Optimizing Aircraft Maintenance For Improved Safety And Cost Reduction

The aim of this assignment is to write a scientific paper on a particular topic – of your own choice – within the scope of aircraft maintenance. In this paper, you will either need to explore a novel contribution to the state of the art, or critically examine, synthesize and reflect upon the current academic state of the art.

Note that you are free to suggest your own topic of choice (or a specific topic within the provided set below). Be sure to align with the lecturer. As a guideline, the following topics can be considered:

• Maintenance planning optimization
• Composites
  • Training
  • Access to knowledge about products and processes
• Condition-based Maintenance (CBM) / predictive maintenance / Prognostics and Health Management (PHM)
  • Applicability of CBM / PHM (technical & economic feasibility)
  • Diagnostics / prognostics for systems and/or structures
• Big data
  • System availability versus diagnostic & prognostic capability
  • Fleet data management
  • Interoperability issues
• Paper-based to paperless organisation
  • Legacy tools & systems
  • Phase-out of aircraft
  • System interoperability
• VR / AR in maintenance

The Importance of Aircraft Maintenance

Air transport sector has been a great contributor to global economy in various ways. While facilitating global movement of people and their commodities, it has improved economies in various nations including improving living standards of many people who work with airways both directly and indirectly. Distant countries have been bridged by well-constructed and developed airways and this has brought efficiency in the movement of different persons and their products. This has further been facilitated by well-developed aircrafts which carry passengers of varying numbers depending on the size of the aircraft involved. As compared to other forms of transport, air transport has proven to be the fastest of all and has fewer accidents. However, weather conditions greatly affect its operability and this has caused many challenges to this sector. Despite having fewer accidents than its competitors, its accidents are deadliest when they occur and cause many fatalities at a glance (Al-Thani, Ahmed &Haouari, 2016).

Therefore, it is very critical to ensure there is proper aircraft maintenance to ensure such fatalities related to this sector are minimized and to increase the trust of customers in this kind of transport sector. Many accidents from this sector have been attributed to technical –related issues, for instance, the most current accidents on Boeing models. This calls for a better and informed approach of maintenance which involves early planning optimizations. This will assist in early identification of technical flaws, curbing of such flaws as well as in making informed decisions concerning the future of aviation and the entire safety of the crews (Bozoudis, Lappas & Kottas, 2018). This will help in reducing costs at various levels of the industry.

Programs for aircraft have evolved over time with the new approaches building on the incapability of the previous approach (Buyya & Dattel Ph D, 2018). The information gathered on routine checks is used as a basis of deciding on the extensions of interval required for inspections. All the maintenance approaches are tied to the following objectives;

• Ensuring there is overall  and reliable safety through maintenance
• Ensure that there is restoration of safety once flaws are detected
• Ensure adequate information is available for making decisions
• To ensure all the above considerations are accomplished at very low costs.

Aircraft maintenance planning is one of the roles of Aviation Production-Management. This is an implication that it is the production management that is involved in arranging the right time as well as the desired sequence and the venue of conducting maintenance task (Callewaert, Verhagen & Curran, 2018). Furthermore, this will aid in ensuring that the plans for the original flight are executed normally as well as ensuring that there is minimization of maintenance costs simultaneously. Global aviation has experienced stiff competition and this has consequently increased the costs of operation thus greatly impacting the industry in a negative way. This has left many aviation enterprises struggling to lower such costs for them to at least have an access to a newer version of profit-growth.

Challenges in Aircraft Maintenance

In perception of aircraft maintenance-costs, the cost has almost reached two-thirds of the original price used in purchasing and this therefore, makes it essential to optimize aircraft maintenance in general. There are various optimization models which can be applied here which include integer-programing model, whose objective is to maximize value, graph theory which uses a polynomial algorithm (De Bruecker, Beliën, Van den Bergh &Demeulemeester, 2018).

Scheduling in airline logistics occurs by paying attention to  the following  points;

• Preparation: identification of arrival time as well as time to depart
• Fleet-assignment problem: identification  of the type flight which takes a given flight-leg
• Routing areas; aircrafts are assigned particular routes
• Disruption-recovery problem: provides reaction over operational disruption by adverse conditions of weather.

Therefore, it is through the first top three points where long-term planning of aircraft maintenance comes from. However, due to daily uncertainties, policies for maintenance planning become ineffective sometimes. Direct operating costs will vary according to the size of the fleet, age as well as usability. In that connection therefore, when such unplanned maintenance occur, costly delays are encountered and even cancellations can occur  if no immediate solution is offered (Eltoukhy, Wang,  Chan & Chung, 2018).

