Maintenance failure mode

Parsons/Honeywell used the PHA for Tooele as a resource document. Eor systems that are the same as or similar to those at Tooele, the level of detail in the PHA is high. Eor systems unique to the Parsons/Honeywell technology provider s package, such as the projectile RWM, the level of detail is low. Maintenance failures are addressed primarily on a generic basis (general work sheets), but a few maintenance failures modes are addressed in process-specific work sheets. 

The failure mode of an equipment item describes the reason for the failure, and is often determined by analysing what causes historic failures in the particular item. This is another good reason for keeping records of the performance of equipment. For example, if it is recognised that a pump typically fails due to worn bearings after 8,000 hours in operation, a maintenance strategy may be adopted which replaces the bearings after 7,000 hours if that pump is a critical item. If a spare pump is available as a back-up, then the policy may be to allow the pump to run to failure, but keep a stock of spare parts to allow a quick repair. 

A suitable maintenance strategy should be developed for equipment by considering the criticality and failure mode, and then applying a mixture of the forms of maintenance described above. In particular, the long-term cost of maintenance of an item of equipment should be estimated over the whole life of the project and combined with its capital cost to select both the type of equipment and form of maintenance which gives the best full lifecycle cost on a discounted basis), while of course meeting the technical, safety and environmental specifications. 

If a plant had periodic inspections, the impeller corrosion in No. 5 might be detected before it became a significant problem, thereby altering the failure mode from catastrophic to a degraded or an incipient failure. In a plant with routine maintenance, it is possible that Nos. 1 and 5 may be eliminated completely by routine seal and impeller changes. 

Data from an existing collection system were analyzed for failure modes and distribution. The results of Pareto analyses indicate the principal causes of failure. A few values of mean times to maintenance action (MTBM) are given for ethylene plant pumps (85 electric driven centrifugal pumps over a 19-month period), and ethylbenzene-styrene monomer plant equipment from 10 months data 4 gas compressors, 3 screw conveyors, 121 pumps, and 235 other items... 

Repair and maintenance records were analyzed to determine failure rates and distribution of failure modes. Preliminary findings are reported which include the Weibull distribution characteristics. Failure mode distributions are approximate. Overall mean-time-between-failure is given for the kiln, leach tank, screwfeeder, tank pump, tank gearbox, and kiln gearbox. The study was confined to an analysis of unscheduled repairs and failures. 

The main objective of the In-Plant Reliability Data System (IPRDS) was to develop a comprehensive and component-specific data base for PRA and other component reliability-related statistical analysis. Data base personnel visited selected plants and copied all the plant maintenance wor)c requests. They also gathered plant equipment lists and plant drawings and in some cases interviewed plant personnel for Information on component populations and duty cycles. Subsequently, the maintenance records were screened to separate out the cases of corrective maintenance applying to particular components these were reviewed to determine such things as failure modes, severity, and, if possible, failure cause. The data from these reports were encoded into a computerized data base. 

For both failure modes, terminations caused by conditions other than the DG and its immediate support systems were not counted. Conditions that invalidated tests or demands for this study Include any operating errors that would not have prevented the DG from being restarted and brought to load in a few minutes without corrective maintenance incorrect trip signals that would not have been operative in the emergency mode and minor water or oil lea)cs that would not have precluded operation of the DG in an emergency. 

Other reports used within facilities record failures of particular interest because of failure mode and system or equipment affected. Some facilities may issue special reports when the plant experiences a shutdown (outage report) or when the occurrence is sufficiently unique or troublesome to warrant further investigation (unusual event report). In general, these reports can be characterized by their relatively restrictive focus (when compared to the maintenance records) and their smaller number. 

Sometimes it is necessary to review the narrative in raw data records to determine whether a failure has occurred, to establish failure modes and severities, and to see if a record is a duplicate or new failure. Often, the narrative section is the only way the data analyst can determine if the document, especially a work order, is for a legitimate failure, routine maintenance, or a specified test. 

Spectrographic analysis allows accurate, rapid measurements of many of the elements present in lubricating oil. These elements are generally classified as wear metals, contaminates, or additives. Some elements can be listed in more than one of these classifications. Standard lubricating oil analysis does not attempt to determine the specific failure modes of developing machine-train problems. Therefore, additional techniques must be used as part of a comprehensive predictive maintenance program. 

