: Use of Live Aircraft Data in Aircraft Maintenance Management
Table of Contents:
The use of live aircraft performance data has been demonstrated in recent years to be a very valuable tool to observe how individual components are performing and to estimate their remaining usefulness before their performance deteriorates to an unacceptable level. If a component is allowed to continue to operate below a given threshold, damage can occur to both the system and other nearby services. The resulting unplanned maintenance is very costly for airlines relying on their aircraft being available throughout the working day to carry revenue, passengers and freight.
This chapter includes an overview of the maintenance philosophies and provides an explanation of where these documents originate, and why reliability has commercial value. The sources of performance data are discussed. The use of wireless communications are explored, as are the advantages of the presence of big data predictive algorithms to estimate performance.
Aircraft Maintenance Management and Its Commercial Importance
Large commercial aircraft are complex devices that contain a very large number of complex components. In order for an aircraft to be dispatched into commercial revenue service, the aircraft must perform or operate to the approved tolerances that the original manufacturer has defined. For example, the engine manufacturer defines how many take-offs and landings an engine can make before it requires components to be changed.
Everything mechanical that moves, requires maintenance to remain serviceable. Applying this understanding to a large wide-body commercial aircraft with upwards of 10b individual components means, in the most basic terms, the minimum standards of performance must be reached before the aircraft can take-off to earn revenue.
Aircraft operators must fly these aircraft as frequently as possible to gain as much revenue during the working permittable flying hours of the aircraft each day If night-time airport curfews occur (e.g. midnight), the industry practice is to conduct routine maintenance activities during the hours of darkness. This allows the aircraft to be ready for commercial service when the flying day commences (e.g. 5 am), making full use of the natural downtime of the aircraft.
Maintenance Planning by the Original Equipment Manufacturer (OEM)
When a new aircraft is designed, produced and flown for the first time, the National Aviation Authority of the country of manufacture becomes responsible for the certification of the new aircraft. In Europe, the European Aviation Safety Agency has legal powers derived from EU legislation, which is devolved to the NAA's. As part of this new certification process, the aircraft manufacturer must demonstrate to the Authority that a realistic minimum standard of engineering performance can be achieved. The aircraft OEM must also define the ongoing scheduled maintenance activities for a new aircraft with the certifying authority. These scheduled maintenance activities are usually conducted at predetermined time or use-based intervals. The defined maintenance activity is based on how the manufacturer foresees the aircraft being operated, and the type of inspection varies significantly. For example, a daily maintenance inspection may only have 20 or so checklist items, with weekly inspections being more detailed in terms of inspection items and complexity of test. An 'A check' inspection occurs at 4-month intervals, taking several days to complete, along with 'C check' hangar activities taking place every 18 months and being very complex. A typical 'C check' would see the engines removed and returned to the relevant manufacturer for strip and overhaul; the aircraft seats removed; floors removed; all the systems checked/tested, implementing any maintenance upgrades or modifications necessary; and the rectification of any other defects so that the aircraft is free from any Tech Log defects. A typical 'C check' duration would be 4 to 6 weeks of high-intensity maintenance in a hanger (working continuous shifts).
The maintenance schedule forms part of the aircraft's certification; the operator relies heavily on the OEM and documentation to define all the maintenance and the ongoing upgrades. The OEM business has many similarities to that of the automotive industry, with financial drivers being the principle factor. An aircraft OEM will communicate with all aircraft operators to offer maintenance, modifications and upgrades. The airlines have the task of categorising these communications from the OEMs, to prioritise modifications that will enhance safety and implement them quickly; this activity is likely to include the Airworthiness Directives of the relevant Aviation Authority to address known safety issues specific to a type of aircraft. The next type of modification will be cost saving, to allow an operator to run the aircraft more efficiently, and to make financial savings. The lowest level of modification will include all other possibilities to change the state of the aircraft, for example, the inclusion of onboard Wi-Fi data communication for passengers.
