A method for establishing a forecasting model for aviation maintenance support resource demand

Abstract

The invention discloses a method for establishing an aviation maintenance and support resource demand prediction model, which relates to the field of supply and demand guarantee of aviation maintenance and support resources. By analyzing the characteristic types of aviation maintenance and support equipment, a prediction demand quantity model calculation method of four types of aviation maintenance and support equipment, namely, irrelevant life distribution, exponential life distribution, normal life distribution and Weibull life distribution, is established. By comparing the characteristics of the four model calculation methods and combining the actual needs of aviation maintenance and support resource demand, an algorithm step design and implementation are performed using a prediction demand quantity calculation method of aviation maintenance and support equipment of the normal life distribution type. Prediction-related input data of aviation maintenance and support equipment is input, and calculation is performed using an algorithm of a prediction model of aviation maintenance and support equipment of the normal life distribution type. The result of an aviation maintenance and support equipment demand prediction scheme is output, including a name of a support equipment, a time period, a use location and a predicted quantity.

Description

Technical Field

[0001] The present invention relates to the field of supply and demand guarantee of aviation maintenance support resources, and in particular to a method for establishing an aviation maintenance support resource demand prediction model. Background Art

[0002] Aviation maintenance support resources collectively refer to the manpower, material resources, funds, information, and time required for aviation maintenance support, primarily encompassing material resources, human resources, financial resources, and information resources. Aviation maintenance support resources play a vital role in ensuring flight safety. Aviation maintenance support equipment generally refers to the material resources within aviation maintenance support resources, primarily referring to the equipment and tools used for the operation and maintenance of aviation equipment. With the rapid development of the aviation industry, the types, models, and quantities of new technologies and materials used in aviation maintenance equipment are increasing. The time period for the use of aviation maintenance equipment is being increasingly refined, and its use is expanding in a wider range of locations. Consequently, the demand for aviation maintenance equipment is increasing, and ensuring the timely supply of both the supply cycle and the quantity of aviation maintenance equipment is becoming increasingly important in the aviation maintenance support sector.

[0003] Traditional methods for supplying aviation maintenance equipment typically include: reserved inventory, cross-location transfers, and centralized, on-demand batch purchasing. Due to the diverse types and models of aviation maintenance equipment, using reserved inventory can overwhelm maintenance warehouse resources and lead to a large backlog of underutilized equipment, wasting maintenance resources. Cross-location transfers can disrupt the equipment's useful lifecycle and make management difficult. And centralized, on-demand batch purchasing can lead to lengthy procurement times for some scarce equipment, reducing maintenance efficiency. Summary of the Invention

[0004] Based on the technical problems existing in the background technology, the present invention proposes a method for establishing an aviation maintenance and support resource demand forecasting model, which relates to the field of supply and demand guarantee of aviation maintenance and support resources. By analyzing the characteristic types of aviation maintenance and support equipment, four types of forecast demand quantity model calculation methods for aviation maintenance and support equipment are established, namely, independent life distribution, exponential life distribution, normal life distribution and Weibull life distribution. By comparing the characteristics of the four model calculation methods and combining them with the actual needs of aviation maintenance and support resource demand, the forecast demand quantity calculation method of aviation maintenance and support equipment of the normal life distribution type is adopted to design and implement the algorithm steps, input the prediction-related input data of the aviation maintenance and support equipment, calculate through the calculation algorithm of the aviation maintenance and support equipment prediction model of the normal life distribution type, and output the results of the aviation maintenance and support equipment demand forecasting plan, including the name of the support equipment, time period, usage location and forecast quantity.

[0005] The present invention provides the following technical solution: a method for establishing an aviation maintenance support resource demand forecasting model, comprising the following steps:

[0006] S1, based on the characteristic types of aviation maintenance and support equipment, classify the life distribution of aviation maintenance and support equipment;

[0007] S2, based on the life distribution type of aviation maintenance and support equipment, performing calculations according to the methods or steps for each life distribution type to obtain a calculation model for the predicted demand quantity of aviation maintenance and support equipment for each life distribution type;

[0008] S3, by analyzing the characteristics of the calculation models for the predicted demand quantity of aviation maintenance and support equipment of each life distribution type, select the normal life distribution type as the calculation model for the predicted demand quantity of aviation maintenance and support equipment;

[0009] S4, design and implement algorithm steps based on the calculation model of the predicted demand quantity of aviation maintenance and support equipment with normal life distribution type;

[0010] S5 provides pre-set standard data as fixed data for the algorithm, inputs prediction-related input data of aviation maintenance and support equipment, calculates through the algorithm of the prediction model, and outputs the results of the prediction plan.

