Aerial material demand prediction and scheduling method, system, equipment and medium

Abstract

The invention discloses an aerial material demand prediction and scheduling method, system and device, and a medium. The method comprises the following steps: collecting use data and multiple influence factor data of a plurality of aerial materials within a preset time length; for any air material, calculating a trend similarity factor between each piece of influence factor data and the use data of the air material, and calculating a variation amplitude similarity factor between each piece of influence factor data and the use data of the air material; adjusting a preset lag order of each piece of influence factor data based on the trend similarity factor and the variation amplitude similarity factor to obtain an improved lag order of each piece of influence factor data, and constructing a vector autoregression model according to the improved lag order to obtain a predicted demand of the aviation material; and completing aerial material scheduling based on the predicted demand quantity of each aerial material. According to the method, the operation efficiency of the prediction model is improved while the influence of different influence factor data on aviation material demand prediction is considered.

Description

Technical Field

[0001] The present application relates to the field of aviation material management technology, and in particular to an aviation material demand forecasting and scheduling method, system, equipment and medium. Background Art

[0002] Forecasting demand for aviation materials is a key link in aviation industry management. Accurate demand forecasting can help airlines plan their inventory rationally, reduce the risk of surplus and shortage, ensure timely access to required materials during peak periods, and ensure flight safety.

[0003] The demand for aviation materials is often affected by multiple factors, such as the number of flights, aircraft utilization rate and economic conditions. The vector autoregression model (VAR) can effectively capture the complex relationship between the demand for aviation materials and other related variables, thereby providing more accurate demand forecasts. When using the traditional vector autoregression model (VAR) to forecast the demand for aviation materials, it is usually based on information criteria such as the Akaike Information Criterion (AIC) or the Bayesian Information Criterion (BIC) to gradually verify different lag orders and then determine the optimal lag order. However, when the amount of data is too large, the time and computing resources required to gradually verify different lag orders are relatively large, and the efficiency is low. At the same time, the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC) tend to choose lower lag orders, which leads to the exclusion of important historical information and the decline of the prediction ability of the prediction model.

[0004] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of this application is to provide a method, system, device and medium for forecasting and scheduling aviation material demand, aiming to solve the technical problems that when the traditional vector autoregression model (VAR) is used to forecast aviation material demand, the process of setting the lag order is inefficient and overly dependent on the information criterion, resulting in inaccurate prediction values.

[0006] To achieve the above-mentioned purpose, the present application provides a method for forecasting and scheduling aviation material demand, including: collecting usage data and multiple influencing factor data of several types of aviation materials for a preset time length; for any type of aviation material, calculating the trend similarity factor between each influencing factor data and the usage data of the aviation material, and calculating the change amplitude similarity factor between each influencing factor data and the usage data of the aviation material; adjusting the preset lag order of each influencing factor data based on the trend similarity factor and the change amplitude similarity factor to obtain an improved lag order of each influencing factor data, constructing a vector autoregression model based on the improved lag order to obtain the predicted demand for the aviation material; and completing aviation material scheduling based on the predicted demand for each aviation material.

[0007] Optionally, the calculation of the trend similarity factor between each influencing factor data and the aviation material usage data includes: respectively calculating a first difference sequence of each influencing factor data and a second difference sequence of the aviation material usage data; and determining the trend similarity factor between the aviation material usage data and each influencing factor data based on each of the first difference sequences and the second difference sequences.

[0008] Optionally, determining the trend similarity factor between the aviation material usage data and each influencing factor data based on the first difference sequence and the second difference sequence includes: for any influencing factor data, comparing the data differences at the same time point in the second difference sequence of the aviation material usage data and the first difference sequence of the influencing factor data to obtain the trend similarity of each time point; determining the trend similarity factor between the aviation material usage data and the influencing factor data based on the trend similarity of each time point.

[0009] Optionally, the calculation of the change amplitude similarity factor between each influencing factor data and the aviation material usage data includes: respectively calculating the attention sequence between each influencing factor data and the aviation material usage data; and determining the change amplitude similarity factor between each influencing factor data and the aviation material usage data based on multiple attention sequences.

[0010] Optionally, the separately calculating the attention sequence between each influencing factor data and the aviation material usage data includes: for any influencing factor data, calculating the absolute value of the data difference between the influencing factor data and the aviation material usage data at the same time point in chronological order, and constructing the attention sequence between the influencing factor data and the aviation material usage data in chronological order.

