in the long-term has been realized, until recently little attention has been given to modeling maintenance systems (Kelly 1989). Duffuaa and Andijani (1999) listed the possible reasons for this delay:

Traditionally maintenance has been regarded as a necessary evil and at best as a non-productive support function,


Page 8

The maintenance function has a complex relationship and interaction with other functions in the organization, Maintenance outputs are difficult to define and measure.

It has been noted that the lack of success of many mathematical maintenance and replacement models is due to the simplicity of the models compared to the complex environment where the applications occur (Wang 2002). This is due to the fact that most of todays practical maintenance problems can not be fitted into the existing mathematical models addressing those problems, owing to the simplified assumptions made in the later (Bala Krishnan 1992). As maintenance systems can be complicated, valid models of them often need to be complicated too. For such models, there may not be exact mathematical solutions worked out, and this is where stochastic simulation becomes a useful tool (Kelton 2002).

In service industries (e.g. the airline industry), it is crucial to treat operation and maintenance as one system due to the high degree of dependency between them (Duffuaa and Andijani 1999). The modeling of operations and maintenance as one system is a prohibitively complex task as far as mathematical modeling is concerned. This is due to the fact that (Duffuaa and Andijani 1998; Duffuaa and Andijani 1999):

Maintenance interacts with other technical and engineering functions in a complex manner, There is a strong inter-dependency between maintenance factors, Maintenance has many uncertain elements (e.g. work order content, work done, time to complete the work order, availability of equipment and spare parts, etc).

Stochastic simulation therefore offers a practical alternative for the modeling and analysis of the maintenance function. In addition, stochastic simulation allows sensitivity analysis on input parameters to be carried out allowing an understanding on how those variables would affect the model output. Besides the actual model use, sensitivity analysis provides


Page 9

information on the accuracy with which the input parameters have to be estimated (Raivio et al. 2001).

It has been proven that stochastic simulation can be successfully implemented in the asset management and planning area. This is because it can effectively model (El Hayek et al. 2005):

Human resources, material availability, materials handling, and industrial processes. Machine breakdowns, delays, and bottlenecks. Estimated machine failure times, levels of damage, productivity losses and the required level of maintenance, and thus, resources. Engineering knowledge and experience. Costing and investment data associated with the above activities and events.

1.4 Proposed MethodologyThe steps in carrying out the analysis are shown in Figure 1.4.1.

Define critical modules

Data Collection and Analysis

Simulation

Sensitivity Analysis

Analyze results and report

Figure 1.4.1 The sequence for the proposed methodology

1. Data collection and analysis of each piece part is prohibitively time consuming and expensive. This is due to the fact that complex machinery consists of tens of thousands of piece parts, and the reliability and performance data of the vast majority of those piece parts is limited and hard to obtain. Module (subsystem) criticality factors were calculated based on two key factors:


Page 10

a. What modules are frequent machine downtime causes, b. What modules are most expensive to maintain.

2. For applications where automated online data acquisition systems are not yet deployed, data collection is the most difficult and time consuming part of model development. Despite the fact that machines are composed of many modules and mini-modules, analysis will focus on those with high criticality factors. This is due to the fact that detailed life-cycle data is available only in hard-copy format, and was manually entered into an electronic database. Statistical analysis performed used best-fit distributions to describe the collected data. For the time-to-failure data, it was the Weibull distribution that was always the best choice. No specific distribution was noted for the analysis of different cost data sets. 3. Stochastic simulation was used as the tool to model the maintenance function in order to meet the objectives set in 1.2. The simulation model inputs the statistical parameters to estimate the reliability behavior as well as its maintenance cost. The model is divided into three sub models, each addressing one of three key aspects in maintenance engineering and Asset Management: a. Maintenance and replacement strategy, b. Overhaul strategy, and c. Spares strategy.

