WO2024125485A1 - 旅客托运行李信息预测方法及相关设备 - Google Patents

旅客托运行李信息预测方法及相关设备 Download PDF

Info

Publication number
WO2024125485A1
WO2024125485A1 PCT/CN2023/138046 CN2023138046W WO2024125485A1 WO 2024125485 A1 WO2024125485 A1 WO 2024125485A1 CN 2023138046 W CN2023138046 W CN 2023138046W WO 2024125485 A1 WO2024125485 A1 WO 2024125485A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
checked baggage
passenger
rule
departing
Prior art date
Application number
PCT/CN2023/138046
Other languages
English (en)
French (fr)
Inventor
张彧龙
李阳
赵中星
陈铮
韩跃
梁亚中
牛冰倩
李文孝
李超
李昕冉
于萌
任嘉勉
Original Assignee
中国民航信息网络股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国民航信息网络股份有限公司 filed Critical 中国民航信息网络股份有限公司
Publication of WO2024125485A1 publication Critical patent/WO2024125485A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06314—Calendaring for a resource

Definitions

  • the present disclosure relates to the field of aviation transportation technology, and in particular to a passenger checked baggage information prediction method and related equipment.
  • Flight loading balance is an important part of flight departure, and the weight of checked baggage of passengers on departing flights is one of the core data of loading business.
  • the weight of checked baggage of passengers on a loading flight In order to obtain the actual weight of checked baggage of passengers on a loading flight, it is necessary to calculate the actual value of the weight of checked baggage after all passengers have completed check-in and baggage check-in.
  • the weight of checked baggage of flight passengers needs to be obtained in advance, that is, the data is obtained before the passengers have completed the check-in procedures, or even before any passengers have checked in. Therefore, it is necessary to estimate the weight of checked baggage of passengers on loading flights in advance.
  • the existing solution to estimating the weight of checked baggage for passengers on loaded flights is for operators to directly estimate the weight based on their personal historical experience.
  • This solution has the following disadvantages: operators rely solely on their own experience to make estimates, which has a low reference value. This results in a large difference between the estimated value and the actual value, which brings risks and hidden dangers to production safety; in addition, the estimated values given by different operators are inconsistent and have large individual differences, which also brings risks and hidden dangers to production safety.
  • the present disclosure provides a passenger checked baggage information prediction method and related equipment, which automatically determines the checked baggage weight of the departing passengers of the current flight according to the departing passenger data corresponding to the current flight and the historical data of the actual checked baggage of the departing passengers, as well as the related rule data, so as to overcome a series of drawbacks of the existing manual estimation method.
  • a method for predicting passenger checked baggage information comprising:
  • the weight of checked baggage of the departing passengers of the current flight is predicted according to the departing passenger data, the historical data of checked baggage of the actual departing passengers and the target rule data, so as to perform load balancing processing on the current flight based on the predicted weight of checked baggage of the departing passengers.
  • a passenger checked baggage information prediction device comprising:
  • the data acquisition module is used to obtain the departing passenger data corresponding to the current flight and the historical data of checked baggage of the actual departing passengers;
  • a rule determination module used to determine target rule data required for predicting passenger checked baggage information for the current flight
  • the information prediction module is used to predict the weight of the checked baggage of the departing passengers of the current flight according to the departing passenger data, the historical data of the checked baggage of the actual departing passengers and at least part of the target rule data, so as to perform load balancing processing on the current flight based on the weight of the checked baggage of the departing passengers.
  • a computer-readable medium stores a computer program, wherein the computer program includes a program code for executing the passenger checked baggage information prediction method provided by the present disclosure.
  • a computer program product includes a computer program carried on a non-transitory computer-readable medium, wherein the computer program contains program codes for executing the passenger checked baggage information prediction method provided by the present disclosure.
  • the passenger checked baggage information prediction method and related equipment obtain the departing passenger data and the actual departing passenger checked baggage history data corresponding to the current flight, determine the target rule data required for predicting the passenger checked baggage information of the current flight, and predict the weight of the departing passenger checked baggage of the current flight based on the departing passenger data of the current flight, the actual departing passenger checked baggage history data and at least part of the target rule data, so as to perform load balancing processing on the current flight based on the predicted departing passenger checked baggage weight.
  • the present disclosure can effectively improve the various drawbacks of the existing manual estimation method by automatically predicting the departing passenger checked baggage weight of the current flight based on relevant data and rules.
  • the reference data based on which the prediction is based is relatively comprehensive, which improves the reference value of the predicted value of the departing passenger checked baggage weight, makes the predicted value closer to the actual value, and does not cause deviations in the prediction results due to individual differences, thus avoiding risks and hidden dangers to production safety.
  • FIG1 is a flow chart of a method for predicting passenger checked baggage information provided by the present disclosure
  • FIG2 is a structural diagram of a device for estimating checked baggage for passengers on a flight in an application example provided by the present disclosure
  • FIG3 is another schematic diagram of a flow chart of a passenger checked baggage information prediction method provided by the present disclosure.
  • FIG4 is another schematic diagram of a flow chart of a passenger checked baggage information prediction method provided by the present disclosure.
  • FIG5 is a schematic diagram of another flow chart of the passenger checked baggage information prediction method provided by the present disclosure.
  • FIG6 is a schematic diagram of the component relationship of each component included in the device in FIG2 provided by the present disclosure.
  • FIG. 7 is a structural diagram of the passenger checked baggage information prediction device provided by the present invention.
  • the present invention provides a method, device, computer readable medium and computer program product for predicting passenger checked baggage information.
  • Computer program product
  • the passenger checked baggage information prediction method provided by the present disclosure at least includes the following processing flow:
  • Step 101 Obtain the departing passenger data corresponding to the current flight and the historical data of checked baggage of the actual departing passengers.
  • FlightInfo includes but is not limited to part or all of the information including airline, flight number, flight date, departure station, arrival station, aircraft registration number, aircraft cabin layout, weight unit, etc.
  • the corresponding flight information is as follows:
  • Departure passenger information includes but is not limited to all or part of the information including airline, flight number, flight date, departure station, arrival station, cabin class, number of passengers booking, number of passengers checked in, etc.
  • the number of passengers booking is fixed information, and the number of passengers checked in changes dynamically as passengers check in.
  • the departure station is Beijing (PEK)
  • the arrival stations of the passengers include Shanghai (SHA) and Guangzhou (CAN), which means that some passengers are from Beijing to Shanghai, and some passengers are from Beijing to Guangzhou.
  • the cabins include J cabin and Y cabin.
  • the historical data of the actual checked baggage of the departing passengers corresponding to the current flight can be queried.
  • the actual historical data of checked baggage of departing passengers (BaggageHistory), including but not limited to part or all of the information of airline, flight number, flight date, departure station, arrival station, actual number of passengers, actual weight of checked baggage of passengers, predicted weight of checked baggage of passengers, average baggage weight per person, prediction error, etc.
  • the airline, flight number, flight date, departure station, and arrival station correspond to the airline, flight number, flight date, departure station, and arrival station of the above-mentioned FlightInfo;
  • the actual number of passengers is the number of passengers who have checked in of the above-mentioned PassengerInfo collected after the flight is closed;
  • the actual weight of the passenger's checked baggage is the total weight of the checked baggage collected after the flight is closed;
  • the predicted weight of the passenger's checked baggage is the weight value calculated by this device;
  • the per capita luggage weight actual weight of passenger checked baggage/actual number of passengers;
  • prediction error
  • an application example of the method of the present disclosure which implements a device for estimating checked baggage for passengers on a loaded flight based on the method of the present disclosure, which includes at least a departing passenger data collection component, a departing actual passenger checked baggage historical data extraction component, an estimation rule data maintenance component, and a departing passenger checked baggage calculation component, so as to implement the processing process of the method of the present disclosure based on the components of the device.
  • Departure passenger data collection component its main function is to collect departure passenger information in real time based on the input flight information
  • the component for extracting historical data of checked baggage of actual departing passengers its main function is to extract relevant data based on the historical data of checked baggage of actual departing passengers;
  • Estimation rule data maintenance component its main function is to store and maintain estimation rule related data
  • Departing passenger checked baggage calculation component The main function is to calculate the weight of departing passenger checked baggage based on the departing passenger data, the actual historical data of departing passenger checked baggage, and the estimation rule parameters.
  • a departing passenger data collection component may be used to collect departing passenger information in real time based on input flight information.
  • Step 102 Determine target rule data required for predicting passenger checked baggage information for the current flight.
  • the present disclosure stores and maintains a corresponding estimation rule data set, which includes passenger checked baggage prediction rules and related parameter data.
  • the estimation rule data maintenance component can be used to store and maintain the estimation rule related data to form an estimation rule data set, so as to query the required rule data for the weight prediction of the checked baggage of departing passengers.
  • estimation rule-related data stored and maintained by the estimation rule data maintenance component includes but is not limited to:
  • Baggage Predict Rule including airline, flight number, departure station, arrival station, and prediction rule.
  • the relevant data of Baggage Predict Rule is input by the user, and there are two options for prediction rule, namely history data prediction (HistoryPredict) and time prediction.
  • Time series prediction (ARIMAPredict) represents directly using the relevant data of BaggageHistory (the historical data of the actual checked baggage of departing passengers) for prediction and using the ARIMA model (Autoregressive Integrated Moving Average model) for time series prediction.
  • ARIMA model Automatic Integrated Moving Average model
  • the arrival stations of the baggage include Shanghai (SHA) and Guangzhou (CAN), as shown below:
  • the above-mentioned time series prediction rules include: prediction rules represented by a pre-built ARIMA (Autoregressive Integrated Moving Average) model; the ARIMA model is obtained by model training based on time series data samples of the weight of checked baggage of departing passengers.
  • ARIMA Automatic Integrated Moving Average
  • Static data of passenger checked baggage weight (StaticBaggageWeight), including airline, departure station, arrival station, cabin, and per capita baggage weight, which are input by the user.
  • StaticBaggageWeight including airline, departure station, arrival station, cabin, and per capita baggage weight, which are input by the user.
  • the premise of using prediction rules for prediction is that there is sufficient historical data. If there is not enough historical data, it is necessary to directly use the data of StaticBaggageWeight (static data of passenger checked baggage weight) for prediction.
  • the corresponding airline for the above flight xx1111 is xx. If the departure station is Beijing (PEK), then the arrival stations for the baggage include Shanghai (SHA) and Guangzhou (CAN), and the cabins include J and Y.
  • PEK departure station
  • SHA Shanghai
  • CAN Guangzhou
  • J and Y The specific examples are as follows:
  • Static data of passenger checked baggage density (StaticBaggageDensity), including airlines and average baggage density, the data is input by the user.
  • ULD static data including airline, aircraft registration number, whether it is a containerized aircraft, default ULD type, default ULD volume, default ULD deadweight, and the data is input by the user.
  • the corresponding airline is xx.
  • the departure station is Beijing (PEK)
  • the aircraft registration number is B1234.
  • the flight xx1111 mentioned above has a flight date of March 1, 2022.
  • the specific examples are as follows:
  • the historical data prediction parameters (HistoryPredictParameters) should be included, including But not limited to airline, flight number, departure station, arrival station, date type, historical data percentage within thirty days, historical data percentage within seven days, historical data percentage before the seventh day, minimum valid days of historical data, data is entered by the user.
  • the date types include workday (Monday to Friday, excluding statutory holidays), weekend (Saturday and Sunday, excluding statutory holidays), and holiday (statutory holidays).
  • the input data must meet the following conditions: each flight (same airline, flight number, departure station, arrival station) needs to contain all date types; the sum of the historical data percentage within thirty days, the historical data percentage within seven days, and the historical data percentage on the seventh day in each record must be 100%; the minimum valid days of historical data must be greater than 0 and less than or equal to 30.
  • the arrival stations of the baggage include Shanghai (SHA) and Guangzhou (CAN).
  • SHA Shanghai
  • CAN Guangzhou
  • the prediction rule selected in BaggagePredictRule is ARIMAPredict
  • the time series prediction parameters (ARIMAPredictParameters) should be included, including airline, flight number, departure station, The arrival station, date type, daily forecast percentage, weekly forecast percentage, and minimum number of valid days for historical data are input by the user.
  • the date types include workday (Monday to Friday, excluding statutory holidays), weekend (Saturday and Sunday, excluding statutory holidays), and holiday (statutory holidays).
  • each flight (same airline, flight number, departure station, and arrival station) must contain all date types; the sum of the daily forecast percentage and weekly forecast percentage in each record must be 100%; the minimum number of valid days for historical data must be greater than or equal to 15 and less than or equal to 30.
  • the arrival stations of the baggage include Shanghai (SHA) and Guangzhou (CAN).
  • SHA Shanghai
  • CAN Guangzhou
  • this step 102 can specifically use the departing passenger checked baggage calculation component to determine the target rule data required for predicting the passenger checked baggage information of the current flight from the estimation rule data set, so as to subsequently predict the weight of the departing passenger checked baggage of the current flight based on the target rule data.
  • This process can be further implemented as follows:
  • the target prediction rule is obtained by searching the passenger checked baggage prediction rule (BaggagePredictRule) based on the airline, flight number, departure station, and arrival station.
  • the target prediction rule is a historical data prediction rule in the passenger checked baggage prediction rule, determine whether the historical data prediction conditions are met. If so, determine that the target rule data includes the historical data prediction rule and its corresponding rule parameters in the estimation rule data set. If not, the target rule data includes the passenger checked baggage weight static data in the estimation rule data set.
  • the prediction rule is a historical data prediction rule HistoryPredict
  • whether the historical data prediction condition is met may be determined according to the following process, and the corresponding target rule data may be determined for the satisfied or unsatisfied condition:
  • DateType Find Holiday according to the flight date of the current flight information (FlightInfo) and determine whether it is a holiday. If not, determine whether it is a weekday or a weekend.
  • ValidDataMin Get the minimum valid days of historical data: Find HistoryPredictParameters according to the airline, flight number, departure station, arrival station, and date type (dateType) of the current flight information (FlightInfo) to get the minimum valid days of historical data;
  • the prediction is made based on the historical data, that is, in this case, the target rule data includes the historical data prediction rule and its corresponding rule parameters in the estimation rule data set. Otherwise, the prediction is made based on the passenger checked baggage weight static data (StaticBaggageWeight), that is, in this case, the target is determined
  • the rule data includes the static data of passenger checked baggage weight in the estimation rule data set.
  • the prediction rule is HistoryPredict, and the following processing is further performed:
  • Get effective days Query historical data based on xx, 1111, PEK, SHA, 2022/1/30-2022/2/28.
  • the target prediction rule is a time series prediction rule in the passenger checked baggage prediction rule, determine whether the time series prediction conditions are met. If so, determine that the target rule parameters include the time series prediction rule and its corresponding rule parameters in the estimation rule data set; if not, the target rule data includes the passenger checked baggage weight static data in the estimation rule data set.
  • the prediction rule is the time series prediction rule ARIMAPredict
  • it can be determined whether the time series prediction condition is met according to the following process, and the corresponding target rule data can be determined according to the situation of being met or not being met:
  • DateType Find Holiday according to the flight date of the current flight information (FlightInfo), and determine whether it is a holiday. If not, determine whether it is a weekday or a weekend;
  • ValidDataMin Get the minimum valid days of historical data: Find ARIMAPredictParameters according to the airline, flight number, departure station, arrival station, and date type (dateType) of the current flight information (FlightInfo) to get the minimum valid days of historical data;
  • the prediction rule is ARIMAPredict, and the following processing is further performed:
  • Get the effective days of daily forecast query historical data based on xx, 1111, PEK, CAN, 2022/1/30 to 2022/2/28.
  • Step 103 predict the weight of checked baggage of the departing passengers of the current flight according to the departing passenger data, the historical data of checked baggage of the actual departing passengers and at least part of the target rule data, so as to perform load balancing processing on the current flight based on the predicted weight of checked baggage of the departing passengers.
  • the weight of checked baggage of the departing passengers of the current flight is predicted based on the departing passenger data corresponding to the current flight, the historical data of checked baggage of the actual departing passengers and at least part of the target rule data.
  • the predicted value is specifically the predicted total weight of checked baggage of the departing passengers of the current flight.
  • target rule data includes the historical data prediction rule and its corresponding rule parameters, extract the historical parameter data required by the historical data prediction rule from the actual passenger checked baggage historical data, and predict the checked baggage weight of the departing passengers of the current flight based on the departing passenger data, the historical parameter data, and the historical data prediction rule and its rule parameters.
  • a component for extracting historical data of checked baggage of actual departing passengers can be used to extract historical parameter data required by the historical data prediction rule from the historical data of checked baggage of actual passengers corresponding to the current flight, so as to obtain the required historical parameter data, including but not limited to:
  • Average per capita baggage weight of historical data within 30 days (avgWeight30): Based on the airline, flight number, departure station, arrival station, [flight date - 30, flight date - 1] of the current flight information (FlightInfo), use the historical data extraction component of the actual checked baggage of departing passengers to extract the average per capita baggage weight;
  • Valid days of historical data within 7 days (effective7): according to the airline, flight number, departure station, arrival station, [flight date - 7, flight date - 1] of the current flight information (FlightInfo), use the departing passenger checked baggage historical data extraction component to extract the valid days;
  • Average per capita baggage weight of historical data within 7 days (avgWeight7): Based on the airline, flight number, departure station, arrival station, [flight date - 7, flight date - 1] of the current flight information (FlightInfo), use the historical data extraction component of the actual checked baggage of departing passengers to extract the average per capita baggage weight;
  • the effective days of historical data before the seventh day (effectiveAt7): based on the current flight information (FlightInfo)
  • Average per capita luggage weight in historical data before the seventh day (avgWeightAt7): Based on the airline, flight number, departure station, arrival station, flight date - 7 of the current flight information (FlightInfo), use the historical data extraction component of the actual checked luggage of departing passengers to extract the average per capita luggage weight.
  • Historical data prediction parameters According to the airline, flight number, departure station, arrival station, and date type (dateType) of the current flight information (FlightInfo), search for HistoryPredictParameters to obtain the percentage of historical data within thirty days (percentage30), the percentage of historical data within seven days (percentage7), and the percentage of historical data before the seventh day (percentageAt7).
  • the departing passenger checked baggage calculation component is used to calculate the weight of the departing passenger checked baggage of the current flight based on various data obtained based on the historical data prediction rules, so as to realize the prediction of the weight of the departing passenger checked baggage.
  • An exemplary prediction process is as follows:
  • bagEstWeight (avgWeight30*percentage30+avgWeight7*percentage7)/(percentage30+percentage7)*bookNum;
  • bagEstWeight (avgWeight30*percentage30+avgWeight7*percentage7+avgWeightAt7*percentageAt7)*bookNum.
  • the passenger checked baggage weight (bagEstWeight) needs to be calculated based on historical data, that is, the weight of the departing passenger checked baggage of the current flight is predicted based on the historical data prediction rules.
  • the process is as follows:
  • avgWeightAt7 Get the average luggage weight per capita (avgWeightAt7) of historical data before the seventh day: Query historical data based on xx, 1111, PEK, SHA, and 2022/2/22. There is no historical data that meets the conditions, and avgWeightAt7 has no value.
  • target rule data includes the time series prediction rule and its rule parameters
  • extract the historical time series parameter data required by the time series prediction rule from the actual passenger checked baggage history data, and predict the checked baggage weight of the departing passengers of the current flight based on the departing passenger data, the historical time series parameter data, and the time series prediction rule and its rule parameters.
  • the method 32) can specifically use the departure actual passenger checked baggage historical data extraction component to extract the historical time series parameter data required by the time series prediction rule from the actual passenger checked baggage historical data corresponding to the current flight, and in the case of missing historical time series parameter data, use the linear difference method to complete the time series, including but not limited to:
  • timeSeriesByDay According to the current flight information (FlightInfo) of airline, flight number, departure station, arrival station, [flight date - 30, Flight date - 1], extract the time series of per capita luggage weight from the above-mentioned historical data extraction component of actual passenger checked luggage, and use the linear difference method to complete the time series. If the data is missing at the beginning and end of the time series, directly use the most recent and valid data.
  • FlightInfo current flight information of airline, flight number, departure station, arrival station, [flight date - 30, Flight date - 1]
  • the extracted time series is as follows: data from 2020/1/1 to 2020/1/6 needs to be obtained, but there are only two days of data in the database, namely 2020/1/2 and 2020/1/5, so other data needs to be completed.
  • the data after the head and tail data are filled in are as follows:
  • timeSeriesByWeek Based on the current The airline, flight number, departure station, arrival station, and the corresponding dates of the thirty identical weeks before the flight date of the flight information (FlightInfo) are extracted from the above-mentioned historical data extraction component of the actual checked baggage of departing passengers, and the time series is supplemented by the linear difference method. If the data is missing at the beginning and end of the time series, the most recent and valid data is used directly.
  • the extracted time series is as follows, and the data of ⁇ 2022/3/1, 2022/3/8, 2022/3/15, 2022/3/22, 2022/3/29 ⁇ needs to be obtained.
  • time series forecast parameters Search ARIMAPredictParameters according to the airline, flight number, departure station, arrival station, and date type (dateType) of the current flight information (FlightInfo), and get the daily forecast percentage (percentageByDay) and weekly forecast percentage (percentageByWeek).
  • the departing passenger checked baggage calculation component is used to calculate the weight of the departing passenger checked baggage of the current flight based on various data obtained based on the time series prediction rule, so as to realize the prediction of the weight of the departing passenger checked baggage.
  • An exemplary prediction process is as follows:
  • Calculate the ARIMA model parameters P1, D1, and Q1 for daily forecasts Set the D1 value range to 1 and 2, the P1 value range to 1-5, and the Q1 value range to 1-5. Traverse all values of P1, D1, and Q1, and calculate the value according to The first 25 days of timeSeriesByDay data and the corresponding model parameters P1, D1, Q1 are used to predict the passenger checked baggage weight for the next 5 days in the ARIMA model. The predicted values are compared with the actual values to find the set of P1, D1, Q1 with the smallest sum of squared errors.
  • Calculate the weekly forecast ARIMA model parameters P2, D2, and Q2 set the D2 value range to 1 and 2, the P2 value range to 1-5, and the Q2 value range to 1-5. Traverse all values of P2, D2, and Q2, and use the first 25 days of timeSeriesByWeek data and the corresponding model parameters P2, D2, and Q2 to predict the passenger checked baggage weight for the next 5 days. Compare the results with the actual values to find the set of P2, D2, and Q2 with the smallest sum of squared errors.
  • Calculate the daily predicted average luggage weight per person (avgWeightByDay): Based on the timeSeriesByDay data and the corresponding model parameters P1, D1, and Q1, use the ARIMA model to predict the average luggage weight per person for the current flight.
  • Calculate the weekly average luggage weight per person (avgWeightByWeek): Based on the timeSeriesByWeek data and the corresponding model parameters P2, D2, and Q2, use the ARIMA model to predict the average luggage weight per person for the current flight.
  • bagEstWeight (avgWeightByDay*percentageByDay+avgWeightByWeek*percentageByWeek)*bookNum.
  • target rule data includes the passenger checked baggage weight static data, predict the departing passenger checked baggage weight of the current flight based on the departing passenger data and the passenger checked baggage weight static data.
  • the required static data of checked baggage weight and departing passenger data can be obtained, including but not limited to:
  • the departing passenger checked baggage calculation component is used to calculate the departing passenger checked baggage weight bagEstWeight of the current flight according to the acquired departing passenger data and the static data of the passenger checked baggage weight, so as to realize the departing passenger checked baggage weight prediction, as follows:
  • the passenger checked baggage information prediction method obtains the departing passenger data and the actual departing passenger checked baggage history data corresponding to the current flight, determines the target rule data required for predicting the passenger checked baggage information of the current flight, and predicts the weight of the departing passenger checked baggage of the current flight based on the departing passenger data of the current flight, the actual departing passenger checked baggage history data and at least part of the target rule data, so as to perform load balancing processing on the current flight based on the predicted departing passenger checked baggage weight.
  • the present disclosure can effectively improve the various drawbacks of the existing manual estimation method by automatically predicting the departing passenger checked baggage weight of the current flight based on relevant data and rules.
  • the reference data on which the prediction is based is relatively comprehensive, which improves the reference value of the predicted value of the departing passenger checked baggage weight, makes the predicted value closer to the actual value, and does not cause deviations in the prediction results due to individual differences. Avoid risks and hidden dangers to production safety.
  • the passenger checked baggage information prediction method provided by the present disclosure may further include:
  • Step 104 Determine the total weight of the checked baggage and the container of the departing passengers of the current flight according to the predicted weight of the checked baggage of the departing passengers.
  • the total weight of the departing passenger's checked baggage and the container is the sum of the departing passenger's checked baggage weight and the container weight; if the current flight is a non-containerized aircraft, the total weight of the departing passenger's checked baggage and the container is the departing passenger's checked baggage weight.
  • the corresponding airline is xx.
  • the departure station is Beijing (PEK)
  • the aircraft registration number is B1234.
  • ULDConfiguration based on xx and B1234, the following record is obtained.
  • the aircraft of this flight is a containerized aircraft, so the container weight needs to be calculated additionally.
  • baggageDensity Get the passenger checked baggage density (baggageDensity), find the passenger checked baggage density static data (StaticBaggageDensity) according to the airline of the current flight information (FlightInfo), and get the corresponding average baggage density;
  • uldNum (bagEstWeight/baggageDensity)/uldPerVolumn, and round up the result;
  • This embodiment can provide data basis for the loading balance of the current flight by determining the total weight of the checked baggage and the container of the departing passengers of the current flight.
  • the weight of the checked baggage of the departing passengers of the current flight is automatically predicted according to relevant data and rules, thereby improving the reference value of the predicted value of the weight of the checked baggage of the departing passengers, and correspondingly improving the reference value of the total weight of the checked baggage and the container of the departing passengers, making the predicted value closer to the actual value, and avoiding risks and hidden dangers to production safety.
  • the passenger checked baggage information prediction method provided by the present disclosure may further include:
  • Step 105 Acquire the checked baggage data of the actual departing passengers of the current flight, and update and store the acquired data into the checked baggage history data of the actual departing passengers of the current flight after the current flight is closed.
  • a device for estimating checked baggage for passengers on loaded flights implemented based on the method disclosed herein may further include a component for processing data on checked baggage of actual departing passengers, as specifically shown in FIG2 .
  • This component may be used to collect checked baggage data of passengers in real time, and store the relevant data in the historical data on checked baggage of actual departing passengers after the flight is closed.
  • BaggageHistory data is as follows:
  • This embodiment obtains the checked baggage data of the actual departing passengers of the current flight, and updates and stores the obtained data in the historical data of the checked baggage of the actual departing passengers after the current flight is closed, thereby continuously updating the historical data of the checked baggage of the actual departing passengers corresponding to the current flight over time, thereby facilitating the provision of historical data support for the checked baggage information of passengers of the flight at future times.
  • the passenger checked baggage information prediction method provided by the present disclosure may further include:
  • Step 106 determining the deviation between the predicted value and the actual value of the checked baggage weight of the departing passengers of the current flight
  • Step 107 When the deviation reaches a preset threshold, early warning processing is performed to adjust the corresponding rule parameters in the estimation rule data set based on the early warning processing, and the weight of checked baggage of departing passengers of the current flight is predicted based on the adjusted rule parameters at a subsequent time.
  • a real-time warning component may be further provided in the device for estimating the checked baggage of passengers on a loaded flight implemented based on the method disclosed herein, as specifically shown in FIG2 .
  • This component may be used accordingly to execute the processing of this embodiment, so as to implement real-time warning processing when the deviation between the predicted value and the actual value of the checked baggage weight of departing passengers reaches a preset threshold.
  • FIG. 6 a schematic diagram of the relationship between the components of the device for estimating checked baggage for passengers on a flight in FIG. 2 is further provided.
  • the real-time warning component is equipped with warning parameters (PrewarningParameters), including but not limited to airline, flight number, departure station, arrival station, number of days, maximum average error, and last warning date.
  • the last warning date is automatically generated by the system (each time a record is generated, the current date is automatically written into the last warning date. Subsequently, as long as a warning is generated for the corresponding flight, the corresponding warning date will be written). The remaining data is entered by the user.
  • the baggage arrival stations include SHA and CAN.
  • An exemplary warning process based on the real-time warning component is as follows (each time a flight is closed, the process will be automatically executed after the departing passenger checked baggage data processing component updates the BaggageHistory):
  • the baggage arrival stations include SHA and CAN, which need to be judged separately:
  • avgDiff 7.46%
  • avgDiffMaxValue 5%
  • avgDiff > avgDiffMaxValue so it is necessary to send an early warning message to the user, and at the same time change the last early warning date to the current flight date in the corresponding PrewarningParameters record.
  • the modified record is as follows ("2022/3/1" is the modified part):
  • the user is promptly informed to modify the estimation rule parameters based on the early warning processing. After the user modifies the relevant rule parameters in the estimation rule data set based on the early warning information, the updated rule parameters will be used for the subsequent prediction of the checked baggage information of departing passengers, and the deviation between the predicted value and the actual value can be effectively reduced.
  • the present disclosure also provides a passenger checked baggage information prediction device, referring to FIG7 , the device comprises:
  • the data acquisition module 10 is used to acquire the departing passenger data corresponding to the current flight and the historical data of checked baggage of the actual departing passengers;
  • a rule determination module 20 used to determine target rule data required for predicting passenger checked baggage information for the current flight
  • the information prediction module 30 is used to predict the weight of the checked baggage of the departing passengers of the current flight according to the departing passenger data, the historical data of the checked baggage of the actual departing passengers and at least part of the target rule data, so as to perform load balancing processing on the current flight based on the weight of the checked baggage of the departing passengers.
  • the rule determination module 20 is specifically configured to:
  • estimation rule data set includes a passenger checked baggage prediction rule and related parameter data
  • the target prediction rule is a historical data prediction rule in the passenger checked baggage prediction rule, determine whether the historical data prediction condition is met, and if so, determine that the target rule data includes the historical data prediction rule and its corresponding rule parameters in the estimation rule data set; if not, the target rule data includes the passenger checked baggage weight static data in the estimation rule data set;
  • the target prediction rule is a time series prediction rule in the passenger checked baggage prediction rule, determine whether the time series prediction conditions are met; if so, determine that the target rule parameters include the time series prediction rule and its corresponding rule parameters in the estimation rule data set; if not, the target rule data includes the passenger checked baggage weight static data in the estimation rule data set.
  • the information prediction module 30 is specifically configured to:
  • target rule data includes the historical data prediction rule and its rule parameters, extract the historical parameter data required by the historical data prediction rule from the actual passenger checked baggage historical data, and predict the weight of the departing passenger checked baggage of the current flight according to the departing passenger data, the historical parameter data, and the historical data prediction rule and its rule parameters;
  • target rule data includes the time series prediction rule and its rule parameters, extract the historical time series parameter data required by the time series prediction rule from the actual passenger checked baggage history data, and predict the weight of the departing passenger checked baggage of the current flight according to the departing passenger data, the historical time series parameter data, and the time series prediction rule and its rule parameters;
  • the departing passenger checked baggage weight of the current flight is predicted based on the departing passenger data and the passenger checked baggage weight static data.
  • the time series prediction rules include: prediction rules represented by a pre-built ARIMA model;
  • the ARIMA model is obtained by training the model based on time series data samples of the weight of checked baggage of departing passengers.
  • the information prediction module 30 is further used to: determine the total weight of the checked baggage and the container of the departing passengers of the current flight according to the predicted weight of the checked baggage of the departing passengers;
  • the total weight of the departing passenger's checked baggage and container is The sum of the weight of the departing passenger's checked baggage and the weight of the container. If the current flight is a non-containerized aircraft, the total weight of the departing passenger's checked baggage and the container is the departing passenger's checked baggage weight.
  • the above device further comprises:
  • the historical data updating module is used to obtain the checked baggage data of the actual departing passengers of the current flight, and update and store the obtained data into the historical data of the actual departing checked baggage of the passengers of the current flight after the current flight is closed.
  • the above device further comprises:
  • the early warning module is used to: determine the deviation between the predicted value and the actual value of the checked baggage weight of the departing passengers of the current flight; when the deviation reaches a preset threshold, perform early warning processing, adjust the corresponding rule parameters in the estimation rule data set based on the early warning processing, and perform the prediction of the checked baggage weight of the departing passengers of the current flight based on the adjusted rule parameters at a subsequent time.
  • the units/modules involved in the embodiments described in the present disclosure may be implemented by software or hardware.
  • the names of the units/modules do not limit the units themselves in some cases.
  • the first acquisition unit may also be described as a "unit for acquiring at least two Internet Protocol addresses".
  • exemplary types of hardware logic components include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip (SOCs), complex programmable logic devices (CPLDs), and the like.
  • FPGAs field programmable gate arrays
  • ASICs application specific integrated circuits
  • ASSPs application specific standard products
  • SOCs systems on chip
  • CPLDs complex programmable logic devices
  • the present disclosure also provides a computer-readable medium having a computer program stored thereon, wherein the computer program includes a program code for executing the passenger checked baggage information prediction method disclosed in the above method embodiment.
  • a computer readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device.
  • a machine readable medium may be a machine readable signal medium or a machine readable storage medium.
  • a machine readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine readable storage media would include electrical connections based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable and removable hard disk, or a computer program product. Programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above.
  • the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two.
  • the computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above.
  • Computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.
  • a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device.
  • a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried.
  • This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above.
  • the computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device.
  • the program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.
  • the computer-readable medium may be included in the electronic device, or may exist independently without being installed in the electronic device.
  • the present disclosure also provides a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, wherein the computer program contains program codes for executing the passenger checked baggage information prediction method disclosed in the above method embodiment.
  • the processes described in the above reference flow charts can be implemented as computer software programs.
  • the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device, or installed from a ROM.
  • the computer program is executed by a processing device, the above functions defined in the method of the embodiment of the present disclosure are executed.
  • the present disclosure provides a method for predicting passenger checked baggage information, the method comprising:
  • the weight of checked baggage of the departing passengers of the current flight is predicted according to the departing passenger data, the historical data of checked baggage of the actual departing passengers and the target rule data, so as to perform load balancing processing on the current flight based on the predicted weight of checked baggage of the departing passengers.
  • determining the target rule data required for predicting passenger checked baggage information for the current flight includes:
  • estimation rule data set includes a passenger checked baggage prediction rule and related parameter data
  • the target prediction rule is a historical data prediction rule in the passenger checked baggage prediction rule, determine whether the historical data prediction condition is met, and if so, determine that the target rule data includes the historical data prediction rule and its corresponding rule parameters in the estimation rule data set; if not, the target rule data includes the passenger checked baggage weight static data in the estimation rule data set;
  • the target prediction rule is a time series prediction rule in the passenger checked baggage prediction rule, determine whether the time series prediction conditions are met; if so, determine that the target rule parameters include the time series prediction rule and its corresponding rule parameters in the estimation rule data set; if not, the target rule data includes the passenger checked baggage weight static data in the estimation rule data set.
  • predicting the weight of checked baggage of departing passengers of the current flight according to the departing passenger data, the historical data of checked baggage of actual departing passengers and at least part of the target rule data includes:
  • target rule data includes the historical data prediction rule and its rule parameters, extract the historical parameter data required by the historical data prediction rule from the actual passenger checked baggage historical data, and predict the weight of the departing passenger checked baggage of the current flight according to the departing passenger data, the historical parameter data, and the historical data prediction rule and its rule parameters;
  • the target rule data includes the time series prediction rule and its rule parameters
  • the historical time series parameter data required by the time series prediction rule is extracted from the actual passenger checked baggage history data
  • the target rule data is extracted based on the departing passenger data, the historical time series parameter data, and the time series Prediction rules and their rule parameters, predicting the weight of checked baggage of departing passengers of the current flight;
  • the departing passenger checked baggage weight of the current flight is predicted based on the departing passenger data and the passenger checked baggage weight static data.
  • the time series prediction rule includes: a prediction rule characterized by a pre-built ARIMA model
  • the ARIMA model is obtained by training the model based on time series data samples of the weight of checked baggage of departing passengers.
  • the above method after predicting the weight of checked baggage of departing passengers of the current flight, further includes:
  • the total weight of the departing passenger's checked baggage and the container is the sum of the departing passenger's checked baggage weight and the container weight; if the current flight is a non-containerized aircraft, the total weight of the departing passenger's checked baggage and the container is the departing passenger's checked baggage weight.
  • the above method further includes:
  • the actual departing passenger checked baggage data of the current flight is obtained, and after the current flight is closed, the obtained data is updated and stored in the actual departing passenger checked baggage historical data of the current flight.
  • the above method further includes:
  • early warning processing is performed to adjust the corresponding rule parameters in the estimation rule data set based on the early warning processing, and the weight of checked baggage of departing passengers of the current flight is predicted based on the adjusted rule parameters at a subsequent time.
  • the present disclosure further provides a passenger checked baggage information prediction device, comprising:
  • the data acquisition module is used to obtain the departing passenger data corresponding to the current flight and the historical data of checked baggage of the actual departing passengers;
  • a rule determination module used to determine target rule data required for predicting passenger checked baggage information for the current flight
  • the information prediction module is used to predict the number of passengers departing from the airport based on the passenger data and the actual number of checked baggage of the departing passengers.
  • the historical data and at least a part of the target rule data are used to predict the checked baggage weight of the departing passengers of the current flight, so as to perform load balancing processing on the current flight based on the checked baggage weight of the departing passengers.
  • the present disclosure further provides a computer-readable medium having a computer program stored thereon, wherein the computer program includes a program code for executing the passenger checked baggage information prediction method as described above.
  • the present disclosure also provides a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program codes for executing the passenger checked baggage information prediction method as described above.
  • the passenger checked baggage information prediction method, device, computer-readable medium and computer program product provided by the present disclosure have at least the following technical advantages:
  • the disclosure needs to be calculated based on the historical data of checked baggage of actual passengers on the flight and the data of passengers departing from the current flight, which refers to both the historical data of baggage and the existing passenger data.
  • the data reference is comprehensive, and the calculated value is more convincing and has a higher reference value, making the predicted value closer to the actual value, avoiding risks and hidden dangers to production safety;
  • the present disclosure uses the historical data of checked baggage of actual passengers on the flight and the data of passengers departing from the current flight as input, and brings in the corresponding estimation rules for calculation.
  • the calculated value of baggage weight obtained is fixed, and there is no situation where different operators use the system and obtain different calculated values, thus eliminating the hidden dangers caused by individual differences;
  • the present disclosure is provided with an early warning function.
  • the relevant parameters of the estimation rule are not properly set, resulting in a large deviation between the calculated value and the actual value, the user will be informed to modify the estimation rule parameters. After the user modifies the relevant parameters, the new rule parameters will be used for subsequent calculations, effectively reducing the deviation between the calculated value and the actual value.