[0011] As a further improvement of the present invention, the life distribution classification types of the aviation maintenance and support equipment described in step S1 include: independent life distribution type, exponential life distribution type, normal life distribution type and Weibull life distribution type.

[0012] As a further improvement of the present invention, the calculation methods of the respective lifetime distribution types described in step S2 include: a calculation method of an independent lifetime distribution type, a calculation method of an exponential lifetime distribution type, a calculation method of a normal lifetime distribution type, and a calculation method of a Weibull lifetime distribution type.

[0013] As a further improvement of the present invention, the calculation steps of the type of irrelevant lifetime distribution in step S2 are as follows:

[0014] A01, taking availability P as the standard reliability, calculate the corresponding reliable life t according to the life distribution type of the support equipment model k kp , the mathematical expression of availability P is as follows:

[0015]

[0016] MTTF (Mean Time To Failure) refers to the average time a device maintains trouble-free operation. For multiple devices of the same type, the average value of the time period from the start of normal operation to the occurrence of a failure is taken, generally expressed in years.

[0017] MTBF (Mean Time Between Failure) refers to the average time between two failures of a device. It is the average of multiple devices of the same type and is generally measured in years.

[0018] A02, with t kp is the update cycle (usually calculated in units of one year), and the number of updates nk required within the guaranteed time period T is calculated;

[0019] A03,n k Multiply by the number of devices that need this protection in a certain type of machine i (number of machines) M i And the total number of aircraft of this type N in preset point j j , that is, the number S of updated security equipment model k required for model i is obtained ki ;

[0020] A04, classify and accumulate all S ki , obtain the predicted demand quantity of support equipment.

[0021] Following the above technical solution, the calculation method of the exponential life distribution type described in step S2 is as follows:

[0022] B01, the distribution function expression of exponential distribution is as follows:

[0023] F(t)=1–e -λt (2)

[0024] Where: λ is the update rate of the support equipment, t is the update period; its availability R(t) is expressed as follows:

[0025] R(t)=1–F(t)=e -λt (3)

[0026] B02, from expressions (2) and (3), we can get:

[0027] R(t p )=e -λtP =P (4)

[0028] B03, solve t from expression (4) p The expression is as follows:

[0029]

[0030] B04, the calculation model expression for the equipment forecast demand quantity of the exponential distribution type is as follows:

[0031]

[0032] Where: M i N is the number of devices that need this protection in a certain type of machine i (number of machines used), j is the total number of aircraft of this type in the preset point j, λ is the update rate of the support equipment, T is the support time period, and P is the availability.

[0033] As a further improvement of the present invention, the calculation method of the normal life distribution type described in step S2 is as follows:

[0034] C01, the distribution function expression of normal distribution is as follows:

[0035]

[0036] in:

[0037] μ is the mean of the normal distribution, which contains a statistical set of N samples and is expressed as:

[0038]

[0039] σ is the standard deviation, which contains the statistical set of N samples and is expressed as:

[0040]

[0041] The expression of its availability is as follows:

[0042]

[0043] C02, from expressions (7) and (10), we can get:

[0044]

[0045] C03, solve t from expression (11) p The expression is as follows:

[0046] t p =μ+u p ·σ (12)

[0047] Where: u p is the standard normal distribution quantile, which can be found from the statistical distribution value table. The commonly used up values ​​are shown in the following table:

[0048]

[0049] C04, the calculation model expression for the equipment forecast demand quantity of the normal distribution type is as follows:

[0050]

[0051] Where: M i N is the number of devices that need this protection in a certain type of machine i (number of machines used), j is the total number of aircraft of this model at preset point j.

[0052] As a further improvement of the present invention, in step S2, the calculation method of the Weibull distribution type is as follows:

[0053] D01, the distribution function expression of Weibull distribution is as follows:

[0054]

[0055] Where: t is the life span (can be Weibull distribution with frequency or time), t0 is the minimum life span, t a is the characteristic life, b is the shape parameter;

[0056] The expression of its availability is:

[0057]

[0058] D02, from expressions (14) and (15), we can get:

[0059]

[0060] D03, solve t from expression (16) p The expression is as follows:

[0061]

[0062] D04, the calculation model expression of the equipment forecast demand quantity obtained by Weibull distribution is as follows:

[0063]

[0064] Where: M i N is the number of devices that need this protection in a certain type of machine i (number of machines used), j is the total number of aircraft of this type in the preset point j, T is the guarantee period, P is the availability, t0 is the minimum life, t a Characteristic lifespan.