Line maintenance as a process occurs within Turn-around-Time (TAT).  This occurs flanked by two flights so as to ensure that there is aircraft dispatch that is reliable and timely. This maintenance takes in routine checks, carrying out post-flight inspection as well as correction of flaws which are performed en-route and points of departure. The decision used in TAT is simply based on the next flight of the concerned aircraft and such decision is on a GO/NOGO basis. A GO-based decision, which permits the next flight to take place, is reached upon the satisfaction of all constraints for MMEL (Manufacturer Minimum Equipment List) (Feng, Bi, Zhao,  Chen & Sun, 2017).

In a contradicting way to TAT, decision-support process in the current times has reactivity to handle unscheduled maintenance. With this process troubleshooting is performed so as to identify the root-cause to permit necessary arrangements of maintenance to be carried out instantly.

Airline planning challenges have been addressed, which include fleet-assignment, maintenance management among others mainly in operational research works. Among the various publications is a publication dealing with aircraft maintenance-routing issue which uses a branch-and-price system of algorithms. The main aim of this algorithm points towards minimization of legal hours of flight especially on the unused legal flight-hours. Comprising the formulation of problem is the management of resource constraints’ availability and the branch-and-price algorithm will play a role of exploring efficiency of the solutions to routing problems. Other advancements in maintenance models have been echoed in areas of operation research for instance fleet scheduling are provided in many areas of aircraft maintenance (Fioriti, Vercella & Viola, 2018).

Optimization Models for Aircraft Maintenance

Mathematical approaches have been investigated as well as approximation tools which rely on Lagrangian relaxation for the purpose of capacity planning by Dijkstra et al. Dynamic programming approaches and heuristic approaches have been combined to provide solutions to embedded scheduling challenge (Grima, 2018).

 Despite all these efforts towards maintenance information and its sophistication as well as implementation of decision-support systems, a lot of limitations on have made the entire process of aircraft maintenance ineffective at some points.  For instance, GO/NOGO decisions only rely on assumptions that delayed flights might occur, cost-related consequences and this makes these approaches to have limited intelligence (Qin, Chan,Chung & Qu,  2017). Furthermore, such approaches take much time and efforts of the maintenance personnel who are supposed to check on available information so as to make decisions on the next step to be taken. This argument is further reinstated by the fact that there is no evidence of communication going on flanked by the operational and maintenance sites of planning which has eventually resulted in sky-rocketed operational costs.

Regardless of the existing similarities, no much weight has been given to this discussion the way this paper has done. This is because many of the studies have given weight to long-term planning while this paper attaches some weights to short-term decisions for operations. This paper highlights a framework to support maintenance decision by keenly dissecting short-term decisions for operational maintenance. It aims at improving fleet operability as well as lowering maintenance costs. Short-term planning is executed by assessing healthy information as well as on economic controls at both the aircraft and fleet-based levels.

 This planning framework has been developed to fit the decision support system and its future is directly dependent on diagnostics as well as Condition-Based Maintenance (CBM) systems.

Planning and scheduling line maintenance requires a combination of health-assessment acquired information and flight operation-related data, the costs of maintenance, presence of maintenance resources in addition to the whole maintenance program. This approach uses the knowledge obtained from multi-criteria models for making decisions in addition to simulation to enable production and evaluation of varying plans of maintenance (Safaei & Jardine, 2018).  While handling the process of making decisions, this approach will apply the following points at each step of the process;

• Identification of the required tasks for maintenance
• Determination of the criteria to make decisions to be used in the evaluation of alternatives.
• Developing  alternatives (Maintenance plans)
• Weighing the consequences related to various alternatives

The process of line maintenance kicks off upon the arrival of aircraft. This comprises of data acquisition, assessing the status of the aircraft in addition to making decisions for the tasks to be executed. Assuming that all the relevant constraints meet the satisfactory considerations, next flight is permitted to commence because the status has become a GO. This proposed approach is keen on supporting decisions to be taken on successive airports pending maintenance-events. This means that tasks can be undertaking at the current position of the aircraft or decisions can be reached for executing them at successive positions.

Scheduling and Planning in Aircraft Maintenance

The proposed method supports that maintenance decision ought to be done anytime, anywhere and feasible alternatives should be produced containing a decision-matrix. Such alternatives will offer possibility of allocating pending maintenance-tasks. This allocation is done on suitable resources any point (either current or the successive airports) within the timeframe prediction. An airport can be that suitable resource for it is capable to dispatch a given maintenance task whereas the maintenance can be a single or multiple activities for example replacement for a given subsystem of an aircraft. This approach will apply two parameters for searching i.e., the maximum alternative’s (MNA) number and sampling rate (SR).

Whenever a decision-point is gotten, i.e. when there is a component that needs to be maintained in an already-landed aircraft, there is generation of MNA-alternatives and this is followed by simulating each alternative at SR times. The following figure shows short-term planning;

 

Fig 1: Generation of alternatives at a decision point

The above figure shows short-term planning and its implementation has been done in a software system-format.  Identification of feasible alternatives has been based on events that require decisions to be taken upon them. The simulation of identified alternatives is done at SR times and estimation is done on the average-performance against every criterion in a format of decision-matrix. Therefore, an alternative whose utility is the best becomes preferable to the proposed-system.

