An aircraft state-based spare parts dynamic programming method and system

Abstract

The application relates to an aircraft state-based spare part dynamic planning method and system, and belongs to the technical field of aircraft spare part planning. The fleet scale and the fleet annual utilization rate of a machine type to which a to-be-predicted component belongs are acquired, and historical fault data of the to-be-predicted component are acquired, so as to calculate the working component quantity, the fault rate, the repair rate and the spare part support rate of the to-be-predicted component. Then, the working component quantity, the fault rate, the repair rate and the spare part support rate are taken as inputs, a genetic algorithm is used to optimize and solve a spare part quantity calculation model, and the spare part demand quantity of the to-be-predicted component in a prediction time period is obtained. The minimum spare part demand quantity of the to-be-predicted component in the prediction time period that meets a given spare part support rate can be obtained, and the prediction precision is high.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of aircraft spare parts planning, in particular to a spare parts dynamic planning method and system based on aircraft state. BACKGROUND

[0002] At present, spare parts quantity calculation methods can be divided into statistical methods and analytical methods. Statistical methods are based on a large amount of historical data to analyze spare parts demand, derive statistical rules of spare parts demand, and then combine component related information to form spare parts quantity calculation methods by using technical solutions such as regression analysis, exponential smoothing, probability statistical theory and Bayesian method. Analytical methods usually use engineering theory and reliability theory methods to regard the system as a cold reserve system, analyze the work, maintenance and ordering process of spare parts, and finally determine the spare parts demand quantity.

[0003] The existing spare parts inventory calculation research focuses on forming a spare parts quantity calculation method based on life distribution theory or using some analysis method, and does not fully consider the influence of the aircraft operation state of the airline on the spare parts demand, so the calculation accuracy of the spare parts demand quantity is low.

[0004] Therefore, there is an urgent need for a spare parts dynamic planning technology based on aircraft state. SUMMARY

[0005] The purpose of the present application is to provide a spare parts dynamic planning method and system based on aircraft state, which can predict the spare parts demand quantity based on the aircraft state and has high prediction accuracy.

[0006] To achieve the above purpose, the present application provides the following solutions:

[0007] In a first aspect, the present application provides a spare parts dynamic planning method based on aircraft state, which comprises the following steps:

[0008] Obtaining the fleet size and fleet annual utilization rate of the aircraft model to which the to-be-predicted component belongs, and obtaining historical failure data of the to-be-predicted component; the historical failure data includes the self-repairing use time and repair cycle of each failure of the to-be-predicted component;

[0009] Calculating the number of working components of the to-be-predicted component according to the fleet size;

[0010] Calculating the failure rate of the to-be-predicted component according to the self-repairing use time of each failure of the to-be-predicted component;

[0011] Calculating the repair rate of the to-be-predicted component according to the fleet annual utilization rate and the repair cycle of each failure of the to-be-predicted component;

[0012] Determining the spare parts support rate according to the type of the to-be-predicted component;

[0013] The working component quantity, the failure rate, the repair rate and the spare part guarantee rate are taken as inputs, a genetic algorithm is used to optimize and solve a spare part quantity calculation model, and a spare part demand quantity of the to-be-predicted component in a prediction time period is obtained; the spare part quantity calculation model is determined according to a Markov birth-death process; the spare part quantity calculation model comprises that a reliability corresponding to the spare part demand quantity is greater than or equal to the spare part guarantee rate and a reliability corresponding to a difference between the spare part demand quantity and 1 is less than the spare part guarantee rate.

[0014] A spare part dynamic programming system based on an aircraft state, the programming system comprising:

[0015] A first data acquisition module is configured to acquire a fleet size and a fleet annual utilization rate of a model to which a to-be-predicted component belongs, and acquire historical failure data of the to-be-predicted component; the historical failure data comprises a self-repairing after use time and a repair cycle of each failure of the to-be-predicted component;

[0016] A calculation module is configured to calculate a working component quantity of the to-be-predicted component according to the fleet size; calculate a failure rate of the to-be-predicted component according to the self-repairing after use time of each failure of the to-be-predicted component; calculate a repair rate of the to-be-predicted component according to the fleet annual utilization rate and the repair cycle of each failure of the to-be-predicted component; and determine a spare part guarantee rate according to a type of the to-be-predicted component;

[0017] A first prediction module is configured to take the working component quantity, the failure rate, the repair rate and the spare part guarantee rate as inputs, use a genetic algorithm to optimize and solve a spare part quantity calculation model, and obtain a spare part demand quantity of the to-be-predicted component in a prediction time period; the spare part quantity calculation model is determined according to a Markov birth-death process; the spare part quantity calculation model comprises that a reliability corresponding to the spare part demand quantity is greater than or equal to the spare part guarantee rate and a reliability corresponding to a difference between the spare part demand quantity and 1 is less than the spare part guarantee rate.

[0018] In a second aspect, the present application provides a spare part dynamic programming method based on an aircraft state, the programming method comprising:

[0019] A fleet size and a fleet annual utilization rate of a model to which a to-be-predicted component belongs are acquired, and historical failure data and a spare part unit price of the to-be-predicted component are acquired; the historical failure data comprises a self-repairing after use time and a repair cycle of each failure of the to-be-predicted component;

[0020] A working component quantity of the to-be-predicted component is calculated according to the fleet size;

[0021] A failure rate of the to-be-predicted component is calculated according to the self-repairing after use time of each failure of the to-be-predicted component;

[0022] calculate the repair rate of the component to be predicted according to the fleet annual utilization rate and the repair period of each failure of the component to be predicted;

[0023] determine the spare part guarantee rate according to the type of the component to be predicted;

[0024] use the genetic algorithm to optimize and solve a spare part quantity dynamic programming model with the spare part unit price, the working component quantity, the failure rate, the repair rate and the spare part guarantee rate as inputs to obtain the spare part demand quantity of the component to be predicted in the prediction time period; the spare part quantity dynamic programming model comprises a target function and a constraint condition; the target function is to minimize the total guarantee cost; and the constraint condition is that the reliability corresponding to the spare part demand quantity is greater than or equal to the spare part guarantee rate.

