Definition of Maintenance and Maintenance Types with Due Care on Preventive Maintenance

Abstract

In this chapter maintenance concept is defined and maintenance types are classified with regard to implementing maintenance policies toward preventive maintenance. Especially achieving planned maintenance policies toward preventive (planned) maintenance and condition based maintenance policies toward predictive maintenance is taken into consideration primarily. Maintenance concept is defined and maintenance types are classified. Due care is given for ‘Preventive Maintenance’ in this chapter. In general ‘Maintenance’ term could be defined as; The integration of all possible technical and administrative actions, including planning, supervising, monitoring and controlling toward retaining an item, a system, a machine to restore their original functional state in which they can perform the intended functions. In addition maintenance, include protective and corrective actions to keep the plant operational system in intended conditions or to maintain the acceptable manufacturing conditions. Optimum maintenance policies aims to sustain system reliability and robustness within minimum cost. In line with the progress of industry, increase in the system, material and manpower costs, increasing demand for robustness and the complex structure of the machines increases also the importance of maintenance policies. Maintenance types could be divided into two main parts namely; 1. Preventive Maintenance, 2. Corrective Maintenance. Preventive Maintenance is also classified into following sub groups; 1.a- Planned Maintenance, 1.b- Predictive Maintenance, 1.c- Advanced Maintenance Implementations; 1.c.1 Reliability Centered Maintenance 1.c.2 Risk Based Maintenance. All of these maintenance types elaborated with relevant figures in the chapter. In this chapter Comparison of Planned and Unplanned (corrective) Maintenance (With regard to transaction and output) is defined with a table. Additionally, Comparison of Planned and Unplanned (corrective) Maintenance (With regard to infra structure) is also tabulated. ‘Benefits of Preventive & Predictive Maintenance’ and ‘Predictive Maintenance Methods’ are defined with relevant descriptive figures in the chapter. For Corrective Maintenance, basic definitions and corrective maintenance steps, types of corrective Maintenance, improvement strategies in corrective maintenance effectiveness are also given. In the final part of ‘Results and conclusions’ expected and verified benefits of implementing maintenance policies for planned and predictive maintenance are explained. Comparisons in some maintenance policies is given.

1. Introduction

In this chapter maintenance concept is defined and maintenance types are classified with regard to implementing maintenance policies toward preventive maintenance. Especially achieving planned maintenance policies toward preventive maintenance and condition based maintenance policies toward predictive maintenance is taken into consideration primarily.

In the progressing parts of this chapter it is attempted to define that planned maintenance policies and condition based maintenance policies provide beneficial results in terms of overall preventive maintenance.

As the result of this chapter evaluation it is questioned whether to obtain below defined fruitful results via achieving maintenance policies such as;

Maintenance policy ensures that: machinery & equipment are in available and reliable condition, company capable of responding usual and sudden customer demands with regard to utilization of equipment, machinery & equipment is stable and consistent enough to manufacture good quality products, better maintained machinery & equipment is a key for succeeding strong competition, better maintained machinery & equipment does not allow sudden or long standing breakdowns, this result is end up with less inventory loss, higher market share and with better maintained machinery & equipment longer MTBF (Mean Times Between Failures) and shorter MTTR (Mean Time To Repair) scores are obtained, and compliying with JIT (just in time) production approach, better maintained machinery & equipment eases overall cost control [1].

Due care is given for ‘Preventive Maintenance’ in this chapter. In general ‘Maintenance’ term could be defined as; The integration of all possible technical and administrative actions, including planning, supervising, monitoring and controlling toward retaining an item, a system, a machine to restore their original functional state in which they can perform the intended functions. In addition maintenance, include protective and corrective actions to keep the plant operational system in intended conditions or to maintain the acceptable manufacturing conditions.

With the final part of this chapter expected and obtained outcomes of maintenance policies are defined with regard to planned and predictive maintenance policies that will cover the overall preventive maintenance benefiits. Comparison is also made between preventive (planned) and predictive (condition based) maintenance policies.

2. Overview of the existing related works

There are various research studies in the recent years on the ‘Preventive Maintenance’ and Total Productive Maintenance (TPM) issues. In a work by [2] Brankovic Dejan, Milovanovic Zdravko, The Role and Importance of Planning of Maintenance in Industrial Practice, importance of maintenance planning and types of mintenance is explained. In the study of [3] Tran Duc, Dabrovsky Karol, Skrzypek Katarzyna, The Predictive Maintenance Concept in the Maintenance Department of the “Industry 4.0”, they have pointed out the importance of predictive maintenance to achieve ‘Industry 4.0’ and to be competent in the market. In a work of [4] Ötleş S, Çolak, UC, Ötleş O. Artificial Intelligence for Industry, with the aid of machine learning and Internet of Things (IoT) approach predictive maintenance is investigated in order to manage the maintenance management potentials and trends. In another study [5] Paresh Girdhar BEng (Mech. Eng), Girdhar and Associates, Practical Machinery Vibration Analysis and Predictive Maintenance, predictive maintenance techniques, maintenance philosophies, principles of predictive maintenance is explained. In another work [6] Tiena Gustina Amran and Leonardus Sujarto, Early Warning System in Preventive Maintenance as a Solution to Reduce Maintenance Cost, importance of early warning system and The Early Warning System application for computer based Preventive Maintenance implementations is studied. In the work of [7]. Dr. S. J. Lacey, The Role of Vibration Monitoring in Predictive Maintenance, importance of predictive maintenance via advanced vibration monitoring techniques is explained in order to detect the equipment failures before its happening.