[0011] Optionally, the determining the change amplitude similarity factor between each influencing factor data and the aviation material usage data based on multiple focus sequences includes: determining a key influencing sequence between each influencing factor data and the aviation material usage data based on each focus sequence; determining the change amplitude similarity factor between each influencing factor data and the aviation material usage data according to the degree of numerical fluctuation of the key influencing sequence between each influencing factor data and the aviation material usage data.

[0012] Optionally, adjusting the preset lag order of each influencing factor data based on the trend similarity factor and the change amplitude similarity factor to obtain the improved lag order of each influencing factor data includes: adjusting the preset lag order of each influencing factor data based on the trend similarity factor and the change amplitude similarity factor using the following formula (1) to obtain the improved lag order of each influencing factor data:

[0013] in, represents the preset lag order, Indicates The variation range between the influencing factor data and the target aviation material usage data is similar to the factor, Indicates The trend similarity factor between the influencing factor data and the target aviation material usage data, Indicates the preset maximum value of the lag order. represents the first preset coefficient, Indicates the second preset coefficient.

[0014] In addition, to achieve the above-mentioned purpose, the present application also provides an aviation material demand forecasting and scheduling system, including: a data acquisition module, used to collect usage data and multiple influencing factor data of several types of aviation materials of a preset time length; each improvement factor calculation module, used to calculate, for any type of aviation material, the trend similarity factor between each influencing factor data and the usage data of the aviation material, and to calculate the change amplitude similarity factor between each influencing factor data and the usage data of the aviation material; an aviation material demand prediction module, used to adjust the preset lag order of each influencing factor data based on the trend similarity factor and the change amplitude similarity factor to obtain the improved lag order of each influencing factor data, and construct a vector autoregression model based on the improved lag order to obtain the predicted demand for the aviation material; an aviation material scheduling module, used to complete aviation material scheduling based on the predicted demand for each aviation material.

[0015] The present application also provides an aviation material demand forecasting and scheduling device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned aviation material demand forecasting and scheduling method.

[0016] The present application also provides a computer-readable storage medium, comprising: a computer program stored therein, and when the computer program is executed by a processor, the above-mentioned aviation material demand forecasting and scheduling method is implemented.

[0017] The present application proposes a method, system, device and medium for forecasting and scheduling aviation material demand. By calculating the trend similarity factor between the data of each influencing factor and the usage data of the aviation material, and using the trend similarity factor to adjust the preset lag order, it is possible to set a larger lag order for influencing factors such as fuel prices, so as to better capture the market effect brought about by fuel prices, that is, the long-term impact on the forecast of aviation material demand.

[0018] In addition, by calculating the similarity factor of the change amplitude between each influencing factor data and the aviation material usage data, and using the similarity factor of the change amplitude to adjust the preset lag order, it can be achieved that when the change amplitude of a certain influencing factor data and the change amplitude of the aviation material usage data are relatively similar (that is, there is a strong correlation between the influencing factor data and the aviation material usage), and if the change of the influencing factor data in the short term is relatively small, then the lag effect of the influencing factor data on the aviation material demand forecast is also relatively short, so that a smaller lag order can be selected, so as to improve the operating efficiency of the prediction model while considering the impact of the influencing factor data on the aviation material demand forecast. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a flow chart of an aviation material demand forecasting and scheduling method according to one implementation of the present application; Figure 2 A structural block diagram of an aviation material demand forecasting and scheduling system according to an implementation mode of the present application; Figure 3 It is a schematic diagram of the structure of an aviation material demand forecasting and scheduling device according to one implementation mode of the present application.

[0020] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0021] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0022] Because the existing technology uses the traditional vector autoregression model (VAR) to predict the demand for aviation materials, it usually determines the optimal lag order by gradually verifying different lag orders based on information criteria such as the Akaike Information Criterion (AIC) or the Bayesian Information Criterion (BIC). However, when the amount of data is too large, the time and computing resources required to gradually verify different lag orders are relatively large, and the efficiency is low. At the same time, the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC) may tend to select a lower lag order, which leads to the exclusion of important historical information, the reduction of the predictive ability of the prediction model, and the large deviation of the predicted value of the demand for aviation materials.

[0023] In order to solve the above problems, the present application provides a method for forecasting and scheduling aviation material demand. The present application scheme is introduced in detail below.

[0024] Figure 1 is a flowchart of an aviation material demand forecasting and scheduling method according to an embodiment of the present application. The aviation material demand forecasting and scheduling method can be executed by an aviation material demand forecasting and scheduling device with data processing capability. Figure 1The aviation material demand forecasting and scheduling method may include the following steps: Step S1, collecting usage data of several types of aviation materials and multiple influencing factor data for a preset time length.