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Educational Administration (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种旅客托运行李信息预测方法及相关设备，获取当前航班的离港旅客数据和离港实际旅客托运行李历史数据，确定对当前航班进行旅客托运行李信息预测需基于的目标规则数据，并根据当前航班的离港旅客数据、离港实际旅客托运行李历史数据和目标规则数据中的至少部分，预测当前航班的离港旅客托运行李重量，以对当前航班进行航班配载平衡处理。通过根据相关数据及规则自动预测当前航班的离港旅客托运行李重量，可有效改善现有人工估算方式存在的各种弊端，预测所基于的参考数据较为全面，提升了离港旅客托运行李重量预测值的参考价值，且不会因个体差异带来预测结果的偏差，避免了为生产安全带来风险和隐患。

Description

旅客托运行李信息预测方法及相关设备
本公开要求于2022年12月15日提交中国专利局、申请号为202211613120.X、发明名称为″旅客托运行李信息预测方法及相关设备″的中国专利申请的优先权，其全部内容通过引用结合在本公开中。
技术领域
本公开涉及航空运输技术领域，尤其涉及一种旅客托运行李信息预测方法及相关设备。
背景技术
航班配载平衡是航班离港的重要环节，离港航班旅客托运行李重量，是配载业务核心数据之一。为获取某一配载航班实际旅客托运行李重量，需要在所有旅客办理完值机和行李托运后，方可累加计算出托运行李重量实际值。但是根据实际配载业务流程，需提前获取航班旅客托运行李重量，也就是旅客尚未办理完值机手续，甚至还未有旅客办理值机的情况下，就获取到这一数据。因此，需要提前估算出配载航班旅客托运行李重量。
对于配载航班旅客托运行李重量的估算，现有解决方式是，由操作人员根据个人历史经验，直接估算出这一重量。该解决方式存在如下弊端：操作人员仅仅依靠自身经验进行估算，参考价值较低，这就导致人工给出的估算值往往和实际值有较大的差异，从而给生产安全带来风险和隐患；另外，不同的操作人员给出的估算值不一致，个体差异较大，这同样会给生产安全带来风险和隐患。
发明内容
有鉴于此，本公开提供一种旅客托运行李信息预测方法及相关设备，通过根据当前航班对应的离港旅客数据和离港实际旅客托运行李历史数据，以及相关规则数据，自动确定当前航班的离港旅客托运行李重量，以克服现有的人工估算方式存在的一系列弊端。
具体方案如下：
一种旅客托运行李信息预测方法，包括：
获取当前航班对应的离港旅客数据和离港实际旅客托运行李历史数据；
确定对当前航班进行旅客托运行李信息预测所需基于的目标规则数据；
根据所述离港旅客数据、所述离港实际旅客托运行李历史数据和所述目标规则数据中的至少部分，预测当前航班的离港旅客托运行李重量，以基于预测的离港旅客托运行李重量对当前航班进行配载平衡处理。
一种旅客托运行李信息预测装置，包括：
数据获取模块，用于获取当前航班对应的离港旅客数据和离港实际旅客托运行李历史数据；
规则确定模块，用于确定对当前航班进行旅客托运行李信息预测所需基于的目标规则数据；
信息预测模块，用于根据所述离港旅客数据、所述离港实际旅客托运行李历史数据和所述目标规则数据中的至少部分，预测当前航班的离港旅客托运行李重量，以基于所述离港旅客托运行李重量对当前航班进行配载平衡处理。
一种计算机可读介质，其上存储有计算机程序，所述计算机程序包含用于执行本公开所提供的旅客托运行李信息预测方法的程序代码。
一种计算机程序产品，其包括承载在非暂态计算机可读介质上的计算机程序，所述计算机程序包含用于执行本公开所提供的旅客托运行李信息预测方法的程序代码。
根据以上方案可知，本公开提供的旅客托运行李信息预测方法及相关设备，获取当前航班对应的离港旅客数据和离港实际旅客托运行李历史数据，确定对当前航班进行旅客托运行李信息预测所需基于的目标规则数据，并根据当前航班的离港旅客数据、离港实际旅客托运行李历史数据和目标规则数据中的至少部分，预测当前航班的离港旅客托运行李重量，以基于预测的离港旅客托运行李重量对当前航班进行配载平衡处理。本公开通过根据相关数据及规则自动预测当前航班的离港旅客托运行李重量，可有效改善现有人工估算方式存在的各种弊端，预测所基于的参考数据较为全面，提升了离港旅客托运行李重量预测值的参考价值，使预测值与实际值更为接近，且不会因个体差异带来预测结果的偏差，避免了为生产安全带来风险和隐患。
附图说明
结合附图并参考以下具体实施方式，本公开各实施例的上述和其他特征、优点及方面将变得更加明显。贯穿附图中，相同或相似的附图标记表示相同或相似的元素。应当理解附图是示意性的，原件和元素不一定按照比例绘制。
图1是本公开提供的旅客托运行李信息预测方法的一种流程示意图；
图2是本公开提供的一应用示例中配载航班旅客托运行李估算装置的组成结构图；
图3是本公开提供的旅客托运行李信息预测方法的另一种流程示意图；
图4是本公开提供的旅客托运行李信息预测方法的又一种流程示意图；
图5是本公开提供的旅客托运行李信息预测方法的再一种流程示意图；
图6是本公开提供的图2中装置所包含的各组件的组件关系示意图；
图7是本公开提供的旅客托运行李信息预测装置的组成结构图。
具体实施方式
下面将参照附图更详细地描述本公开的实施例。虽然附图中显示了本公开的某些实施例，然而应当理解的是，本公开可以通过各种形式来实现，而且不应该被解释为限于这里阐述的实施例，相反提供这些实施例是为了更加透彻和完整地理解本公开。应当理解的是，本公开的附图及实施例仅用于示例性作用，并非用于限制本公开的保护范围。
本文使用的术语″包括″及其变形是开放性包括，即″包括但不限于″。术语″基于″是″至少部分地基于″。术语″一个实施例″表示″至少一个实施例″；术语″另一实施例″表示″至少一个另外的实施例″；术语″一些实施例″表示″至少一些实施例″。其他术语的相关定义将在下文描述中给出。
需要注意，本公开中提及的″第一″、″第二″等概念仅用于对不同的装置、模块或单元进行区分，并非用于限定这些装置、模块或单元所执行的功能的顺序或者相互依存关系。
需要注意，本公开中提及的″一个″、″多个″的修饰是示意性而非限制性的，本领域技术人员应当理解，除非在上下文另有明确指出，否则应该理解为″一个或多个″。
本公开提供一种旅客托运行李信息预测方法、装置、计算机可读介质及计 算机程序产品。
参见图1提供的旅客托运行李信息预测方法，本公开提供的旅客托运行李信息预测方法，至少包括以下处理流程：
步骤101、获取当前航班对应的离港旅客数据和离港实际旅客托运行李历史数据。
具体可根据当前航班的航班信息查询当前航班的离港旅客数据。
其中，航班信息(FlightInfo)，包括但不限于航空公司、航班号、航班日期、起飞站、到达站、飞机注册号、飞机舱位布局、重量单位等中的部分或全部信息。
比如一个航空公司的多航段航班(1111)，从北京(PEK)-上海(SHA)-广州(CAN)，其包含两个航节：北京-上海、上海-广州，对应航班信息如下：
表1
离港旅客信息(PassengerInfo)，包括但不限于航空公司、航班号、航班日期、起飞站、到达站、舱位、旅客订座人数、旅客已值机人数等中的全部或部分信息。在航班配载过程中，旅客订座人数是已确定的信息，旅客已值机人数是随着旅客办理值机而动态变化的。
比如上述xx1111航班，如果此时始发站在北京(PEK)，那么此时旅客的到达站就包含上海(SHA)和广州(CAN)，代表有的旅客是从北京到上海，有的旅客是从北京到广州，舱位包含J舱和Y舱，旅客信息举例如下：
表2

并可根据所需的历史数据查询条件，查询得到当前航班对应的离港实际旅客托运行李历史数据。
离港实际旅客托运行李历史数据(BaggageHistory)，包括但不限于航空公司、航班号、航班日期、起飞站、到达站、实际旅客人数、旅客托运行李实际重量、旅客托运行李预测重量、人均行李重量、预测误差等中的部分或全部信息。
其中，所述航空公司、航班号、航班日期、起飞站、到达站对应于上述FlightInfo的航空公司、航班号、航班日期、起飞站、到达站；所述实际旅客人数为航班关闭后，采集到的上述PassengerInfo的旅客已值机人数；所述旅客托运行李实际重量为航班关闭后，采集到的托运行李总重；所述旅客托运行李预测重量为通过本装置计算得到的重量值；所述人均行李重量＝旅客托运行李实际重量/实际旅客人数；所述预测误差＝|旅客托运行李预测重量-旅客托运行李实际重量|/旅客托运行李实际重量。
比如上述xx1111航班，如果此时始发站在北京(PEK)，那么此时行李的到达站就包含上海(SHA)和广州(CAN)，举例历史数据如下：
表3

可选的，参见图2，提供了本公开方法的一应用示例，该示例实现了一基于本公开方法的配载航班旅客托运行李估算装置，其中至少包括离港旅客数据采集组件、离港实际旅客托运行李历史数据提取组件、估算规则数据维护组件和离港旅客托运行李计算组件，以基于该装置的各组件实现本公开方法的处理过程。
各组件的功能如下：
离港旅客数据采集组件：主要功能是根据输入的航班信息，实时采集离港旅客信息；
离港实际旅客托运行李历史数据提取组件：主要功能是根据离港实际旅客托运行李历史数据提取相关数据；
估算规则数据维护组件：主要功能是存储与维护估算规则相关数据；
离港旅客托运行李计算组件：主要功能是根据离港旅客数据、离港实际旅客托运行李历史数据、估算规则参数，计算出离港旅客托运行李重量。
在该示例中，具体可采用离港旅客数据采集组件，基于输入的航班信息，来实时采集得到离港旅客信息。
步骤102、确定对当前航班进行旅客托运行李信息预测所需基于的目标规则数据。
本公开存储并维护有对应的估算规则数据集，估算规则数据集包括旅客托运行李预测规则及相关参数数据。
针对图2的示例，具体可基于估算规则数据维护组件存储与维护估算规则相关数据，形成估算规则数据集，以便从中查询所需的规则数据，用于离港旅客托运行李重量预测。
可选的，基于估算规则数据维护组件所存储与维护的估算规则相关数据，包括但不限于：
11)旅客托运行李预测规则(BaggagePredictRule)，包括航空公司、航班号、起飞站、到达站、预测规则。BaggagePredictRule相关数据由用户进行输入，其中预测规则有两个可选项，分别是历史数据预测(HistoryPredict)和时 间序列预测(ARIMAPredict)，分别代表直接使用BaggageHistory(离港实际旅客托运行李历史数据)的相关数据进行预测和采用ARIMA模型(AutoregressiveIntegratedMovingAveragemodel，差分整合移动平均自回归模型)进行时间序列预测，这是两套各自独立的预测规则，每套预测规则都有各自独立的预测参数。
比如上述xx1111航班，如果此时始发站在北京(PEK)，那么此时行李的到达站就包含上海(SHA)和广州(CAN)，举例如下：
表4
本公开中，上述时间序列预测规则包括：预先构建的ARIMA(AutoregressiveIntegratedMovingAverage，差分整合移动平均自回归)模型所表征的预测规则；ARIMA模型通过基于离港旅客托运行李重量的时间序列数据样本进行模型训练而得到。
12)旅客托运行李重量静态数据(StaticBaggageWeight)，包括航空公司、起飞站、到达站、舱位、人均行李重量，数据由用户进行输入。采用预测规则进行预测的前提是有足够的历史数据，如果没有足够的历史数据，需要直接采用StaticBaggageWeight(旅客托运行李重量静态数据)的数据进行预测。
比如上述xx1111航班，对应的航空公司是xx，如果此时始发站在北京(PEK)，那么此时行李的到达站就包含上海(SHA)和广州(CAN)，舱位包含J舱和Y舱，具体举例如下：
表5