[0065] As a further improvement of the present invention, the algorithm steps for predicting the required quantity of aviation maintenance and support equipment of the normal life distribution type described in step S4 are as follows:

[0066] E01, initialization, sets the standard data required for the algorithm to run, including:

[0067] Construct a standard normal distribution quantile table,

[0068] Read in the preset quota standards for security equipment;

[0069] E02, read in the necessary data required for prediction, including:

[0070] The time period for entering historical data,

[0071] Enter the training plan and actual configuration values ​​of each equipment to be predicted at each airport in the past years within the time period in chronological order;

[0072] E03, preprocessing, generates the necessary parameters required for prediction, including:

[0073] For a certain pre-installed equipment, calculate the theoretical demand value for each year based on historical data.

[0074] Calculate the difference between the theoretical demand value and the actual configuration value for each year for this specific pre-installed equipment.

[0075] Normalize all annual differences by dividing them by the annual demand value.

[0076] The average value of the normalized data of all airports in each year for this specific preset equipment within the historical data period is summarized on an annual basis.

[0077] Calculate the mean μ and standard deviation σ of the normal distribution,

[0078] Temporarily store the normal distribution parameters μ and σ of the specific device;

[0079] E04, set the time period to be predicted and the parameters of the prediction task, including:

[0080] Set the time period to be predicted and the type of prediction task,

[0081] Enter the training plan for the predicted time period.

[0082] Check the quotas of all support equipment at all airports in the plan and calculate the theoretical demand value R;

[0083] E05, enter the availability P of each device to be predicted i ;

[0084] E06, repeat the following process for all airports in the plan within the forecast time period, including:

[0085] Calculate the standard normal distribution quantile u from the equipment availability Pi lookup table Pi ,

[0086] The difference in future demand is predicted based on the equipment update calculation model of the normal distribution type.

[0087] According to the task type, check the margin table to correct the predicted difference Δi,

[0088] Add the predicted value of the equipment difference to the theoretical demand value R of the equipment i ;

[0089] E07, summarize the predicted total amount of all types of pre-installed equipment by airport classification and output.

[0090] As a further improvement of the present invention, the pre-set standard data in step S5 includes prefabrication location information, pre-placed support equipment quota information and predicted task type information.

[0091] As a further improvement of the present invention, the prediction-related input data described in step S5 include a training plan for prediction, historical preset details, and the type of prediction task; the result of the prediction scheme output through the algorithm calculation of the prediction model described in step S5 includes intermediate data and output data, and the output data includes the name of the security equipment, time period, usage location, and predicted quantity.

[0092] Compared with the prior art, the present invention has the following beneficial effects:

[0093] First, the method for establishing an aviation maintenance and support resource demand forecasting model disclosed herein can output the results of an aviation maintenance and support resource demand forecasting plan based on the forecasted input data for aviation maintenance and support equipment. The forecasting plan includes the name of the equipment, time period, location of use, and predicted quantity. Based on the one-to-one correspondence between the predicted quantity, usage time, and location of use in the output results, the quantity of maintenance and support equipment required for that location and time of use can be pre-ordered, eliminating the problem of excessive inventory usage in maintenance warehouses.

[0094] Secondly, the advance determination of the time and location of use effectively solves the technical problems of confusion in the usage cycle and difficulty in management caused by the mutual location calling method; in the case of insufficient inventory, it can timely purchase and prepare materials in advance according to the predicted quantity in the output results, solving the problem of reduced efficiency of aviation equipment maintenance due to long procurement time of maintenance equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0096] Figure 1 A flow chart of the steps for implementing the present invention;

[0097] Figure 2 A flowchart for calculating the forecast demand quantity model regardless of the life distribution type according to the present invention;

[0098] Figure 3 A flowchart for calculating the forecast demand quantity model of the exponential life distribution type according to the present invention;

[0099] Figure 4 This is a flowchart for calculating the predicted demand quantity model of the normal life distribution type described in the present invention;

[0100] Figure 5 A flowchart for calculating the predicted demand quantity model of the Weibull life distribution type according to the present invention;

[0101] Figure 6 A flowchart of the algorithm implementation steps for predicting the required quantity of aviation maintenance and support equipment of the normal life distribution type according to the present invention;

[0102] Figure 7 A schematic diagram of pre-set standard data for a specific embodiment of the present invention;

[0103] Figure 8 This is a schematic diagram of input data for a specific embodiment of the present invention;

[0104] Figure 9Schematic diagram of result data of a specific embodiment of the present invention.

