Condition Based Maintenance Optimization: Solution Implementation in Practice - Coursework Example

? Condition Based Maintenance Optimization Solution Implementation in Practice and Condition Based Maintenance Methods This text analyses the move towards CBM (condition based monitoring) as well as its major methods, advantages, disadvantages and implementation in the practical arena. Contents Abstract 2 Contents 2 Maintenance Methods 3 Corrective Maintenance 3 Preventive Maintenance 4 RCM Based Methods 4 Introduction to CBM 5 Methods Utilised in CBM 6 Advantages of CBM 7 Disadvantages of CBM 8 Implementing CBM 9 Rotating Mechanical System 9 Pumps and Compressors 9 Rolling Element Bearings 10 Pipelines and Static Equipment 11 Induction Motors and Gearboxes 11 Heat Exchangers 12 Conclusion 12 Bibliography 14 Maintenance Methods Maintenance refers to various techniques that are utilised to ensure that equipment and machinery is available for work. The prime objective is to maximise the uptime of machines and other equipment. Maintenance techniques have evolved over the years and have been adapted to various situations. Some programs are customised to maximise uptime. Others are designed to minimise the cost of spares while others are optimised to minimise human interference. The term maintenance is taken as the combination of technical, administrative and managerial actions that are taken throughout the lifecycle of equipment and machines. Broadly maintenance can be classified into two categories which are corrective maintenance and preventive maintenance. (European Federation of National Maintenace Societies, 2011) Corrective Maintenance When a machine or equipment breaks down and maintenance is then carried out, it is known as corrective maintenance. Generally corrective maintenance has been the most extensive maintenance technique that has been applied throughout the years. Previously corrective maintenance was the preferred technique. It was largely held that maintenance should be carried out when the parts of a machine or equipment complete their effective lifetime. This belief emerged as a solution to optimise the life of parts. The replacement of parts before need arose was not a highly preferred technique. However, as industrial enterprises expanded and competition arose, the need for maximising uptime increased. This was only possible if parts were replaced before break down as the failure of one part would endanger the other workings parts. The development of HSE (health, safety and environment) requirements meant that machine and equipment failure was not tolerable anymore. In case that machine and equipment failed, there were significant chances for the discharge of lethal products such as ammonia, cyclo hexane, lead based compounds etc. (Lockheed Martin, 2011) Eventually, maintenance enterprises moved over to the practice of replacing defective parts before failure. These methods are better known as preventive maintenance. Preventive Maintenance The contention behind preventive maintenance is to avoid failure in the first place. This is necessitated by objectives that define the maximisation of production times and the minimisation of safety incidents. Most equipment and machines are installed in such a complicated manner that monitoring all equipment and machinery at all times in not practicable. Therefore, certain intervals for machinery and equipment monitoring must be defined. These techniques tend to be designed in different ways that maximise different offerings from maintenance. While preventive maintenance strategies offer benefit in the longer run but migration to preventive maintenance systems is often troublesome in the shorter run. Often the problem lies with maintenance costs and human factors. Moreover, preventive maintenance demands that machines and equipment be opened up even if troublesome symptoms are not present. This often reduces the effective lifetimes of machines as certain parts need to be replaced with every maintenance operation. Preventive maintenance strategies display certain disadvantages if implemented purely. These disadvantages can be covered up by utilising a mix of both corrective and preventive maintenance strategies that are optimised to reduce cost and effort. This is better known as RCM (reliability centred maintenance). A RCM analysis is crucial to implement preventive maintenance strategies effectively. RCM analysis ensures that maintenance is optimised as per cost and other factors. (MTain, 2009) RCM Based Methods Reliability centred maintenance can be considered to be an optimal mix of reactive maintenance, interval or time based maintenance, condition based maintenance and proactive maintenance strategies. If these maintenance strategies are implemented independent of each other, there are grey areas left that dissuade optimisation. This problem can be dealt with by integrating the various maintenance strategies together so that their advantages and strengths can be maximised by ensuring equipment reliability while minimising life cycle costs at the same time. Reliability centred maintenance can be seen to be a combination of the elements listed below. Moreover, factors responsible for the implementation of each strategy are listed as well. Reliability Centred Maintenance Reactive Maintenance small items non critical items inconsequential items equipment unlikely to