[0025] A spare part dynamic programming system based on an aircraft state, the programming system comprising:

[0026] The second data acquisition module is configured to acquire the fleet size and the fleet annual utilization rate of the aircraft model to which the component to be predicted belongs, and acquire historical failure data and a spare part unit price of the component to be predicted; the historical failure data comprises self-repairing use time and a repair period of each failure of the component to be predicted;

[0027] The calculation module is configured to calculate the working component quantity of the component to be predicted according to the fleet size; calculate the failure rate of the component to be predicted according to the self-repairing use time of each failure of the component to be predicted; calculate the repair rate of the component to be predicted according to the fleet annual utilization rate and the repair period of each failure of the component to be predicted; and determine the spare part guarantee rate according to the type of the component to be predicted;

[0028] The second prediction module is configured to use the genetic algorithm to optimize and solve a spare part quantity dynamic programming model with the spare part unit price, the working component quantity, the failure rate, the repair rate and the spare part guarantee rate as inputs to obtain the spare part demand quantity of the component to be predicted in the prediction time period; the spare part quantity dynamic programming model comprises a target function and a constraint condition; the target function is to minimize the total guarantee cost; and the constraint condition is that the reliability corresponding to the spare part demand quantity is greater than or equal to the spare part guarantee rate.

[0029] According to the specific embodiments of the present application, the following technical effects are provided:

[0030] The application provides an aircraft state-based spare part dynamic planning method and system, obtains the fleet size and fleet annual utilization rate of a machine type to which a component to be predicted belongs, and obtains historical failure data of the component to be predicted, so as to calculate the working component quantity, failure rate, repair rate and spare part support rate of the component to be predicted, then uses the working component quantity, failure rate, repair rate and spare part support rate as input, uses a genetic algorithm to optimize and solve a spare part quantity calculation model, and obtains the spare part demand quantity of the component to be predicted in a prediction time period, so that the minimum spare part demand quantity of the component to be predicted in the prediction time period that meets a given spare part support rate can be obtained, and the prediction accuracy is high. BRIEF DESCRIPTION OF DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0032] Figure 1 The method flowchart of the planning method provided in Embodiment 1 of the present application;

[0033] Figure 2 The principle diagram of the planning method provided in Embodiment 1 of the present application;

[0034] Figure 3 The transfer state diagram provided in Embodiment 1 of the present application;

[0035] Figure 4 The solution flowchart of the genetic algorithm provided in Embodiment 1 of the present application;

[0036] Figure 5 The system block diagram of the planning system provided in Embodiment 2 of the present application;

[0037] Figure 6 The method flowchart of the planning method provided in Embodiment 3 of the present application;

[0038] Figure 7 The system block diagram of the planning system provided in Embodiment 4 of the present application. DETAILED DESCRIPTION

[0039] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments only constitute some embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the present application.

[0040] The application aims to provide an aircraft state-based spare part dynamic planning method and system, which can predict spare part demand based on aircraft state and has high prediction accuracy.

[0041] In order to make the above-mentioned purposes, features and advantages of the application more obvious and easy to understand, the application will be further described in detail below in combination with the drawings and specific embodiments.

[0042] Embodiment 1

[0043] This embodiment is used to provide an aircraft state-based spare part dynamic planning method, as shown in Figure 1 and Figure 2 The planning method comprises the following steps:

[0044] S1: Obtain the fleet size and fleet annual utilization rate of the aircraft model to which the component to be predicted belongs, and obtain the historical failure data of the component to be predicted; the historical failure data comprises the time to service after failure and the repair cycle of each failure of the component to be predicted;

[0045] Before performing the spare part dynamic planning, the application first analyzes the spare part support influencing factors related to the aircraft state, and the main influencing factors of the spare part support related to the aircraft state include historical failure data, seasonality, daily utilization rate, fleet size, etc. Each influencing factor is analyzed as follows.

[0046] (1) Influence of historical failure data. The historical failure data includes failure date, failure cause, time to service after repair (TSR), repair cycle, repair cost, etc. The historical failure data of each component on the aircraft is helpful for spare part demand prediction and can exclude differences caused by region, route, technology and management. Specifically, the failure date can determine whether the spare part failure is affected by season; the time to service after repair (TSR) can calculate the average time between failures and the service life characteristics of the component, and calculate the failure rate; the repair cycle determines the repair rate of the repairable system; and the repair cost can provide an economic reference.

[0047] (2) Influence of season. Due to the influence of season, the number of times of spare part failure in different months is different. For example, in regions with distinct seasons, the temperature is cold in winter and warm in summer, and the annual temperature difference is relatively large. By counting the number of times of spare part failure in different months, it is found that the number of times of spare part failure in different months is quite different. By counting the number of times of spare part failure in spring and summer and in autumn and winter, it is found that the failure of the spare part shows seasonality.

[0048] (3) Influence of daily utilization rate. Airlines mainly engage in passenger transportation. Since the daily utilization rate of the aircraft in the peak season is higher than that in the off-season, and the passenger seat rate in the peak season is high, the loss caused by the lack of spare parts is also high, so the influence of the off-season and the peak season needs to be considered in the spare part support.

[0049] According to the statistics of daily utilization rate in the slack season and the peak season, the daily utilization rate of the aircraft in the peak season is about 1.1 times of the average daily utilization rate, and the daily utilization rate of the aircraft in the slack season is about 0.9 times of the average daily utilization rate, so the average daily utilization rate correction factor is introduced The aircraft parking loss caused by the shortage of spare parts is higher in the peak season than in the slack season, so the aircraft parking loss correction factor is introduced

[0050] (4) Influence of fleet size. According to the fleet size, many domestic airlines can be classified into large, medium-large, medium, medium-small, small airlines. The influence of fleet size on spare parts is obvious, the larger the fleet size, the more the demand for spare parts, and the lower the average spare parts support cost.

[0051] Considering the above influencing factors, when planning spare parts, the fleet size, fleet annual utilization rate and historical failure data need to be obtained first, and the spare parts demand under the influence of these parameters is calculated.

[0052] The component to be predicted described in the embodiment can be any component on the aircraft, such as a display component.

[0053] S2: calculating the working component quantity of the component to be predicted according to the fleet size;

[0054] Specifically, S2 can include determining the number of components to be predicted installed on each aircraft of the aircraft model to which the component to be predicted belongs, and then calculating the product of the fleet size and the number to obtain the working component quantity of the component to be predicted. For example, the component to be predicted is a component of a certain aircraft model, the fleet size of the aircraft model is s, and each aircraft in the fleet has m components installed on it. Then n = s*m components will be in work during normal operation, that is, the working component quantity of the component to be predicted is n.

[0055] S3: calculating the failure rate of the component to be predicted according to the use time after self-repair of each failure of the component to be predicted;

[0056] S3 can include calculating the average value of the use time after self-repair of each failure of the component to be predicted to obtain the mean time between failures, and then calculating the reciprocal of the mean time between failures to obtain the failure rate of the component to be predicted.