[0105] In the figure: 1. Pre-set standard data; 2. Prefabrication location information; 3. Prefabrication support equipment quota information; 4. Forecast task type information; 5. Prefabrication data name; 6. Prefabrication point location name; 7. Prefabrication area; 8. Prefabrication point capacity; 9. Prefabrication quota table; 10. Prefabrication equipment table; 11. Prefabrication task type; 12. Prefabrication coefficient; 13. Quota number; 14. Aircraft model; 15. Prefabrication equipment; 16. Unit of measurement; 17. Number of aircraft; 18. Number of equipment; 19. Equipment name; 20. Model; 21. Price; 22. Volume; 23. Input data; 24. Forecast training plan; 25. Historical preset details; 26. Forecast task type; 27. Plan editing No.; 28. Time period; 29. ​​Prefabrication area; 30. Airport number; 31. Aircraft type; 32. Quantity; 33. Historical plan table; 34. Historical preset table; 35. Historical number; 36. Historical plan time period; 37. Historical plan airport; 38. Historical plan aircraft type; 39. Historical plan quantity; 40. Historical preset time period; 41. Historical preset airport; 42. Historical preset equipment name; 43. Historical preset quantity; 44. Result data; 45. Intermediate data; 46. Output data; 47. Equipment name; 48. Mean μ; 49. Standard deviation σ; 50. Forecast number; 51. Time period; 52. Airport; 53. Equipment name; 54. Forecast quantity. DETAILED DESCRIPTION

[0106] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0107] Example 1:

[0108] See also Figure 1 The present invention provides a method for establishing an aviation maintenance support resource demand forecasting model, comprising the following steps:

[0109] S1. Classify the lifespan distribution of aviation maintenance and support equipment based on its characteristic types;

[0110] S2. Classify aviation maintenance and support equipment according to their lifespan distribution types, perform calculations according to the methods for each lifespan distribution type, and obtain a calculation model for the predicted demand quantity of aviation maintenance and support equipment for each lifespan distribution type;

[0111] S3. Analyze the characteristics of the calculation models for the predicted demand quantity of aviation maintenance and support equipment of each life distribution type, and select the normal life distribution type as the calculation model for the predicted demand quantity of aviation maintenance and support equipment;

[0112] S4. Design and implement algorithmic steps based on the calculation model of the predicted demand quantity of aviation maintenance and support equipment with normal life distribution type;

[0113] S5. Provide pre-set standard data as fixed data for the algorithm, input prediction-related input data of aviation maintenance and support equipment, calculate through the algorithm of the prediction model, and output the results of the prediction plan.

[0114] Example 2:

[0115] The difference between the embodiment 1 and the embodiment 1 is as follows, which details the implementation process of the prediction demand quantity model calculation in step S2 of the embodiment 1 regardless of the life distribution type:

[0116] See also Figure 2 The method for establishing an aviation maintenance support resource demand forecasting model of the present invention implements the calculation of the forecast demand quantity model regardless of the life distribution type described in step S2 as follows:

[0117] A01, taking availability P as the standard reliability, calculate the corresponding reliable life t according to the life distribution type of the support equipment model k kp , the mathematical expression of availability P is as follows:

[0118]

[0119] MTTF (Mean Time To Failure) refers to the average time a device maintains trouble-free operation. For multiple devices of the same type, the average value of the time period from the start of normal operation to the occurrence of a failure is taken, generally expressed in years.

[0120] MTBF (Mean Time Between Failure) refers to the average time between two equipment failures. It is the average of multiple similar devices and is usually measured in years.

[0121] A02, with t kp is the update cycle (usually calculated in years), and the number of updates nk required within the guaranteed time period T is calculated;

[0122] A03,n k Multiply by the number of devices that need this protection in a certain type of machine i (number of machines) M i And the total number of aircraft of this type N in preset point jj , that is, the number S of updated security equipment model k required for model i is obtained ki ;

[0123] A04, classify and accumulate all S ki , obtain the predicted demand quantity of support equipment.

