Handbook of Reliability, Availability, Maintainability and Safety in Engineering Design - Part 33 pdf

4.2 Theoretical Overview of Availability and Maintainability in Engineering Design 303 of systems availability. Maintainability, on the other hand, is similar to reliability in that both relate the occurrence o f a single type of event over time. It is thus neces- sary to consider in closer detail the various de finitions of availability (Conlon et al. 1982). Inherent availability can be defined as “the p rediction of expected system per- formance or system operability over a period which includes the predicted system operating time and the predicted corrective maintenance down time”. Achieved availability can be defined as “the assessment of system operability or equipment usage in a simulated environment, over a period which includes its predicted operating time and active maintenance down time”. Operational availability can be defined as “the evaluation of potential equip- ment usage in its intended operationalenvironment, over a period which includes its predicted operating time, standby time, and active and delayed maintenance down time”. These definitions indicate that the availability of an item o f equipment is con- cerned either with expected system performance over a period of expected opera- tional time, or with equipment usage over a period of expected operational time, and that the expected utilisation of the item of equipment is its expected usage over an accountable period of total time inclusive of downtime. This aspect of usage over an accountable period relates the concepts of availability to utilisation of an item of equipment, where the accountable period is a measure of the ratio of the actual input to the standard input during the operational time of successful system perfor- mance. The process measure of operational input is thus included in the concept of availability. By grouping selected availability techniques into these three differ- ent qualitative definitions, it can be readily discern ed wh ich techniques, relating to each of the three terms, can be logically applied in the different stages of the design process, either independently or in conjunction with reliability and maintainability analysis. As with reliability prediction, the techniques for predicting inherent availabil- ity would be more appropriate during conceptual or preliminary design,whenal- ternative systems in their gener al context are being identified in preliminary block diagrams, such as first-run process flow diagrams (PFDs), and estimates of the prob- ability of successful performance or operation of alternative designs are necessary. Techniques for the assessment of achieved availability would be more appro- priate during schematic design, when the PFDs are frozen, process functions de- fined with relevant specifications relating to specific process performance criteria, and process availability assessed according to expected equipment usage over an ac- countable period of operating time, inclusive of predicted active maintenance down- time. Techniques for the evaluation of operational availability would be more appro- priate during detail design, when components of equipment detailed in pipe and instrument drawings(P&IDs) are being specified according to equipment design cri- teria, and equipment reliability, availability an d maintainability ar e evaluated from a determination of the frequencies with which failures occur over a predicted period of operating time, based on known component failurerates, and the frequencies with 304 4 Availability and Maintainability in Engineering Design which component failures are repaired during active corrective maintenance down- time. This must also take into account preventive maintenance d owntime, as well as delayed maintenance downtime. Maintainability analysis is a further method of determining the integrity of engi- neering design by considering all the relevant maintainability characteristics of the system and its equipment. This would include an analysis of the following (MIL- STD-470A; MIL-STD-471A): • Quantitative characteristics • Physical characteristics. Quantitative characteristics considered for a system design are its specific main- tainability pe rformance characteristics, which inclu de aspects such as mean time to repair, maximum time to repair, built-in-test and health status and monitoring: • Mean time to repair (MTTR ): This is calculated by considering the times needed to implement the corrective maintenance and preventive maintenance tasks for each level of maintenance ap- propr iate to the respective systems hierarchical levels. • Maximum time to repair: This is an important part of the qua ntitative characteristics of maintainability performance, in that it gives an indication of the ‘worst-case’ scenario. • Built-in-test (BIT): The establishment of a BIT capability is important. For example, the principal means of fault detection and isolation at the component level requires