It can be seen from Figure 1.4.2 that the three sub-models are implemented using a common simulation engine. The turbine engine is divided into a number of critical modules, each characterized by its empirically derived life-time distribution. The simulation model consist of a number of stages including operational up-time, maintenance downtime, logistic support, and a user interface for specifying decisions such as the maintenance and replacement strategy, combination of modules that make up an engine, and the number of spares of the scenario that will be tested. The advantage of this simulation engine is that it provides maintenance


Page 11

managers with a decision making tool in regards to critical parameters related to the life-cycle of the asset in a short time-frame, and based on the available data. The accuracy of the model can be improved by increasing the level of detail, thereby approaching the results obtained from pure statistical analysis.

Maintenance and replacement strategies

Reliability Forecasts

Spares strategy

User Interface Analysis and report

Statistical pre-analysis

Simulation EngineAccess database

Simulation of up-timeImport SVC for all modules fitted onto the engine from the database Import module Weibull Parameters corresponding with SVC from database Draw a random number from each module distribution. Find Minimum Module Failure time and equate to engine failure time

Simulation of downtimeApply simulation conditions on what modules require recondition.Engine Enters Shop and the level of maintenance is declared. Wait for the next failure time.

Remove all modules that require recondition

Replace the removed modules with new or refurbished ones

Wait for next failure, then place engine OnWing

User Interface

Refurbish all replaced modules, and then place in a spares pool

Update module SVCs in the database

Figure 1.4.2 Model overview


Page 12

4. Sensitivity analysis is performed on key decision variables such as a components soft-time, component age, and the number of spares. Not only does this step

provide an understanding on how those variables affect the system output, but also provides information on the accuracy with which the input parameters have to be estimated. 5. Analyzing results obtained from the simulation model is the final step towards achieving the required output. The optimal maintenance strategy, replacement strategy, module age combination, number of spares are reported.

1.5 Thesis LayoutThe thesis contains six main chapters and the conclusion.

Chapter Two presents a detailed literature survey covering a number of relevant topics. Different aspect of maintenance management including: definition, practices and approaches, costs and profits, and its impact on business performance are all discussed in Section 2.1. Section 2.2 addresses the maintenance policies for degrading assets including perfect, imperfect, and non-effect maintenance. Since this thesis deals mainly with imperfect maintenance, a detailed literature review which covers the treatment models for imperfect maintenance is presented. In addition, this section covers the effect of imperfect maintenance on asset reliability for different maintenance strategies. Section 2.3 presents a detailed review about Reliability-Centered Maintenance (RCM): definition, purpose, benefits, views, and implementation. In Section 2.4, the concept of maintenance simulation and applications are introduced. In addition, different existing maintenance and

replacement simulation models as well as inventory levels are presented. The Weibull distribution: definition, characteristics, applications, and advantages are presented in Section 2.5.


Page 13

Chapter Three presents the preliminary analysis, which is an important step towards model development. Section 3.1 presents an overview of a typical complex machine, the aircraft turbine engine RB211-524G, which will be used for the case studies presented in this thesis. In Section 3.2 the engine maintenance including: piece parts classifications, engine shop-visits, as well as life-cycle data are presented. Time-to-failure data statistical analysis on system (engine) level is presented in Section 3.3. The parts criticality analysis including the analysis of shop-visit cost data as well as shop-visit causes are presented in Section 3.4. Section 3.5 presents the statistical time-to-failure data analysis for all critical subsystems (engine modules).

Chapter Four presents a methodology that optimizes the maintenance and overhaul strategy for a fleet of turbine engines. In Section 4.1, the maintenance strategy was optimized based on the existing models that optimize CPUT. Section 4.2 presents a detailed methodology based on stochastic simulation that optimizes the replacement strategies for complex machinery. The methodology compares between different corrective, preventive, as well as predictive strategies with opportunistic replacements based on their values for CPUT. In addition, the optimal replacement strategy was obtained by performing a sensitivity analysis on different modules soft-time. Finally, the validation of the model was performed by comparing results obtained from simulation and conventional statistical methods.