[0124] Example 3:

[0125] The difference between the first embodiment and the second embodiment is as follows, which details the method for calculating the forecast demand quantity model of the exponential life distribution type in step S2 of the first embodiment:

[0126] See also Figure 3 The method for establishing an aviation maintenance support resource demand forecasting model of the present invention implements the calculation method of the exponential life distribution type described in step S2 as follows:

[0127] B01, the distribution function expression of exponential distribution is as follows:

[0128] F(t)=1–e -λt (2)

[0129] Where: λ is the update rate of the support equipment, t is the update period; its availability R(t) is expressed as follows:

[0130] R(t)=1–F(t)=e -λt (3)

[0131] B02, from expressions (2) and (3), we can get:

[0132]

[0133] B03, solve t from expression (4) p The expression is as follows:

[0134]

[0135] B04, the calculation model expression for the equipment forecast demand quantity of the exponential distribution type is as follows:

[0136]

[0137] Where: M i N is the number of devices that need this protection in a certain type of machine i (number of machines used), j is the total number of aircraft of this type in the preset point j, λ is the update rate of the support equipment, T is the support time period, and P is the availability.

[0138] Example 4:

[0139] The differences between Example 1, Example 2 and Example 3 are as follows, which details the implementation method of calculating the predicted demand quantity model of the normal life distribution type in step S2 of Example 1:

[0140] See also Figure 4 The method for establishing an aviation maintenance support resource demand forecasting model according to the present invention implements the calculation method of the forecast demand quantity model of the normal life distribution type described in step S2 as follows:

[0141] C01, the distribution function expression of normal distribution is as follows:

[0142]

[0143] in:

[0144] μ is the mean of the normal distribution, which contains a statistical set of N samples and is expressed as:

[0145]

[0146] σ is the standard deviation, which contains the statistical set of N samples and is expressed as:

[0147]

[0148] The expression of its availability is as follows:

[0149]

[0150] C02, from expressions (7) and (10), we can get:

[0151]

[0152] C03, solve t from expression (11) p The expression is as follows:

[0153] tp=μ+u p ·σ (12)

[0154] Where: u p is the standard normal distribution quantile, which can be found from the statistical distribution value table. The commonly used u p The values ​​are shown in the following table:

[0155]

[0156] C04, the calculation model expression for the equipment forecast demand quantity of the normal distribution type is as follows:

[0157]

[0158] Where: Mi N is the number of devices that need this protection in a certain type of machine i (number of machines used), j is the total number of aircraft of this model at preset point j.

[0159] Embodiment 5:

[0160] The differences between the first, second, third and fourth embodiments are as follows, which describe in detail the implementation method of calculating the forecast demand quantity model of the Weibull distribution type in step S2 of the first embodiment:

[0161] See also Figure 5 The method for establishing an aviation maintenance resource demand forecasting model according to the present invention implements the calculation method of the Weibull distribution type forecasting demand quantity model described in step S2 as follows:

[0162] D01, the distribution function expression of Weibull distribution is as follows:

[0163]

[0164] in:

[0165] t is the life span (can be Weibull distribution with frequency or time), t0 is the minimum life span, t a is the characteristic life, b is the shape parameter;

[0166] The expression of its availability is:

[0167]

[0168] D02, from expressions (14) and (15), we can get:

[0169]

[0170] D03, solve t from expression (16) p The expression is as follows:

[0171]

[0172] D04, the calculation model expression of the equipment forecast demand quantity obtained by Weibull distribution is as follows:

[0173]

[0174] Where: M i N is the number of devices that need this protection in a certain type of machine i (number of machines used), j is the total number of aircraft of this type in the preset point j, T is the guarantee period, P is the availability, t0 is the minimum life, t a Characteristic lifespan.

[0175] Example 6:

[0176] The technical differences between Example 1, Example 2, Example 3, Example 4, and Example 5 are as follows: by comparing and analyzing the implementation methods of calculating the predicted demand quantity model of the four distribution types in Example 2, Example 3, Example 4, and Example 5, combined with the characteristics of aviation maintenance and support equipment itself, the normal life distribution type is selected as the calculation model for the predicted demand quantity of aviation maintenance and support equipment.

[0177] See also Figure 6 The method for establishing an aviation maintenance support resource demand forecasting model of the present invention implements the algorithm steps of predicting the demand quantity of aviation maintenance support equipment of the normal life distribution type described in step S4 as follows:

[0178] E01, initialization, sets the standard data required for the algorithm to run, including:

[0179] Construct a standard normal distribution quantile table,

[0180] Read in the preset quota standards for security equipment;

[0181] E02. Read in the necessary data for prediction, including:

[0182] The time period for entering historical data,

[0183] Enter the training plan and actual configuration values ​​of each equipment to be predicted at each airport in the past years within the time period in chronological order;

[0184] E03, preprocessing, generates the necessary parameters required for prediction, including:

[0185] For a certain pre-installed equipment, calculate the theoretical demand value for each year based on historical data.