fail redundant equipment Interval Based Maintenance equipment subject to wear and tear consumable replacement items equipment where failure patterns are known Condition Based Maintenance items with random failure patterns items not subjected to wear failures induced by preventive maintenance Proactive Maintenance maintenance conducted after RFCA (Root Cause Failure Analysis) age exploration items FMEA (Failure Modes and Effects Analysis) acceptance testing An effective RCM program is composed of the various maintenance strategies listed above. (Pride, Reliability-Centered Maintenance (RCM), 2010) This text explores the various condition based maintenance strategies and methods both theoretically and practically. A critical analysis will be presented of condition based monitoring in the passages that follow. Introduction to CBM In generalised texts the terms “predictive maintenance” and “condition based maintenance” after often used in place of each other. However, there are certain subtle differences between predictive and condition based maintenance. Predictive programs rely on data gathering that is periodic and often manual. On the other hand, condition based maintenance (CBM) relies on data that is gathered continuously and is analysed in real time. This has far reaching consequences as CBM programs rely on sensors that gather data in real time. Along with sensors, the CBM programs rely on transmitters to move sensory information. This information is transmitted and stored on computing platforms and means must be made available to analyse the captured information. CBM has been shown to be more suited to critical facilities such as process industries like power plants, refineries, fertilizer plants, gas compression stations etc. Methods Utilised in CBM CBM can be initiated and implemented only if a degraded state of the particular system is already defined. The various characteristics of the system are monitored through various techniques which describe the state of the system in real time. As soon as the degradation characteristics cross a specific defined threshold, appropriate maintenance actions are triggered. The degradation characteristics that are defined should effectively relate the system’s state (or the state of a particular component) to (Rausch, 2008): the remaining useful life of the system; decision on the failure threshold; feasibility to implement condition monitoring techniques and technology. The most common condition based monitoring methods in practice today at most industrial facilities include (Neelamkavil, Condition based Maintenance Management in Critical Facilities, 2010): Vibration Monitoring: This is utilised to detect misalignment, wear, fatigue as well as loose assemblies for rotary machines which include engines, motors, bearings, pumps, gear boxes and the like. Vibration data is collected over reasonable amounts of time and these are compared to previously established baselines. If the degradation characteristics threshold is crossed then alarms are actuated. Based on these alarms, maintenance personnel are alerted. Process Parameter Monitoring: This involves the use of methods to track a variety of operational characteristics. Typical operational characteristics include process efficiency, electrical current, system temperature and pressure etc. which can directly be related to the health of the system. Thermography: This method captures infrared emissions of a component in order to determine if the operating temperatures and other related conditions fluctuate outside normal operating conditions. Any abnormal temperature changes whether positive or negative can be taken as symptoms of expected failures. Tribology: This is the study of the effects of friction displayed between two surfaces in contact with each other. Frictions causes the generation of particulate matter after surface wear. These particles can be monitored using wear particulate analysis. Moreover, lubrication analysis can also be performed to ascertain appropriate lubricant change frequency in a system. Visual Inspection: This method is the easiest to implement and simple visual analysis is utilised. Defects such as loose components or assemblies, structural aberrations such as sways and cracks etc. are identified and appropriate action is taken. NDT (Non Destructive Testing): This method employs classical NDT techniques such as dye penetrate tests, ultrasonic testing, radiographic analysis etc. to determine aberrations in materials that are not visually identifiable. Wear and tear especially related to piping and structures can be identified with relative ease. Ultrasonic techniques are more frequently applied to vessels, critical piping as well as shafts to ensure that they are sturdy enough for service. Advantages of CBM There is little doubt the CBM techniques ensure that maintenance uncertainty is reduced substantially. Equipment condition monitoring in real time ensures that active analysis can be conducted and remedial action can be appropriately applied. The monitoring process in CBM involves collecting and analysing relevant equipment parameters so that the system’s state can be ascertained. Any deviations and aberrations can easily be extracted from the analysis. These equipment parameters represent the set of characteristics that reflect the state of the equipment. Any abnormality that is detected can be used