 The extract below shows variables used by this approach;

 

Fig 2: Indices used

 This mechanism is activated at every decision-point for the tasks that are deferred for the purpose of identifying the point of each task’s dispersal.

Risks associated with economy as well as operations are likely to emanate upon executing each alternative. As mentioned in this discussion, a good definition of the word alternative is the scenario of allocating suitable resources to maintenance tasks which are pending at current and/or sequential stations. In order to calculate utility (i.e. the performance of alternative), an assumption is made that there are K decision-alternatives whose denotation is given as Al_1 …to Al_k... Therefore, in order to evaluate alternative process, there are four points to be used which are: flight-delay, useful-life remaining, risks associated with operations as well as costs (Spirtovic, Bekker, Poppinga, de Jong, 2018). Maintenance engineer has to aim at minimizing all the mentioned four points, if there is possibility of doing so. Here, the average figures of SR samples are considered, those of each alternative, in measuring the assignments-related consequences. The criterion to be used here is assigned C. Normalization of these values is done as shown below;

Line Maintenance and Decision Support Process

 And combination of the entire sets of consequences can be used in calculating utility as shown below;

This means that when an alternative whose utility is the best of all has allocated task at current-airport, then maintenance engineer can select respective tasks to be executed at current decision-point (at the current airport). But if the above has not been met, the next decision-point can be used in allocation of tasks.

This approach supports the extension of the existing models of constraints for accounting the various components of maintenance requirements (Verhoeff, Verhagen, Curran, 2015). This will also facilitate aspects of time to be accounted, which comprises of the departure as well as the time for exact arrival and flight delays. A finite quantity of maintenance stations exists whereas there is a limitation of person-power and this apply to various organizations operating airlines together with their aircrafts too.

There are various which owe their relations to aircraft maintenance-costs and can fall under the following categories;

• Overhead costs: this category represents costs which indirectly attributed to operations of maintenance but they are of necessity.
• Equipment costs: costs required for purchasing and maintaining those facilities which are essential for the processes of maintenance tasks. Include costs used to host equipment, infrastructural maintenance, hire equipment among others.
• Supplies/logistics costs: costs incurred in general components of transport.
• Personnel costs: costs to meet provision of labor towards executing the various tasks of maintenance for instance overtime costs, costs for acquiring a maintenance engineer and other related tasks(Vianna, Yoneyama, 2017).

 In order to coordinate all these costs, a ‘techno-economic model’ will be used to model a harmonized way which is equal to operating time;

Maintainance cost/per task/per component =(( rate of equipment+labour rate+overhead rate) * operating time per componet) + component procuremement costs)

 Whereas the cost is calculated by the following formula;    repersenting  cost-value consuquence in regards of cost-criterion.

Operational risk is a representation of risk that results in disruption operational plan and because extra costs because of maintenance tasks which are unscheduled (Vianna, Rodrigues & Yoneyama, 2015). It is assessed in order to provide an estimate of the cost as well as the likelihood of maintaining events which are unscheduled and are likely to impact fleet operational-plan. In order to model function of operational-risk, expected values are used. Principally, these values are the summations of every possible outcome of an experiment for a randomized variable. This total summation is them multiplied by the pay off and this is mathematically expressed as shown in the figure below;

 

Fig 3: Mathematical expression of operational risk

Flight delay is one of the best criteria to be used in the process of making relevant decisions for planning. Timely flights are paramount and in case delays are experienced, they should be much minimized. Measuring delays for aircrafts due to maintenance helps to assess the substitutes of performance. The duration for flight, maintenance events as well as delays in departure are exhibited as variables of stochastic nature. The alternative flight’s delay-values are expressed as depicted in the figure below;

Advancements in Fleet Operations and Maintenance Models

 

Fig 4: equation to calculate flight delay value

Conclusion

In conclusion, it is clear that the current systems for decision making are very reactive regarding their principles. The common the role they is that of ensuring there is proper planning to assist in  the maintenance of unscheduled events as well as short-term planning to curb such events.  It is therefore, very paramount to have decision support approaches based on the above discussed criteria i.e. cost, operational risks, flight delay as well as the RUL.

As discussed above, line maintenance as a process occurs within Turn-around-Time (TAT).  This occurs flanked by two flights so as to ensure that there is aircraft dispatch that is reliable and timely. This maintenance takes in routine checks, carrying out post-flight inspection as well as correction of flaws which are performed en-route and points of departure. The decision used in TAT is simply based on the next flight of the concerned aircraft and such decision is on a GO/NOGO basis. Therefore, the proposed method supports that maintenance decision ought to be done anytime, anywhere and feasible alternatives should be produced containing a decision-matrix. Such alternatives will offer possibility of allocating pending maintenance-tasks.