The airlines engineering office team must review all these communications from all the OEMs in a timely fashion, and quickly decide what is a must have, nice to have or a don't need. The OEM, being the manufacturer, will be more profitable if all the suggested modifications are implemented, whereas the airlines will bear the financial burden of both the activities and the loss of revenue availability of the aircraft.
If the aircraft develops a technical fault during the day, or if a fault is identified during the night time maintenance activity, these defects are entered in the aircraft Technical Log (Tech Log), a mandated record of all items that have arisen during the day or flight sector, and every registered aircraft must carry a copy of this document. Previously these Tech logs were 4 or 5 sheets of coloured carbon paper that were torn out before the flight with a white top sheet that remained in the book. However, in recent years, with the advent of cheap handheld computing and applications, the data is directly added by the person reporting the fault onto an electronic handheld device (e.g. an iPad or similar, containing specific applications), removing the need for large numbers of administration staff in the Engineering offices who previously would type out all the paper-based data into a computing maintenance database.
When faults arise, the details of the fault are recorded by the pilot or engineer into the electronic Tech Log. The airline that operates the aircraft must address the fault before the next flight in order to remain legally compliant with legislation. All maintenance activities must be performed by approved maintenance repair organisations (e.g. FAR 145/Part 145 approved companies) employing licensed maintenance engineers.
Minimum Standards of Equipment of Systems – Master Minimum Equipment List
When an aircraft is certified by a National Aviation authority, one important OEM document known as the Master Minimum Equipment List (MMEL) is produced and approved, and this document specifies what systems can be unserviceable, i.e. the plane could continue to fly safely with some defined defects. Some potential defects are quite trivial, such as a tail logo light filament that illuminates the painted artwork on the vertical stabiliser, since the maximum timeframe to defer and later perform the repair could be tens of days from the time of entry. However, other defects may be sufficiently serious and affect safety so that only one or two sectors can be flown.
If a defect occurs during the flying daytime, when the aircraft lands an engineer must evaluate the problem (for safety and legal compliance) and perform a recorded maintenance action in the aircraft technical log before the next flight. If the defect can be rectified quickly and the engineer has the correct spare part(s) available, the favoured option is to do the repair on the 'ramp' next to the terminal so as not to cause delays to the flying schedule. However, often repairs and testing take time and resources and a delay to the remaining days' flight schedule would have a significant impact on the operator's revenue, so a deferred maintenance action may be chosen by the engineer.
All defects are categorised by the ATA 100 code, an American 00 to 79 code classification of all systems for fixed-wing large aircraft. The engineer would check the restrictions written in the Minimum Equipment List (MEL), an airline-specific document that is more restrictive than the OEM's MMEL. Providing the defective item can be deferred, the necessary administrative actions are undertaken and the 'open' failure is 'closed' by the engineer, and an Acceptable Deferred Defect (ADD) is raised. The number of ADDs is carefully monitored by the Engineering management and closely scrutinised by the National Aviation Authority, as the emphasis is to use this activity only where appropriate. Typically, if the MEL permits, AADs can last up to seven consecutive days, and if the defect cannot be fixed, the item can be 'closed' and 'reopened' twice more, giving a total timeframe of up to 21 days from the initial entry.
If the aircraft defect can be deferred to a 'C check' the operator will defer the rectification until the scheduled aircraft planned hangar maintenance check (because the aircraft will be in the hanger for up to 6 weeks), and this is known as a Base Deferred Defect (BDD).
If the defective system has an effect on the safe operation of the aircraft or is not permitted to be a deferrable item within the MEL (i.e. the permissible defects and limits), then the aircraft is grounded until such time that the defect is fixed, and the engineer approves the repairs to a standard In Accordance With the Aircraft Maintenance Manual (IAW AMM). The industry now categorises the grounding of the defective aircraft as Aircraft On Ground (AOG), and this code is the highest level of maintenance priority.
Component Reliability and Maintenance Strategy
Unscheduled defects pose a real problem to the efficient operation of the aircraft because it is very challenging to 'predict' when the next component might fail.