[0057] Specifically, the Mean Time Between Failures (MTBF) can be obtained by the TSR recorded in the actual operation of the airline. Assuming that n times of failure data TSR of a component are counted, the sum of the TSR of n times of failure divided by the number n is the average of the TSR of the component after each repair, which is also the average failure interval time MTBF of the component under the current aircraft state. The failure rate calculation formula used in this embodiment is:

[0058]

[0059] wherein λ is the failure rate; MTBF is the average failure interval time; TSR i is the TSR after the i-th failure; i = 1, 2, …, n; n is the total number of failures in the historical maintenance data.

[0060] After obtaining the historical failure data of the component to be predicted, the failure rate of the component to be predicted based on the current aircraft state can be determined by using the above formula.

[0061] In S1, it is found that the failure of some components is affected by the season. Based on this, this embodiment proposes an optimized failure rate calculation method, which considers the influence of the season in the failure rate calculation process. At this time, S3 can include:

[0062] (1) According to the failure date of each failure of the component to be predicted included in the historical failure data, it is determined whether the component to be predicted is a seasonal high-incidence failure component;

[0063] (2) If yes, the failure high-incidence season of the component to be predicted is determined, the failures with the failure date within the failure high-incidence season are selected as the calculation failures, and the average of the TSR of all the calculation failures is calculated to obtain the average failure interval time;

[0064] For example, when analyzing the historical failure data, it is found that 20 failures of the component to be predicted occurred in spring and summer (March to August), and only 2 failures of the component to be predicted occurred in other months. At this time, it can be considered that the component to be predicted is a seasonal high-incidence failure component, and spring and summer are the failure high-incidence season of the component to be predicted. At this time, the failures with the failure date within March to August in the historical failure data are selected as the calculation failures, and the average of the TSR of these calculation failures is calculated to obtain the average failure interval time. It should be noted that 20 times and 2 times are only an example of a seasonal high-incidence failure component in this embodiment. In actual application, those skilled in the art can determine whether the component to be predicted is a seasonal high-incidence failure component according to the needs, and select the failure high-incidence season of the component to be predicted.

[0065] (3) If no, calculate the average value of the self-repairing after time of each failure of the component to be predicted, to obtain the mean time between failures;

[0066] (4) Calculate the reciprocal of the mean time between failures, to obtain the failure rate of the component to be predicted.

[0067] The embodiment aims to solve the optimal spare parts demand, that is, to solve the minimum spare parts under the requirement of meeting the given spare parts guarantee rate. For the target, the embodiment should fully consider the actual operation situation, so the seasonal factor needs to be considered. Therefore, for the seasonal high-failure components, the average time between failures will use seasonal data instead of annual average data. This is because the data of the seasonal high-failure components in the season with high failure rate is more representative than the annual average data, which can more accurately determine the spare parts demand.

[0068] S4: Calculate the repair rate of the component to be predicted according to the fleet annual utilization rate and the repair cycle of each failure of the component to be predicted;

[0069] The unit of TSR monitored by the airline is flight hour (FH), so the unit of failure rate is FH -1 , and the unit of repair rate is also FH -1 . Since the unit of repair cycle RTAT is calendar day, the unit conversion is done by fleet annual utilization rate AUR. Thus, S3 can include: calculating the average value of the repair cycle of each failure of the component to be predicted, to obtain the average repair cycle; using the fleet annual utilization rate and the average repair cycle as inputs, calculating the repair rate of the component to be predicted by using the first repair rate calculation formula.

[0070] The first repair rate calculation formula is as follows:

[0071]

[0072] Wherein, μ is the repair rate; MTTR (Mean Time to Restoration) is the average pre-recovery time; RTAT is the average repair cycle; AUR is the fleet annual utilization rate.

[0073] As another optional implementation, the influencing factors of flight state also include the difference between the daily utilization rate of the aircraft in peak season and off-season. Based on this, S4 can include:

[0074] (1) Determine the average daily utilization rate correction factor according to the prediction period;

[0075] Specifically, determine whether the prediction period belongs to peak season or off-season, and determine the value of the average daily utilization rate correction factor according to the peak season and off-season.

[0076] (2) Calculate the average value of the repair cycle of each failure of the component to be predicted, and obtain the average repair cycle;

[0077] (3) Taking the fleet annual utilization rate, the average daily utilization rate correction factor and the average repair cycle as inputs, the repair rate of the component to be predicted is calculated by using the second repair rate calculation formula.

[0078] The second repair rate calculation formula is as follows:

[0079]

[0080] Wherein, μ is the repair rate; MTTR is the average pre-recovery time; RTAT is the average repair cycle; AUR is the fleet annual utilization rate; θ is the average daily utilization rate correction factor.

[0081] S5: determining the spare part support rate according to the type of the component to be predicted;

[0082] Before S5, the planning method of the embodiment further comprises: setting the spare part support rate of the spare part in multiple levels according to the safety level and the repair time limit requirement, to obtain the spare part support rate corresponding to different types of spare parts.

[0083] Specifically, the multi-level setting of the spare part support rate level can include:

[0084] The existing spare part support rate requirement is usually 0.9, and the spare part support rate level setting is relatively extensive, which is easy to cause spare part redundancy, resulting in a large amount of capital occupation. Therefore, the embodiment combines the classification of civil aircraft spare parts safety level and the different requirements of GO IF part failure repair time limit to set the spare part support rate in multiple levels. According to the safety level, the spare parts can be divided into non-release parts (NO GO parts), conditional release parts (GO IF parts) and release parts (GO parts). According to the repair time limit requirement, it can be divided into four categories: "A", "B", "C" and "D". "A" type: no standard time limit; "B" type: it must be implemented within three consecutive calendar days; "C" type: it must be implemented within ten consecutive calendar days; "D" type: it must be implemented within one hundred and twenty consecutive calendar days. When the spare part is a NO GO part, the spare part support demand is the highest, and the spare part support rate level is the highest. When the spare part is an A type GO IF part, the spare part support demand is higher, and the spare part support rate level is higher. The B type GO IF part is less, and the D type GO IF part has the lowest spare part support rate level among the GO IF parts. In the GO part, in order to meet the passenger demand, the spare parts related to the passenger cabin service should be paid attention to, and the spare part support rate level of the spare parts irrelevant to the passenger cabin service is the lowest.