[0025] It should be noted that the demand for aviation materials is affected by many factors, such as the number of flights and the number of aircraft, which directly affect the demand for aircraft parts. At the same time, factors such as oil prices, seasonal changes, and laws and regulations will also affect the operating plans of airlines, thereby affecting the changes in the demand for aviation materials. Therefore, it is difficult to ensure the accuracy of aviation material demand forecasting based only on historical aviation material usage data. This embodiment uses a vector autoregression model to forecast aviation material demand, which can consider the impact of multiple factors on aviation material demand and increase the reliability of aviation material demand forecasting.

[0026] In a specific implementation process, the usage data of several types of aviation materials and multiple influencing factor data of a preset time length are collected in the inventory management system of the airline. The multiple influencing factor data may be flight quantity data, aircraft quantity data, fuel price data, passenger load factor data, and flight sortie data. It should be noted that in this embodiment, the usage data of each aviation material is the usage amount of each aviation material.

[0027] Step S2, for any type of aviation material, calculating the trend similarity factor between each influencing factor data and the usage data of the aviation material, and calculating the change range similarity factor between each influencing factor data and the usage data of the aviation material.

[0028] It should be noted that this embodiment uses the vector autoregression model (VAR) to predict the demand for aviation materials, and the lag order of each influencing factor data needs to be manually preset. The lag order in the vector autoregression model refers to the number of past observations included in the prediction model, i.e., the vector autoregression model. For example, if the lag order is selected as 2, the observations at the past two time points need to be considered in the prediction model.

[0029] It should be further explained that for different aviation materials, the influence of various influencing factors on the demand forecast varies. For example, the demand for aircraft seats is greatly affected by the passenger load factor. When the passenger load factor is high, the aircraft seats are more worn out and used. The demand for wings is less affected by the passenger load factor. Therefore, for aircraft seats and wings, that is, for different aviation materials, the lag order set for the same influencing factor, that is, the passenger load factor, is also different, and thus different aviation materials need to be analyzed and processed separately.

[0030] In the specific implementation process, any type of aviation material is recorded as the target aviation material. This embodiment is described by taking the target aviation material as an example.

[0031] In one embodiment, in step S2, calculating the trend similarity factor between each influencing factor data and the aviation material usage data may specifically include: S21, respectively calculating a first difference sequence of each influencing factor data and a second difference sequence of the aviation material usage data; S22: Determine trend similarity factors between the aviation material usage data and each influencing factor data based on the first difference sequences and the second difference sequences.

[0032] In the specific implementation process, the first difference sequence of the target material usage data is calculated and obtained, and the kth data in the first difference sequence of the target material usage data is the difference between the k+1th data and the kth data in the target material usage data; the second difference sequence of each influencing factor data is calculated and obtained according to the calculation method of the first difference sequence.

[0033] In one embodiment, in step S22, determining the trend similarity factor between the use data of the aviation material and each influencing factor data in each of the first difference sequence and the second difference sequence may specifically include: S221, for any influencing factor data, comparing the data difference at the same time point in the second difference sequence of the aviation material usage data and the first difference sequence of the influencing factor data to obtain the trend similarity at each time point; S222: Determine a trend similarity factor between the aviation material usage data and the influencing factor data based on the trend similarity at each time point.

[0034] In the specific implementation process, taking any influencing factor data as an example, the trend similarity of the two data corresponding to the same time point in the first difference sequence of the influencing factor data and the second difference sequence of the target aviation material usage data is calculated in chronological order, wherein, when the two data corresponding to the same time point are both positive or both negative, the trend similarity is 1, and when the two data corresponding to the same time point are one positive and one negative, the trend similarity is 0. It should be noted that when the two data corresponding to the same time point are 0 and a positive number, 0 and a negative number, or 0 and 0, the trend similarity is 1.

[0035] Furthermore, taking any influencing factor data as an example, the trend similarities between the two data corresponding to all the same time points in the first difference sequence of the influencing factor data and the second difference sequence of the target aviation material usage data are accumulated to obtain the trend similarity factor between the influencing factor data and the target aviation material usage data.