In all of these studies different useful aspects of preventive and predictive maintenance implementations are pointed out and in some cases advantages of these advanced maintenance techniques over reactive-corrective maintenance are also mentioned.

3. Methodology

In this chapter main concern is given for preventive maintenance activities. With consideration of recent publications and references a research is made on preventive maintenance. Research methodology depends on the definition of maintenance types and the relevant maintenance policies. Throughout the chapter implementation results of the maintenance policies are questioned. Relevant comparisons are made between maintenance policies. The findings are discussed in the part of ‘Results and conclusions’.

4. Definition of maintenance and maintenance types

In general ‘Maintenance’ term could be defined as; The integration of all possible technical and administrative actions, including planning, supervising, monitoring and controlling toward retaining an item, a system, a machine to restore their original functional state in which they can perform the intended functions [8].

In addition maintenance, include protective and corrective actions to keep the plant operational system in intended conditions or to maintain the acceptable manufacturing conditions. Optimum maintenance policies aims to sustain system reliability and robustness within minimum cost. In line with the progress of industry, increase in the system, material and manpower costs, increasing demand for robustness and the complex structure of the machines increases also the importance of maintenance policies. One of the main reasons in inefficiencies and inconsistincies of production systems is lack of proper maintenance policies or lack of their implementations. However in the recent years, the importance of maintenance policies have been perceived by the industrial, engineering sectors and academic disciplines within operational management and due care have been given [9]. Maintenance types could be divided into two main parts namely; 1. Preventive Maintenance, 2. Corrective Maintenance. Corrective maintenance could be divided into two groups; 2.a- Unplanned repair and Change, 2.b-Foreseen repair and change.

Preventive Maintenance is also classified into following sub groups; 1.a- Planned Maintenance, 1.b- Predictive Maintenance, 1.c- Advanced Maintenance Implementations; 1.c.1 Reliability Centered Maintenance 1.c.2 Risk Based Maintenance.

A schematic view is given with Figure 1 for Work Flow of Preventive Maintenance [10].

Figure 1.

Work flow of preventive Maintenance [10].

In Table 1 Comparison of Planned and Unplanned (corrective) Maintenance (With regard to transaction and output) is given. It is clearly evident from the table that planned maintenance has a wide range of supremacy over the unplanned maintenance with regard to transactions and output.

Serial No.	Parameters	Unplanned maintenance	Planned maintenance
01	Failures	High	Low
02	Down time/stoppage	High	Low
03	Product output	Low	High
04	Maintenance costs	High	Low
05	Reliability of equipment/plant	Low	High
06	Availability of equipment/plant	Low	High
07	Percantage usage of equipment/plant	Low	High
08	Spare parts and inventory control	No	Yes
09	Failure Warning / Early Failure Warning	No	Possible

Table 1.

Comparison of Planned and Unplanned (corrective) Maintenance (With regard to transaction and output). (Own compiled).

In Table 2 Comparison of Planned and Unplanned (corrective) Maintenance (With regard to infrastructure) is given. It is clearly evident from the table that planned maintenance has a wide range of supremacy over the unplanned maintenance with regard to infrastructure. However as it is indicated with (*) marked items planned maintenance would require extra financial allocations.

Serial No.	Parameters	Unplanned maintenance	Planned maintenance
01	Maintenance manpower size	Small	Wide
02	Technical competence of manpower	Low	High
03	Required special equipment	No	Yes*
04	Required expert services	No	Yes*
05	Need for establishment of a special laboratory	No	Yes*
06	Computer back-up is necessary	No	Yes
07	Personel spesific training	No-insufficient	Yes
08	Infrastructure establishment cost	Low	High

Table 2.

Comparison of Planned and Unplanned (corrective) Maintenance (With regard to infra structure). (Own compiled).

^*

Marked items require extra financial resources.

5. Preventive maintenance

5.1 Planned maintenance

As a part of preventive maintenance, planned maintenance is to plan the maintenance of machinery, equipment, buildings and plants in advance of their wearing or unplanned stoppage and to prevent them to face breakdown. The time frame of Maintenance plan is usually defined as 6 month or 1 year. Depending on the conditions and structure of the machinery, equipment or plants maintenance intervals defined in weeks, months, quarters or yearly basis. In most cases an initiating advice is taken from machine or equipment manufacturer maintenance manuals for definition of regular maintenance interval. Time based planned maintenance renders the machinery to restore their own original proper functioning state and defers the unexpected breakdowns. During the planned maintenance certain parts of machinery or equipment are replaced with regard to their service life capacity [2].

Benefits of planned (preventive) maintenance policies;

Planned maintenance offers five basic outcomes in general namely;

1. Maintenance cost reduction- By constructing a planned (preventive) maintenance plan small scale failures of the equipment can be detected in advance they turn into bigger problems hence substantial maintenance costs are avoided.

2. Extended Asset Life: Timely and regularly maintaining assets prolongs expected life span and early breakdown in life cycle is avoided.

3. Workplace safety improvement: With the outcome of planned (preventive) maintenance proper functioning of equipment provides also work place safety, hence operators and workers will avoid potential accidents.

4. Improving awereness and company culture: Apart from reducing equipment downtime and failures, planned (preventive) maintenance will also reduce employee absentees due to improved working environment team spirit and moral attitudes.

5. Decreased downtime due to planned (preventive) maintenance: Planned maintenance enables the maintenance team to resolve minor equipment failures before they turn into bigger problems. Due to gathering valuable data on the equipment history with planned maintenance will enable the responsible persons to take preventive actions toward improving life cycle span of the equipment [11]

However in order to obtain above defined benefits, advanced digital infra-structure must be established and a platform of IoT (Internet of Things) must be placed. Futhermore in order to analyze the machinery data, machine learning and forecasting based modeling statistical techniques must be utilized [12].