[0186] Calculate the difference between the theoretical demand value and the actual configuration value for each year for this specific pre-installed equipment.

[0187] Normalize all annual differences by dividing them by the annual demand value.

[0188] The average value of the normalized data of all airports in each year for this specific preset equipment within the historical data period is summarized on an annual basis.

[0189] Calculate the mean μ and standard deviation σ of the normal distribution,

[0190] Temporarily store the normal distribution parameters μ and σ of the specific device;

[0191] E04, set the time period to be predicted and the parameters of the prediction task, including:

[0192] Set the time period to be predicted and the type of prediction task,

[0193] Enter the training plan for the predicted time period.

[0194] Check the quotas of all support equipment at all airports in the plan and calculate the theoretical demand value R;

[0195] E05, enter the availability P of each device to be predicted i ;

[0196] E06, repeat the following process for all airports in the plan within the forecast time period, including:

[0197] The equipment availability P i Look up the table to calculate the standard normal distribution quantile u Pi ,

[0198] The difference in future demand is predicted based on the equipment update calculation model of the normal distribution type.

[0199] According to the task type, check the margin table to correct the predicted difference Δi,

[0200] Add the predicted value of the equipment difference to the theoretical demand value R of the equipment i ;

[0201] E07, summarize the predicted total amount of all types of pre-installed equipment by airport classification and output.

[0202] Embodiment seven:

[0203] Compared with the first embodiment, the distinguishing technology is as follows, which introduces in detail the pre-provided set of pre-set standard data, which is used as fixed data for the algorithm.

[0204] Before implementing the prediction algorithm, some standard data needs to be set in advance. Figure 7 The pre-set standard data (1) described in step S5 includes prefabrication location information (2), pre-installed support equipment quota information (3) and prediction task type information (4), which are used as fixed data of the algorithm for query when used. When the algorithm is used, the prediction time period is input, and after the model calculation, the prediction plan is output.

[0205] The prefabrication location information (2) is the physical data of each prefabrication point itself, which is used for data management in the prediction algorithm. Some of the data also includes the prefabrication data name (5), the prefabrication point location name (6), the prefabrication area (7) and the prefabrication point capacity (8) in order to increase the readability and clarity of the output results.

[0206] The pre-installed support equipment quota information (3) is the support equipment data required for each aircraft model to implement support maintenance. It is generally imported into the model before the algorithm is implemented. When generating the pre-installed plan, it generates all the detailed requirements of the support equipment after querying according to the input training plan. It is the basic data for the implementation of the algorithm model and is generally composed of the pre-installed quota table (9) and the pre-installed equipment table (10). The pre-installed quota table (9) includes the quota number (13), aircraft model (14), pre-installed equipment (15), measurement unit (16), number of aircraft (17) and number of equipment (18). The pre-installed equipment table (10) includes the equipment name (19), model (20), price (21) and volume (22);

[0207] The forecast task type information (4) is used to modify the final forecast result according to the task type of the forecast plan. It is imported into the model before the algorithm is implemented and consists of a prefabricated task type (11) and a prefabricated coefficient (12).

[0208] Embodiment 8:

[0209] Compared with the first embodiment and the seventh embodiment, the difference between the two technologies is as follows, and the prediction-related input data is introduced in detail.

[0210] See also Figure 8 ,The input data (23) consists of the training plan for prediction (24), ,historical preset details (25) and the type of prediction ,task (26).

[0211] The prediction training plan (24) is the input data for generating the prediction scheme, which is used to generate the algorithm model to implement all the support equipment requirements, including the plan number (27), time period (28), prefabrication area (29), airport number (30), aircraft type (31) and quantity (32);

[0212] The historical preset details (25) are used to count and generate all parameters for the prediction algorithm model implementation according to the training plan of previous years (or a certain time period) and the actual matching preset plan data, including the historical plan table (33) and the historical preset table (34). The historical plan table (33) includes the historical number (35), the historical plan time period (36), the historical plan airport (37), the historical plan aircraft type (38) and the historical plan quantity (39); the historical preset table (34) includes the historical preset time period (40), the historical preset airport (41), the historical preset equipment name (42) and the historical preset quantity (43).

[0213] The type of prediction task (26) is handled by direct setting or input.