Chapter Five introduces the concept of Shop Visit Count (SVC). This concept was used to estimate the reliability characteristics of machinery with rotable modules. A methodology based on stochastic simulation was developed by considering the reliability trend of modules as a function of SVC. The combination of module ages (SVCs) which optimizes the Mean-Time-On-Wing (MTOW) was obtained. In addition, a sensitivity analysis that estimates the effect of module age on the MTOW was performed. Finally, a case study for replacements limited by spares proved that this model may improve system up-time by up to 16.4 per cent.


Page 14

Chapter Six presents a methodology also based on stochastic simulation that optimizes the inventory levels of complex machinery. The model relates the number of spare engines to the engine availability expressed in terms of the fill rate. In addition, the average delay in instances where a spare engine is unavailable is estimated. The repair capacity expressed in terms of the engine turn-around time (TAT) is also estimated for different spares configurations. Engine availability at different levels of inventory was estimated for the real-life and optimal replacement strategies discussed in Chapter Four. Finally, a case study that uses the simulation model to find the Present Value (PV) for different engine lease agreements is presented.

Chapter Seven presents a conclusion to this work. It discussed the limitations, and the proposed future research avenues.

A comprehensive set of appendices that includes the research journal publications has been included.

Chapter Two Theoretical Frame of Reference

In this chapter, a detailed literature survey which reviews existing concepts and methodologies which are relevant to this thesis is presented. The author has tried to make this review complete; however, those papers which are excluded were considered either not directly relevant or covered in other articles.

Chapter 2 Literature Review

2.1 Fundamental knowledge

2.2 Reliability-Centered Maintenance (RCM)

2.3 Key Supporting Concepts

2.4 The Weibull Method

2.5 Discrete Simulation

2.6 Literature Shortages

Figure 2.1 Chapter layout

An overview of this chapter is shown in Figure 2.1. Section 2.1 presents a synopsis of the fundamental knowledge regarded as the foundation of the presented work. It is aimed at providing a general idea about its magnitude as well as its significance for industry. Section 2.2 presents a detailed view on Reliability-Centered Maintenance (RCM): its purpose, method, benefits, view, and implementation. Key supporting concepts addressing maintenance and replacement strategies, which are the two main topics of interest, are presented in Section 2.3. In Section 2.4, an overview about the Weibull method, advantages and applications are presented. The concept of discrete simulation and its applications in the maintenance industry are presented in Section 2.5. Section 2.6 presents the shortages


Page 16

observed in literature, which is used not only as a guideline for this project, but as an avenue for future projects as well.

2.1 Fundamental Knowledge

Different people understand Asset Management differently. The understanding of Asset Management depends mainly on the type of industry and organizations they are involved in (van Voorthuysen 2005). Woodhouse (2001) defines Asset Management as a set of business processes, disciplines, and professional practices. It is an integrated, holistic, performance focused, whole life- costed, data-based, people inclusive, and risk managed spectrum of modern methods. Sultan (2000) states that Asset Management is an evolving discipline which in recent years has become the focus of leading-edge organizations looking for a sustainable competitive advantage. An Asset Management plan is a means of taking control of your assets and understanding the what, when, and how. The consequences of not keeping track of the assets and understanding the behavior means relying on intuition and hit or miss management. It has been mentioned that even though Asset Management and maintenance have traditionally been used interchangeably, it was not until recently that maintenance was realized to be a key subset of Asset Management. So what is maintenance? The British Standards institution (BS 3811: 1993) defines maintenance as the combination of all technical and administrative actions intended to retain an item in or restore it to a state in which it can perform its required function. The Royal Australian Air Force (RAAF) defines maintenance as all technical and administrative actions required to retain an item in, or restore it to a state in which it can perform its required function. Moubray (1997) defines the scope of maintenance as insuring that physical assets continue to do what their users want them to do. Tsang et al. (1999) stated that the scope of maintenance management should cover every stage of the life cycle of technical systems (plant, machinery, equipment, and facilities: specification, acquisition, planning, operation, performance evaluation improvement, replacement, and disposal. Therefore, different


Page 17

authors have defined maintenance and its scope differently; see among others (Pintelon and Gelders 1992; Al-Najjar 1997; Organ et al. 1997). The definitions differ in scope, i.e. the target of maintenance varies from concentrating on an item only to include the whole process (Alsyouf 2004).