the use of self-diagnostics or built-in-testing. This capability, in terms of its effectiveness, may need to be quantified. • Health status and monitoring (HSM): Incorporated into the design of the system could be a HSM capability.This could be a relatively simple concept, such as monitoring the temperature of the shaft of a turbine to safeguard against the main bearings overheating. Other HSM sys- tems may employ a multitude of sensors, such as strain gauges, thermal sensors, accelerometers, etc., to measure electrical and mechanical stresses on a particular component of the assembly or system. Physical characteristics take into consideration issues and characteristics that will accommodate ease of maintenance, such as ergonomics and visibility, testability, accessibility and interchangeability: • Ergonomics: Ergonomics addresses the physical characteristics of concern to the maintenance function. This could range from the weight of components and required lifting points to the clearancebetweenelectricalconnectors,to the overalldesign config- uration of assemblies and componentsfor maximum visibility during inspections and maintenance. Visibility is an element of maintainability design that allows the m aintenance function visual access to assemblies and components for ease of maintenance action. Even short-duration tasks can increase downtime if the 4.2 Theoretical Overview of Availability and Maintainability in Engineering Design 305 component is blocked from view. Designing for visibility greatly reduces main- tenance times. Human engineering design criteria, as well as human engineering requirements, are well establishe d for military systems and equipment, as pre- sented in the different military standards for systems, equipment and facilities (MIL-STD-1472D; MIL-STD-46855B). • Testability: Testability is a measure of the ability to detect system faults and to isolate these at the lowest replaceable component. The speed with which faults are d iagnosed can greatly influence downtime and maintenance costs. As technology advances continue to increase the capability and complexity of systems, the use of auto- matic diagnostics as a means of fault detection, isolation and recovery (FDIR) substantially reduces the need for highly trained maintenance personnel and can decrease maintenance costs by reducing the need to replace components. FDIR systems include both internal diagnostic systems, referred to as built-in-test(BIT) or built-in-test-equipment (BITE), and external diagnostic systems, referred to as automatic test equipment (ATE), or offline test equipment. The equipment are used as part of a reduced support system, all of which will minimise downtime and cost over the operational life cycle. • Test point: Test points must be interfaced with the testability engineering effort. A system may require some manual diagnostic interaction, where specific test points will be required for fault diagnostic and isolation purposes. • Test equipment: Test equipment assessment is of how test instrumentation would interface with the process system or equipment. • Accessibility: Accessibility is perhaps the most important attribute. With complex integration of systems, the design o f a single system must avoid the need to remove another system’s equip ment to gain access to a failed item. Furthermore, the ability to permit the use of standard hand tools must be observed throughout. Accessi- bility is the ease with which an item can be accessed during maintenance, and can greatly impact maintenance times if not inherent in the design, especially on systems where in-process maintenance is required. When accessibility is poor, other failures are often caused by isolation/disconnection/removal and installa- tion of other items that might hamper access, causing rework. Accessibility of all replaceable, maintainable items will provide time and energy savings. • Interchangeability: Interchangeability refers to the ability and ease with which a component can be replaced with a similar component without excessive time or undue retrofit or recalibration. This flexibility in design reduces the number of maintenance pro- cedures and, consequently, reduces maintenance costs. Interchangeab ility also allows for system expansion with minimal associated costs, due to the use of standard or common end-items. 