Reliability is often considered as the measure of serviceable performance with time. If a component performs in such a way that is outside of the original designed specification performance, it is considered to have failed. Reliability as an engineering field is a mature subject, with roots stretching back to the Second World War. During the Second World War, mechanical systems were expected to have a finite lifespan, due to assumed damage that would naturally result from conflict. The logic was that a system would require less regular maintenance because the asset, be it an aircraft, land vehicle, etc., would have a given probability that it damaged and potentially lost or unrepairable. Consequently, emphasis was given to manufacturing items in very large volumes rather than for long-life performance. When maintenance was performed, all components on a vehicle removed down to a single-component level and inspected. If necessary, additional maintenance and new components would be fitted. This type of philosophy, i.e. very labour-intensive maintenance, tear-downs, inspection and rebuilds continued up to the early 1970s. This is reflected in the duration of a licensed Aircraft Engineers maintenance course for a Boeing 747-100, which took some 3 months for the engine and airframe elements. By comparison, a maintenance course of similar scope for a B747-400 series is around 4 to 5 weeks. The difference is explained by the change in the maintenance philosophy between the very early 1970s and the present day: the regular maintenance activity of the present time has been optimised for the use and the operation of the system. The need to 'pull everything apart' has been superseded with a more mathematical approach, based on performance data for large numbers of components.
Bathtub Curve for Reliability and Mathematical Predictions
The concept of reliability considers that individual components all have differing levels of wear and serviceability. A mathematical evaluation of a large number of components in service and their time to failure has been established for many years, known as the 'bathtub curve' as illustrated in Figure 5.1, where the Y-axis represents the failure rate and the X-axis the time elapsed to failure.
Figure 5.1 shows three distinct regions of the reliability Bathtub shape, namely the curve to the far left, the curve to the far right and lastly the flat line between the two curves. The left zone (Figure 5.1) is known as the infant mortality region and the decreasing curve value represents the manufacturing defects that might result in an earlier than expected failure. Conversely, the right-hand side (Figure 5.1) is known as the wear out region and is attributed to the increasing failure rate due to the item wearing out more rapidly, and the curve illustrates the exponential increase. The flat line region (Figure 5.1) illustrates the random failure rate of the majority of components, as the parts are engineered to perform for a minimum time duration and reach the right side of the curve.
Mathematical evaluation of the time to component reliability can be performed, such as the Mean Time To Failure (MTTF) or Mean Time Between
Reliability Bathtub curve of component reliability.
Overhaul (MTBO). The mathematical models are valuable because they are based on a large data set for identical components. If the data set is not sufficiently large and several infant mortality events are experienced, the data arising from the Bathtub curve would show the mathematical MTTF as being much smaller than the real-world large data set. The objective of the reliability calculation is to predict when the average or MTTF will occur, and ultimately to remove the component before possible catastrophic failure, overhauling it to like new performances.
The use of these mathematical tools to evaluate performance permitted the exchange of components to be factored into the overall maintenance strategy, which is dictated by the OEM. If an item cannot be reasonably repaired, then it is classified as a consumable part, and after use is disposed of by the appropriate means. However, many of the components are complex and expensive, so, the justification to repair them at an approved company is both justified and necessary. These components are known as rotables - i.e. they can be repaired by a third-party approved overhaul organisation, recertified as good as new and re-fitted onto live aircraft. Clearly if the rotable parts are operated until catastrophic failure, the repair and recertification is not going to be physically or economically possible, especially if the components have structurally failed. Accordingly, reliability tools are utilised to predict the optimal usage time for each component.
An aircraft OEM will optimise the maintenance programme for their aircraft, including the use of reliability software to predict the given components' failure rates, as the overarching goal is to keep the aircraft flying as long as possible. The aircraft OEM also minimises the maintenance to acceptable levels by the certifying authority such as the Federal Aviation Administration (with the goal to minimise unscheduled maintenance) and to schedule all component changes, where possible, to planned scheduled maintenance, namely the 'base C checks in the aircraft hangers.
This complex situation of planning, predicting and monitoring the status of both the aircraft and the tens of thousands of individual rotables and components is only possible with the use of computing.