2.1.1 Maintenance Tasks and ActivitiesThe fact that effective maintenance plans and practices are essential for world class service is beyond question (Alsyouf 2004). Maintenance practices such as hard-time maintenance, preventive maintenance, condition monitoring, technical analysis, planning, control, training, teamwork, multitasking and others are key factors to achieve the desired performance (Mitchell et al. 2002; Mitchell 2002). Traditionally maintenance has been divided into routine and non-routine (Figure 2.1.1). The non-routine is divided further into deferred (non-routine tasks carried out at a convenient time), and emergency. Even though emergency tasks are mostly following breakdowns, there can be occasions where they are required even if no breakdown occurs (Hastings 2001).

Maintenance

Routine

Non Routine

Servicing / Preventive

Deferred

Emergency

Inspection

Non-Breakdown

Breakdown

Adjustment / Calibration

Condition Monitoring

Figure 2.1.1 Types of maintenance activity (Hastings 2001)


Page 18

In the past few decades, maintenance concepts and strategies have undergone major developments (Figure 2.1.2). Consequently, maintenance approaches including strategies, methodologies, or philosophies have been implemented in practice or proposed in literature. See among others (Bala Krishnan 1992; Dekker 1996; Al-Najjar 1997; Moubray 1997; McKone and Weiss 1998; Kumar et al. 1999a; Sherwin 2000; Swanson 2001; Mitchell 2002).

??? Proactive/Predictive Diagnostic/Prognostic Preventive Maintenance Reactive MaintenanceFigure 2.1.2 Maintenance concepts development (Alsyouf 2004).

The type of maintenance (inspection, repair, or replacement) is triggered by events (failure, condition, service time) according to the defined maintenance strategy (Alsyouf 2004). Maintenance strategy involves the identification, researching and implementation of decisions involving inspections, repairs, and replacements. Its aim is to estimate the optimal maintenance schedule for a plant by formulating the best life-plan for its individual units (Kelly 1997). It may consist of a mixture of different strategies which may vary from facility to facility (Dekker 1996; Al-Najjar 1997; Zeng 1997). Maintenance strategy depends on several factors including goals of maintenance, the nature of the plant, work flow pattern, and the work environment (Pintelon and Gelders 1992; Al-Najjar 1997; Gallimore and Penlesky 1998). The aim of maintenance is to control failures of industrial plants, machinery, and equipment (Alsyouf 2004). Maintenance actions can take several forms and make the use of several approaches. These include corrective maintenance (breakdown maintenance), preventive maintenance (replacement maintenance at periodic intervals) using various


Page 19

statistical models based on time-to-failure data, or condition-based maintenance using lifecycle data collected from different condition monitoring techniques. In all cases, the decision maker is required to choose from all appropriate strategies the most cost-effective one for each component or module enabling the optimization of the maintenance strategy for the plant (Alsyouf 2004). An accurate identification, optimization, and implementation of a maintenance strategy will cause significant savings in terms of reducing premature replacement costs, maintaining stable production capabilities, and reduce system deterioration rate (Vineyard et al. 2000). Knowing the optimal maintenance plan is not an easy task. In most cases, companies are either perform too much maintenance too early or too little too late, all of which have a significant impact on the asset life-cycle cost (Liptrot and Palarchio 2000).