References

Al-Thani, N.A., Ahmed, M.B. and Haouari, M., 2016. A model and optimization-based heuristic for the operational aircraft maintenance routing problem. Transportation Research Part C: Emerging Technologies, 72, pp.29-44.

Bozoudis, M., Lappas, I. and Kottas, A., 2018. Use of Cost-Adjusted Importance Measures for Aircraft System Maintenance Optimization. Aerospace, 5(3), p.68.

Buyya, R. and Dattel Ph D, A., 2018. Assessment of Virtual Technology as a Teaching Method in Aircraft Maintenance.

Callewaert, P., Verhagen, W.J. and Curran, R., 2018. Integrating maintenance work progress monitoring into aircraft maintenance planning decision support. Transportation Research Procedia, 29, pp.58-69.

Clark, P.J., Newcomer, J.M. and Jones, A.M., 2018. Overcoming Gender Barriers in Aircraft Maintenance: Women’s Perceptions in the United States. The Collegiate Aviation Review International, 33(2).

De Bruecker, P., Beliën, J., Van den Bergh, J. and Demeulemeester, E., 2018. A three-stage mixed integer programming approach for optimizing the skill mix and training schedules for aircraft maintenance. European Journal of Operational Research, 267(2), pp.439-452.

Eltoukhy, A.E., Wang, Z.X., Chan, F.T. and Chung, S.H., 2018. Joint optimization using a leader–follower Stackelberg game for coordinated configuration of stochastic operational aircraft maintenance routing and maintenance staffing. Computers & Industrial Engineering, 125, pp.46-68.

Ewald, V., Goby, X., Jansen, H., Groves, R.M. and Benedictus, R., 2018. Incorporating Inductive Bias into Deep Learning: A Perspective from Automated Visual Inspection in Aircraft Maintenance. In Proc. 10th Intl Symposium on NDT in Aerospace, Dresden (pp. 1-9).

Feng, Q., Bi, X., Zhao, X., Chen, Y. and Sun, B., 2017. Heuristic hybrid game approach for fleet condition-based maintenance planning. Reliability Engineering & System Safety, 157, pp.166-176.

Fioriti, M., Vercella, V. and Viola, N., 2018. Cost-estimating model for aircraft maintenance. Journal of Aircraft, 55(4), pp.1564-1575.

Gratton, G. and Shukri, M., 2017. The potential risk of communication media in conveying critical information in the aircraft maintenance organisation: A case study.

Grima, R., 2018. Effort estimation for aircraft maintenance using machine learning techniques (Bachelor's thesis, University of Malta).

Kim, K., Shao, Q.G., Araki, R., Earl, A., Dong, B., Guo, W. and Shi, E., Honeywell International Inc, 2019. Aircraft maintenance systems and methods for ECS fouling predictions. U.S. Patent 10,242,508.

Qin, Y., Chan, F.T., Chung, S.H. and Qu, T., 2017, April. Development of MILP model for integrated aircraft maintenance scheduling and multi-period parking layout planning problems. In 2017 4th International Conference on Industrial Engineering and Applications (ICIEA) (pp. 197-203). IEEE.

Rodrigues, L.R., Embraer SA, 2016. Maintenance planning optimization for repairable items based on prognostics and health monitoring data. U.S. Patent 9,424,693.

Safaei, N. and Jardine, A.K., 2018. Aircraft routing with generalized maintenance constraints. Omega, 80, pp.111-122.

Spirtovic, D., Bekker, R., Poppinga, G. and de Jong, A., 2018. Aircraft Preventive Maintenance: Task Packaging & Scheduling.

Stadnicka, D. and Ratnayake, R.C., 2017. Enhancing aircraft maintenance services: a VSM based case study. Procedia Engineering, 182, pp.665-672.

Vali, N., Gillenwater, K., Mattingly, W. and Wilson, J., Taleris Global LLP, 2018. Display panel portion with a graphical user interface component for an aircraft maintenance interface. U.S. Patent Application 29/554,809.

Verhoeff, M., Verhagen, W.J.C. and Curran, R., 2015. Maximizing operational readiness in military aviation by optimizing flight and maintenance planning. Transportation Research Procedia, 10, pp.941-950.

Vianna, W.O.L. and Yoneyama, T., 2017. Predictive maintenance optimization for aircraft redundant systems subjected to multiple wear profiles. IEEE Systems Journal, 12(2), pp.1170-1181.

Vianna, W.O.L., Rodrigues, L.R. and Yoneyama, T., 2015. Aircraft line maintenance planning based on PHM data and resources availability using large neighborhood search. In Proceedings of the Annual Conference of the Prognostics and Health Management Society, San Diego.