5.2 Predictive maintenance

As it is understood from the headings, this type of maintenance points out the prediction of breakdown probability of an equipment by automated computurized monitoring and assessment steps and provide a new maintenance plan for failure prevention [3].

For a successfull implementation of predictive maintenance and to ascertain the machinery-equipment proper functioning conditions it is recommended to provide the following data; a. Equipment Operational Records, b. Past data and records on downtimes, breakdowns, performance. c. Condition of machinery-equipment with regard to operating parameters in the past operating period. d. Artificial Intelligence data from machine learning and data analytics. Upon obtaining the above defined data and having benchmarking experience from similar equipment and similar cases a new maintenance schedule is defined [13]. Predictive maintenance provides information on average performance values of machinery and equipment, their potential failures, maintenance state and schedule, the ways how to repair the equipment,..etc. similar data. Hence it provides valuable information in advance of critical production equipment breakdowns for the maintenance team and guides them for the certain repair and maintenance ways and inform them in order to minimize the equipment down time. Due to this beneficial consequence sustainability and efficiency in production is increased and production and maintenance costs is decreased [4].

6. Predictive maintenance methods

Various tools and ways may be employed in the implementation of ‘Predictive Maintenance’. Some of the main ways are given below;

1. Failure early warning system

2. Vibration Analysis

3. Thermal analysis-thermography

4. Acoustic Emission

5. Oil & Particle Analysis

6. Corrosion Monitoring

7. Performance Monitoring [5].

6.1 Failure early warning system (IEWS)

One of the prime implementation methods of the predictive maintenance is to utilize the early warning system. Early warning system supports the existing computerized maintenance program. With the implementation of early warning system (IEWS), maintenance activities are conducted more effectively and potential failures of machinery and equipment are detected in advance of failure. With the aid of this system, priorities in maintenance needs of the equipment is ascertained and certain parts that are necessary for replacement, are defined. With the availability of (IEWS) records maintenance costs and potential saving of the organization by avoiding the damage is defined. IEWS provide a valuable maintenance data base for the future use of the organization and for establishing an efficient document control system. As a result with the implementation of IEWS overall effectiveness of equipment and plant usage is increased and it paves the way for further innovative and sustainable maintenance actions [6].

One of the implementation area of an early warning system is using new technology for detecting abnormal equipment performance in power plants. Existing technology can reduce derates and forced shutdowns by providing means to plant operators to adjust-repair small problems before they turn into large problems.

Power generation operators are oriented adopting asset management to improve process efficiency and to increase return on assets (ROA). High value equipment and components such as boilers, turbines, generators and auxiliary systems present an attractive target for asset management since they susceptible to cause derates and forced outages when they fail. Some new technologies in this regard calls predictive condition monitoring, reduces forced outages and derates through actionable early warning of failure of critical power plant equipment. Apart from preventative maintenance implementations, which foreseen maintenance based on failure statistics for a type of equipment problems over time, predictive condition monitoring provides equipment-specific, condition-based early warning [14].

6.2 Vibration analysis

Vibration analysis is a common used predictive maintenance technique that is used to define the existing operating condition of a machinery or equipment in advance of developing problems before they become too critical and might cause unexpected downtime via regular monitoring of equipment vibrations. By way of vibration monitoring, deteriorating or damage of equipment bearings, vanes or blades, belts, mechanical looseness and worn or broken gears,..etc. can be detected [7].

In Figure 2 [15] Vibration analysis in predictive maintenance is depicted with Early bearing and Gearbox Fault, Late Stage Bearing and Gearbox Fault and Imbalance, Misalignment, Looseness stages.

Figure 2.

Vibration analysis in predictive maintenance [15].

Basically vibration measurement technique, is an effective, non-intrusive method to monitor machinery or equipment during start-ups, stoppage and regular operation stages. Most frequent usage of vibration measurement is realized on all rotating equipment especially on spindle-bearings, piston-cylinder connections namely various types of gas, steam, and wind turbines, compressors, motors, pumps, vantilating fans, rolling mills, gearboxes,…etc. Main parts of vibration analysis system could be defined as; a. Signal pickup(s), b. Signal anlayzer, c. Analysis software, d. A Computer for data analysis and storage. A confugaration can be made among those four basic parts to establish a permanent online system, a periodic analysis system or a multiplexed system that provides sampling of a certain transducers at advance defined time intervals [5].

7. Benefits of vibration analysis

It is possible with vibration analysis to determine the improper cases in machinery and equipment maintenance or repair. Amon those improper practices, improper bearing installation and replacement, inaccurate spindle alignment, or loose rotor balancing can be cited. Early vibration testing renders maintenance staff predictible information on required repairs and necessary parts, enables them take the faulty equipment away from the operation place to prevent any possible hazard, help to prevent equipment ceasing, fosters extending equipment life capability, helps to reduce unexpected equipment breakdowns and failures. Statistically almost over 75% of common revolving equipment failures are related to misalignment and unbalance, hence vibration analysis becomes a prime tool that can be employed to reduce or mitigate repeating equipment failures and problems. As a result a vibration analysis may be employed as prime segment of generic predictive maintenance program [16].

Main components of predictive maintenance in the electrical engineering is defined with Figure 3 [17].

Figure 3.

The main components of predictive maintenance in the electrical engineering [17].