[0036] It can be understood that the second difference sequence of the use data of the aviation materials can characterize the change trend of the use data of the aviation materials, and accordingly, the first difference sequence of the data of each influencing factor can characterize the change trend of each influencing factor, and then the trend similarity factor can characterize the similarity between the change trend of each influencing factor data and the change trend of the use data of the aviation materials. Since the present embodiment obtains the lag order of each influencing factor data according to the difference between the influence degree of each influencing factor data on the use data of the aviation materials, and the influence degree can be reflected by the trend similarity between the data, taking the aviation materials as aircraft seats and the influencing factor data as passenger load factor and fuel price as an example, since the higher the passenger load factor, the higher the loss of the aircraft seats, the corresponding use data also increases accordingly, and then the trend similarity factor between the passenger load factor and the aircraft seats is larger, and the influence degree of the passenger load factor on the demand for aircraft seats is larger; and the higher the fuel price, the use data corresponding to the aircraft seats can increase or decrease, and the trend similarity factor between the fuel price and the aircraft seats is smaller, and the influence degree of the fuel price on the demand for aircraft seats is also smaller.

[0037] It should be noted that when there is an approximate linear correlation between the data of an influencing factor and the data of aviation material usage, that is, there is a strong correlation between the data of the influencing factor and the usage of aviation materials, and the change of the influencing factor in the short term is relatively small, the lag effect of the influencing factor data on the demand forecast of aviation materials is also relatively short-term. For example, the influencing factor data takes the number of flights as an example, and the aviation materials take aircraft tires as an example. The aircraft tire usage data can reflect the tire wear. If the number of flights in a certain time period is 1200, 1500, 1800, 2000, 2500, and 2700; the aircraft in the same time period The tire usage data are 100, 125, 150, 180, 225, and 240. According to the data analysis, the flight quantity data and the aircraft tire usage data show an approximately linear relationship, and the two most recent flight quantity data are 2500 and 2700, without large fluctuations. Therefore, for aircraft tires, the impact of the flight quantity data on its demand is slightly different from the impact of historical data. Therefore, a smaller lag order can be selected for the flight quantity data in its influencing factor data, so as to improve the operating efficiency of the prediction model while considering the impact of the flight quantity data on the aircraft tire demand forecast.

[0038] Based on this, this embodiment constructs a change amplitude similarity factor based on the correlation strength between each influencing factor data and the aviation material usage data on the basis of the trend similarity factor.

[0039] In one embodiment, in step S2, calculating the variation amplitude similarity factor between each influencing factor data and the aviation material usage data may specifically include: S23, respectively calculating the attention sequence between each influencing factor data and the use data of the aviation material; S24. Determine a similarity factor of the variation range between each influencing factor data and the aviation material usage data based on the plurality of focus sequences.

[0040] In the specific implementation process, taking any influencing factor data as an example, the absolute values ​​of the difference between the two data corresponding to the same time point in the influencing factor data and the target aviation material usage data are arranged in chronological order to form a focus sequence of the influencing factor data and the target aviation material usage data.

[0041] In one embodiment, in step S24, determining the change amplitude similarity factor between each influencing factor data and the aviation material usage data based on the plurality of focus sequences may specifically include: S241, determining a key influence sequence between each influencing factor data and the use data of the aviation material based on each of the focus sequences; S242: Determine a factor for approximating the magnitude of changes between the data of each influencing factor and the data for use of the aviation material according to the degree of numerical fluctuation of a key influencing sequence between the data of each influencing factor and the data for use of the aviation material.

[0042] In a specific implementation process, a first preset number is set, and the first preset number of data is acquired in reverse order in the focus sequence, and the key impact sequence is obtained by arranging them in time sequence.

[0043] It should be noted that, since the future demand for aviation materials is predicted, the recent usage data of aviation materials is more effective than the past usage data of aviation materials. Therefore, this embodiment obtains data in reverse order in the focus sequence, which helps to improve the prediction accuracy of aviation material demand.

[0044] Furthermore, the absolute value of the difference between any data in the key influence sequence and other data is obtained, the minimum absolute value of the difference is taken as the difference of the data, the differences of all data in the key influence sequence are accumulated, and the amplitude difference of the focus influence sequence is obtained, that is, the amplitude difference between the influencing factor data and the usage data of the target aviation material.

[0045] It can be understood that the magnitude difference between a certain influencing factor data and the usage data of the target aviation material can represent the strength of the correlation between the influencing factor data and the usage data of the target aviation material.

[0046] Further, a second preset number of data is obtained from each influencing factor data in reverse order, and the standard deviation of the obtained data set is calculated, which is recorded as the recent change degree of each influencing factor data. It should be noted that the values ​​of the first preset number and the second preset number can be the same or different.