[0214] Embodiment 9:

[0215] Compared with the first embodiment, the seventh embodiment and the eighth embodiment, the difference technology is as follows, which details the algorithm calculation of the prediction model and outputs the result data (44) of the prediction scheme, including intermediate data (45) and output data (46).

[0216] See also Figure 9 , the result data (44) of the prediction scheme includes intermediate data (45) and output data (46).

[0217] Intermediate data (45) is the intermediate result of the algorithm model. It is the statistical parameter of each security device calculated according to the set time period and the input historical data. It can be calculated once and permanently saved for reuse in future predictions without having to calculate again. Intermediate data (45) includes the device name (47), mean μ (48), and standard deviation σ (49).

[0218] The output data (46) is the output result of the algorithm model, which is the estimated preset quantity of support equipment required according to the set time period, mainly including the prediction number (50), time period (51), airport (52), equipment name (53) and predicted quantity (54).

[0219] The present invention can output the results of an aviation maintenance and support resource demand forecasting program based on the forecast-related input data of aviation maintenance and support equipment. The results of the maintenance and support resource demand forecasting program include the name of the support equipment, time period, usage location and predicted quantity. Based on the one-to-one correspondence between the predicted quantity in the output result, the usage time and usage location, the quantity of maintenance and support equipment required at the location and usage time can be reserved in advance, solving the problem of excessive occupation of maintenance warehouse resources by using the method of reserving inventory; at the same time, the usage time and usage location are determined in advance, solving the problem of mutual location calls causing confusion in the usage cycle and difficulty in management; in the case of insufficient inventory, timely purchase and preparation of materials in advance based on the predicted quantity in the output result, solving the problem of reduced efficiency of aviation equipment maintenance due to long maintenance equipment procurement time.

[0220] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A method for establishing an aviation maintenance support resource demand forecasting model, characterized by: The following steps are involved: S1, classify the life distribution of aviation maintenance and support equipment according to the characteristic types of aviation maintenance and support equipment; S2, according to the life distribution classification type of aviation maintenance and support equipment, calculation is performed according to the calculation method of each life distribution type, and a calculation model for the predicted demand quantity of aviation maintenance and support equipment of each life distribution type is obtained; S3, through the characteristic analysis of the calculation model of the predicted demand quantity of aviation maintenance and support equipment of each life distribution type, the normal life distribution type is selected as the calculation model of the predicted demand quantity of aviation maintenance and support equipment; S4, design and implement algorithm steps according to the calculation model of the predicted demand quantity of aviation maintenance and support equipment with normal life distribution type; S5, provides pre-set standard data as fixed data for the algorithm, inputs prediction-related input data of aviation maintenance and support equipment, calculates through the algorithm of the prediction model, and outputs the results of the prediction plan.

2. The method for establishing an aviation maintenance support resource demand forecasting model according to claim 1, characterized in that: The life distribution classification types of the aviation maintenance and support equipment described in step S1 include: an independent life distribution type, an exponential life distribution type, a normal life distribution type and a Weibull life distribution type.

3. The method for establishing an aviation maintenance support resource demand forecasting model according to claim 1, characterized in that: The calculation methods of the respective life distribution types described in step S2 include: a calculation method of an independent life distribution type, a calculation method of an exponential life distribution type, a calculation method of a normal life distribution type and a calculation method of a Weibull life distribution type.

4. The method for establishing an aviation maintenance support resource demand forecasting model according to claim 3, characterized in that: The calculation method of the irrelevant life distribution type is as follows: A01, taking availability P as the standard reliability, calculate the corresponding reliable life t according to the life distribution type of the support equipment model k kp , the mathematical expression of availability P is as follows: Among them: MTTF refers to the average time for equipment to operate without failure. For multiple equipment of the same type, the average value of all time periods from the start of normal operation to the occurrence of failure is taken, usually in years; MTBF refers to the average time period between two equipment failures. The average value of multiple similar equipment is taken, usually in years. A02, with t kp is the update cycle, usually in years, and calculates the number of updates nk required within the guaranteed time period T; A03,n k Multiply by the number M of devices that need the protection in a certain model i i and the total number of aircraft of this model N at preset point j j , that is, the number S of updated security equipment model k corresponding to model i is obtained ki ; A04, classify and accumulate all S ki , obtain the predicted demand quantity of support equipment.