7.1 Thermal analysis

One of the most effective tools of predictive maintenance is to utilize the thermographic analysis. The technique is based on the infrared thermography (IRT). Since the majority of the machinery equipment failures in the industry becomes evident with temperature changes that can be sensed by regular monitoring with an infrared thermographic system. Thermal data from the equipment is collected via thermal sensor which may be a key source of information for diagnosis and enables the maintenance team to define the failure causes in advance of their occurrence. Hence such an early detection prevents potential future problems before occurrence and high and unexpected repair costs are avoided [18].

Thermal scanning test with thermal camera: Thermal scanning process with thermal camera is made with the aid of infrared beams. All the objects having heat level above −273°C, transmit thermal energy. Visible wavelength that is seen by human eyes is between 400 nm and 700 nm range. The beams having values below that level; infra violet beams, x-rays or gama rays and the beams having values above that level; infra red, micro wave beams, radio-tv beams can not be seen or sensed by human eyes. Thermal cameras due to self mounted sensors can sense the infra-red beams that are emitted from the hot surfaces of the objects and evaluate them with the aid of a special software and ascertains the temperature values. Heat measurement with thermal cameras will help to define the equipment failures as it was found with the other predictive maintenance techniques. For example on the locations that loose electrical connections are evident, overheating may arise due to increasing resistance or electric motors due to inefficient working conditions may present higher working temperatures than the normal operating temperatures. Due to friction between rotating components unwanted higher temperatures may arise. This excess temperature indicates a loss of efficiency, on the other hand fire may break out unless a preventive measure is taken. Proper functioning of thermal camera will reflect all these problems on the screen directly [19].

Impairment detection with thermal analysis: Figure 4 shows the thermal camera image for the rotating furnace mantle in cement factory. The aim for thermal analysis is to define the heat distribution in the furnace whether it is uniform or not and to detect the wears of refracter tiles on the mantle interior wall. These analyses are monitored continuously both with mobile thermal camera and on-line thermal scanners. As the result of those analysis life capacity of refracter tiles could be detected and stoppage planning to be made for rotating furnace. Furthermore with the thermal analysis method the following detections could also be made; a. Anomalies in electric panels and connections, b. Failure detection in electric motors, c. Refracter tile wearing on rotating furnace and siclons. d. Leak detection in hydraulic and steam circuits…etc. Figure 5.

Figure 4.

IR thermal images of an equipment currently used in power electrical installations [17].

Figure 5.

Impairment detection with thermal analysis on the rotating furnace mantle [19].

7.2 Acoustic emission

With the aid of Acoustic Emission method, impairments in the machine elements such as bearings and gearboxes and gas leakages on the pipes are detected. In most cases, in acoustic emission monitoring is made with ultrasonic voice detectors. E.g ultrasonic air leak detector, and bearing voice sensing instruments are utilized for this purpose.

Case study for acoustic emission detection: In a cement factory as it is seen from (Figure 6) thickness measurements are made on the cement factory rotating furnace mantle in annual periods and some laminated region is detected on the mantle. For clear definition of laminated region detailed scanning is made with ultrasonic test equipment and laminated area has been detected (Figure 6). Upon further review and investigations, laminated region has been replaced with the new mantle part. After defining the laminated part on the furnace mantle, the effected region is marked with permanent color marker and it has been monitored and controlled in suitable furnace stoppage periods until the mantle part is replaced. As the result of those controls it is found with the ultrasonic tests that laminated region is propagated and a possible future mantle crack and an unexpected furnace stoppage is prevented with the timely measures.

Figure 6.

Impairment detection on the cement factory rotating furnace mantle via ultrasonic analysis [19].

7.3 Oil and particle analysis

Lubrication is necessary for the mechanical parts of the equipment for smooth operation. With the use of proper lubricant, friction among the mechanical parts is minimized. If the quality of the lubricant (oil) is worsened in the course of time, wear and overheat arises due to excess friction. Lubrication among the moving parts of the equipment is very important. The analysis of lubricant between the contact surfaces, is one of the preferred methods in predictive maintenance [20].

Lubrication in equipment and machinery mechanical parts provides two main benfits; firstly, it provides a preserving film between the moving mechanical parts surfaces, and hence reducing the harmful friction, eliminating unwanted seizing, secondly, lubrication provides cooling of mechanical components, protects the metal surfaces from corrosion, and provides a contaminant deposit free surface [21].

Possible changes in the physical and chemical features of the oil affects the performance characteristics of the lubrication oil, which may lead to performance hampering. That is why it is essential to assess the performance parameters of the oil to clerify that if the oil quality is worsened to a critical level that oil can not fulfill its intended function. There are various lubricant oil evaluating and monitoring techniques that may monitor the oil charactersitics fully or partially. Main causes of the lubricant oil degradation could be attributed to, particle contamination, oxidation and or water contamination [22].

Oxidation products hampers the required viscosity state and lead to wear particles formation that also results additional damage to the mechanical system when they contact with the component surfaces. Wear particles may block the filters and or oil holes and hence causing oil shortage and friction and seizing between moving mechanical components. Worst of all wear particles can tear the filter and high level contamination may occur. Resulting study of wear debris in the oil, enables to detect potential harm in advance of the expected failure so that required preventive measure could be taken [23].

At the result of oil analysis, parameters such as; physical and chemical features of the oil, number and size of the contaminant particules and the pollution of the oil are analyzed in order to make interpretation about the possible future failures. Oil analysis gives us clues about the level or magnitude of impairment on the worn parts. Main reason for the friction of mechanical parts is usually attributed to the low viscosity. Hence viscosity of oil is monitored periodically for assessing overall oil quality, oil is replaced on the point that oil loose its intended features and the equipment is taken into immediated maintenance program [24].