[0047] It can be understood that the standard deviation can represent the degree of change in the data, and thus the recent degree of change in the data of a certain influencing factor can represent the recent degree of change in the value of the influencing factor data.

[0048] Furthermore, the recent change degree of each influencing factor data is linearly normalized, and the amplitude difference between each influencing factor data and the usage data of the target aviation material is linearly normalized.

[0049] The difference between 1 and the product of the amplitude difference between the data of each influencing factor and the usage data of the target aviation material and the degree of recent change is recorded as the change amplitude similarity factor between the data of each influencing factor and the target aviation material.

[0050] Step S3, adjusting the preset lag order of each influencing factor data based on the trend similarity factor and the change amplitude similarity factor to obtain the improved lag order of each influencing factor data, constructing a vector autoregression model according to the improved lag order to obtain the predicted demand for the aviation materials.

[0051] It should be noted that when fuel prices continue to rise, airlines need time to adjust their flight numbers and route planning in the face of fuel price changes. If fuel prices continue to rise, for cost considerations, the number of flights will first decrease, and the use of aviation materials will decrease. However, during summer vacations or holidays, under the premise of continued fuel price increases, the number of flights will increase instead, and the use of aviation materials will increase. Since the trend similarity factor can reflect the degree of similarity between the change trend of aviation material usage data and the data of each influencing factor, for example, the change trend of the number of flights data is similar to that of aviation material usage data, that is, the trend similarity factor is large, while the change trend of fuel prices and aviation material usage data is different, that is, the trend similarity factor is small. Therefore, for factors such as fuel prices, increasing their lag order can better capture this adaptation process and the market response that follows the trend.

[0052] It should be further explained that when the change amplitude of an influencing factor data and the aviation material usage data are similar and large, it can be characterized that there is an approximate linear correlation between the influencing factor data and the aviation material usage data, that is, there is a strong correlation between the influencing factor data and the aviation material usage. At the same time, the change of the influencing factor in the short term is relatively small, so its impact on the aviation material demand is slightly different from the historical impact, and its lagged effect on the aviation material demand forecast is also relatively short. A smaller lag order can be selected, so as to achieve the goal of considering the impact of the influencing factor data on the aviation material demand forecast while improving the operating efficiency of the prediction model.

[0053] In one embodiment, in step S3, adjusting the preset lag order of each influencing factor data based on the trend similarity factor and the change amplitude similarity factor to obtain the improved lag order of each influencing factor data may specifically include: Linear normalization is performed on the trend-similar factors between all influencing factor data and target aviation material usage data, and linear normalization is performed on the change amplitude-similar factors between all influencing factor data and target aviation material usage data.

[0054] It should be noted that linear normalization of both trend-similar factors and change amplitude-similar factors can make the influence of trend-similar factors and change amplitude-similar factors on the prediction model more uniform, thus avoiding a certain feature from being too large and dominating the model training.

[0055] Furthermore, based on a trend similarity factor between a certain influencing factor data and the usage data of the target aviation material and a change amplitude similarity factor between the influencing factor data and the usage data of the target aviation material, the preset lag order is adjusted to obtain an improved lag order of the influencing factor data.

[0056] Specifically, As an example, based on the influencing factor data The trend similarity factor between the influencing factor data and the target aviation material usage data and the The change range between the influencing factor data and the target aviation material usage data is similar to the factor that adjusts the preset lag order to obtain the Improved lag order of influencing factor data The calculation formula is:

[0057] in, represents the preset lag order, Indicates The variation range between the influencing factor data and the target aviation material usage data is similar to the factor, Indicates The trend similarity factor between the influencing factor data and the target aviation material usage data, Indicates the preset maximum value of the lag order. represents the first preset coefficient, represents the second preset coefficient. The sum of the first preset coefficient and the second preset coefficient is 1, and the maximum value of the preset lag order must be greater than the preset lag order.

[0058] It should be noted that the improved lag order calculation formula constructed in the present embodiment can achieve that when the trend similarity factor of a certain influencing factor data is small, the influencing factor data obtains a larger lag order, which helps to capture the impact of the influencing factor on the long-term trend of the aviation material demand forecast; when the change amplitude similarity factor of a certain influencing factor data is large, the influencing factor data obtains a smaller lag order, so as to consider the impact of the influencing factor data on the predicted demand for aviation materials while improving the operating efficiency of the prediction model.