306 4 Availability and Maintainability in Engineering Design Maintainability ha s true design characteristics. Attempts to improve the inherent maintainability of a product/item after the design is frozen are usually expensive, in- efficient and ineffective, as demonstrated so often in engineering installations when the first main ten ance effort requires the use of a cutting torch to access the item requiring replacement. In the application of maintainability analysis, there are basically two approaches to predicting the mean time to rep air ( MTTR). The first is a work study method that analyses each repair task into definable work elements. This requires an extensive databank of average times for a wide range of repair tasks for a particular type of equipment. In the absence of sufficient data o f average repair times, the work study method of comparative estimation is applied, whereby repair times are simulated from failures of similar types of equipment. The second app roach is empirical and involvesrating a number of maintainability factors against a checklist. The resulting maintainability scores are converted into an estimated MTTR by means of a nomograph obtained by regression analysis of many different repair times. This second approach is described in detail in the USA military handbook titled ‘Maintainability prediction’ (MIL-HDBK-472), of which the referenced Procedure 3 is considered to be appropriate for general engineering application. In this procedure, the predicted repair time for each task is arrived at by considering a checklist o f maintainability criteria, and by scoring points for each criterion. The score for each criterion increases with the d egree of conformity to an expected standard. The criteria in the checklist are grouped under three head- ings: • Physical design factors • Design dictates—facilities • Design dictates—maintenance skills. The points scored under each heading are appropriately weighted, and relate to the predicted repair time by means of a regression equation that is presented in the form of a nomograph. The checklist is used for accumulating the scores of all the various repair tasks of a particular item, and is reproduced in part below (MIL-HDBK-472). Scoring will app ly to main tainability design concepts for ease o f maintenance. This is concerned with design for maintainability criteria such as visual and manipula- tive actions, which would normally precede maintenance actions. The regression equation to calculate the p redicted d owntime is of the form: Mct = antilog(3.54651−0.02512A−0.03055B−0.01093C) where Mct is corrective maintenance time, and A, B and C are scores o f the relevant checklists. 4.2 Theoretical Overview of Availability and Maintainability in Engineering Design 307 CHECKLIST—MIL 472 PROCEDURE 3 Checklist A—scoring physical design factors: 1. Access (external) 2. Latches and fasteners (external) 3. Latches and fasteners (internal) 4. Access (internal) 5. Packaging 6. Units/parts (failed) 7. Visual displays 8. Fault and operation indicators 9. Te st points availability 10. Test points identification 11. Labelling 12. Adjustments 13. Testing in circuit 14. Protective devices 15. Safety personnel. Checklist B—scoring design dictates—facilities: 1. External test equipment 2. Connectors 3. Jigs and fixtures 4. Visual contact 5. Assistance operations 6. Assistance technical 7. Assistance supervisory. Checklist C—scoring design dictates—maintenance skills: 1. Arm-leg-back strength 2. Endurance and energy 3. Eye-hand coordination 4. Visual requirements 5. Logic application 6. Memory retention 7. Planning 8. Precision 9. Patience 10. Initiative. The nomograph given in Fig. 4.2 includes scales against which scores for the phys- ical design factors and the design dictates are marked. 308 4 Availability and Maintainability in Engineering Design Fig. 4.2 Regression equation of predicted repair time in nomograph form 4.2.1 Theoretical Overview of Availability and Maintainability Prediction in Conceptual Design Availability and mainta inability prediction attempts to quantify the measures of suc- cessful operational performance of systems under conditions of failure and subject to restoration criteria. Although similar but not identical to reliability prediction, availability and maintainability predictions are predominan tly considered in the pre- liminary design phase, and usually extend through to the schematic design phase of the engineering design process, together with estimations of expected failure rates and expected repair rates. The most applicable methodology for availability and maintainability prediction in the conceptual design phase includes basic concepts of mathematical modelling such as: i. Cost modelling for design availability and maintainability ii. Availability modelling based on system performance iii. Inherent availability modelling with uncertainty iv. Preliminary maintainability modelling. 4.2.1.1 Cost Modelling for Design Availability and Maintainability Design availability and maintainability are directly related to ca pital and operating costs of the engineered installation. (In power-generating installations, the conse- quence of low availability is directly related to the cost of replacement-power costs. 