13)旅客托运行李密度静态数据(StaticBaggageDensity)，包括航空公司、行李平均密度，数据由用户进行输入。
比如上述xx1111航班，对应的航空公司是xx，具体举例如下：
表6
14)集装器静态数据(ULDConfiguration)，包括航空公司、飞机注册号、是否为集装化飞机、默认集装器类型、默认集装器容积、默认集装器自重，数据由用户进行输入。
比如上述xx1111航班，对应的航空公司是xx，如果此时始发站在北京(PEK)，则飞机注册号是B1234，具体举例如下：
表7
15)法定节假日数据(Holiday)，包括日期、是否为法定节假日，数据由用户进行输入。每年年底都要输入下一年的节假日数据，用户只需输入法定节假日日期范围，***会自动生成下一年全年Holiday数据。
比如上述xx1111航班，航班日期是2022年3月1日，具体举例如下：
表8
如果BaggagePredictRule(旅客托运行李预测规则)中预测规则选择的是HistoryPredict，则应包括历史数据预测参数(HistoryPredictParameters)，包括 但不限于航空公司、航班号、起飞站、到达站、日期类型、三十天内历史数据占比、七天内历史数据占比、第七天前历史数据占比、历史数据有效天数最小值，数据由用户进行输入。所述日期类型包含workday(周一至周五，不包含法定节假日)、weekend(周六周日，不包含法定节假日)、holiday(法定节假日)。输入数据需满足以下条件：每个航班(相同的航空公司、航班号、起飞站、到达站)需要包含所有的日期类型；每一条记录中的三十天内历史数据占比、七天内历史数据占比、第七天历史数据占比之和必须为100％；历史数据有效天数最小值要大于0小于等于30。
比如上述xx1111航班，如果此时始发站在北京(PEK)，那么此时行李的到达站就包含上海(SHA)和广州(CAN)，具体举例如下：
表9
如果BaggagePredictRule中预测规则选择的是ARIMAPredict，则应包括时间序列预测参数(ARIMAPredictParameters)，包括航空公司、航班号、起飞站、 到达站、日期类型、按日预测占比、按周预测占比、历史数据有效天数最小值，数据由用户进行输入。所述日期类型包含workday(周一至周五，不包含法定节假日)、weekend(周六周日，不包含法定节假日)、holiday(法定节假日)。输入数据需满足以下条件：每个航班(相同的航空公司、航班号、起飞站、到达站)需要包含所有的日期类型；每一条记录中的按日预测占比、按周预测占比之和必须为100％；历史数据有效天数最小值要大于等于15小于等于30。
比如上述xx1111航班，如果此时始发站在北京(PEK)，那么此时行李的到达站就包含上海(SHA)和广州(CAN)，具体举例如下：
表10
基于存储与维护的估算规则数据集，本步骤102具体可使用离港旅客托运行李计算组件，从估算规则数据集中确定对当前航班进行旅客托运行李信息预测所需基于的目标规则数据，以便后续基于目标规则数据对当前航班的离港旅客托运行李重量进行预测。该过程可进一步实现为：
21)根据当前航班的航班信息，确定当前航班在预先配置的估算规则数据集中对应的目标预测规则。
可选的，使用离港旅客托运行李计算组件，根据当前航班信息(FlightInfo) 中的航空公司、航班号、起飞站、到达站查找旅客托运行李预测规则(BaggagePredictRule)，获取目标预测规则。
比如上述xx1111航班，如果此时始发站在北京(PEK)，那么此时行李的到达站就包含上海(SHA)和广州(CAN)，查询结果如下：
表11
22)若目标预测规则为旅客托运行李预测规则中的历史数据预测规则，确定是否满足历史数据预测条件，若是，确定目标规则数据包括历史数据预测规则及其在估算规则数据集中对应的规则参数，若否，所述目标规则数据包括所述估算规则数据集中的旅客托运行李重量静态数据。
示例性的，如果预测规则为历史数据预测规则HistoryPredict，可根据如下过程确定是否满足历史数据预测条件，并针对满足或未满足情况，确定对应的目标规则数据：
2.1获取有效天数(effectiveDays)：根据当前航班信息(FlightInfo)的航空公司、航班号、起飞站、到达站、[航班日期-30，航班日期-1]，从上述离港实际旅客托运行李历史数据提取组件提取有效天数；
2.2获取日期类型(dateType)：根据当前航班信息(FlightInfo)的航班日期查找Holiday，判断其是否为holiday，如果不是，再判断其是workday还是weekend；
2.3获取历史数据有效天数最小值(validDataMin)：根据当前航班信息(FlightInfo)的航空公司、航班号、起飞站、到达站、日期类型(dateType)查找HistoryPredictParameters，获取历史数据有效天数最小值；
2.4如果effectiveDays(有效天数)＞＝validDataMin(有效天数最小值)，则根据历史数据进行预测，也就是说，该情况下确定目标规则数据包括历史数据预测规则及其在估算规则数据集中对应的规则参数。否则，则根据旅客托运行李重量静态数据(StaticBaggageWeight)进行预测，即该情况下，确定目标 规则数据包括估算规则数据集中的旅客托运行李重量静态数据。
比如上述xx1111航班，如果此时始发站在北京(PEK)，那么行李到达站为SHA时符合预测规则是HistoryPredict，则进一步执行以下处理：
获取有效天数：根据xx、1111、PEK、SHA、2022/1/30-2022/2/28查询历史数据，符合条件的历史数据如下，共两条记录，如下表所示，所以effectiveDays＝2。
表12
获取日期类型：航班日期为2022年3月1日，查找Holiday，不是holiday，继续判断，2022年3月1日是星期二，所以最终得到dateType＝workday。
获取历史数据有效天数最小值：根据xx、1111、PEK、SHA、workday查找HistoryPredictParameters，结果如下所示，validDataMin＝1。
表13

effectiveDays＝2，validDataMin＝1，effectiveDays＞validDataMin，因此根据历史数据进行预测，相应确定出目标规则数据包括历史数据预测规则及其在估算规则数据集中对应的规则参数。
23)若目标预测规则为旅客托运行李预测规则中的时间序列预测规则，确定是否满足时间序列预测条件，若是，确定目标规则参数包括时间序列预测规则及其在估算规则数据集中对应的规则参数；若否，所述目标规则数据包括所述估算规则数据集中的旅客托运行李重量静态数据。
示例性的，如果预测规则为时间序列预测规则ARIMAPredict，可根据如下过程确定是否满足时间序列预测条件，并针对满足或未满足情况，确定对应的目标规则数据：
3.1获取按日预测有效天数(effectiveByDay)：根据当前航班信息(FlightInfo)的航空公司、航班号、起飞站、到达站、[航班日期-30，航班日期-1]，从上述离港实际旅客托运行李历史数据提取组件提取有效天数；
3.2获取按周预测有效天数(effectiveByWeek)：根据当前航班信息(FlightInfo)的航空公司、航班号、起飞站、到达站、航班日期之前的三十个相同周目对应日期，从上述离港实际旅客托运行李历史数据提取组件提取有效天数；
3.3获取日期类型(dateType)：根据当前航班信息(FlightInfo)的航班日期查找Holiday，判断其是否为holiday，如果不是，再判断其是workday还是weekend；
3.4获取历史数据有效天数最小值(validDataMin)：根据当前航班信息(FlightInfo)的航空公司、航班号、起飞站、到达站、日期类型(dateType)查找ARIMAPredictParameters，获取历史数据有效天数最小值；
3.5如果effectiveByDay＞＝validData并且effectiveByWeek＞＝validData，则根据时间序列进行预测，也就是说，该情况下确定目标规则参数包括时间序列预测规则及其在估算规则数据集中对应的规则参数。否则根据旅客托运行李重量静态数据(StaticBaggageWeight)进行预测，即该情况下，确定目标规则数据包括估算规则数据集中的旅客托运行李重量静态数据。
比如上述xx1111航班，如果此时始发站在北京(PEK)，那么行李到达站为CAN时符合预测规则是ARIMAPredict，则进一步执行以下处理：
获取按日预测有效天数：根据xx、1111、PEK、CAN、2022/1/30到2022/2/28查询历史数据，符合条件的历史数据如下，共两条记录，如下表所示，所以effectiveByDay＝2。
表14
获取按周预测有效天数：根据xx、1111、PEK、CAN、{2021/8/3，2021/8/10，...2022/2/15，2022/2/22}查询历史数据，没有符合条件的历史数据，所以effectiveByWeek＝0。
获取日期类型：航班日期为2022年3月1日，查找Holiday，不是holiday，继续判断，2022年3月1日是星期二，所以最终得到dateType＝workday。
获取历史数据有效天数最小值：根据xx、1111、PEK、CAN、workday查找ARIMAPredictParameters，结果如下，validDataMin＝20。
表15

effectiveByDay＝2，effectiveByWeek＝0，validDataMin＝20，effectiveByDay＜validDataMin，effectiveByWeek＜validDataMin，因此根据旅客托运行李重量静态数据(StaticBaggageWeight)进行预测，相应确定目标规则数据包括估算规则数据集中的旅客托运行李重量静态数据。
步骤103、根据所述离港旅客数据、所述离港实际旅客托运行李历史数据和所述目标规则数据中的至少部分，预测当前航班的离港旅客托运行李重量，以基于预测的离港旅客托运行李重量对当前航班进行配载平衡处理。
之后，进一步根据当前航班对应的离港旅客数据，离港实际旅客托运行李历史数据和目标规则数据中的至少部分，对当前航班的离港旅客托运行李重量进行预测，预测值具体为预测得到的当前航班的离港旅客托运行李总重。
该过程可进一步实现为：
31)若所述目标规则数据包括所述历史数据预测规则及其相应规则参数，从所述实际旅客托运行李历史数据中提取所述历史数据预测规则所需的历史参数数据，根据所述离港旅客数据、所述历史参数数据，和所述历史数据预测规则及其规则参数，预测当前航班的离港旅客托运行李重量。
其中，针对图2的示例，具体可采用离港实际旅客托运行李历史数据提取组件，从当前航班对应的实际旅客托运行李历史数据中，提取历史数据预测规则所需的历史参数数据，以实现所需历史参数数据的获取，包括但不限于：
30天内历史数据人均行李重量均值(avgWeight30)：根据当前航班信息(FlightInfo)的航空公司、航班号、起飞站、到达站、[航班日期-30，航班日期-1]，使用离港实际旅客托运行李历史数据提取组件提取人均行李重量均值；
7天内历史数据有效天数(effective7)：根据当前航班信息(FlightInfo)的航空公司、航班号、起飞站、到达站、[航班日期-7，航班日期-1]，使用离港实际旅客托运行李历史数据提取组件提取有效天数；
7天内历史数据人均行李重量均值(avgWeight7)：根据当前航班信息(FlightInfo)的航空公司、航班号、起飞站、到达站、[航班日期-7，航班日期-1]，使用离港实际旅客托运行李历史数据提取组件提取人均行李重量均值；
第七天前历史数据有效天数(effectiveAt7)：根据当前航班信息(FlightInfo) 的航空公司、航班号、起飞站、到达站、航班日期-7，使用离港实际旅客托运行李历史数据提取组件提取有效天数；
第七天前历史数据人均行李重量均值(avgWeightAt7)：根据当前航班信息(FlightInfo)的航空公司、航班号、起飞站、到达站、航班日期-7，使用离港实际旅客托运行李历史数据提取组件提取人均行李重量均值。
除此之外，对于该方式下当前航班的离港旅客托运行李重量的预测，还需从估算规则数据集中获取历史数据预测规则的相关规则参数，包括但不限于：
历史数据预测参数：根据当前航班信息(FlightInfo)的航空公司、航班号、起飞站、到达站、日期类型(dateType)查找HistoryPredictParameters，获取三十天内历史数据占比(percentage30)、七天内历史数据占比(percentage7)、第七天前历史数据占比(percentageAt7)。
以及，还需获取当前航班离港旅客信息中的订座人数(bookNum)：根据当前航班信息(FlightInfo)的航空公司、航班号、航班日期、起飞站，以及旅客到达站，在离港旅客信息(PassengerInfo)中查找符合条件的记录，累加所有记录的订座旅客人数。
在此基础上，使用离港旅客托运行李计算组件，基于历史数据预测规则以获取的各种数据为依据，计算当前航班的离港旅客托运行李重量bagEstWeight，实现离港旅客托运行李重量预测，一示例性预测过程如下：
如果effective7＝0，bagEstWeight＝avgWeight30*bookNum；
如果effective7＞0，effectiveAt7＝0，bagEstWeight＝(avgWeight30*percentage30+avgWeight7*percentage7)/(percentage30+percentage7)*bookNum；
如果effective7＞0，effectiveAt7＞0，bagEstWeight＝(avgWeight30*percentage30+avgWeight7*percentage7+avgWeightAt7*percentageAt7)*bookNum。
比如上述xx1111航班，如果此时始发站在北京(PEK)，那么行李到达站为SHA时需要根据历史数据计算旅客托运行李重量(bagEstWeight)，即基于历史数据预测规则对当前航班的离港旅客托运行李重量进行预测，过程如下：
获取30天内历史数据人均行李重量均值(avgWeight30)：根据xx、1111、PEK、SHA、2022/1/30-2022/2/28查询历史数据，符合条件的历史数据如下， avgWeight30＝(10.77+10.71)/2＝10.74。
表16
获取7天内历史数据有效天数(effective7)：根据xx、1111、PEK、SHA、2022/2/22-2022/2/28查询历史数据，符合条件的历史数据和1)中一样，effective7＝2。
获取7天内历史数据人均行李重量均值(avgWeight7)：根据xx、1111、PEK、SHA、2022/2/22-2022/2/28查询历史数据，符合条件的历史数据和1)中一样，avgWeight7＝(10.77+10.71)/2＝10.74。
获取第七天前历史数据有效天数(effectiveAt7)：根据xx、1111、PEK、SHA、2022/2/22查询历史数据，没有符合条件的历史数据，effectiveAt7＝0。
获取第七天前历史数据人均行李重量均值(avgWeightAt7)：根据xx、1111、PEK、SHA、2022/2/22查询历史数据，没有符合条件的历史数据，avgWeightAt7没有值。
获取历史数据预测参数：根据xx、1111、PEK、SHA、workday查找HistoryPredictParameters，得到如下记录，则percentage30＝40％、percentage7＝40％、percentageAt7＝20％。
表17