[0085] Based on this, the embodiment gives a reference setting value of the multi-level spare part guarantee rate requirement based on the safety level classification and repair time limit for the step of setting the spare part guarantee rate level of different types of spare parts, as shown in Table 1 below:

[0086] Table 1

[0087]

[0088] Since the purpose of setting the spare part guarantee rate is to balance safety and benefit, different airlines should set values according to their own business philosophy and actual situation, combined with the operating experience of the airlines, and constantly improve the setting value of the spare part guarantee rate in the actual operation process, so that it meets the ideal operating condition. Therefore, the airlines should set customized spare part guarantee rate levels combined with actual operating experience and business philosophy and constantly improve and modify them in operation to meet the actual ideal operating state, rather than strictly using the values shown in Table 1.

[0089] S6: using the number of working components, the failure rate, the repair rate and the spare part guarantee rate as inputs, using a genetic algorithm to optimize and solve a spare part quantity calculation model to obtain a spare part demand quantity of the to-be-predicted component in a prediction time period; the spare part quantity calculation model is determined according to a Markov birth-death process; the spare part quantity calculation model includes: a reliability corresponding to the spare part demand quantity is greater than or equal to the spare part guarantee rate, and a reliability corresponding to a difference between the spare part demand quantity and 1 is less than the spare part guarantee rate.

[0090] The to-be-predicted component is a component of a certain aircraft type, the fleet size of the aircraft type is s, m components of the component are installed on each aircraft, n components are working in normal operation, the annual utilization rate of the fleet is AUR, unit flight hour (FH), the last repair use time (TSR) and repair cycle (RTAT) of each failure of the component are obtained, λ and μ represent the failure rate and repair rate of the component, and k is set as the purchased spare part quantity of the component. The minimum k value that satisfies the spare part guarantee rate α is solved. The spare part quantity calculation model of the embodiment is used to solve the minimum k value.

[0091] The establishment process of the spare part quantity calculation model of the embodiment includes:

[0092] (1) Consider the spare part failure characteristics of the Markov birth-death process.

[0093] The state of the aircraft system is represented by the number of component failures i, the first n+k+1 states of the system are considered, that is, 0, 1, 2, …, n+k, and the number of working components and the number of inventory spare parts corresponding to each state of the system are as follows:

[0094] State 0: the number of working components is n, and the number of inventory spare parts is k;

[0095] State 1: the number of working components is n, the number of stored spare components is k-1;

[0096] …

[0097] State k: the number of working components is n, the number of stored spare components is 0;

[0098] State k+1: the number of working components is n-1, the number of stored spare components is 0;

[0099] …

[0100] State n+k-1: the number of working components is 1, the number of stored spare components is 0;

[0101] State n+k: the number of working components is 0, the number of stored spare components is 0.

[0102] It can be found that this system can be classified as a Markov birth-death process according to the definition of birth-death process.

[0103] For state i (0≤i≤k), the total failure rate of the system per unit time is nλ; for state i (k+1≤i≤k+n), the total failure rate of the system per unit time is (n+k-i)λ; for state i, the total repair rate of the system is iμ. The transition state diagram of the system is shown in Figure 3 .

[0104] For state 0, the probability of being in the state within Δt time from the state is approximately nλΔt, and the probability of not being transferred is [1-nλΔt]; at t+Δt time, the probability of the system not being transferred is still P0[1-nλΔt], and the probability of being transferred from the state to state 1 is P0nλΔt. The probability of being transferred from the second state to the first state within Δt time is approximately μΔt; at t+Δt time, the probability of the system being transferred from the second state to the first state is P1μΔt. The following formula can be obtained:

[0105] P0(t+Δt)=P0(t)[1-nλΔt]+P1(t)μΔt;

[0106] Let Δt→0, then

[0107]

[0108] Similarly, the following can be obtained:

[0109]

[0110] According to the differential equation, P0(t), P1(t), …, P n+k (t) can be obtained, and Wherein, P0(t) means the probability of the system in state 0, P1(t) means the probability of the system in state 1, P n+k (t) means the probability of the system in state n+k.

[0111] (2) Establish the spare parts quantity calculation model meeting the given spare parts support rate level

[0112] Since the installed quantity and the usage quantity of the component to be predicted are large, the dimension of the differential equation increases, and the solution becomes difficult. The system state is limited, and there must be a stationary distribution, and the component stable working condition is also concerned in the embodiment. This means that when in the limit case, after the system is in complete steady state, the probability P i is a constant, and its differential is 0.

[0113] For state 0, nλP0=μP1, then

[0114] For state 1, (nλ+μ)P1=nλP0+2μP2, then

[0115] …

[0116] For state k, (nλ+kμ)P k =nλP k-1 +(k+1)μP k+1 , then

[0117]

[0118] For state k+1, [(n-1)λ+(k+1)μ]P k+1 =nλP k +(k+2)μP k+2 , then

[0119]

[0120] …

[0121] For state k+n, (n+k)μP k+n =λP k+n-1 , then

[0122]

[0123] In summary, the probability of each state of the system is:

[0124]

[0125] According to the derivation of the total probability formula, it can be obtained that:

[0126]

[0127] So the reliability of the system is:

[0128]

[0129] In combination with the above formula, the corresponding minimum number of spare parts under a given reliability can be easily obtained, and at the same time, the reliability of the system can also be obtained.

[0130] For a given spare part guarantee rate a, the k value obtained according to the following formula is the minimum spare part quantity that meets the spare part guarantee rate a:

[0131]

[0132] The above formula is the spare part quantity calculation model established in this embodiment. For a given spare part guarantee rate requirement, the minimum spare part quantity that meets the spare part guarantee rate can be calculated. After obtaining the spare part quantity calculation model, the number of working components, repair rate, failure rate and spare part guarantee rate are used as inputs, and the genetic algorithm is used to optimize and solve the spare part quantity calculation model, so that the spare part demand quantity k of the component to be predicted in the prediction period can be obtained.

[0133] The genetic algorithm can effectively solve the problem of solving the spare part quantity calculation model, which is an evolutionary algorithm. The "chromosome" is the solution of the genetic algorithm to solve the problem, that is, the individual. In the algorithm, the problem is coded in a certain way, and a population is initialized before running the algorithm. Then the individuals in a certain environment are selected according to the principle of "survival of the fittest", and are copied, and then are exchanged and mutated according to a certain probability to generate a new generation of population, which will be more suitable for the environment. Finally, after many generations of evolution, the best chromosome is obtained, which is most suitable for the environment, that is, the optimal solution of the problem.