[0059] Furthermore, a preset lag order of the target aviation material is set, and a vector autoregression model is constructed according to the preset lag order of the target aviation material and the improved lag order corresponding to each influencing factor data. Based on the vector autoregression model, the predicted demand for the target aviation material is obtained.

[0060] Step S4, completing the aviation material scheduling based on the predicted demand for each aviation material.

[0061] In one embodiment, in step S4, completing the aviation material scheduling based on the predicted demand for each aviation material may specifically include: S41. Based on the forecast demand for each aviation material, evaluate whether the current inventory can meet the forecast demand and determine whether new materials need to be purchased; S42. Prioritize each aviation material according to the urgency and importance of demand; S43. Setting an aviation material dispatching plan based on the priority of each aviation material.

[0062] In the specific implementation process, if the available inventory of a certain aviation material is greater than or equal to the sum of the forecast demand for the aviation material and the safety stock of the aviation material, no purchase will be made; if the available inventory of a certain aviation material is less than the sum of the forecast demand for the aviation material and the safety stock of the aviation material, the aviation material will be purchased, and the purchase quantity is equal to the forecast demand plus the safety stock minus the available inventory. Among them, the available inventory is the actual inventory of the aviation material, and the safety stock represents the minimum inventory of the aviation material set to cope with demand fluctuations.

[0063] Furthermore, scoring standards for the importance and urgency of the demand for each aviation material are set according to whether each aviation material is a key material in production or operation. For example, the urgency of the demand for aviation materials that are immediately needed for operations is set to 5 points, the urgency of the demand for aviation materials that are needed for operations in the short term is set to 4 points, the urgency of the demand for aviation materials that are needed in the medium term is set to 3 points, the urgency of the demand for aviation materials that can be postponed is set to 2 points, and the urgency of the demand for aviation materials that are not urgently needed for operations is set to 1 point; the importance of the demand for aviation materials that are critical to operations is set to 5 points, the importance of the demand for aviation materials that are important but replaceable for operations is set to 4 points, the importance of the demand for aviation materials that are generally important for operations is set to 3 points, the importance of the demand for aviation materials that are secondarily important for operations is set to 2 points, and the importance of the demand for aviation materials that are not important for operations is set to 1 point.

[0064] Then, the obtained corresponding scores of the urgency of the demand for aviation materials and the importance of the demand for aviation materials can be used to determine the priority score of each aviation material, and then the aviation materials can be sorted according to the priority score. For example, the priority score of a certain aviation material is set as the sum of the urgency of the demand for aviation materials multiplied by the first weight coefficient and the importance of the demand for aviation materials multiplied by the second weight coefficient, and the aviation materials are prioritized according to their priority scores.

[0065] Furthermore, each aviation material is dispatched according to its priority, and the storage space for aviation materials is optimized to ensure that high-priority aviation materials are in the most convenient location.

[0066] Based on the above embodiments, Figure 2 FIG. 1 is a structural block diagram of an aviation material demand forecasting and scheduling system according to an implementation mode of the present application, such as Figure 2 As shown, the aviation material demand prediction and scheduling system may include: a data acquisition module 210, a module for calculating each improvement factor 220, an aviation material demand prediction module 230, and an aviation material scheduling module 240, wherein: The data collection module 210 is used to collect usage data of several types of aviation materials and multiple influencing factor data for a preset time length; Each improvement factor calculation module 220 is used to calculate, for any type of aviation material, a trend similarity factor between each influencing factor data and the aviation material usage data, and a change range similarity factor between each influencing factor data and the aviation material usage data; The aviation material demand prediction module 230 is used to adjust the preset lag order of each influencing factor data based on the trend similarity factor and the change amplitude similarity factor to obtain the improved lag order of each influencing factor data, and construct a vector autoregression model according to the improved lag order to obtain the predicted demand of the aviation material; The aviation material scheduling module 240 is used to complete aviation material scheduling based on the predicted demand of each aviation material.

[0067] In an exemplary embodiment, each improvement factor calculation module 220 can also be used to respectively calculate the first difference sequence of each influencing factor data and the second difference sequence of the aviation material usage data; based on each of the first difference sequence and the second difference sequence, determine the trend similarity factor between the aviation material usage data and each influencing factor data.

[0068] In an exemplary embodiment, each improvement factor calculation module 220 can also be used to compare the data differences at the same time point in the second difference sequence of the aviation material usage data and the first difference sequence of the influencing factor data for any influencing factor data to obtain the trend similarity of each time point; based on the trend similarity of each time point, determine the trend similarity factor between the aviation material usage data and the influencing factor data.