获取订座人数(bookNum)：根据xx、1111、2022/3/1、PEK、SHA查找PassengerInfo，得到如下记录，则bookNum＝3+88＝91。
表18
计算bagEstWeight：effective7＝2，则effective7＞0，effectiveAt7＝0，因此bagEstWeight＝(avgWeight30*percentage30+avgWeight7*percentage7)/(percentage30+percentage7)*bookNum＝(10.74*40％+10.74*40％)/(40％+40％)*91＝977.34。
32)若所述目标规则数据包括所述时间序列预测规则及其规则参数，从所述实际旅客托运行李历史数据中提取所述时间序列预测规则所需的历史时间序列参数数据，根据所述离港旅客数据、历史时间序列参数数据，和所述时间序列预测规则及其规则参数，预测当前航班的离港旅客托运行李重量。
其中，针对图2的示例，该方式32)具体可采用离港实际旅客托运行李历史数据提取组件，从当前航班对应的实际旅客托运行李历史数据中，提取时间序列预测规则所需的历史时间序列参数数据，并在历史时间序列参数数据有缺失情况下，采用线性差值法补全时间序列，包括但不限于：
获取按日预测的人均行李重量时间序列(timeSeriesByDay)：根据当前航班信息(FlightInfo)的航空公司、航班号、起飞站、到达站、[航班日期-30， 航班日期-1]，从上述离港实际旅客托运行李历史数据提取组件提取人均行李重量时间序列，并采用线性差值法补全时间序列，如果是时间序列头尾缺少数据，则直接使用最近日期且有效的数据。
比如提取的时间序列如下，需要获取到2020/1/1-2020/1/6的数据，但是数据库中只有两天有数据，分别是2020/1/2和2020/1/5，那么就需要补全其他数据。
表19
首先把缺少的日期补上，得到如下数据：
表20
然后把头部和尾部缺少的数据补上，Value1是头部数据，距离他最近且有效的数据是2020/1/2的数据，所以Value1＝15，同理Value4＝13，头尾数据补充完的数据如下：
表21
最后把中间缺少的数据使用线性插值法补上，Value2的计算如下：(2020/1/5-2020/1/3)/(2020/1/3-2020/1/2)＝(13-Value2)/(Value2-15)，即2/1＝(13-Value2)/(Value2-15)，计算得到Value2＝43/3＝14.33，同理可算出Value3＝41/3＝13.67，补充完数据后最终的时间序列如下：
表22
获取按周预测的人均行李重量时间序列(timeSeriesByWeek)：根据当前 航班信息(FlightInfo)的航空公司、航班号、起飞站、到达站、航班日期之前的三十个相同周目对应日期，从上述离港实际旅客托运行李历史数据提取组件提取人均行李重量时间序列，并采用线性差值法补全时间序列，如果是时间序列头尾缺少数据，则直接使用最近日期且有效的数据。
比如提取的时间序列如下，需要获取到{2022/3/1，2022/3/8，2022/3/15，2022/3/22，2022/3/29}的数据，但是数据库中只有两天有数据，分别是2022/3/15和2022/3/29，那么就需要补全其他数据。
表23
参考上述按日预测的例子，补全后得到如下数据：
表24
除此之外，对于该方式下当前航班的离港旅客托运行李重量的预测，还需从估算规则数据集中获取时间序列预测规则的相关规则参数，包括但不限于：
获取时间序列预测参数：根据当前航班信息(FlightInfo)的航空公司、航班号、起飞站、到达站、日期类型(dateType)查找ARIMAPredictParameters，获取按日预测占比(percentageByDay)、按周预测占比(percentageByWeek)。
以及，还需获取当前航班离港旅客信息中的订座人数：根据当前航班信息(FlightInfo)的航空公司、航班号、航班日期、起飞站，以及旅客到达站，在离港旅客信息(PassengerInfo)中查找符合条件的记录，累加所有记录的订座旅客人数。
在此基础上，使用离港旅客托运行李计算组件，基于时间序列预测规则以获取的各种数据为依据，计算当前航班的离港旅客托运行李重量bagEstWeight，实现离港旅客托运行李重量预测，一示例性预测过程如下：
计算按日预测的ARIMA模型参数P1、D1、Q1：D1取值范围设为1、2，P1取值范围设为1-5，Q1取值范围设为1-5，遍历P1、D1、Q1的所有取值，根据 timeSeriesByDay的前25天数据和对应的模型参数P1、D1、Q1，带入ARIMA模型预测后5天的旅客托运行李重量，并和实际值进行比对，寻找误差平方和最小的一组P1、D1、Q1。
举一个简单的例子，有如下的时间序列：
表25
假设D1＝1，P1的取值范围是2和3，Q1的取值范围是1和2，遍历P1、D1、Q1的所有取值，根据案例中时间序列的前三个值和对应的模型参数P1、D1、Q1，带入ARIMA模型预测后2天的数据，得到结果如下：
表26
计算每组P1、D1、Q1对应的预测误差平方和，计算结果如下，可以看出当D1＝1，P1＝3，Q1＝1时，预测误差平方和最小，因此最终确定D1＝1，P1＝3，Q1＝1。
表27

计算按周预测的ARIMA模型参数P2、D2、Q2：D2取值范围设为1、2，P2取值范围设为1-5，Q2取值范围设为1-5，遍历P2、D2、Q2的所有取值，根据timeSeriesByWeek的前25天数据和对应的模型参数P2、D2、Q2，带入ARIMA模型预测后5天的旅客托运行李重量，并和实际值进行比对，寻找误差平方和最小的一组P2、D2、Q2。
计算按日预测的人均行李重量(avgWeightByDay)：根据timeSeriesByDay的数据和对应的模型参数P1、D1、Q1，带入ARIMA模型预测当前航班的人均行李重量。
计算按周预测的人均行李重量(avgWeightByWeek)：根据timeSeriesByWeek的数据和对应的模型参数P2、D2、Q2，带入ARIMA模型预测当前航班的人均行李重量。
基于时间序列预测规则，计算bagEstWeight：
bagEstWeight＝(avgWeightByDay*percentageByDay+avgWeightByWeek*percentageByWeek)*bookNum。
33)若所述目标规则数据包括所述旅客托运行李重量静态数据，根据所述离港旅客数据和所述旅客托运行李重量静态数据，预测当前航班的离港旅客托运行李重量。
该方式下，可选的，具体可获取所需的旅客托运行李重量静态数据及离港旅客数据，包括但不限于：
获取不同舱位的人均行李重量(avgWeight)：根据当前航班信息(FlightInfo)的航空公司、起飞站、到达站查找旅客托运行李重量静态数据(StaticBaggageWeight)，获取对应的舱位、人均行李重量；
获取不同舱位的订座人数(bookNum)：根据当前航班信息(FlightInfo)的航空公司、航班号、航班日期、起飞站，以及旅客到达站，查找离港旅客信息(PassengerInfo)中对应的舱位、订座旅客人数。
在此基础上，使用离港旅客托运行李计算组件，根据获取的离港旅客数据和旅客托运行李重量静态数据，计算当前航班的离港旅客托运行李重量bagEstWeight，实现离港旅客托运行李重量预测，具体如下：
计算bagEstWeight：所有相同舱位的avgWeight*bookNum之和。
比如上述xx1111航班，如果此时始发站在北京(PEK)，那么行李到达站为CAN时，需要根据旅客托运行李重量静态数据计算旅客托运行李重量(bagEstWeight)，过程如下：
获取不同舱位的人均行李重量(avgWeight)：根据xx、PEK、CAN查找StaticBaggageWeight，得到如下记录：
表28
获取不同舱位的订座人数(bookNum)：根据xx、1111、2022/3/1、PEK、CAN查找，得到如下记录：
表29
计算bagEstWeight：bagEstWeight＝20*4+16*66＝1136。
根据以上方案可知，本公开提供的旅客托运行李信息预测方法，获取当前航班对应的离港旅客数据和离港实际旅客托运行李历史数据，确定对当前航班进行旅客托运行李信息预测所需基于的目标规则数据，并根据当前航班的离港旅客数据、离港实际旅客托运行李历史数据和目标规则数据中的至少部分，预测当前航班的离港旅客托运行李重量，以基于预测的离港旅客托运行李重量对当前航班进行配载平衡处理。本公开通过根据相关数据及规则自动预测当前航班的离港旅客托运行李重量，可有效改善现有人工估算方式存在的各种弊端，预测所基于的参考数据较为全面，提升了离港旅客托运行李重量预测值的参考价值，使预测值与实际值更为接近，且不会因个体差异带来预测结果的偏差， 避免了为生产安全带来风险和隐患。
在一实施例中，可选的，参见图3所示的旅客托运行李信息预测方法流程图，本公开提供的旅客托运行李信息预测方法，还可以包括：
步骤104、根据预测的离港旅客托运行李重量，确定当前航班的离港旅客托运行李和集装器总重。
其中，如果当前航班为集装化飞机，则离港旅客托运行李和集装器总重为离港旅客托运行李重量与集装器重量之和，如果当前航班为非集装化飞机，则离港旅客托运行李和集装器总重为离港旅客托运行李重量。
可选的，具体可根据当前航班所对应航班信息(FlightInfo)中的航空公司、飞机注册号，查询集装器静态数据(ULDConfiguration)，以确定当前航班是否为集装化飞机，如果当前航班是集装化飞机，还需额外计算集装器重量(uldEstWeight)，并根据预测的离港旅客托运行李重量和集装器重量，计算离港旅客托运行李和集装器总重，即totalEstWeight＝bagEstWeight+uldEstWeight；如果当前航班不是集装化飞机，则离港旅客托运行李和集装器总重为离港旅客托运行李重量，即totalEstWeight＝bagEstWeight。
比如上述xx1111航班，对应的航空公司是xx，如果此时始发站在北京(PEK)，则飞机注册号是B1234，根据xx、B1234查找ULDConfiguration，得到如下记录，该航班飞机是集装化飞机，因此需要额外计算集装器重量。
表30
其中，集装器重量(uldEstWeight)的一示例性计算流程如下：
获取旅客托运行李密度(baggageDensity)，根据当前航班信息(FlightInfo)的航空公司查找旅客托运行李密度静态数据(StaticBaggageDensity)，获取对应的行李平均密度；
获取集装器容积(uldPerVolumn)和自重(uldPerWeight)：根据当前航班信息(FlightInfo)的航空公司、飞机注册号查找集装器静态数据(ULDConfiguration)，获取默认集装器容积、默认集装器自重；
计算所需集装器个数(uldNum)：uldNum＝(bagEstWeight/baggageDensity)/uldPerVolumn，计算结果向上取整；
计算集装器重量(uldEstWeight)：uldEstWeight＝uldNum*uldPerWeight。
本实施例通过确定当前航班的离港旅客托运行李和集装器总重，可实现为当前航班的配载平衡提供数据依据，且在确定离港旅客托运行李和集装器总重时，通过根据相关数据及规则自动预测当前航班的离港旅客托运行李重量，提升了离港旅客托运行李重量预测值的参考价值，相应提升了离港旅客托运行李和集装器总重的参考价值，使预测值与实际值更为接近，避免了为生产安全带来风险和隐患。
在一实施例中，可选的，参见图4所示的旅客托运行李信息预测方法流程图，本公开提供的旅客托运行李信息预测方法，还可以包括：
步骤105、获取当前航班的实际离港旅客托运行李数据，在当前航班关闭后将获取的数据更新存储至当前航班的离港实际旅客托运行李历史数据中。
针对图2的示例，可在基于本公开方法实现的配载航班旅客托运行李估算装置中，进一步增设离港实际旅客托运行李数据处理组件，具体如图2所示，相应可使用该组件，实时采集旅客托运行李数据，并在航班关闭后存储相关数据到离港实际旅客托运行李历史数据。
比如假设当前航班xx1111，日期2022/3/1，则在该航班关闭后，将当前航班在2022/3/1日期下的实际离港旅客托运行李数据更新至BaggageHistory，更新后BaggageHistory数据如下：
表31