[0134] Specifically, as shown in Figure 4 The flowchart of the genetic algorithm is shown. The genetic algorithm is used to optimize and solve the spare part quantity calculation model to obtain the spare part demand quantity of the component to be predicted in the prediction period, which can include:

[0135] (1) randomly determine a plurality of spare part quantities;

[0136] (2) encode each spare part quantity to obtain a plurality of individuals; all individuals form an initial population;

[0137] In the embodiment, "Encoding" represents the encoding mode of the chromosome, 0 in "varTypes" indicates that the decision variable corresponding to the chromosome after decoding is discrete, 1 indicates that the decision variable corresponding to the chromosome after decoding is continuous, "ranges" represents the upper limit of the range matrix of the independent variable, which is greater than the lower limit, and 0 in "borders" indicates that the decision variable does not contain a boundary, and 1 indicates that the decision variable contains a boundary. The embodiment can also create a population chromosome matrix.

[0138] (3) For each individual in the initial population, it is judged whether the individual satisfies the spare part quantity calculation model, and all individuals satisfying the spare part quantity calculation model are selected to form an evolved population;

[0139] It should be noted that the individual satisfying the spare part quantity calculation model means that, after the individual is input into the spare part quantity calculation model, the reliability corresponding to the individual is greater than or equal to the spare part guarantee rate, and the reliability corresponding to the difference between the individual and 1 is less than the spare part guarantee rate.

[0140] (4) The evolved population is selected, exchanged and mutated to obtain an updated population;

[0141] (5) It is judged whether the iteration termination condition is reached;

[0142] In the embodiment, the iteration termination condition can mean that the number of iterations reaches the maximum number of iterations, or the number of individuals in the updated population is 1.

[0143] (6) If yes, the minimum value of the updated population is taken as the spare part demand quantity of the to-be-predicted component in the prediction time period;

[0144] (7) If no, the updated population is taken as the initial population of the next iteration, and the step of "for each individual in the initial population, it is judged whether the individual satisfies the spare part quantity calculation model" is returned.

[0145] After the software runs the genetic algorithm for the guarantee spare part quantity problem, the minimum spare part quantity k satisfying the given spare part guarantee rate requirement can be obtained.

[0146] The embodiment analyzes the influence of seasons on the number of aircraft spare part failures to evaluate the seasonal difference of failure rate and take it into account in the spare part quantity calculation method, analyzes the influence of off-season and peak season on the daily utilization rate and aircraft parking loss to evaluate the seasonal difference of the daily utilization rate and aircraft parking loss and take it into account in the spare part quantity calculation method, analyzes the factors affecting the spare part guarantee of the aircraft state, and evaluates the factors affecting the spare part guarantee in the spare part quantity calculation method, so that the spare part demand quantity can be calculated more accurately.

[0147] Embodiment 2:

[0148] The embodiment is used for an aircraft state-based spare part dynamic programming system, as shown in Figure 5 The programming system comprises:

[0149] A first data acquisition module M1 is configured to acquire a fleet size of a model to which a component to be predicted belongs and a fleet annual utilization rate, and acquire historical failure data of the component to be predicted; the historical failure data comprises a self-repairing after use time and a repair period of each failure of the component to be predicted;

[0150] A calculation module M2 is configured to calculate a working component quantity of the component to be predicted according to the fleet size; calculate a failure rate of the component to be predicted according to the self-repairing after use time of each failure of the component to be predicted; calculate a repair rate of the component to be predicted according to the fleet annual utilization rate and the repair period of each failure of the component to be predicted; and determine a spare part support rate according to a type of the component to be predicted;

[0151] A first prediction module M3 is configured to take the working component quantity, the failure rate, the repair rate and the spare part support rate as inputs, and use a genetic algorithm to optimize and solve a spare part quantity calculation model to obtain a spare part demand quantity of the component to be predicted in a prediction time period; the spare part quantity calculation model is determined according to a Markov birth-death process; and the spare part quantity calculation model comprises: a reliability corresponding to the spare part demand quantity is greater than or equal to the spare part support rate, and a reliability corresponding to a difference between the spare part demand quantity and 1 is less than the spare part support rate.

[0152] Embodiment 3

[0153] In the economic-based spare part programming, the spare part cost usually only covers the procurement cost, without considering the aircraft parking loss cost, or considering the loss cost without considering the factors such as the difference between the off-season and the peak season. Therefore, in combination with the actual work of the aviation spare part management, the embodiment fully considers the influence of the aircraft state on the spare part demand, adopts the Markov birth-death process conforming to the characteristics of the spare part maintenance to construct a spare part quantity calculation model, and constructs a spare part quantity dynamic programming model based on the spare part cost analysis under the premise of meeting a given spare part support rate, so as to calculate the optimal spare part quantity that meets the requirement of the spare part support rate and makes the total support cost lowest, thereby providing technical method support for the spare part programming of the airlines and realizing the control of the economic cost and the improvement of the enterprise benefits.

[0154] Therefore, the embodiment is used for providing an aircraft state-based spare part dynamic programming method, as shown in Figure 6 The programming method comprises:

[0155] T1: Obtain the fleet size and fleet annual utilization rate of the aircraft model to which the component to be predicted belongs, and obtain the historical failure data and spare part unit price of the component to be predicted; the historical failure data includes the self-repairing use time and repair cycle of each failure of the component to be predicted;

[0156] T2: Calculate the number of working components of the component to be predicted according to the fleet size;

[0157] T2 of the present embodiment is the same as S2 of embodiment 1, and will not be repeated here.

[0158] T3: Calculate the failure rate of the component to be predicted according to the self-repairing use time of each failure of the component to be predicted;

[0159] T3 of the present embodiment is the same as S3 of embodiment 1, and will not be repeated here.

[0160] T4: Calculate the repair rate of the component to be predicted according to the fleet annual utilization rate and the repair cycle of each failure of the component to be predicted;

[0161] T4 of the present embodiment is the same as S4 of embodiment 1, and will not be repeated here.

[0162] T5: Determine the spare part guarantee rate according to the type of the component to be predicted;

[0163] T5 of the present embodiment is the same as S5 of embodiment 1, and will not be repeated here.

[0164] T6: Take the spare part unit price, the number of working components, the failure rate, the repair rate and the spare part guarantee rate as inputs, and use genetic algorithm to optimize and solve the spare part quantity dynamic programming model to obtain the spare part demand quantity of the component to be predicted in the prediction period; the spare part quantity dynamic programming model includes an objective function and a constraint condition; the objective function is to minimize the total guarantee cost; the constraint condition is that the reliability corresponding to the spare part demand quantity is greater than or equal to the spare part guarantee rate.