A clear understanding of the operating characteristics and failure modes will provide the answer to which predictive maintenance method should be used. 

There are defect limits that are associated with random failure modes. For example, if there is a leak from a mechanical seal on a pump, where do we decide that the leakage is excessive and requires immediate maintenance Vibration analysis severity levels are also typical examples of when do we have severe enough conditions to warrant equipment shutdown and overhaul. In such circumstances, the defect limit is dependent upon individual subjective judgment. 

Standard life is described as the average lifetime that is acceptable to any plant failure analyst or troubleshooter. Therefore, if we arrive at defect limits in machinery within the maintenance program, we have also reached the standard life of all the failure modes in the plant. Do we now ... 

Experience shows that some machines have more frequent failures than do others. Obviously, different failure modes have different frequencies of occurrence. This is usually described as mean-time-between-failure (MTBF) and expresses the probability of machinery failure and breakdown events as a function of time. This is of particular interest to the maintenance failure analyst and troubleshooter who have to grapple with the realization that some machinery failure modes appear slowly and predictably whilst others occur randomly and unpredictably. In most cases, both types of failures have been encountered. 

While preventive maintenance is concerned with regularly testing, and reconditioning equipment to prevent failures in service and premature deterioration, it follows that predictive maintenance procedures are concerned with the ability to predict when the equipment will fail and then developing schedules to implement timely repairs. Predictive maintenance does not imply that with the use of these techniques, failure modes in equipment can be prevented rather, it suggests that the occurrence of failure can be predicted and thus planned for. An appropriate example would be the inspection and change of a major compressor face-type oil seal where random heat checking (FM) has been observed over the years. 

The existence of breakdown maintenance unquestionable. The maintenance activity is necessary to restore machinery equipment back to service after failure modes developed that were ... 

The third step is to proceed to the component level because more than likely, the cause will be uncovered here. Table 62.2 shows machinery component failure modes commonly encountered in machinery failure analysis together with suggested standard life values, Weibull indices (/J), and responsiveness to preventable or predictable maintenance strategies. Referring to this table will help decision-making. To assist the analyst in documenting bad actor failure modes on the component level. 

Failure Mode Weibull Standard Maintenance Strategy... 

Safety and loss performance in the chemical industry is the result of the interaction of plant design, construction and maintenance with production processes, and trained people applying a well-developed operating discipline. An accident leading to personal injury, property damage, or product loss invariably is the result of the failure of one or more of these elements. Each factor involved in chemical production—equipment, process, product, and people—may be subject to a variety of failure modes which may lead to accidents. 

The Failure Mode and Effect Analysis (FMEA) is based on the systematic analysis of failure modes for each element of a system, by defining the failure mode and the consequences of this failure on the integrity of that system. It was first used in the 1960s in the field of aeronautics for the analysis of the safety of aircraft [15]. It is required by regulations in the USA and France for aircraft safety. It allows assessing the effects of each failure mode of a system s components and identifying the failure modes that may have a critical impact on the operability safety and maintenance of the system. It proceeds in four steps ... 

A new methodology designed to optimize both the planning of preventive maintenance and the amount of resources needed to perform maintenance in a process plant is presented. The methodology is based on the use of a Montecarlo simulation to evaluate the expected cost of maintenance as well as the expected economic loss, an economical indicator for maintenance performance. The Montecarlo simulation describes different failure modes of equipment and uses the prioritization of maintenance supplied, the availability of labour and spare parts. A Genetic algorithm is used for optimisation. The well-known Tennessee Eastman Plant problem is used to illustrate the results. 

To ameliorate all the aforementioned shortcomings, we developed a new maintenance model based on the use of Monte Carlo simulation. The model incorporates three practical issues that have not been considered in previous work i) different failure modes of equipment, ii) ranking of equipments according to the consequences of failure, iii) labor resource constraints and material resource constraints. The maintenance model, which was developed by Nguyen et al. (2008) is integrated here with a GA optimization to optimize the PM frequency. 

The cost of corrective maintenance, the repair time and the economic losses are determined corresponding to the type of failure modes identified. 

---