本实施例通过获取当前航班的实际离港旅客托运行李数据，在当前航班关闭后将获取的数据更新存储至离港实际旅客托运行李历史数据，实现了随时间推进不断对当前航班对应的离港实际旅客托运行李历史数据进行更新，从而便于为该航班未来时间的旅客托运行李信息提供历史数据支持。
在一实施例中，可选的，参见图5所示的旅客托运行李信息预测方法流程图，本公开提供的旅客托运行李信息预测方法，还可以包括：
步骤106、确定当前航班的离港旅客托运行李重量预测值与实际值之间的偏差；
步骤107、在所述偏差达到预设阈值情况下，进行预警处理，以基于预警处理调整所述估算规则数据集中的相应规则参数，并在后续时间基于调整后的规则参数进行当前航班的离港旅客托运行李重量预测。
针对图2的示例，可在基于本公开方法实现的配载航班旅客托运行李估算装置中，进一步增设实时预警组件，具体如图2所示，相应可使用该组件，执行本实施例的处理，实现在离港旅客托运行李重量预测值与实际值之间的偏差达到预设阈值时，实时进行预警处理。
参见图6，进一步提供了图2中配载航班旅客托运行李估算装置各组件的组件关系示意图。
其中，实时预警组件设置有预警参数(PrewarningParameters)，包括但不限于航空公司、航班号、起飞站、到达站、天数、平均误差最大值、上次预警日期，其中，上次预警日期由***自动生成(每生成一条记录自动把当前日期写入上次预警日期，后续只要对应航班产生预警，则写入对应的预警日期)，其余数据由用户输入。
比如上述xx1111航班，如果此时始发站在北京(PEK)，则行李到达站包括SHA和CAN，具体举例如下：
表32
基于实时预警组件的一示例性预警流程如下(每次航班关闭，离港实际旅客托运行李数据处理组件更新完BaggageHistory后，会自动执行该流程)：
41)判断是否到达预警时间：根据当前航班信息(FlightInfo)的航空公司、飞机注册号、航班号、起飞站、到达站查找预警参数(PrewarningParameters)，获取对应的天数(days)、平均误差最大值(avgDiffMaxValue)、上次预警日期，如果当前航班日期＞＝(上次预警日期+days)，表示已到达预警时间，继续进行后续流程，否则表示未到达预警时间，结束后续流程；
42)获取预测误差均值(avgDiff)：根据当前航班信息(FlightInfo)的航空公司、航班号、起飞站、到达站、[航班日期-days+1，航班日期]，从上述离港实际旅客托运行李历史数据提取组件提取预测误差均值；
43)判断是否需要预警：如果avgDiff＞avgDiffMaxValue，则需要预警，给用户发送预警信息，同时在对应的PrewarningParameters记录中把上次预警日期改为当前航班日期，如果avgDiff＜＝avgDiffMaxValue，则不需要做任何处理。
比如上述xx1111航班，如果此时始发站在北京(PEK)，则行李到达站包括SHA和CAN，需要分别进行判断：
判断是否到达预警时间：行李到达站是SHA时，根据xx、1111、PEK、SHA查找PrewarningParameters，得到下表第一条记录，则days＝5，avgDiffMaxValue＝5％，上次预警日期＝2022/1/1，当前航班日期为2022/3/1，(上次预警日期+days)为2022/1/6，因此已达到预警时间，继续进行后续判断；行李到达站是CAN时，根据xx、1111、PEK、CAN查找PrewarningParameters，得到下表第二条记录，则days＝10，avgDiffMaxValue＝5％，上次预警日期＝2022/2/25，当前航班日期为2022/3/1，(上次预警日期+days)为2022/3/7，因此未达到预警时间，结束后续流程。
表33
获取预测误差均值(avgDiff)：只有到达站为SHA时需要后续判断，因此根据xx、1111、PEK、SHA、2022/2/25-2022/3/1查找BaggageHistory，得到如下记录，avgDiff＝(7.14％+6.67％+8.56％)/3＝7.46％。
表34

判断是否需要预警：avgDiff＝7.46％，avgDiffMaxValue＝5％，avgDiff＞avgDiffMaxValue，因此需要给用户发送预警信息，同时在对应的PrewarningParameters记录中把上次预警日期改为当前航班日期，修改后的记录如下(″2022/3/1″为修改部分)：
表35
当估算规则的相关参数设置不合适时，会导致预测结果和实际值存在较大偏差，基于此，本实施例中，基于预警处理及时告知用户修改估算规则参数，在用户基于预警信息修改估算规则数据集中的相关规则参数后，后续会采用更新后的规则参数进行离港旅客托运行李信息预测，并可达到有效降低预测值和实际值之间的偏差的效果。
对应于上述的旅客托运行李信息预测方法，本公开还提供一种旅客托运行李信息预测装置，参见图7，该装置包括：
数据获取模块10，用于获取当前航班对应的离港旅客数据和离港实际旅客托运行李历史数据；
规则确定模块20，用于确定对当前航班进行旅客托运行李信息预测所需基于的目标规则数据；
信息预测模块30，用于根据所述离港旅客数据、所述离港实际旅客托运行李历史数据和所述目标规则数据中的至少部分，预测当前航班的离港旅客托运行李重量，以基于所述离港旅客托运行李重量对当前航班进行配载平衡处理。
在一实施方式中，规则确定模块20，具体用于：
根据当前航班的航班信息，确定当前航班在预先配置的估算规则数据集中对应的目标预测规则；其中，所述估算规则数据集包括旅客托运行李预测规则及相关参数数据；
若所述目标预测规则为所述旅客托运行李预测规则中的历史数据预测规则，确定是否满足历史数据预测条件，若是，确定所述目标规则数据包括所述历史数据预测规则及其在所述估算规则数据集中对应的规则参数；若否，所述目标规则数据包括所述估算规则数据集中的旅客托运行李重量静态数据；
若所述目标预测规则为所述旅客托运行李预测规则中的时间序列预测规则，确定是否满足时间序列预测条件，若是，确定所述目标规则参数包括所述时间序列预测规则及其在所述估算规则数据集中对应的规则参数；若否，所述目标规则数据包括所述估算规则数据集中的旅客托运行李重量静态数据。
在一实施方式中，信息预测模块30，具体用于：
若所述目标规则数据包括所述历史数据预测规则及其规则参数，从所述实际旅客托运行李历史数据中提取所述历史数据预测规则所需的历史参数数据，根据所述离港旅客数据、所述历史参数数据，和所述历史数据预测规则及其规则参数，预测当前航班的离港旅客托运行李重量；
若所述目标规则数据包括所述时间序列预测规则及其规则参数，从所述实际旅客托运行李历史数据中提取所述时间序列预测规则所需的历史时间序列参数数据，根据所述离港旅客数据、历史时间序列参数数据，和所述时间序列预测规则及其规则参数，预测当前航班的离港旅客托运行李重量；
若所述目标规则数据包括所述旅客托运行李重量静态数据，根据所述离港旅客数据和所述旅客托运行李重量静态数据，预测当前航班的离港旅客托运行李重量。
在一实施方式中，所述时间序列预测规则包括：预先构建的ARIMA模型所表征的预测规则；
其中，所述ARIMA模型通过基于离港旅客托运行李重量的时间序列数据样本进行模型训练而得到。
在一实施方式中，信息预测模块30，还用于：根据预测的离港旅客托运行李重量，确定当前航班的离港旅客托运行李和集装器总重；
其中，如果当前航班为集装化飞机，则离港旅客托运行李和集装器总重为 离港旅客托运行李重量与集装器重量之和，如果当前航班为非集装化飞机，则离港旅客托运行李和集装器总重为离港旅客托运行李重量。
在一实施方式中，上述装置还包括：
历史数据更新模块，用于：获取当前航班的实际离港旅客托运行李数据，在当前航班关闭后将获取的数据更新存储至当前航班的离港实际旅客托运行李历史数据中。
在一实施方式中，上述装置还包括：
预警模块，用于：确定当前航班的离港旅客托运行李重量预测值与实际值之间的偏差；在所述偏差达到预设阈值情况下，进行预警处理，以基于预警处理调整所述估算规则数据集中的相应规则参数，并在后续时间基于调整后的规则参数进行当前航班的离港旅客托运行李重量预测。
描述于本公开实施例中所涉及到的单元/模块可以通过软件的方式实现，也可以通过硬件的方式来实现。其中，单元/模块的名称在某种情况下并不构成对该单元本身的限定，例如，第一获取单元还可以被描述为″获取至少两个网际协议地址的单元″。
本文中以上描述的功能可以至少部分地由一个或多个硬件逻辑部件来执行。例如，非限制性地，可以使用的示范类型的硬件逻辑部件包括：现场可编程门阵列(FPGA)、专用集成电路(ASIC)、专用标准产品(ASSP)、片上***(SOC)、复杂可编程逻辑设备(CPLD)等等。
本公开还提供一种计算机可读介质，其上存储有计算机程序，所述计算机程序包含用于执行如上文方法实施例公开的旅客托运行李信息预测方法的程序代码。
在本公开的上下文中，计算机可读介质(机器可读介质)可以是有形的介质，其可以包含或存储以供指令执行***、装置或设备使用或与指令执行***、装置或设备结合地使用的程序。机器可读介质可以是机器可读信号介质或机器可读储存介质。机器可读介质可以包括但不限于电子的、磁性的、光学的、电磁的、红外的、或半导体***、装置或设备，或者上述内容的任何合适组合。机器可读存储介质的更具体示例会包括基于一个或多个线的电气连接、便携式计算机盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦除可 编程只读存储器(EPROM或快闪存储器)、光纤、便捷式紧凑盘只读存储器(CD-ROM)、光学储存设备、磁储存设备、或上述内容的任何合适组合。
需要说明的是，本公开上述的计算机可读介质可以是计算机可读信号介质或者计算机可读存储介质或者是上述两者的任意组合。计算机可读存储介质例如可以是——但不限于——电、磁、光、电磁、红外线、或半导体的***、装置或器件，或者任意以上的组合。计算机可读存储介质的更具体的例子可以包括但不限于：具有一个或多个导线的电连接、便携式计算机磁盘、硬盘、随机访问存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、光纤、便携式紧凑磁盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本公开中，计算机可读存储介质可以是任何包含或存储程序的有形介质，该程序可以被指令执行***、装置或者器件使用或者与其结合使用。而在本公开中，计算机可读信号介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了计算机可读的程序代码。这种传播的数据信号可以采用多种形式，包括但不限于电磁信号、光信号或上述的任意合适的组合。计算机可读信号介质还可以是计算机可读存储介质以外的任何计算机可读介质，该计算机可读信号介质可以发送、传播或者传输用于由指令执行***、装置或者器件使用或者与其结合使用的程序。计算机可读介质上包含的程序代码可以用任何适当的介质传输，包括但不限于：电线、光缆、RF(射频)等等，或者上述的任意合适的组合。
上述计算机可读介质可以是电子设备中所包含的；也可以是单独存在，而未装配入电子设备中。
本公开还提供一种计算机程序产品，其包括承载在非暂态计算机可读介质上的计算机程序，所述计算机程序包含用于执行如上文方法实施例公开的旅客托运行李信息预测方法的程序代码。
特别地，根据本公开的实施例，上文各参考流程图描述的过程可以被实现为计算机软件程序。在这样的实施例中，该计算机程序可以通过通信装置从网络上被下载和安装，或者从存储装置被安装，或者从ROM被安装。在该计算机程序被处理装置执行时，执行本公开实施例的方法中限定的上述功能。
综上所述，根据本公开的一个或多个实施例，本公开提供了一种旅客托运行李信息预测方法，所述方法包括：
获取当前航班对应的离港旅客数据和离港实际旅客托运行李历史数据；
确定对当前航班进行旅客托运行李信息预测所需基于的目标规则数据；
根据所述离港旅客数据、所述离港实际旅客托运行李历史数据和所述目标规则数据中的至少部分，预测当前航班的离港旅客托运行李重量，以基于预测的离港旅客托运行李重量对当前航班进行配载平衡处理。
根据本公开的一个或多个实施例，上述方法中，确定对当前航班进行旅客托运行李信息预测所需基于的目标规则数据，包括：
根据当前航班的航班信息，确定当前航班在预先配置的估算规则数据集中对应的目标预测规则；其中，所述估算规则数据集包括旅客托运行李预测规则及相关参数数据；
若所述目标预测规则为所述旅客托运行李预测规则中的历史数据预测规则，确定是否满足历史数据预测条件，若是，确定所述目标规则数据包括所述历史数据预测规则及其在所述估算规则数据集中对应的规则参数；若否，所述目标规则数据包括所述估算规则数据集中的旅客托运行李重量静态数据；
若所述目标预测规则为所述旅客托运行李预测规则中的时间序列预测规则，确定是否满足时间序列预测条件，若是，确定所述目标规则参数包括所述时间序列预测规则及其在所述估算规则数据集中对应的规则参数；若否，所述目标规则数据包括所述估算规则数据集中的旅客托运行李重量静态数据。
根据本公开的一个或多个实施例，上述方法中，根据所述离港旅客数据、所述离港实际旅客托运行李历史数据和所述目标规则数据中的至少部分，预测当前航班的离港旅客托运行李重量，包括：
若所述目标规则数据包括所述历史数据预测规则及其规则参数，从所述实际旅客托运行李历史数据中提取所述历史数据预测规则所需的历史参数数据，根据所述离港旅客数据、所述历史参数数据，和所述历史数据预测规则及其规则参数，预测当前航班的离港旅客托运行李重量；
若所述目标规则数据包括所述时间序列预测规则及其规则参数，从所述实际旅客托运行李历史数据中提取所述时间序列预测规则所需的历史时间序列参数数据，根据所述离港旅客数据、历史时间序列参数数据，和所述时间序列 预测规则及其规则参数，预测当前航班的离港旅客托运行李重量；
若所述目标规则数据包括所述旅客托运行李重量静态数据，根据所述离港旅客数据和所述旅客托运行李重量静态数据，预测当前航班的离港旅客托运行李重量。
根据本公开的一个或多个实施例，上述方法中，所述时间序列预测规则包括：预先构建的ARIMA模型所表征的预测规则；
其中，所述ARIMA模型通过基于离港旅客托运行李重量的时间序列数据样本进行模型训练而得到。
根据本公开的一个或多个实施例，上述方法，在预测当前航班的离港旅客托运行李重量之后，还包括：
根据预测的离港旅客托运行李重量，确定当前航班的离港旅客托运行李和集装器总重；
其中，如果当前航班为集装化飞机，则离港旅客托运行李和集装器总重为离港旅客托运行李重量与集装器重量之和，如果当前航班为非集装化飞机，则离港旅客托运行李和集装器总重为离港旅客托运行李重量。
根据本公开的一个或多个实施例，上述方法，还包括：
获取当前航班的实际离港旅客托运行李数据，在当前航班关闭后将获取的数据更新存储至当前航班的离港实际旅客托运行李历史数据中。
根据本公开的一个或多个实施例，上述方法，还包括：
确定当前航班的离港旅客托运行李重量预测值与实际值之间的偏差；
在所述偏差达到预设阈值情况下，进行预警处理，以基于预警处理调整所述估算规则数据集中的相应规则参数，并在后续时间基于调整后的规则参数进行当前航班的离港旅客托运行李重量预测。
根据本公开的一个或多个实施例，本公开还提供一种旅客托运行李信息预测装置，包括：
数据获取模块，用于获取当前航班对应的离港旅客数据和离港实际旅客托运行李历史数据；
规则确定模块，用于确定对当前航班进行旅客托运行李信息预测所需基于的目标规则数据；
信息预测模块，用于根据所述离港旅客数据、所述离港实际旅客托运行李 历史数据和所述目标规则数据中的至少部分，预测当前航班的离港旅客托运行李重量，以基于所述离港旅客托运行李重量对当前航班进行配载平衡处理。
根据本公开的一个或多个实施例，本公开还提供一种计算机可读介质，其上存储有计算机程序，所述计算机程序包含用于执行如上文所述的旅客托运行李信息预测方法的程序代码。
根据本公开的一个或多个实施例，本公开还提供一种计算机程序产品，其包括承载在非暂态计算机可读介质上的计算机程序，所述计算机程序包含用于执行如上文所述的旅客托运行李信息预测方法的程序代码。
本公开提供的旅客托运行李信息预测方法、装置、计算机可读介质及计算机程序产品，至少具备以下技术优势：
a)本公开需要根据航班实际旅客托运行李历史数据和当前航班离港旅客数据进行计算，既参考了行李历史数据，同时也参考了现有旅客数据，数据参考全面，得到的计算值更具有说服力，参考价值更高，使预测值与实际值更为接近，避免了为生产安全带来风险和隐患；
b)本公开以航班实际旅客托运行李历史数据和当前航班离港旅客数据作为输入，带入相应的估算规则进行计算，得到的行李重量计算值是固定的，不存在不同的操作人员使用本***，得到不同的计算值这种情况，消除了个体差异带来的隐患；
c)本公开设有预警功能，当估算规则的相关参数设置不合适导致计算结果和实际值存在较大偏差时，会告知用户修改估算规则参数，用户修改相关参数后，后续会采用新的规则参数进行计算，有效降低了计算值和实际值之间的偏差。
需要说明，尽管已经采用特定于结构特征和/或方法逻辑动作的语言描述了本主题，但是应当理解所限定的主题未必局限于上面描述的特定特征或动作。相反，上面所描述的特定特征和动作仅仅是实现所限定主题的示例形式。
虽然在上面论述中包含了若干具体实现细节，但是这些不应当被解释为对本公开的范围的限制。在单独的实施例的上下文中描述的某些特征还可以组合地实现在单个实施例中。相反地，在单个实施例的上下文中描述的各种特征也可以单独地或以任何合适的子组合的方式实现在多个实施例中。
以上描述仅为本公开的较佳实施例以及对所运用技术原理的说明。本领域技术人员应当理解，本公开中所涉及的公开范围，并不限于上述技术特征的特定组合而成的技术方案，同时也应涵盖在不脱离上述公开构思的情况下，由上述技术特征或其等同特征进行任意组合而形成的其它技术方案。例如上述特征与本公开中公开的(但不限于)具有类似功能的技术特征进行互相替换而形成的技术方案。