[0165] The establishment process of the spare part quantity dynamic programming model based on total spare part guarantee cost analysis of the present embodiment includes:

[0166] (1) Total spare part guarantee cost analysis:

[0167] According to the actual operation of the airline, the spare parts will be immediately arranged from the third party for help after the lack of spare parts, the consumption parts are purchased by price allocation, the turnover parts are short-term leased, and the total support cost of the spare parts is composed of the spare parts procurement cost, the spare parts maintenance cost, the spare parts storage cost, the spare parts leasing fee and the aircraft parking loss. The embodiment aims at the spare parts quantity planning, and the correlation between the spare parts maintenance cost and the spare parts quantity is not considered, so that the spare parts maintenance cost is not calculated in the dynamic planning model of the spare parts quantity, but the maintenance cost is not zero. At the same time, the spare parts storage cost is very low and can be ignored, and the transportation cost of the leased spare parts is also very low and can be ignored. Therefore, the total support cost in the dynamic planning model of the spare parts is composed of the procurement cost, the lack of parts cost and the aircraft parking loss. Specifically as follows:

[0168] C T For the total support cost, the objective of the model is to minimize the total support cost.

[0169] C P For the procurement cost, that is, the product of the unit price UP of the spare parts and the quantity k of the spare parts.

[0170] C S For the lack of parts cost, that is, the cost of leasing spare parts after the lack of parts occurs, which is equal to the product of the number of the lack of parts, the probability, the repair cycle and the daily rental. The calculation formula of the lack of parts cost is as follows:

[0171]

[0172] Wherein, R D is the daily rental of the spare parts. According to the mutual aid agreement signed by the known airline and the mutual aid unit, the benchmark price of the leased available parts (non-new) is 72% of the catalog price, and the daily rental paid by the lessee is 0.45% of the benchmark price, that is, R D = UP x 72% x 0.45%.

[0173] C O is the parking loss, when the NO GO parts are in the lack of parts condition, the aircraft will be parked and the parking loss will be caused. For other types of spare parts, the cost of this item is zero. The aircraft parking loss caused by the lack of spare parts in the peak season is also higher than that in the off-season. When calculating, a correction factor ψ is multiplied, and the value of the aircraft parking loss correction factor ψ is determined according to the time period to be guaranteed.

[0174] The calculation formula of the parking loss is:

[0175]

[0176] Wherein, C g is the ground delay loss; C a is the air delay loss; C m is the accommodation compensation loss; C w is the reputation loss; C mbT is the main business cost; T d T is the flight delay length; T is the total flight and delay length; x c T is the flight delay probability caused by the carrier; P is the net profit; V is the average ticket value of passengers; x d T is the flight delay probability; F is the passenger transport volume.

[0177] (2) A dynamic spare parts quantity planning model is constructed: the model takes the minimum total support cost as the objective function, and satisfies the given spare parts support rate as the constraint condition.

[0178] The dynamic spare parts quantity planning model is as follows:

[0179]

[0180] The above formula is expanded as:

[0181]

[0182]

[0183] The embodiment utilizes the genetic algorithm to optimize and solve the dynamic spare parts quantity planning model, and the spare parts demand quantity of the to-be-predicted component in the prediction time period can include:

[0184] (1) A plurality of spare parts quantities are randomly determined;

[0185] (2) Each spare parts quantity is coded to obtain a plurality of individuals; all the individuals form an initial population;

[0186] (3) For each individual in the initial population, the objective function value of the individual is calculated, and the individuals satisfying the constraint condition are selected to form an evolution population;

[0187] (4) The evolution population is selected, exchanged and mutated to obtain an updated population;

[0188] (5) It is judged whether the iteration termination condition is reached;

[0189] (6) If yes, the individual with the minimum objective function value and satisfying the constraint condition in the updated population is taken as the spare parts demand quantity of the to-be-predicted component in the prediction time period;

[0190] (7) If no, the updated population is taken as the initial population of the next iteration, and the step of “for each individual in the initial population, the objective function value of the individual is calculated” is returned.

[0191] After the genetic algorithm of the above spare parts quantity dynamic planning problem is run through software, the optimal spare parts quantity k that makes the total support cost minimum under the satisfaction of the given support rate requirement can be obtained.

[0192] In this embodiment, the display assembly of the E190 aircraft model of H Airlines is taken as an example to calculate and optimize the spare parts quantity by using the above model.

[0193] The display assembly of the E190 aircraft model is subject to an exponential distribution, each aircraft is equipped with 5 display assemblies, the fleet size of this aircraft model is 6, the annual utilization rate of the fleet is 3000 FH, the historical failure data is shown in Table 2, the display assembly is a GO IF part, the repair time limit is A, the unit price of the equipment is 146000 US dollars, and the display assembly spare parts quantity during the summer operation needs to be determined.

[0194] Table 2: Historical failure data of display assembly

[0195]

[0196] According to the date of failure, the above failure data is divided into spring and summer season and autumn and winter season (April to September is spring and summer season, October to next March is autumn and winter season), and the summary can be obtained that the number of failures of the display assembly in spring and summer season is 13, and the MTBF is 6923.64 FH; the number of failures of the display assembly in autumn and winter season is 2, and the MTBF is 8962.20 FH. From the above historical failure data, it can be seen that the failure of the display assembly occurs mainly in spring and summer season (April to September), the failure interval time in spring and summer season is relatively short, and the failure rate is relatively high. Considering the demand during the summer operation, the failure rate in spring and summer season is used for spare parts quantity calculation, that is, the failure rate

[0197]

[0198] The average repair cycle RTAT is 45 days, and considering that the summer operation period belongs to the peak season, the daily utilization rate correction factor is added to the calculation, and the repair rate is:

[0199]

[0200] Since the display assembly belongs to a GO IF part with a repair time limit of A, referring to Table 1, the spare parts support rate a is set to 0.95. The number of working components is:

[0201] n = s x m = 30;

[0202] The above data is substituted into the Markov birth and death process spare parts quantity calculation model, and the genetic algorithm is run by using software to obtain that 4 spare parts of the display assembly are purchased, which can meet the requirement of support rate a ≥ 0.95 in the peak season.

