Airport transportation path optimization and scheduling management system based on big data

Abstract

The invention discloses an airport transportation path optimization and scheduling management system based on big data, and relates to the technical field of transportation path optimization. The airport transportation path optimization and scheduling management system based on big data comprises an airport luggage transportation data acquisition and preprocessing module used for acquiring airport luggage transportation data in real time and preprocessing the airport luggage transportation data; the path delay evaluation and normal path screening module is used for constructing a passing path sequence, evaluating a path operation state and outputting a normal path set; the transportation task scheduling and path selection module is used for analyzing the scheduling priorities of the candidate paths and completing optimal allocation of the candidate paths; and the transportation execution feedback and path state updating module is used for quantifying the deviation degree of the path during actual execution, correcting a path delay evaluation value and node parameters and updating scheduling input. The problems of transportation accumulation and deadlock caused by the fact that queuing and waiting mechanisms between nodes are not considered in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of transport route optimization, and in particular to an airport transport route optimization and scheduling management system based on big data. Background Art

[0002] As demand for air travel continues to grow, major hub airports face multiple pressures during peak hours, including intensive baggage transportation tasks, limited routing resources, and demanding scheduling timelines. From passenger check-in to loading onto flights, baggage undergoes multiple steps, including sorting, transshipment, and handover, creating a highly complex network of transportation routes. To meet the increasing demands of modern aviation logistics for operational efficiency and system coordination, corresponding intelligent management mechanisms have been established to achieve efficient organization and dynamic scheduling of the baggage transportation process.

[0003] For example, patent application CN117521924A discloses a multi-conditional path selection method for airport transportation. This invention proposes a continuous-time search algorithm based on minimum vertex cover. Cardinality conflicts derived from the mutual exclusion principle are used to construct a cardinality conflict graph. A minimum vertex cover algorithm is then used to decompose AMRs with independent cardinality conflicts. The conflict graph is then solved to identify the most influential conflict pairs and heuristic values. This method breaks high-level search symmetry, reduces the number of high-level node splits, improves algorithm speed, and minimizes conflicts.

[0004] However, existing methods often focus on static factors like route length and transport speed, ignoring the dynamic changes in queuing mechanisms and processing resources between nodes. This results in the selection of theoretically the shortest routes, but in practice, they are prone to backlogs, delays, and even task deadlocks. Regarding task scheduling, there is generally a lack of an adaptive model between the transport task itself and the real-time status of the route, making it impossible to implement a joint scheduling strategy based on task urgency, route load capacity, and historical scheduling data.

[0005] Therefore, in response to the above problems, there is an urgent need for an airport transportation route optimization and scheduling management system based on big data. Summary of the Invention

[0006] Technical problems solved In response to the shortcomings of the existing technology, the present invention provides an airport transportation route optimization and scheduling management system based on big data, which solves the problem of transportation accumulation and deadlock caused by the existing technology not considering the queuing and waiting mechanism between nodes.

[0007] Technical Solution To achieve the above objectives, the present invention is implemented through the following technical solutions: an airport transportation path optimization and scheduling management system based on big data, including: an airport baggage transportation data acquisition and preprocessing module, which is used to collect multi-source data from airport transportation path maps, sensor networks, check-in systems and flight scheduling systems in real time, extract airport baggage transportation data from the multi-source data, and preprocess the airport baggage transportation data; a path delay evaluation and normal path screening module, which is used to construct a pass path sequence based on the airport baggage transportation path, and evaluate the operating status of each pass path in the pass path sequence based on the preprocessed airport baggage transportation data, and screen the pass path sequence based on the evaluation results. The transport task scheduling and path selection module calls the normal path set, filters the paths that overlap with the passable paths, builds the candidate path set, analyzes the scheduling priority of the candidate paths, completes the optimal allocation of the candidate paths, generates scheduling instructions, and records the transport task execution log; the transport execution feedback and path status update module is used to extract data from the transport task execution log, establish a path execution feedback data set, quantify the degree of deviation of the path during actual execution, and adjust the running time evaluation value of the passable path and the processing capacity configuration of the processing point based on the quantification result. The adjustment result is used as the update basis for path optimization and task scheduling to build a closed-loop scheduling mechanism.