2.1.2 Maintenance Costs and ProfitsTraditionally, maintenance has been regarded as cost center (Gatland et al. 1997). Many authors have regarded it as a non-productive support function, i.e. a necessary evil. See among others Bamber et al. (1999), Duffuaa and Andijani (1999), Al-Najjar (2000), Ralph (2000), Sherwin (2000), Al-Najjar et al. (2001), and Alsyouf (2004). Maintenance cost can be divided into direct and indirect. Examples of direct costs include labor, material, and overheads such as tools, transportation, and training. Indirect costs arise from factors such as planned and unplanned maintenance. This includes loss of production, accidents, unavailability of logistics, and loss of customers goodwill (Ahlmann 1984; Blanchard 1986; Al-Najjar 1997; Blanchard 1997; Shonder and Hughes 1997; Wilson 1999; Al-Najjar et al. 2001; Mirghani 2001; Saranga 2004). In the last few years, researchers have focused their efforts on turning maintenance into a profit-gathering function. See among others Ahlmann (1984), Gatland et al. (1997), Jonsson (1999), Wilson (1999), and Sherwin (2000). Savings and gains resulting from a cost-effective maintenance plan have been discussed by many authors. See among others Ahlmann (1984), Maggard and Rhyne (1992), Foelkel (1998), Kutucuoglu et al. (2001), and Swanson (2003). These advantages will be expanded on in the next section.


Page 20

2.1.3 Impact on Business PerformanceIt has been reported that capital investment ise influenced by many factors including equipment useful life, equipment redundancy, inventory levels, equipment damage due to breakdown, and others. On the other hand, the capability of machines to provide quality services is deeply influenced by maintenance effectiveness (Henning 1989; Taguchi et al. 1989; Oakland 1995; Al-Najjar 1997; Edwards et al. 1998; Alsyouf 2004). Maintenance efficiency contributes through better resource utilization, improved service quality, and reduced rework. In addition, it can avoid the need for additional investments and man-hour by expanding the capacity of existing resources (Ahlmann 1984; Gits 1992; Gits 1994; Ben-Daya and Duffuaa 1995; Al-Najjar 1997; Dunn 1998; Ralph 2000; Swanson 2001). Coetzee (1999) proved that the use of maintenance methodologies, philosophies, and strategies to optimize maintenance in an organization has enabled it to effectively deal with the increased importance of this function. However, it has been realized that since maintenance is service function, its merits are not realized in the short run (Pintelon and van Puyvelde 1997). In some occasions, this makes the importance of maintenance to an organization paradoxical. This is due to the fact that on one hand, the more maintenance positively affects the overall strategic goals of an organization the less recognizable it becomes to top management as a value adding activity. A poor maintenance program, on the other hand, can obstruct the addition of value, retard the advantage of a capital resource, which may lead to the destruction of the business strategy (Dunn 1998; Mcgrath 1999; Alsyouf 2004). Kutucuoglu et al. (2001) noted that in the past few years, there have been evident examples of maintenance as a profit-generating function. This is due to factors such as the effect of equipment maintenance on flexibility, service quality, operational costs, and environmental and employee safety. Effective maintenance and system reliability are key factors when quality and timely services need to be ahead of competition. See among others Cooke and Paulsen (1997), Madu (1999), Cooke (2000), and Madu (2000).


Page 21

2.2 Reliability-Centered Maintenance (RCM)

Moubray (1997) defines RCM as a process used to determine the maintenance requirements of any physical asset in its operating context. The same author also defined it as a process used to determine what must be done to ensure that any physical asset continues to do whatever its users want it to do in its present operating context. Another definition is a systematic approach for identifying effective and efficient maintenance tasks for items in accordance with a specific set of procedures and for establishing intervals between maintenance tasks (IEC60300-3-11 1999).

2.2.1 Purpose of RCMThe principles of RCM rose from the thorough examination of certain questions that were often blurred (Nowlan and Heap 1978): How do failures occur? What are their consequences? What good can preventive maintenance do?

Maintenance is primarily aimed at preserving system failure (Nowlan and Heap 1978; MIL-STD-2173 1986; Moubray 1997; Rausand 1998), not preserve equipment (Backlund 2003). Therefore, it is essential to understand what the expected outcome should be, and that the primary task is preserving that outcome or function (Smith 1993). Failure prevention is avoiding and reducing the consequence of failure more than preventing failure itself (Horton 1993). Anderson and Neri (1990) and Smith (1993) summarized the RCM method in four main features: Preserve functions, Identify failure modes that can affect the functions, Prioritize function need via the failure modes,


Page 22

Select only applicable and effective preventive maintenance tasks.