[0203] Now the display assembly of the E190 aircraft model is dynamically planned based on economy as follows. For H Airlines, the display assembly of the E190 aircraft is a GO IF part, which can be handled for failure reservation, and the loss of parking is not considered, that is, C O = 0; the daily rental is R D= UP x 72% x 0.45%, substituting into the economic spare parts quantity dynamic programming model, and using software to run genetic algorithm can obtain that the total guarantee cost is the lowest when the spare parts quantity is 4 under the condition of meeting the given guarantee rate constraint, that is, when the spare parts quantity is 4, the guarantee rate requirement is met and the guarantee cost is minimized, which is the optimal spare parts quantity obtained by the spare parts quantity dynamic programming.

[0204] In fact, the current spare parts quantity of the E190 display assembly of H Airlines is 4, and the guarantee state is ideal, so it can be known that the calculation result of the embodiment is consistent with the actual operation of the airlines. Through the application of the spare parts planning case of the embodiment, it is illustrated that the spare parts dynamic programming method based on the aircraft state proposed in the embodiment can improve the spare parts demand prediction accuracy, thereby effectively controlling the spare parts cost and improving the efficiency of the airlines.

[0205] The spare parts dynamic programming method of the embodiment sets multiple levels of spare parts guarantee rate according to the safety level and fault repair time limit of civil aircraft spare parts, considers the spare parts fault characteristics of Markov birth-death process, constructs a spare parts quantity calculation model meeting the spare parts guarantee rate level, analyzes the spare parts guarantee influencing factors related to the aircraft state, evaluates the influencing factors in the spare parts quantity calculation method, and then establishes a spare parts quantity dynamic programming model based on spare parts cost analysis. Compared with the prior art, the spare parts planning method improves the spare parts demand prediction accuracy, makes the spare parts guarantee planning more consistent with the actual situation, effectively controls the economic cost under the premise of ensuring the spare parts guarantee rate requirement, improves the efficiency of the airlines, and thus can provide technical method support for the spare parts planning of the airlines.

[0206] Embodiment 4:

[0207] The embodiment is used to provide a spare parts dynamic programming system based on an aircraft state, as shown in Figure 7 The programming system comprises:

[0208] The second data acquisition module M4 is used to acquire the fleet size and fleet annual utilization rate of the aircraft model to which the to-be-predicted component belongs, and acquire the historical fault data and spare parts unit price of the to-be-predicted component; the historical fault data comprises the self-repairing after use time and repair cycle of each fault of the to-be-predicted component;

[0209] The calculation module M2 is used to calculate the number of working components of the to-be-predicted component according to the fleet size; calculate the failure rate of the to-be-predicted component according to the self-repairing after use time of each fault of the to-be-predicted component; calculate the repair rate of the to-be-predicted component according to the fleet annual utilization rate and the repair cycle of each fault of the to-be-predicted component; and determine the spare parts guarantee rate according to the type of the to-be-predicted component;

[0210] The second prediction module M5 is configured to, taking the spare part unit price, the number of working components, the failure rate, the repair rate and the spare part guarantee rate as inputs, use a genetic algorithm to optimize and solve a spare part quantity dynamic programming model to obtain the spare part demand quantity of the component to be predicted in a prediction time period; the spare part quantity dynamic programming model comprises a target function and a constraint condition; the target function is to minimize the total guarantee cost; and the constraint condition is that the reliability corresponding to the spare part demand quantity is greater than or equal to the spare part guarantee rate.

[0211] The various embodiments are described in a progressive manner in the specification, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be mutually referred to. For the system disclosed by the embodiments, since it corresponds to the method disclosed by the embodiments, the description is relatively simple, and the relevant parts can be referred to the method part.

[0212] The principles and implementation manners of the present application are described by using specific examples in the specification. The above description of the embodiments is only used to help understand the method of the present application and its core idea. Meanwhile, for those skilled in the art, the specific implementation manners and application ranges can be changed according to the idea of the present application. In conclusion, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A dynamic programming method for spare parts based on aircraft state, characterized in that, The planning method includes: Obtain the fleet size and annual fleet utilization rate of the aircraft type to which the component to be predicted belongs, and obtain the historical fault data of the component to be predicted; the historical fault data includes the usage time and repair cycle after each fault of the component to be predicted. Calculate the number of working parts of the component to be predicted based on the fleet size; The failure rate of the component to be predicted is calculated based on the self-repair usage time after each failure of the component to be predicted. Specifically, this includes: determining whether the component to be predicted belongs to a seasonally high-incidence failure component based on the failure date of each failure of the component to be predicted included in the historical failure data; if so, determining the high-incidence season of the component to be predicted, selecting failures with failure dates within the high-incidence season as calculation failures, and calculating the average self-repair usage time of all the calculation failures to obtain the average failure interval time; if not, calculating the average self-repair usage time of each failure of the component to be predicted to obtain the average failure interval time; and calculating the reciprocal of the average failure interval time to obtain the failure rate of the component to be predicted. The repair rate of the component to be predicted is calculated based on the annual fleet utilization rate and the repair cycle of each failure of the component to be predicted. Specifically, this includes determining the average daily utilization rate correction factor based on the prediction period, specifically determining whether the prediction period belongs to the peak season or the off-season, in order to determine the average daily utilization rate correction factor. Calculate the average repair cycle for each failure of the component to be predicted, and obtain the average repair cycle; using the fleet annual utilization rate, the average daily utilization rate correction factor, and the average repair cycle as inputs, utilize... Calculate the repair rate of the component to be predicted, wherein, For repair rate; MTTR This represents the average time before recovery. RTAT Mean repair cycle; AUR For fleet annual utilization rate; This is a correction factor for average daily utilization. Based on safety levels and repair time limits, the spare parts availability rate is set at multiple levels to obtain the spare parts availability rate corresponding to different types of spare parts. The spare parts availability rate is determined according to the type of the component to be predicted. The safety levels include non-released parts, conditionally released parts, and releaseable parts. The repair time limits include categories A, B, C, and D. Category A refers to a time limit without a specified standard. Category B refers to the corresponding repair work that must be completed within three consecutive calendar days. Category C refers to the corresponding repair work that must be completed within ten consecutive calendar days. Category D refers to the corresponding repair work that must be completed within one hundred and twenty consecutive calendar days. Using the number of working parts, the failure rate, the repair rate, and the spare parts availability rate as inputs, a genetic algorithm is used to optimize and solve the spare parts quantity calculation model to obtain the spare parts demand of the component to be predicted during the prediction period. The spare parts quantity calculation model is determined based on a Markov birth-death process. The spare parts quantity calculation model includes: the reliability corresponding to the spare parts demand is greater than or equal to the spare parts availability rate, and the reliability corresponding to the difference between the spare parts demand and 1 is less than the spare parts availability rate. The spare parts quantity calculation model is as follows: ； in, k For spare parts demand; n Number of working parts; Failure rate; For repair rate; To ensure spare parts availability.