[0069] In an exemplary embodiment, each improvement factor calculation module 220 can also be used to respectively calculate the focus sequence between each influencing factor data and the aviation material usage data; and determine the change amplitude similarity factor between each influencing factor data and the aviation material usage data based on multiple focus sequences.

[0070] In an exemplary embodiment, each improvement factor calculation module 220 can also be used to calculate the absolute value of the data difference between the influencing factor data and the aviation material usage data at the same time point in chronological order for any influencing factor data, and construct a focus sequence between the influencing factor data and the aviation material usage data in chronological order.

[0071] In an exemplary embodiment, each improvement factor calculation module 220 can also be used to determine the key influence sequence between each influencing factor data and the aviation material usage data based on each of the focus sequences; and determine the change amplitude similarity factor between each influencing factor data and the aviation material usage data based on the degree of numerical fluctuation of the key influence sequence between each influencing factor data and the aviation material usage data.

[0072] In an exemplary embodiment, the aviation material demand forecasting module 230 may use the following formula (1) to obtain the improved lag order of each influencing factor data:

[0073] in, represents the preset lag order, Indicates The variation range between the influencing factor data and the target aviation material usage data is similar to the factor, Indicates The trend similarity factor between the influencing factor data and the target aviation material usage data, Indicates the preset maximum value of the lag order. represents the first preset coefficient, Indicates the second preset coefficient.

[0074] Those skilled in the art should understand that the division of the various modules in the embodiment is merely a division of logical functions, and in actual application, all or part of them can be integrated into one or more actual carriers, and these modules can be implemented entirely in the form of software called through a processing unit, or entirely in the form of hardware, or in the form of a combination of software and hardware. It should be noted that each module in an intelligent aviation material inventory management system in this embodiment corresponds one-to-one to each step in an intelligent aviation material inventory management method in the aforementioned embodiment. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned intelligent aviation material inventory management method, which will not be repeated here.

[0075] Based on the above embodiments, Figure 3 FIG. 1 is a schematic diagram of the structure of an aviation material demand forecasting and scheduling device according to an implementation mode of the present application, such as Figure 3 As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330 and a communication bus 340, wherein the processor 310, the communication interface 320 and the memory 330 communicate with each other through the communication bus 340. The processor 310 may call the logic instructions in the memory 330 to execute a method for forecasting and scheduling demand for aviation materials, the method comprising: collecting usage data and multiple influencing factor data of several aviation materials of a preset time length; for any aviation material, calculating the trend similarity factor between each influencing factor data and the usage data of the aviation material, and calculating the change amplitude similarity factor between each influencing factor data and the usage data of the aviation material; adjusting the preset lag order of each influencing factor data based on the trend similarity factor and the change amplitude similarity factor to obtain the improved lag order of each influencing factor data, constructing a vector autoregression model based on the improved lag order to obtain the predicted demand of the aviation material; and completing aviation material scheduling based on the predicted demand of each aviation material.

[0076] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0077] On the basis of the above embodiments, on the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the aviation material demand forecasting and scheduling method provided by the above methods, which method includes: collecting usage data and multiple influencing factor data of several types of aviation materials for a preset time length; for any type of aviation material, calculating the trend similarity factor between each influencing factor data and the aviation material usage data, and calculating the change amplitude similarity factor between each influencing factor data and the aviation material usage data; adjusting the preset lag order of each influencing factor data based on the trend similarity factor and the change amplitude similarity factor to obtain the improved lag order of each influencing factor data, constructing a vector autoregression model based on the improved lag order to obtain the predicted demand for the aviation material; and completing aviation material scheduling based on the predicted demand for each aviation material.

[0078] On the basis of the above embodiments, on another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the aviation material demand forecasting and scheduling method provided by the above methods, the method comprising: collecting usage data of several types of aviation materials and multiple influencing factor data of a preset time length; for any type of aviation material, calculating the trend similarity factor between each influencing factor data and the usage data of the aviation material, and calculating the change amplitude similarity factor between each influencing factor data and the usage data of the aviation material; adjusting the preset lag order of each influencing factor data based on the trend similarity factor and the change amplitude similarity factor to obtain an improved lag order of each influencing factor data, constructing a vector autoregression model based on the improved lag order to obtain the predicted demand for the aviation material; and completing aviation material scheduling based on the predicted demand for each aviation material.