Claims (10)

  1. 一种旅客托运行李信息预测方法，其特征在于，包括：
    获取当前航班对应的离港旅客数据和离港实际旅客托运行李历史数据；
    确定对当前航班进行旅客托运行李信息预测所需基于的目标规则数据；
    根据所述离港旅客数据、所述离港实际旅客托运行李历史数据和所述目标规则数据中的至少部分，预测当前航班的离港旅客托运行李重量，以基于预测的离港旅客托运行李重量对当前航班进行配载平衡处理。
  2. 根据权利要求1所述的方法，其特征在于，所述确定对当前航班进行旅客托运行李信息预测所需基于的目标规则数据，包括：
    根据当前航班的航班信息，确定当前航班在预先配置的估算规则数据集中对应的目标预测规则；其中，所述估算规则数据集包括旅客托运行李预测规则及相关参数数据；
    若所述目标预测规则为所述旅客托运行李预测规则中的历史数据预测规则，确定是否满足历史数据预测条件，若是，确定所述目标规则数据包括所述历史数据预测规则及其在所述估算规则数据集中对应的规则参数；若否，所述目标规则数据包括所述估算规则数据集中的旅客托运行李重量静态数据；
    若所述目标预测规则为所述旅客托运行李预测规则中的时间序列预测规则，确定是否满足时间序列预测条件，若是，确定所述目标规则参数包括所述时间序列预测规则及其在所述估算规则数据集中对应的规则参数；若否，所述目标规则数据包括所述估算规则数据集中的旅客托运行李重量静态数据。
  3. 根据权利要求2所述的方法，其特征在于，所述根据所述离港旅客数据、所述离港实际旅客托运行李历史数据和所述目标规则数据中的至少部分，预测当前航班的离港旅客托运行李重量，包括：
    若所述目标规则数据包括所述历史数据预测规则及其规则参数，从所述实际旅客托运行李历史数据中提取所述历史数据预测规则所需的历史参数数据，根据所述离港旅客数据、所述历史参数数据，和所述历史数据预测规则及其规则参数，预测当前航班的离港旅客托运行李重量；
    若所述目标规则数据包括所述时间序列预测规则及其规则参数，从所述实际旅客托运行李历史数据中提取所述时间序列预测规则所需的历史时间序列 参数数据，根据所述离港旅客数据、历史时间序列参数数据，和所述时间序列预测规则及其规则参数，预测当前航班的离港旅客托运行李重量；
    若所述目标规则数据包括所述旅客托运行李重量静态数据，根据所述离港旅客数据和所述旅客托运行李重量静态数据，预测当前航班的离港旅客托运行李重量。
  4. 根据权利要求2所述的方法，其特征在于，所述时间序列预测规则包括：预先构建的ARIMA模型所表征的预测规则；
    其中，所述ARIMA模型通过基于离港旅客托运行李重量的时间序列数据样本进行模型训练而得到。
  5. 根据权利要求1所述的方法，其特征在于，在预测当前航班的离港旅客托运行李重量之后，还包括：
    根据预测的离港旅客托运行李重量，确定当前航班的离港旅客托运行李和集装器总重；
    其中，如果当前航班为集装化飞机，则离港旅客托运行李和集装器总重为离港旅客托运行李重量与集装器重量之和，如果当前航班为非集装化飞机，则离港旅客托运行李和集装器总重为离港旅客托运行李重量。
  6. 根据权利要求1所述的方法，其特征在于，还包括：
    获取当前航班的实际离港旅客托运行李数据，在当前航班关闭后将获取的数据更新存储至当前航班的离港实际旅客托运行李历史数据中。
  7. 根据权利要求2所述的方法，其特征在于，还包括：
    确定当前航班的离港旅客托运行李重量预测值与实际值之间的偏差；
    在所述偏差达到预设阈值情况下，进行预警处理，以基于预警处理调整所述估算规则数据集中的相应规则参数，并在后续时间基于调整后的规则参数进行当前航班的离港旅客托运行李重量预测。
  8. 一种旅客托运行李信息预测装置，其特征在于，包括：
    数据获取模块，用于获取当前航班对应的离港旅客数据和离港实际旅客托运行李历史数据；
    规则确定模块，用于确定对当前航班进行旅客托运行李信息预测所需基于的目标规则数据；
    信息预测模块，用于根据所述离港旅客数据、所述离港实际旅客托运行李 历史数据和所述目标规则数据中的至少部分，预测当前航班的离港旅客托运行李重量，以基于所述离港旅客托运行李重量对当前航班进行配载平衡处理。
  9. 一种计算机可读介质，其特征在于，其上存储有计算机程序，所述计算机程序包含用于执行如权利要求1-7任一项所述的旅客托运行李信息预测方法的程序代码。
  10. 一种计算机程序产品，其特征在于，其包括承载在非暂态计算机可读介质上的计算机程序，所述计算机程序包含用于执行如权利要求1-7任一项所述的旅客托运行李信息预测方法的程序代码。
PCT/CN2023/138046 2022-12-15 2023-12-12 旅客托运行李信息预测方法及相关设备 WO2024125485A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211613120.XA CN115630833B (zh) 2022-12-15 2022-12-15 旅客托运行李信息预测方法及相关设备
CN202211613120.X 2022-12-15

Publications (1)

Publication Number Publication Date
WO2024125485A1 true WO2024125485A1 (zh) 2024-06-20

Family

ID=84911214

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/138046 WO2024125485A1 (zh) 2022-12-15 2023-12-12 旅客托运行李信息预测方法及相关设备

Country Status (2)

Country Link
CN (1) CN115630833B (zh)
WO (1) WO2024125485A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115630833B (zh) * 2022-12-15 2023-04-14 中国民航信息网络股份有限公司 旅客托运行李信息预测方法及相关设备
CN117252402B (zh) * 2023-11-17 2024-02-06 民航成都电子技术有限责任公司 机场值机柜台航司分配规划方法、装置、设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111784049A (zh) * 2020-06-30 2020-10-16 中国民航信息网络股份有限公司 一种旅客流失时间预测方法及装置
CN111915123A (zh) * 2020-06-04 2020-11-10 中国南方航空股份有限公司 一种航班全流程自动化预配载方法
US20200361633A1 (en) * 2019-05-19 2020-11-19 Air Black Box Technologies Llc Managed connecting service for mass transit baggage
US20210035029A1 (en) * 2019-07-31 2021-02-04 International Business Machines Corporation Dynamically updating an automated luggage handling system based on changing reservations
CN115630833A (zh) * 2022-12-15 2023-01-20 中国民航信息网络股份有限公司 旅客托运行李信息预测方法及相关设备

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107977809B (zh) * 2017-11-17 2021-05-04 黄光乾 一种国内客运物流运输的方法
CN107730187B (zh) * 2017-11-17 2021-05-04 黄光乾 一种国内/国际航空物流运输的方法
US20190318441A1 (en) * 2018-04-13 2019-10-17 International Business Machines Corporation Indirect luggage weight identification
US11820534B2 (en) * 2020-06-09 2023-11-21 Accenture Global Solutions Limited Baggage weight prediction
WO2022020825A1 (en) * 2020-07-24 2022-01-27 Nance Kirk C Automated survey process to determine average passenger weight and average checked-bag weight used in determining aircraft weight

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200361633A1 (en) * 2019-05-19 2020-11-19 Air Black Box Technologies Llc Managed connecting service for mass transit baggage
US20210035029A1 (en) * 2019-07-31 2021-02-04 International Business Machines Corporation Dynamically updating an automated luggage handling system based on changing reservations
CN111915123A (zh) * 2020-06-04 2020-11-10 中国南方航空股份有限公司 一种航班全流程自动化预配载方法
CN111784049A (zh) * 2020-06-30 2020-10-16 中国民航信息网络股份有限公司 一种旅客流失时间预测方法及装置
CN115630833A (zh) * 2022-12-15 2023-01-20 中国民航信息网络股份有限公司 旅客托运行李信息预测方法及相关设备

Also Published As

Publication number Publication date
CN115630833B (zh) 2023-04-14
CN115630833A (zh) 2023-01-20

Similar Documents

Publication Publication Date Title
WO2024125485A1 (zh) 旅客托运行李信息预测方法及相关设备
Ghelichi et al. Logistics for a fleet of drones for medical item delivery: A case study for Louisville, KY
Liang et al. On a new rotation tour network model for aircraft maintenance routing problem
Reinhardt et al. Synchronized dial-a-ride transportation of disabled passengers at airports
US10665114B2 (en) Aircraft fuel optimization analytics
CN104751681B (zh) 一种基于统计学习模型的停机位分配方法
CN111915046B (zh) 用于输出信息的方法和装置
CN111401601B (zh) 一种面向延误传播的航班起降时间预测方法
US20150276410A1 (en) Journey planning method and system
CN112330983B (zh) 不正常航班一体化智能恢复方法
CN111798079A (zh) 航班调整方法、装置、电子设备及存储介质
CN112241405B (zh) 航班计划自动生成方法及装置、存储介质及电子设备
CN113486031A (zh) 一种离港航班数据的更新方法及相关设备
CN112184117A (zh) 行李托运目的地的确定方法及装置、存储介质及电子设备
CN112200625A (zh) 一种航班资源推荐方法及装置
CN112966846A (zh) 旅客签转方法及装置、存储介质及电子设备
CN110751359B (zh) 一种自动化航线网络评估方法、电子设备及存储介质
CN111915123A (zh) 一种航班全流程自动化预配载方法
Scardaoni et al. Aircraft turnaround time estimation in early design phases: Simulation tools development and application to the case of box-wing architecture
Chan et al. Agent-based flight planning system for enhancing the competitiveness of the air cargo industry
CN115204526B (zh) 航班燃油数据的采集与处理方法、装置及计算机可读介质
CN112651673A (zh) 一种资源规划方法及相关设备
CN114239325B (zh) 机场值机托运柜台配置规划方法、装置、设备和存储介质
CN105654340A (zh) 旅客真实航程的确定方法及***
CN115049100A (zh) 机场资源分配优化方法、装置、设备及可读存储介质