2. The planning method according to claim 1, characterized in that, The optimization and solution of the spare parts quantity calculation model using a genetic algorithm to obtain the spare parts demand of the component to be predicted during the prediction period specifically includes: Randomly determine the quantity of multiple spare parts; Each of the aforementioned spare parts quantities is encoded to obtain multiple individuals; all of the aforementioned individuals constitute an initial population. For each individual in the initial population, determine whether the individual satisfies the spare parts quantity calculation model, and select all individuals that satisfy the spare parts quantity calculation model to form the population to be evolved. The population to be evolved is selected, exchanged, and mutated to obtain an updated population; Determine whether the iteration termination condition has been met; If so, the minimum value of the updated population shall be used as the spare parts requirement of the component to be predicted during the prediction period. If not, then the updated population is used as the initial population for the next iteration, and the step of "for each individual in the initial population, determine whether the individual satisfies the spare parts quantity calculation model" is returned.

3. A spare parts dynamic programming system based on aircraft state, used to implement the spare parts dynamic programming method based on aircraft state as described in any one of claims 1-2, characterized in that, The planning system includes: The first data acquisition module is used to acquire the fleet size and annual fleet utilization rate of the aircraft model to which the component to be predicted belongs, and to acquire the historical fault data of the component to be predicted; the historical fault data includes the usage time and repair cycle after each fault of the component to be predicted. The calculation module is used to calculate the number of working parts of the component to be predicted based on the fleet size; calculate the failure rate of the component to be predicted based on the self-repair usage time after each failure of the component to be predicted; calculate the repair rate of the component to be predicted based on the annual utilization rate of the fleet and the repair cycle of each failure of the component to be predicted; and determine the spare parts availability rate based on the type of the component to be predicted. The first prediction module is used to optimize and solve the spare parts quantity calculation model using a genetic algorithm, taking the number of working parts, the failure rate, the repair rate, and the spare parts availability rate as inputs, to obtain the spare parts demand of the part to be predicted during the prediction period. The spare parts quantity calculation model is determined according to the Markov birth-death process. The spare parts quantity calculation model includes: the reliability corresponding to the spare parts demand is greater than or equal to the spare parts availability rate, and the reliability corresponding to the difference between the spare parts demand and 1 is less than the spare parts availability rate.

4. A dynamic programming method for spare parts based on aircraft status, characterized in that, The planning method includes: Obtain the fleet size and annual fleet utilization rate of the aircraft model to which the component to be predicted belongs, and obtain the historical fault data and spare parts unit price of the component to be predicted; the historical fault data includes the usage time and repair cycle after each fault of the component to be predicted. Calculate the number of working parts of the component to be predicted based on the fleet size; The failure rate of the component to be predicted is calculated based on the self-repair usage time after each failure of the component to be predicted. Specifically, this includes: determining whether the component to be predicted belongs to a seasonally high-incidence failure component based on the failure date of each failure of the component to be predicted included in the historical failure data; if so, determining the high-incidence season of the component to be predicted, selecting failures with failure dates within the high-incidence season as calculation failures, and calculating the average self-repair usage time of all the calculation failures to obtain the average failure interval time; if not, calculating the average self-repair usage time of each failure of the component to be predicted to obtain the average failure interval time; and calculating the reciprocal of the average failure interval time to obtain the failure rate of the component to be predicted. The repair rate of the component to be predicted is calculated based on the annual fleet utilization rate and the repair cycle of each failure of the component to be predicted. Specifically, this includes determining the average daily utilization rate correction factor based on the prediction period, specifically determining whether the prediction period belongs to the peak season or the off-season, in order to determine the average daily utilization rate correction factor. Calculate the average repair cycle for each failure of the component to be predicted, and obtain the average repair cycle; using the fleet annual utilization rate, the average daily utilization rate correction factor, and the average repair cycle as inputs, utilize... Calculate the repair rate of the component to be predicted, wherein, For repair rate; MTTR This represents the average time before recovery. RTAT Mean repair cycle; AUR For fleet annual utilization rate; This is a correction factor for average daily utilization. Based on safety levels and repair time limits, the spare parts availability rate is set at multiple levels to obtain the spare parts availability rate corresponding to different types of spare parts. The spare parts availability rate is determined according to the type of the component to be predicted. The safety levels include non-released parts, conditionally released parts, and releaseable parts. The repair time limits include categories A, B, C, and D. Category A refers to a time limit without a specified standard. Category B refers to the corresponding repair work that must be completed within three consecutive calendar days. Category C refers to the corresponding repair work that must be completed within ten consecutive calendar days. Category D refers to the corresponding repair work that must be completed within one hundred and twenty consecutive calendar days. Using the spare parts unit price, the number of working parts, the failure rate, the repair rate, and the spare parts availability rate as inputs, a genetic algorithm is used to optimize and solve the spare parts quantity dynamic programming model to obtain the spare parts demand of the component to be predicted during the prediction period. The spare parts quantity dynamic programming model includes an objective function and constraints. The objective function is to minimize the total support cost. The constraint is that the reliability corresponding to the spare parts demand is greater than or equal to the spare parts availability rate.

5. A spare parts dynamic programming system based on aircraft state, used to implement the spare parts dynamic programming method based on aircraft state as described in claim 4, characterized in that, The planning system includes: The second data acquisition module is used to acquire the fleet size and annual fleet utilization rate of the aircraft model to which the component to be predicted belongs, and to acquire the historical fault data and spare parts unit price of the component to be predicted; the historical fault data includes the self-repair usage time and repair cycle of each fault of the component to be predicted. The calculation module is used to calculate the number of working parts of the component to be predicted based on the fleet size; calculate the failure rate of the component to be predicted based on the self-repair usage time after each failure of the component to be predicted; calculate the repair rate of the component to be predicted based on the annual utilization rate of the fleet and the repair cycle of each failure of the component to be predicted; and determine the spare parts availability rate based on the type of the component to be predicted. The second prediction module is used to optimize and solve the spare parts quantity dynamic programming model using a genetic algorithm, taking the spare parts unit price, the number of working parts, the failure rate, the repair rate, and the spare parts availability rate as inputs, to obtain the spare parts demand of the component to be predicted during the prediction period. The spare parts quantity dynamic programming model includes an objective function and constraints. The objective function is to minimize the total maintenance cost. The constraints are that the reliability corresponding to the spare parts demand is greater than or equal to the spare parts availability rate.

Images