[0079] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for forecasting and scheduling aviation material demand, characterized in that: The aviation material demand forecasting and scheduling method comprises: Collect usage data of several types of aviation materials and multiple influencing factors for a preset time period; For any type of aviation material, calculate the trend similarity factor between each influencing factor data and the usage data of the aviation material, and calculate the change amplitude similarity factor between each influencing factor data and the usage data of the aviation material; Adjusting the preset lag order of each influencing factor data based on the trend similarity factor and the change amplitude similarity factor to obtain an improved lag order of each influencing factor data, constructing a vector autoregression model according to the improved lag order to obtain the predicted demand for the aviation material; Complete aviation material scheduling based on the predicted demand for each aviation material.

2. The method for forecasting and scheduling aviation material demand according to claim 1, characterized in that: The calculating of trend similarity factors between the data of each influencing factor and the usage data of the aviation material includes: respectively calculating a first difference sequence of each influencing factor data and a second difference sequence of the usage data of the aviation material; A trend similarity factor between the use data of the aviation material and each influencing factor data is determined based on each of the first difference sequences and the second difference sequences.

3. The method for forecasting and scheduling aviation material demand according to claim 2, characterized in that: The determining, based on each of the first difference sequence and the second difference sequence, a trend similarity factor between the use data of the aviation material and each influencing factor data comprises: For any influencing factor data, compare the data difference at the same time point in the second difference sequence of the aviation material usage data and the first difference sequence of the influencing factor data to obtain the trend similarity at each time point; A trend similarity factor between the aviation material usage data and the influencing factor data is determined based on the trend similarity at each time point.

4. The method for forecasting and scheduling aviation material demand according to claim 1, characterized in that: The calculation of the variation range approximation factor between the influencing factor data and the aviation material usage data includes: respectively calculating the attention sequence between each influencing factor data and the usage data of the aviation material; Based on the plurality of focus sequences, a change magnitude similarity factor between each influencing factor data and the aviation material usage data is determined.

5. The method for forecasting and scheduling aviation material demand according to claim 4, characterized in that: The respectively calculating the attention sequence between each influencing factor data and the aviation material usage data includes: For any influencing factor data, the absolute values ​​of the data differences between the influencing factor data and the aviation material usage data at the same time point are calculated in chronological order, and a focus sequence between the influencing factor data and the aviation material usage data is constructed in chronological order.

6. The method for forecasting and scheduling aviation material demand according to claim 4, characterized in that: The determining of the change amplitude similarity factor between each influencing factor data and the aviation material usage data based on the plurality of focus sequences includes: Determine a key influence sequence between each influencing factor data and the use data of the aviation material based on each of the focus sequences; The variation amplitude similarity factor between the data of each influencing factor and the usage data of the aviation material is determined according to the numerical fluctuation degree of the key influencing sequence between the data of each influencing factor and the usage data of the aviation material.

7. The method for forecasting and scheduling aviation material demand according to claim 1, characterized in that: The step of adjusting the preset lag order of each influencing factor data based on the trend similarity factor and the change amplitude similarity factor to obtain the improved lag order of each influencing factor data includes: Based on the trend similarity factor and the change amplitude similarity factor, the preset lag order of each influencing factor data is adjusted using the following formula (1) to obtain the improved lag order of each influencing factor data: in, represents the preset lag order, Indicates The variation range between the influencing factor data and the target aviation material usage data is similar to the factor, Indicates The trend similarity factor between the influencing factor data and the target aviation material usage data, Indicates the preset maximum value of the lag order. represents the first preset coefficient, Indicates the second preset coefficient.

8. An aviation material demand forecasting and scheduling system, characterized in that: The aviation material demand forecasting and scheduling system includes: A data collection module is used to collect usage data of several types of aviation materials and multiple influencing factor data for a preset time period; Each improvement factor calculation module is used to calculate, for any type of aviation material, a trend similarity factor between each influencing factor data and the use data of the aviation material, and a change range similarity factor between each influencing factor data and the use data of the aviation material; An aviation material demand prediction module is used to adjust the preset lag order of each influencing factor data based on the trend similarity factor and the change amplitude similarity factor to obtain an improved lag order of each influencing factor data, and construct a vector autoregression model according to the improved lag order to obtain the predicted demand of the aviation material; The aviation material scheduling module is used to complete aviation material scheduling based on the predicted demand for each aviation material.

9. An aviation material demand forecasting and scheduling device, characterized in that: The aviation material demand forecasting and scheduling equipment includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the aviation material demand forecasting and scheduling method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for forecasting and scheduling aviation material demand according to any one of claims 1 to 7 is implemented.

Images