It is important to realize that RCM is a structured way of utilizing the best maintenance methods and disciplines (Sandtorv and Rausand 1991; Sutton 1995) and therefore is not aimed towards finding new principles for performing maintenance tasks . RCM in many aspects can be compared with some kind of quality assurance of a maintenance performance, defined as all systematic actions required to plan and verify that the efforts spent on maintenance are applicable and cost effective (Sandtorv and Rausand 1991; Backlund 2003). If a maintenance program already exists, the result of an RCM analysis will often be to eliminate inefficient preventive maintenance tasks (Rausand 1998).

2.2.2 RCM BenefitsRCM has many benefits and advantages, and can have a significant impact on business operations, safety, logistics, configuration and administration (Smith 1993). Several authors discussed those benefits, see among others Anderson and Neri (1990), Ryan (1992), Bowler and Leonard (1994), Harris and Moss (1994), Pintelon et al. (1999), Hardwick and Winsor (2001), Backlund (2003) and others. Benefits gathered from those authors are summarized in Table 2.2.1.

Table 2.2.1 - RCM potential benefits (Backlund 2003)

Maintenance Improved system availability, Less corrective maintenance tasks, More condition monitoring tasks, More methodical maintenance,

Logistics Cross-discipline usage of experience, A broader way of working, Better and more comprehensive documentation, Reduced inventory levels,

Costs Reductions in preventive maintenance costs, expenditure, Improved long-term budgeting.

Better operational feedback, Reduced maintenance


Page 23

Less maintenance hours, Improved maintenance plans, Maintenance optimization, Assess hidden failure modes (less corrective maintenance),

Better availability of maintenance history data,

Improved maintenance planning, Better teamwork, exchange among different departments (engineering, maintenance, management), Better communication between representatives of maintenance and operations functions, Better System safety. On-line information

Uniform and consistent asset maintenance, Improved system reliability.

2.2.3 The RCM MethodRCM has been described extensively by different authors. See among others Nowlan and Heap (1978), MIL-STD-2173 (1986), Smith (1993), Vatn et al. (1996), Moubray (1997), Rausand (1998), Rausand and Vatn (1998), and NASA 2000). Moubray (1997) developed seven questions that are regarded as the core of RCM theory: 1. What are the functions and associated performance standards of equipment in its present operating context? 2. In what ways does the system fail to fulfill its functions? 3. What are the causes of each functional failure? 4. What happens when each failure occurs? 5. In what ways does each failure matter? 6. What can be done to prevent each failure? 7. What should be done if a suitable preventive task can not be found?


Page 24

Figure 2.2.1 - RCM analysis steps

Those questions are answered by working through a number of structured steps described by MIL-STD-2173 (1986), Moubray (1997), Rausand (1998), Rausand and Vatn (1998), IEC60300-3-11 (1999), NASA (2000), Backlund (2002), Backlund (2003) and shown in Figure 2.2.1. The order and the number of steps can differ among different authors recommendations, however, their approaches are all quite similar (Backlund 2003), and some of them can be overlapping in time (Vatn 1996; Rausand 1998). More about each step is discussed in detail by Rausand (1998).

2.2.4 Views on RCMEven though the definitions of RCM in different textbooks are quite similar, different authors had different opinions regarding what RCM is (Backlund 2003). RCM has been described as: A logical and systematic methodology (Bassile et al. 1995; Ben-Daya 2000; Hipkin and Decook 2000; Rajotte and Jolicoeur 2000). A power full tool (Geraghty 1996; August 1997). A maintenance technique (Bowler and Malcom 1994; Rotton 1994; Page 2001). A maintenance management strategy (Briggs 1994; Harris and Moss 1994). A process (Adjaye 1994; Hipkin and Lockett 1995; Hardwick and Winsor 2001).