7.4 Corrosion monitoring

One of the important conditional based monitoring method in predictive maintenance is corrosion monitoring. Corrosion is defined as ‘metals losing their metallic features by getting into chemical and electrochemical reactions with the surrounding environment’. Corrosion is very important for a country’s economy. Recent research indicates that corrosion causes a loss of ¼ of total steel production. Climatic conditions in the cities, rain waters (traces of sulfuric acid or nitric acid), and sea waters are the prime causes of the corrosion.

The most encountered materials type for corrosion are metals since they have a higher tendency with electrochemical reactions. In metals corrosion oxigen is the prime reason. However there are some side effects for corrosion along with oxygen. For example aluminum external surface oxidized very quickly and after surface oxidizing is finished, a resistant protective coating is formed that prevents oxidizing the deeper surfaces. That means external surface is coated with oxygen (corrosion) resistant (Al₂O₃). During corrosion, anodic (electron donating-oxidation) reactions and cathodic (electron receiving-reduction) reactions occur together.

7.4.1 Factors effecting corrosion

Effect of the environment: The rate of corrosion of metals is largely related to the environment in which they are found. The amount of humidity in the environment, acidity-basicity, the ability of air, oxygen or water to pass through the environment, leakage currents and various bacteria appear as initiating and accelerating factors.

Effect of temperature: Increasing ambient temperature increases the rate of corrosion by increasing ion movement. The soil, whose ambient temperature varies between −50 and + 50 0C, freezes at 0 0C and the ion movement speed decreases to a minimum. Increasing the temperature also has the effect of lowering the oxygen concentration. However, this effect is rather weak compared to the reactions caused by increased ion movement.

Effect of material selection: One of the factors that cause corrosion is the use of metals that have potential differences with each other. This is an initiating and accelerating factor of corrosion. For example, stainless steel bolts and gaskets placed on panels made of steel sheet, as a common mistake, cause galvanic corrosion in the area where they are located. In such cases, bolts or gaskets to the main surface should be isolated with plastic.

Differences in properties between grains: As a result of the differences between the grain sizes of the metals and the different concentrations in the two grains, the boundary of the two grains creates a suitable environment for the initiation of corrosion. As a very common mistake is to corrode the welding areas in tanks and similar structures made of stainless steel materials, even though it is not expected by the manufacturer. The way to prevent this corrosion is either to use electrode welding or to apply a galvanic anode cathodic protection system as a preventative.

System design: In systems where corrosive materials are stored, designs should be applied to prevent the accumulation of corrosive medium (water etc.). Also, very thin gaps that can cause liquid accumulation between them should be avoided.

Oxygen concentration of the environment where the system is located: In the same type of soil, the dissolved air concentration may not be the same everywhere. In systems with different ventilation conditions, the system standing next to each other is the anode in one area, while it can act as a cathode in the area next to it, causing electrochemical corrosion.

Effect of soil electrical resistivity: High conductivity in low electrical resistivity regions causes the ionic medium to be more active. Therefore, the corrosion mechanism develops faster.

7.4.2 Corrosion types

Homogeneous (uniform) corrosion: It is the type of corrosion that occurs on the metal surface at an equivalent severity. As a result of corrosion, the metal thickness decreases by the same amount at every point. Metals that are produced from the same type of material in the atmosphere and are not affected by any external factors undergo homogeneous corrosion.

Galvanic corrosion: It is the type of corrosion caused by the use of two materials with different potentials together or the difference in the ground structure. Corrosion caused by the use of different materials creates a galvanic cell between two metals at different potentials when they are in contact with each other, and the active metal acts as the anode and the noble metal acts as the cathode, causing corrosion in the active metal. For example, if copper and steel come into contact, steel will corrode due to copper.

Electrolyte: Solution or moist materials containing ions that conduct electric current. In summary, it is a corrosion phenomenon that occurs on the more electronegative metal surface when two different metals immersed in an electrolyte are in contact with each other.

Galvanic anod: The galvanic anode is the electrode that is used to protect a structure cathodically and that provides current production by dissolving it as a positive ion in the environment. If a more active metal (galvanic anode) is to be attached to a corroding metal, then the electrons required for the cathode reaction are provided by the self-propelled oxidation reaction of the metal connected as the galvanic anode. Thus, all anodic reactions on the protected metal surface are completely stopped. Galvanic anode cathodic protection is also based on this basic principle. In order to cathodically protect a steel pipeline with galvanic anodes, a more active metal (magnesium anode, etc.) is connected to the pipeline. Thus, magnesium becomes the anode in the galvanic battery and the cathode in the steel pipe. Magnesium dissolves at the anode, releasing electrons. These electrons supply the electron requirement of the cathodic reaction. In order for the system to work spontaneously, there must be a potential difference between the anode and cathode enough to overcome the circuit resistance. Types of galvanic anodes. There are three types of galvanic anodes. 1. Magnesium anode, 2. Zinc anode, 3. Aluminum anode.

Crack Corrosion: It is a type of corrosion that occurs in a crack on the metal surface or in a narrow gap. The main cause of this corrosion is oxygen between the crack and the surrounding electrolyte concentration or the difference of metal ion concentration. Since the outer parts of the crack will be the cathode, corrosion does not occur in this region.

Pitting corrosion: Corrosion that occurs in the form of deep and narrow cavities as a result of the concentration of corrosion on very narrow areas is called pitting corrosion. The depth of these pits is approximately the size of its diameter. The mouth areas of the pits are often filled with corrosion products. It is a dangerous local damage with the appearance of tingling on the metal surface.