to indicate a failure. On the one hand this ensured prompt action which reduces the chances for damage to equipment and on the other hand this approach ensures that any future equipment deterioration can be adequately predicted. This allows for substantial time to be available for fault rectification to occur. The application of a CBM based approach allows maintenance costs to be reduced and optimised in the longer run. (Neelamkavil, 2010) One of the best advantages offered through CBM strategy is that workforce is utilised efficiently. The condition monitoring data is routed directly to the appropriate workforce and this aids in concentrating maintenance activities where and when required. The development of ICTs (information and communication technologies) allows the larger corporations to develop central maintenance hubs that are actively monitoring data from remote facilities. The data is analysed and appropriate personnel in equipment vicinity are alarmed. Moreover, trouble shooting and diagnosis can be carried out remotely too. This lowers workforce costs as a small team can be used to accomplish maintenance analysis and implementation. (Campos, 1-20) This advantage becomes all the more apparent in applications such as offshore oil and gas production facilities where personnel cannot be located in deep water regions but can be moved as required. (Web, 1981) Disadvantages of CBM The initial implementation of CBM methods is often dissuaded by costs. Data collection is the cornerstone of any CBM implementation. However, the installation of myriad sensors presents a new challenge as well as large costs. Similarly, the equipment required to transmit and analyse the generated data is often expensive. The application of CBM must often be justified because the large costs must be offset within reasonable periods of time. Although data in real time can enhance the total confidence level in any model that is utilised but only a few operators can reliably let components run to failure. However, if components are replaced too early on then replacement costs may be too significant. Finding a balance between these two states using CBM is often tricky and involves major risks that could have financial as well as safety related risks attached. (Scarf, 2007) Data collection is in itself a new cost too although most implementations fail to take appropriate notice. Installation costs aside, the regular functioning of installed components for data collection must also be guaranteed. This involves both maintenance and replacements of installed sensors as well as transmitters which represent the secondary costs in a CBM system. Again the cost of data collection must be balanced against the gains achieved from data collection. Optimisation to such a state represents another tedious balancing act that may not always be optimised. (Rao & Naikan, 2006) Furthermore, maintenance resulting from CBM must be dynamically scheduled. This is necessary as the execution times of certain activities are being updated on a frequent basis as more and more conditional monitoring information is collected. This necessitates the implementation of appropriate cost benefit analysis along with any CBM strategy. (Grall, Berenguer, & Dieulle, 2002) Implementing CBM Various kinds of equipment are utilised in industrial, military and other applications where CBM or other forms of maintenance are required. Typical components include rotary and stationary equipment. The implementation of CBM to individual components in these systems is described below to bring out practical connotations of CBM implementation. Rotating Mechanical System Most major processing and military applications feature rotating mechanical systems. The implementation of CBM techniques to these systems is often carried out in the shape of vibration monitoring. This implementation aids in identifying problems such as misalignment, metal fatigue, unbalanced forces, improper lubrication of ball bearings as well as cracks or other defects that appear in welded as well as constructed parts. The tendency of rolling element bearings to seize can also be predicted reliably using CBM techniques as rolling element bearings reveal tell tale signs of failure during monitoring. Two approaches can be utilised. Expert systems can be coalesced with vibration analysis to prevent losses by timely detection. On the other hand, artificial neural network classifiers can be utilised for condition monitoring of rotating mechanical systems. Pumps and Compressors Myriad pumping devices are utilised in critical applications for the bulk movement as well as controlled delivery of liquids. Various types of pumps installed include rotary, reciprocating, centrifugal and diaphragm pumps. CBM has been utilised in the industry with substantial success over the years and continues to be an active research area too. Most CBM systems have begun to classify pump based problems into two broad categories: Hydraulic Problems: failure of the pump to deliver liquid, delivery of insufficient capacity, losing priming at start up as well as the development of insufficient pressure Mechanical Problems: consuming excessive power, mechanical difficulties at seat chambers and / or bearings, noise, vibration or breakage Fatigue has been identified as a common cause for pump failure. Vibration monitoring has proved to be highly suitable for CBM