4.2 Theoretical Overview of Availability and Maintainability in Engineering Design 309 However, replacement power costs contribute mostly to overall costs in cases of unplanned low availabilities, whereas capital and operating costs are the dominant contributors in designing for availability and maintainability.) The basic assumption underlying a reliability, availability, maintainability and safety (RAMS)analysisof engineering design is that there is an optimal availability or availability range for which capital and operating costs are at a minimum. Designing for availability affects the engineering design’s installation capital costs with respect to perf ormance relating to process capability, functional effec- tiveness, and operational condition. Designing for maintainability has an effect on the engineering design’s installa- tion capital costs with respect to systems configuration, equipment selection, main- tenance, and the in itial provision of contract spares. Capital costs are influenced by systems configuration, such as provision for equipment redundancy where standby equipment is required to increase reliability, or provision for parallel systems where maintainability is increased through an increase in the modes of operation whereby selective maintenance shutdowns are accommodated without decreasing produc- tivity. Equipment that is more reliable i s generally more costly because of higher strength and corrosion resistant materials, or because of more stringent manufactur- ing specifications. Equipment selection helps to differentiate between critical sub-systems where highly reliable equipment are required, and non-critical sub-systems where the use of less reliable equipment might reduce capital costs without appreciably sacrific- ing system availability. Strategic application of scheduled mainten a nce, particularly partial and total shutdowns and provisioning of initial contract spares in the form of complete assemblies, improves maintenance downtime but also increases capital costs. Availability and maintainability prediction methodologies in designingfor avail- ability and designing for maintainability can assist in the prediction of the engineer- ing design’s installation capital costs (with respect to systems performance, con- figuration, equipment selection, maintenance, an d the initial pr ovision of contract spares), to balance unavailability costs against excessive capital costs. While pro- visions for maintenance and initial contract sp ares are part o f an engineerin g de- sign’s installation capital costs, scheduled maintenance and replenishment o f con- tract spares inventory beyond the installation’s warranty period usually becomes part of operating costs, particularly operating and maintenance (O&M) costs. A well-developed preventive maintenance plan, established during the engineer- ing design stage, will reduce overall life-cycle costs by improving equipment op- erational reliability and availability for periods of high demand, thereby r educing operating costs. For high-demand equipment (particular to continuous processes), the consequence of low availability is normally high replacement costs. This cost may be minimised by strategically scheduled preventive maintenance in the form of shutdowns during periods of low demand, if possible. 310 4 Availability and Maintainability in Engineering Design a) Economic Loss and the Cost of Dependency Loss in production is due to the unavailability of plant and equipment as a result of the need for scheduled maintenance shutdowns, or for unplanned shutdowns because of economic operational and physical consequences of functional failure. Costs due to the unavailability of a plant as a result of unplanned shutdowns at times of high demand generally incur higher replacement costs, as they occur when the replacement cost is still considered to be less than the loss. Loss in production de- pends upon the type of process, type of equipment, the design layout or equipment configuration, the process capacity of equipment, as well as the capacity/demand relationship. The cost of a loss in production (i.e. loss in product and in production effort) during the period of lost production time is fundamentally the cost of waste in dependent productive resources. The cost of waste is the cost of the loss incurred as aresultofdependency on these productive resources. The cost of the loss incurred as a result o f the dependency on productive resources constitutes an economic loss. Economic loss can be quantified as the cost of dependency on productive resources (Huggett and Edmundson 1986). Systems economic loss in production can be quantified as the cost of relying upon the system or equipment with regard to its systems configuration, the system’s pro- cess output, the system’s capacity surplus, and the demand on the system. Systems dependency can be formulated as Dependency= Output−Surplus Demand ×100% . (4.2) The measure of systems dependency is the system’s output minus the system’s ca- pacity surplus as a ratio to the demand on the system, expressed as a percentage. The system’s capacity surplus is the system’s design capacity minus the demand on the system. The same principle is valid at equipment level, or for the process as a whole. This can be formulated as Capacity Surplus = Desing Capacity−Demand . (4.3) The measure of design capacity of a series of systems, sub-systems or equipment in a process is