Stratification corrosion: If intergranular corrosion occurs parallel to the extrusion or rolling surface, it is called stratification corrosion. In this type of corrosion seen in aluminum and its alloys, damage occurs from the grain boundary elongated in the rolling direction. Corroded metal layers are separated from each other and the corrosion products formed cause the material to separate in layers.

Erosion corrosion: In this type of corrosion, which is especially common in pipe systems and ports, the wear rate of the metal increases due to the relative movement between the metal and the corrosive medium. Holes, grooves and trenches form on the metal surface. It manifests itself in many structures in motion in water. The presence of solid particles in the environment further increases the corrosion rate.

Leakage current corrosion: The leakage current of rail vehicles such as trains, trams and subways in the soil causes very severe and rapid corrosion in underground pipes. At every point of the line, there is a current toward the ground. Metal corrodes according to Faraday’s Law. In particular, the leakage current emitted from the rail vehicle returns from the pipe to the rail around the point where the negative pole is connected to the rail and creates the risk of corrosion.

Coating failure corrosion: The potential of a coated metal is different from the potential of an uncoated metal. In case of deterioration or perforation of some parts of the coating due to workmanship errors, these regions will become anodes and will corrode. This type of corrosion is a corrosion that concentrates in very small areas on the metal surface [25].

Corrosion monitoring in predictive maintenance is one of the important methods. The damage on the metal bodies or on the metal surfaces realized in different forms as explained in details above. In most cases corrosion in metal surfaces ends up with reduced wall thickness or tears or holes on the metal surfaces [26].

Corrosion mechanism usually defined as an electrochemical process, including charge transfer between anodic and cathodic parts of the system. Corrosion measurement in an electrochemical reaction is usually made by setting up two working electrodes (WE1 and WE2). The current is measured via a zero resistance ammeter. (ZRA). As depicted with (Figure 7) [27].

Figure 7.

Practical layout of an EN measurement cell for corrosion monitoring [27].

Cathodic Protection (CP) monitoring: One of the most important forms of corrosion protection for submerged/underground structures (such as tanks or pipelines) is cathodic protection. In Figure 8 [28], a typical cathodic protection architecture is depicted [28].

Figure 8.

Practical EN setup using embedded electrodes for corrosion monitoring [28].

7.5 Performance monitoring

One of the most important predictive condition monitoring technique is the performance monitoring. In our days high technology performance monitoring instruments are being designed and manufactured. A common example in the industry is on-line and/or off-line performance monitoring instruments for electrical motors. By way of those instruments it is possible to monitor and detect insulation defects, broken rotor rods, torque problems, load problems and power problems (Figure 9) [29].

Figure 9.

SKF – Baker AWA IV – Offline Test Instrument SKF – Baker (left), The Explorer – Online Test Instrument (right), for electric motor performance monitoring [29].

On the above given figures (Figure 10) [31] two instruments are the sample for on-line vibration performance monitoring by, SKF Multilog IMX system is designed for on-line monitoring and it is capable of making simultaneous analysis of a potential failure. SKF Multilog DMX system is designed for protecting the turbo machinery or turbines and is capable of analysis [31].

Figure 10.

SKF MULTILOG DMX (left) – SKF MULTILOG IMX-S (right), monitoring & analysers, for monitoring vibration performance [30].

8. The benefits of predictive maintenance policies

Experience and feedback data from the field indicates that ‘Predictive maintenance’ provides the following benefits;

1. Maintenance costs of the production department is decreased.

2. Equipment life time capacity is increased.

3. More precise detectability of falures in advance of occurrence.

4. Higher gains in ROI (Return of Investment).

5. The equipment breakdown durations is decreased.

6. Production quality, sustainability and outputs is increased due to pro-active and advance intervention into maintenance process.

7. Increases safety.

8. Reduces interruptions of services.

9. Greater customer satisfaction.

However in order to obtain above defined benefits, advanced digital infra-structure must be established and a platform of IoT (Internet of Things) must be placed. Futhermore in order to analyze the machinery data, machine learning and forecasting based modeling statistical techniques must be utilized [12].

8.1 Advanced maintenance implementations

Advanced Maintenance Implementations are also divided into two groups;

1. 1.c.1 Reliability Centered.

2. 1.c.2 Risk Based Maintenance.

8.2 Reliability centered maintenance

(RCM)-As a part of advanced maintenance technique, ‘reliability-centered maintenance (RCM)’ is the optimized mix of reactive, time or interval-based, condition-based, and proactive maintenance types.

The objective of RCM, to generate optimized maintenance plans. Since the maintenance is defined as the sum of technical and administrative activities in order to protect the equipment integrity and it is made for enabling equipment to fulfill its intended functions. On the other hand RCM is the maintenance activities that are implemented for an equipment to fulfill its functions in a manner that is technically in compliance, feasible and approved economically.

Those defined duties; must protect equipment functions, prevent the unexpected premature breakdowns and mitigate the effects of those failures when they happened. Overall objective of a RCM is to intersect the plant reliability and profitability with pro-active maintenance amount on an optimum point. In this way with RCM an optimized maintenance plan is generated for the concerned equipment. RCM is an indicator to show the compliance with company policies and standards. RCM adjusts the maintenance levels and required resources.

In the RCM approach, there are seven basic questions that are implemented with the main lines;

• What kind of functions and performance standards must be fullfilled for a physical entity in the plant existing operation conditions?

• What kind of obstacles are there to prevent fulfilling those functions?