Page 25

A complete maintenance philosophy based on maintenance engineering (Worledge 1993).

Even though the different views on RCM can be, to some extent, very similar, however, the many concepts can be extremely confusing to both researchers as well as practitioners. In addition, those different views indicate how people look upon RCM, what it can achieve, and consequently, what the requirements are when introducing it to an organization (Backlund 2003).

2.2.5 RCM ImplementationEven though some authors have stated that many companies have successfully introduced RCM (Moubray 1997; Hipkin 1998; Rausand 1998), some companies have faced severe difficulties and in some occasions implementation has failed (Smith 1993; Worledge 1993; Worledge 1993; Bowler and Leonard 1994; Bowler and Malcom 1994; Schawn and Khan 1994; Moubray 1997). The consequences of failed implementation can be rather severe since it may lead to a less effective maintenance performance despite expanded resources. In addition, it may cause the personnel involved to resist future improvement projects (Backlund 2002). The main criteria identified by (Backlund 2002; Backlund 2003) as being the main factors behind unsuccessful implementations are high initial costs, timely economic return, high resource availability requirements, and poor data availability: Data availability: Full benefit of RCM can only be achieved if life-cycle data (LCD) is available (Madu 2000; Backlund 2003). The company should be ready to collect real data throughout the life of the equipment (Nowlan and Heap 1978; Vatn et al. 1996) in order to obtain optimal values for maintenance intervals (Backlund 2003). In cases where no, or very poor data is available, RCM can only be used to assess the type of suitable maintenance tasks (Sandtorv and Rausand 1991).


Page 26

Costs: Even though the implementation of RCM may lead to substantial benefits, high initial costs may delay, and in some occasions cause the failure to adopt this philosophy. This is due to the fact that management support may vanish with unexpected escalation in costs (Bowler et al. 1995). A number of nuclear plants have withdrawn the implementation of RCM due to the high initial costs involved (Worledge 1993).

Time: RCM is regarded as a long-term plan with short term expectations (Latino 1999). A main drawback of RCM is it being too time-consuming (August 1997; Hipkin 1998). A challenge is to adopt RCM to achieve faster economic return (Jones 1995; Rausand and Vatn 1998).

Resources: The introduction of RCM requires a large amount of resources to be successful (Schawn and Khan 1994; Jones 1995; Moubray 1997; Latino 1999). It is not always possible for resources and operator training according to RCM to be available (Smith 1993).

Examples of successful implementations have been reported in many industries. It was noted that in those industries, RCM caused significant improvements in system reliability and availability. In addition, it reduced the amount of preventive maintenance activities, and increased system safety. Those industries include: Aircraft industry (Nowlan and Heap 1978) Offshore oil industry (Sandtorv and Rausand 1991) Nuclear industry (Srikrishna et al. 1996) Robot and automobile industry (Pintelon and Gelders 1992) The conditions for implementing RCM in safety related and complex machinery involving leading edge technologies have some specific characteristics. The successful implementation of RCM in the aircraft industry among other reasons is due to the fact that RCM was applied early in the design stage with a few resource constraints, and to specialist


Page 27

personnel performing the analysis (Harris and Moss 1994; Backlund 2002). In other industries like power, processing and manufacturing, RCM is mainly applied to preexisting plants that are individually designed to meet a wide range of output requirements. Another condition is the available resources, which are usually established by custom and usage, and the time of introduction of RCM in times of restraint and rationalization (Harris and Moss 1994; Backlund 2002). Literature findings on comprehensive studies on RCM implementation were very few. Backlund and Akersten (2003) presented a detailed study on managing the introduction of RCM in a hydropower company. The author listed the deficiencies that became obstacles to the progress of the RCM implementation. They include: Lack of Computerized Maintenance Management System (CMMS), which made the gathering of information and data needed to support the analysis difficult. Lack of RCM computer system, which was required to handle the many analyses made. Lack of plant register, which made it difficult to develop an RCM co

weibull model

Documents