5. The method for establishing an aviation maintenance support resource demand forecasting model according to claim 3, characterized in that: The exponential life distribution type is calculated as follows: B01, the distribution function expression of exponential distribution is as follows: F(t)=1–e -λt (2) Where: λ is the update rate of the support equipment, t is the update period; the expression of its availability R(t) is as follows: R(t)=1–F(t)=e -λt (3) B02, from expressions (2) and (3), we can get: B03, solve expression (4) to get t p The expression is as follows: B04, the calculation model expression of the equipment forecast demand quantity of exponential distribution type is as follows: Where: M i is the number of the required protection equipment in a certain model i, N j is the total number of aircraft of this type in the preset point j, λ is the update rate of the support equipment, T is the support time period, and P is the availability.

6. The method for establishing an aviation maintenance support resource demand forecasting model according to claim 3, characterized in that: The calculation method of the normal life distribution type is as follows: C01, the distribution function expression of normal distribution is as follows: Where: μ is the mean of the normal distribution, which contains a statistical set of N samples, and its expression is: σ is the standard deviation, which contains the statistical set of N samples and its expression is: The expression of its availability is as follows: C02, from expressions (7) and (10), we can get: C03, solve the expression (11) to get t p The expression is as follows: t p =μ+u p ·s (12) Where: u p is the standard normal distribution quantile, which can be found from the statistical distribution value table. The commonly used u p The values ​​are shown in the following table: C04, the calculation model expression of the equipment forecast demand quantity of the normal distribution type is as follows: Where: M i is the number of the required protection equipment in a certain model i, N j is the total number of aircraft of this model in preset point j.

7. According to the method for establishing an aviation maintenance support resource demand forecasting model according to claim 3, it is characterized in that: The calculation method of the Weibull distribution type is as follows: D01, the distribution function expression of Weibull distribution is as follows: Where: t is the life span, which is the number or time of Weibull distribution, t0 is the minimum life span, t a is the characteristic life, b is the shape parameter; The expression of its availability is: D02, from expressions (14) and (15), we can get: D03, solve expression (16) to get t p The expression is as follows: D04, the calculation model expression of the equipment forecast demand quantity obtained by Weibull distribution is as follows: Where: M i is the number of the required protection equipment in a certain model i, N j is the total number of aircraft of this type in the preset point j, T is the guarantee time period, P is the availability, t0 is the minimum life, t a Characteristic life span.

8. The method for establishing an aviation maintenance support resource demand forecasting model according to claim 1, characterized in that: The algorithm steps for predicting the required quantity of aviation maintenance and support equipment of the normal life distribution type described in step S4 are as follows: E01, initialization, sets the standard data required for the algorithm to run, including: Construct a standard normal distribution quantile table, Read in the preset quota standards for security equipment; E02, read in the necessary data required for prediction, including: The time period for entering historical data, Enter the training plan and actual configuration value of each equipment to be predicted at each airport in the past years in the time period in chronological order; E03, preprocessing, generates the necessary parameters required for prediction, including: For a certain preset equipment, calculate the theoretical demand value for each year based on historical data. Calculate the difference between the theoretical demand value and the actual configuration value for each year for this specific pre-installed equipment. Divide all annual differences by the annual demand value for normalization. The average value of the normalized data of all airports for each year of the specific preset equipment within the historical data period is summarized by year, Calculate the mean μ and standard deviation σ of the normal distribution. Temporarily store the normal distribution parameters μ and σ of the specific device; E04, set the time period to be predicted and the parameters of the prediction task, including: Set the time period to be predicted and the type of prediction task. Enter the training plan for the time period to be predicted. Check the quotas of all support equipment for all airports in the plan and calculate the theoretical demand value R; E05, enter the availability P of each equipment to be predicted i ; E06, repeat the following process for all airports in the plan within the forecast period, including: The equipment availability P i Look up the table to calculate the standard normal distribution quantile u Pi , The difference in future demand is predicted based on the equipment update calculation model of the normal distribution type. According to the task type, check the margin table to correct the predicted difference Δi. Add the predicted value of the equipment difference to the theoretical demand value R of the equipment. i ; E07, summarize the predicted total amount of all types of pre-installed equipment by airport classification and output.

9. The method for establishing an aviation maintenance support resource demand forecasting model according to claim 1, characterized in that: The pre-set standard data in step S5 includes prefabrication location information, pre-installed support equipment quota information and predicted task type information.

10. The method for establishing an aviation maintenance support resource demand forecasting model according to claim 1, characterized in that: The prediction-related input data described in step S5 include the prediction training plan, historical preset details and the type of prediction task; the result of the prediction scheme output by the algorithm calculation of the prediction model described in step S5 includes intermediate data and output data, and the output data includes the name of the security equipment, time period, usage location and predicted quantity.

Images