• What are the basic causes for functional failures?

• What are the consequences if a failure is realized?

• How the failure is happened?

• What actions could be taken for preventing and detecting failures before it happens?

• What have to be done if a pro-active maintenance method is not available?

RCM analysis are made with a team that is selected from the representatives of Operations, Process technologies, Maintenance group, Special engineering units (material, equipment, etc) Basic steps of RCM could be defined as; 1. Equipment information, 2. Prominent failure modes, 3. Failure scenario and criticality, 4. Failure modes features & duties, 5. Economic verification, 6. Grouping maintenance duties and implementation, 7. Analysis, feedback &review 8. Equipment Selection.

For RCM analysis, by considering criticality equipment selection is made. The criticality in here is lack of maintenance for an equipment or maintenance in case of failure and the associated risk for this situation. Risk analysis is made with evaluating the failure results and probability; Health and Working Safety (Fire, toxic wastes, and gases), Economic losses (production loss, maintenance cost), Environmental (Leakages, un-controlled emissions,..), With the criticality evaluation here, Equipment list will be defined for RCM analysis.

Equipment Information: In order to detect the failure mechanism (failure mode) and realizing equipment information is needed. Failure mode consists of a model for defining the failure type for a part of equipment. A failure mode is defined as such; Object + Failure Definition Prominent failure modes: Prominent failure modes must contain all failures that could be progressed in systematic ways. Frequency of these failures must be 20 years or more frequent. In listing the Prominent failure modes, investigating operational conditions of equipment and evaluation of local conditions are important along with the knowledge and experience. In here Pareto approach is utilized for distinguishing the most important few failures from less important many failures. In reality most of the reasons of breakdown one or two failure mode [32].

Reliability-centered maintenance components: The components of RCM program are shown in Figure 11, [33]. This figure showing that RCM program consists of (reactive maintenance, preventive maintenance, condition based maintenance, and proactive maintenance (System Root Cause Failure Analysis (RCFA), Failure mode effect analysis (FMEA), Acceptance testing).

Figure 11.

Components of reliability centered maintenance (RCM) program [33].

Basic steps of RCM are defined as follows:

1. Step1: selection of system and data collection.

2. Step2: definition of system boundary.

3. Step3: description of system and functional block.

4. Step4: system function functional failures.

5. Step5: implementation of failure mode effect analysis.

6. Step6: logic tree diagram.

7. Step7: task selection [34].

Determining the list of the basic system components is one of the first stage in definition of RCM. The criticality analysis requires different kind of data of each component that build up the system. The effect of failure of the system main components may effect system productivity and maintenance cost. The factors effecting selection of critical system are as follows:

1. Mean-time between failures (MTBF).

2. Total maintenance cost.

3. Mean time to repair (MTTR).

4. Availability.

In the implementation of RCM, some of the well known reliability analysis methods are utilized such as Logic Tree Analysis (LTA), Failure Mode Effect Analysis (FMEA), Failure Mode Effect Criticality Analysis (FMECA).

In usual applications in RCM, in order to perform failure modes, effects and criticality analysis (FMEA/FMECA) the identification of the following basic information has to be defined as indicated in Table 3 [35].

Criteria	Weight	Levels
Impact on production P	30%	(3) Very important
		(2) Important
		(1) Normal
Impact on safety S	30%	(3)Very important
		(2) Important
		(1) Normal
Availability of standby A	25%	(3) Without standby
		(2) With stand by and medium availability, and
		(1) With standby and high availability
Equipment value V	15%	(3)High value
		(2) Normal, and
		(1) Low value

Table 3.

Criticality analysis in Reliability Centered Maintenance (RCM) [35].

9. Risk based maintenance

The importance of maintenance function is becoming increased in the recent days in various industrial sectors. Risk Based maintenance (RBM) is a maintenance policy and strategy that combines the advantages of traditional maintenance methods with Risk Based Inspection (RBI) methods and focuses on the mechanical integrity context. RBM helps to select the most effective maintenance strategy depending on the equipment state score and reliability parameters for the whole equipment in the plant. The construction of RBM necessitates the recording of equipment failure and maintenance history in order to define the existing condition of equipments. The data that is collected via RBI, while added into equipment or system history will provide technical inputs for risk analysis. Natural result of this process is to generate the control & maintenance plans for each equipment in order to achieve a reliable and safe plant.

Risk Based Inspection, sheds the lights on the areas of having mechanical integrity that defines the risks that are not defined with other organizational risk analysis methods. In this context RBI becomes a tool for risk analysis and risk management.

To integrate the RBI studies into corporate risk management activities, will be the key factor for the success of risk management program.

To maintain the equipments maintenance and safety in the plant with economic and technical competency, and to consider and manage the problems such as corrosion, erosion, operational and environmental impacts, that might be the causes of breakdowns and stoppages, reducing maintenance costs and down times are the issues that are not fully focused yet by the maintenance sector [36].

Decision-making is required in establishing an optimum maintenance plan, and RBM can play a significant contribution in this stage. But even before RBM was introduced, experienced old personnel in the company had probably made their own decisions on items such as how to obtain optimum result, what inspections are more suitable and what kind of parts/components needs to be prioritized to maintain safety. Presence of such kind of highly skilled old craftsmen could lead to the high level of reliability for devices and or equipment [37] (Figure 12).

Figure 12.

Risk based maintenance framework [38].

Risk-based maintenance schemes favors low maintenance costs while maintaining a high reliability of the grid, because maintenance measures/actions are planned according to the technical condition of the equipment and the consequences in case of failures only. An investigation in this regard of the individual loss of energy due to failures at the ring main unit of Medium Voltage (MV)/Low Voltage (LV) substations is selected to evaluate the importance of the characteristics of the grid and the station [39].