the value of the smallest design capacity of the individual capacities in the process, measured as the process output. The measure of design capacity of parallel systems, sub-systems or equipment in a process is the sum of the individual capacities in the process, measur ed as the process output. The measure of process output can be quantified in the form o f system, sub-system or equipment output basedonitsproduction cycle time, sequencing and utilisation. The economic loss of production can be quantified as the cost of dependency Economic loss = Cost of dependency (4.4) The cost of dependency is the cost of a loss in production during the period that the system or equipment is down due to total or partial shutdowns. This cost of 4.2 Theoretical Overview of Availability and Maintainability in Engineering Design 311 dependency is, in fact, the relative lost time cost due to total or partial shutdown of the system or equipment at its relative value of dependency Cost of dependency = Relative lost time cost (4.5) The relative lost time cost is calculated as the product of the lost time multiplied by the cost of a loss in production during the period of lost time of the system or equipmentat its relative value of dependency.The cost of a loss in production during the period of lost time of the system or equipment at its relative value of dependency is determined by the product of the cost of the loss in production at the dependency of 100%, m ultiplied by the dependency of the system or equipment. Thus, relative lost time co st can be formulated as Relative lost time cost = Lost time (4.6) ×Cost of production loss at 100% dependency ×System or equipment dependency . The cost of production loss at 100% dependency is considered to be the value of lost time of the system or equipment at 100% dependency Relative lost time cost = Lost time (4.7) ×Value of lost time ×System or equipment dependency . Example problem In the illustration below (Fig. 4.3), three systems are in parallel configuration with a total parallel system process capacity of 1,500 tons (t) of prod- uct. System A1 has a design capacity of 600t, system A2 has a design capacity of 500t, and system A3 has a design capacity of 400t. The total demand on the parallel system process is 1,000t. What would be the economic loss of production,orcost of dependency, in the event of system A1 being down for 5 days as a result of shut- down, and then systems A1 and A2 being down for 5 days as a result of shutdown? The value of process lost time is estimated at $20,000 per day. System A1 600 tons capacity System A2 500 tons capacity System A3 400 tons capacity 1000 tons process output demand Fig. 4.3 Three-system parallel configuration system 312 4 Availability and Maintainability in Engineering Design Solution In the three-system parallel configuration system: The dependency on system A1 is: Dep. A1 = 600−(1,500−1,000) 1,000 ×100% = 10% . The dependency on system A2 is: Dep. A2 = 500−(1,500−1,000) 1,000 ×100% = 0% . The dependency on system A3 is: Dep. A3 = 400−(1,500−1,000) 1,000 ×100% = −10% . The negative value for the dependency on system A3 is an indication of it being superfluous or redundant in the three-system configuration process, as there already exists surplus capacity from system A1 and system A2. What is the economicloss of productionin the eventof system A1 being downfor 5 days as a result of downtime? The economic loss of productioncan be quantifiedas the cost of dependency on the systems, where the cost of dependency is the relative lost time cost due to downtime of the systems at their relative value of dependency: Economic loss = Cost of dependency = Relative lost time cost . The relative lost time cost is calculated as the product of the actual lost time mul- tiplied by the value of the actual lost time of the systems at 100% dependency, multiplied by the systems’ dependency: Relative lost time cost = Lost time ×Value of lost time ×System or equipment dependency Relative lost time cost = 5days×$20,000/day×10% for system A1: = $10,000 . In the event that system A1 experiences downtime, the dependencies on systems A2 and A3 change drastically. With system A1 down, now the p arallel system capacity of systems A2 and A3 is 500 plus 400 t. The process output demand still remains at 1,000t. What is the dependency on each of systems A2 and A3 in the two-system parallel configuration process? . capital costs. Availability and maintainability prediction methodologies in designingfor avail- ability and designing for maintainability can assist in the prediction of the engineer- ing design s installation. use of standard or common end-items. 306 4 Availability and Maintainability in Engineering Design Maintainability ha s true design characteristics. Attempts to improve the inherent maintainability. Preliminary maintainability modelling. 4.2.1.1 Cost Modelling for Design Availability and Maintainability Design availability and maintainability are directly related to ca pital and operating costs