monitoring of pumps. This is because most pumps have a large number of integrated rotating parts that tend to amplify movement as soon as a fault develops. Another recent innovation is the utilisation of ultrasonic sensors. Ultrasonic measurements are based on the acoustic emissions from high pressure process pumps. Condition monitoring sensors are permanently fixed and are well suited for use in corrosive, inaccessible settings, submerged applications as well as hazardous areas. (Hansford, 2002) Pressure is the most common process parameter that is actively monitored because it varies in relation to the requirements of the operation or process. Compressors are used to enhance gas pressures and typically two types of compressors are utilised that are the centrifugal and reciprocating compressors. Techniques utilised for pumps can be suitably used with additional monitoring of pressures. In case the compressor surges or begins to leak, the first signs appear in the pressures of the compression system. CBM can be used with ease and reliability over here as well. (Carnero, 2005) Rolling Element Bearings Rolling element bearings are pervasively distributed throughout rotary mechanical systems. Their load carrying capacity and low friction characteristics as well as simple operation warrants their use. Bearings come in a great variety of sizes and range from the small to the very large. CBM presents a number of choices to monitor the behaviour of rolling element bearings. Vibration monitoring has been employed widely within the industry to monitor rolling element bearings. Various techniques used in the industry. Electrostatic sensors may be utilised along with vibration measurement, lubricant temperature measurement as well as the use of eddy current testing to sense debris within lubrication recirculation systems. An alternative technique is to employ electrostatic wear site sensors that identify charge through surface wear. Vibration accelerometers are also used as well as thermocouples as well as inductive and ferromagnetic particle counters. A newly developed technique is to use time frequency processing which is better known as “basis pursuit”. Features are extracted from signals collected from fault rolling element bearings. The integration of the techniques presented above can provide a much clearer picture on which maintenance decisions can be based. (Yang & al., 2005) Pipelines and Static Equipment Both pipelines and other stationary equipment represent a significant amount of investment within any industrial enterprise where critical maintenance techniques are desired. Pipelines are used to transport liquids and gases in any processing facilities. Vessels are utilised to process and store various materials in a processing facility. However, most maintenance programs pay scant attention to both pipelines and other stationary equipment. CBM presents an unparalleled methodology to actively monitor vessels and pipelines that are subject to corrosive and hazardous service as well as subject to high temperatures and pressures. New fibre optic sensors have been developed that are capable of measuring both temperature as well as strain in a pipeline, vessel or tank. (Yan & Chyan, 2010) CBM systems provide detailed structural health information that can be critical to providing early signs of degradation such as corrosion, formation of hydrates etc. that could lead to disastrous failures such as leakage or explosions. Ultrasonic condition monitoring is widely utilised to detect failures in pipelines. (Bandes, 2009)However, it is not feasible enough to place sensors far and wide throughout the entire network given the large associated costs. Therefore, it is advisable to install condition monitoring sensors in places where the greatest rates of corrosion are historically detected or to place sensors where a failure could yield disastrous consequences. Induction Motors and Gearboxes Induction motors represent a core installation in any industry. The versatility and rugged nature of induction motors makes them a natural choice. Popular implementation of CBM techniques in the industry depend on monitoring the current signature of induction motors. These techniques are known collectively as MCSA (motor current signature analysis). Any variation in the current drawn can serve as an important indicator of problems that may have arisen. (Rodriguez & Negrea, 2008) Gear boxes are widely used throughout the industry to vary both speed and torque. In any large train, the gear boxes have an important role to play. They not only deal with small misalignments between driver and driven machines but also help to provide axial displacement. Typically gear boxes are monitored for defects using vibration monitoring as well as debris analysis. Vibration techniques may also be utilised with non intrusive methods for detection of variations in gear loads that depend on MCSA. (Ebersbach & Peng, 2006) Heat Exchangers A leading component of all stationary equipment within a process industry is the heat exchangers. Appropriate heat transfer ensures that the process is carried out effectively and efficiently. Heat recovery greatly aids in reducing the final price of the product and affects profitability directly. Fouling is a major problem experienced by heat exchangers. Its consequences are both thermal and hydraulic in nature. Until a few years