10. Corrective maintenance

10.1 Corrective (Reactive) maintenance (CM)

Corrective maintenance is also called as reactive maintenance. Corrective maintenance is realized upon observing or detecting a breakdown. In most cases Corrective Maintenance is made after the equipment-machinery breakdown-failure or detecting any equipment problem.

Corrective maintenance is usually encountered in the companies that planned maintenance is not regularly adopted or embraced.

Some examples for corrective Maintenance (CM):

• During the normal production flow a sudden breakdown may occur, the equipment is stopped, and urgent intervention may be needed. Such maintenance is called as corrective-reactive maintenance.

• An important wear that could cause further failure could be detected during a routine equipment inspection, in that case a corrective maintenance is realized in order to prevent further damages and production delays the worn out part is replaced.

• In some cases a simple and cheap equipment part is detected as faulty, but replacement is not made until the part is broken down completely.

In most cases for the corrective maintenance (CM) implementation cases usual score for Mean Time to Repair (MTTR) is realized in longer duration than the expection. Furthermore.

During the CM processes root causes for failures are not dealt with hence mean time between failure (MTBF) parameter could result in lower durations than expection. As a result within a short period of maintenance sequence a lot of repeated failures are encountered [40].

Corrective Maintenance Procedure Work Flow is given with (Figure 13) [41].

Figure 13.

Corrective maintenance procedure work flow [41].

10.2 Types of corrective maintenance

Corrective maintenance may be classified under the following categories.

Corrective repair: This kind of equipment repair is made after detecting/observing the failure in order to recover the problem to normal functioning state.

Its operational state.

Basic overhaul: This kind of repair is made to restore the equipment overall parts to their normal functioning state for over-burdened equipment regardless of detecting any specific failure.

11. Method

Salvage: This kind of repair is made usually for worn out aged equipment that is not feasible using after reapair, usually corrective maintenance for that kind equipment is made for selling the equipment with reasonable price.

Servicing: This kind of corrective maintenance may require external expert supplier maintenance intervention such as engine cylinder & piston repairing or replacement.

Rebuild: This is a rather costly CM maintenance operation. If a critical equipment/machine can not be replaced easily in that case rebuilding complete parts and body ciuld be inevitable. In that case by considering original equipment maintenance/service manual re-building the whole parts is attempted regardless of the higher maintenance costs [42].

11.1 Improvement strategies in corrective maintenance effectiveness

In order to improve corrective maintenance performance, corrective maintenance duration has to be reduced. Some of the useful measures to improve corrective maintenance effectiveness is given below:

• Proper design has to be made in order to reach the equipment components easily,

• As less parts as possible to be dismantled to reach the repair location,

• There must be sufficient room to enable operator working properly in the maintenance operator working space,

• Vision convenience has to be provided during corrective maintenance,

• Standard and/or interchangeable parts has to be preferred in the equipment body to enable demounting with various tools, and to reduce corrective maintenance duration,

• Hole lids are to be provided with openable as minimum 180° or to be complete demountable,

• Ergonomic working height to be preferred during corrective maintenance,

• Proper lubrication holes to be provided to enable easy maintenance,

• Detailed proper maintenance work instructions to be provided to detect the faults and failures easily and to react for repair properly [1, 11, 43].

At the end of the chapter preventive and predictive maintenance activities are compared with Table 4 [44].

Maintenance parameters	Preventive maintenance	Predictive maintenance
Maintenance cost	HIGHER – planned interval must be regularly implemented	LOWER-maintenance is implemented just before the breakdown occurs or when needed. US Department of Energy research indicated that predictive maintenance is extremely cost-effective. 25–30% reduction of maintenance costs
Failure detection ability	LOWER- Failure detection can be made during the regular maintenance time	HIGHER-Since high technology equipment and/or sensors are utilized, detection will be mostly precise and in earlier times
Return on investment financial gains	LOWER-	HIGHER-A US Department of Energy research indicated that predictive maintenance is extremely cost-effective. Implementing a predictive maintenance software can deliver notable financial gains with a significant ROI
Number of break-downs	HIGHER	LOWER-US Department of Energy research indicated that 70–75% fewer breakdowns
Reduction of downtime	LOWER	HIGHER-US Department of Energy research indicated that 35–45% downtime decline
Infra structure cost	LOWER	HIGHER-Set up of infrastructure, relevant hardware and software, provision of sensors and training relevant operators will be extra costly [44]

Table 4.

Comparison of preventive and predictive maintenance policies [44].

Source: [44]

12. Results and conclusions

In this chapter basic definitions are made for preventive (planned) maintenance, predictive (condition based) maintenance and assessment of policies for those maintenance types.

Additionally advanced maintenance types (with reliability and risk based maintenance) and reactive (corrective) maintenance types are also defined. And comparison of preventive and predictive maintenance is made in terms of the Maintenance Cost, Failure Detection Ability, Return on Investment Financial Gains, Number of break-downs, Reduction of downtime, Infra structure cost,….etc.

It is concluded that in most of the parameters, predictive maintenance have superior features over the other maintenance policies.

We could argue that with regard to Total Productive Maintenance (TPM) approach predictive maintenance policy is the most effective type. Followed by preventive maintenance and advanced maintenance (reliability based and risk based) policies. Companies willing to adopt Total Quality Management approach should switch from reactive (corrective) maintenance into preventive (planned) and predictive (condition based) maintenance policies.