ago, there were limited methods to investigate fouling such as boroscopy for larger heat exchangers and disassembly of smaller heat exchangers. Generally it is difficult and rather expensive to mount sensors into the nooks and crannies of heat exchangers for condition monitoring. Therefore, generally the inlet and outlet temperatures of heat exchangers are monitored. The stage to stage inter pass temperatures can also be monitored such as for heat exchangers on large process compressors for example process air compressors. Fouling through process deposition as well as corrosion can be detected easily. However, long term baselines must exist for a reliable analysis to take effect. (Sikos & Klemes, 2010) Conclusion CBM can be seen as a maintenance technique that is maturing well with the passage of time. CBM offers real time data to analyse and interpret maintenance situations for optimal benefit. However, the large volumes of data as well as the associated costs of generating, storing and analysing such data present unique costs both as primary and secondary costs. Major CBM techniques include vibration monitoring, process parameters monitoring, thermography, tribology, visual inspection and NDT. These techniques have proven successful in detecting problems before hand thus enabling operators to deal with the situation more efficiently. Furthermore, CBM offers options to reduce workforce numbers as well as remote monitoring. This tends to cut down costs for maintenance in the shorter term. However, the installation of sensors, transmitters and data storage and analysis equipment presents long term costs that must be justified judiciously. Overall, CBM can be used to deal effectively with items that present randomised failure patterns, items that are not subject to regular wear and tear and to items that induce failures as a result of overdoing preventive maintenance. At this point in time, CBM methods are too expensive to implement system wide especially for older installations that require extensive modifications to place sensors and transmitters. Moreover, operators need to be better trained to deal with large volumes of data as well as cost benefit analysis and dynamic maintenance scheduling. However, as technology becomes cheaper, there is little doubt that CBM will expand its role beyond critical industrial applications alone. Bibliography Bandes, A. (2009). Ultrasonic Condition Monitoring. Retrieved July 23, 2011, from UE Systems: http://www.uesystems.com Campos, J. (1-20). Development in the application of ICT in condition monitoring and maintenance. Computers in Industry 60(1) , 2009. Carnero, M. C. (2005). Selection of diagnostic techniques and instrumentation in a predictive maintenance program: a case study. Decision Support System 38(4) , 539-555. Ebersbach, S., & Peng, Z. e. (2006). The investigation of the condition and faults of a spur gearbox using vibration and wear debris analysis techniques. Wear 260(1-2) , 16-24. European Federation of National Maintenace Societies. (2011). What does EFNMS stand for? Retrieved July 20, 2011, from European Federation of National Maintenace Societies: http://www.efnms.org/What-EFNMS-stands-for/m13l2/What-EFNMS-stands-for.html Grall, A., Berenguer, C., & Dieulle, L. (2002). A condition-based maintenance policy forstochastically deteriorating systems. Reliability Engineering & System Safety 76(2) , 167-180. Hansford, C. (2002). Condition monitoring: Combating down time with vibration sensors. World Pumps 48 , 50-53. Lockheed Martin. (2011). KELLY AVIATION CENTER SIGNS SECOND AGREEMENT WITH OC-ALC. Retrieved July 20, 2011, from Lockheed Martin: http://www.lockheedmartin.com/news/press_releases/2008/0122ae_u2sanantonio.html MTain. (2009). Reliability Centred Maintenance Analysis. Retrieved July 20, 2011, from MTain: http://www.mtain.com/logistics/logrcm.htm Neelamkavil, J. (2010). Condition based Maintenance Management in Critical Facilities. Institute for Research in Construction. Pride, A. (2010, June 17). Reliability-Centered Maintenance (RCM). Retrieved July 20, 2011, from Whole Building Design Program: http://www.wbdg.org/resources/rcm.php Rao, P. N., & Naikan, V. A. (2006). An Optimization Methodology for Condition Based Minimal and Major Preventive Maintenance. Economic Quality Control 21(1) , 127-141. Rausch, M. T. (2008). Condition based Maintenance of a single system under Spare Part Inventory Constraints. Wichita: Wichita State University. Rodriguez, P. V., & Negrea, M. e. (2008). A simplified scheme for induction motor condition monitoring. Mechanical Systems and Signal Processing 22(5) , 1216-1236. Scarf, P. A. (2007). A Framework for Condition Monitoring and Condition Based Maintenance. Quality Technology and Quantitative Management 4(2) , 301-312. Sikos, L., & Klemes, J. (2010). Reliability, availability and maintenance and optimisation of heat exchanger networks. Applied Thermal Engineering 30(1) , 63-69. Web, G. D. (1981). Inspection and Repair of oil and gas production installations in deep water . Ocean Management 7(1-4) , 313-326. Yan, S. Z., & Chyan, L. S. (2010). Performance enhancement of BOTDR fibre optic sensor for oil and gas pipeline monitoring. Optical Fibre Technology 16(2) , 100-109. Yang, H. J., & al., M. e. (2005). Fault diagnosis of rolling element bearings using basis pursuit. Mechanical Systems and Signal Processing 19(2) , 341-356. Read More