[0008] Furthermore, multi-source data from the airport transportation route map, sensor network, check-in system and flight scheduling system are collected in real time. The specific steps for extracting airport baggage transportation data from the multi-source data are as follows: by performing structured analysis on the airport transportation route map, the connection relationship and numbering information of each transportation processing point are extracted; the sensor network is synchronously connected to periodically read the real-time values of the conveyor belt speed and the number of baggage on the current path; the baggage check-in time and the corresponding processing point identifier are obtained from the check-in system; at the same time, the mapping relationship between the flight departure time and the baggage target segment is extracted from the flight scheduling system; the above-mentioned source data are aligned according to a unified timestamp format, and the airport baggage transportation data including the path length, conveyor belt speed, the number of baggage on the current path, the number of baggage arriving at the node, the maximum baggage processing capacity of the node, the flight departure time and the baggage check-in time are extracted.

[0009] Furthermore, the specific steps for preprocessing airport baggage transportation data are as follows: outliers in path length and conveyor belt speed are identified through outlier detection and correction algorithms to eliminate data drift caused by erroneous entry; the airport baggage transportation data is denoised through a sliding window median filter algorithm to smooth out data jumps caused by system jitter and instantaneous repeated recordings of sensors; the airport baggage transportation data is statistically standardized through the Z-score normalization method; and the airport baggage transportation data is uniformly range-adjusted through decimal calibration normalization to achieve normalization of the airport baggage transportation data.

[0010] Furthermore, the specific steps of constructing a pass path sequence based on the airport baggage transportation path and evaluating the operating status of each pass path in the pass path sequence based on the pre-processed airport baggage transportation data are as follows: in the airport baggage transportation path, a number of physical transportation processing points are set, each transportation processing point is used to perform baggage sorting, transfer and handover operations, and a node number is assigned to each transportation processing point; multiple consecutive transportation processing points are connected in sequence to construct a pass path sequence from the check-in entrance to the flight loading area; based on the pre-processed airport baggage transportation data, The operating status of each accessible path is evaluated: the path length is divided by the conveyor belt speed to obtain the basic transportation time; the number of bags on the current path is added to the number of bags arriving at the node, and then divided by the maximum baggage handling capacity of the node plus the minimum term to obtain the processing load time; the basic transportation time and the processing load time are added to obtain the total time consumption for transportation and processing; the flight departure time is subtracted from the baggage check-in time and then added to the natural constant, and the logarithm is taken. The inverse of this logarithm value is added and one is obtained; the total time consumption term is multiplied by the urgency adjustment factor to obtain the path delay evaluation value.

[0011] Furthermore, based on the evaluation results, a passable normal path is screened out from the passable path sequence, and the specific steps of outputting a normal path set are as follows: a path delay evaluation value and a path delay threshold are compared in real time; if the path delay evaluation value is less than the path delay threshold, it is marked as a normal path, all normal paths are extracted, and a normal path set is constructed; if the path delay evaluation value is greater than or equal to the path delay threshold, it is marked as a congested path, and the scheduling of the path is suspended; a fixed scheduling cycle is set, and the path delay evaluation value is regularly updated based on the latest collected airport baggage transportation data, and the path label is dynamically updated according to the updated path delay evaluation value.

[0012] Furthermore, the normal path set is called to screen the passable paths of the current task and the repeated paths in the normal path set, and the specific steps for constructing the candidate path set are as follows: for each baggage transportation task to be scheduled, the number of times the current task has been delayed in the past cycle is counted and recorded as the scheduling wait number; and the starting processing point number of the current task and the loading processing point number corresponding to the target flight are determined, and all passable paths between the two are extracted; the normal path set is called to screen the passable paths of the current task and the repeated paths in the normal path set to construct the candidate path set.

[0013] Furthermore, the specific steps for analyzing the scheduling priority of candidate paths are as follows: Based on real-time airport baggage transportation data and path delay evaluation values, the scheduling priority of candidate paths is analyzed: the number of scheduling waits is increased by one and then divided by the path delay evaluation value to obtain the path scheduling compensation item; the flight departure time is subtracted from the baggage check-in time and then the minimum item is added, and the inverse of the value is taken to obtain the task urgency factor; the conveyor belt speed is divided by the path length to obtain the unit transportation efficiency factor; the path scheduling compensation item, the task urgency factor and the unit transportation efficiency factor are multiplied to obtain the scheduling adaptation evaluation value.

[0014] Furthermore, the specific steps for completing the optimal allocation of candidate paths, generating scheduling instructions, and recording the transport task execution log are as follows: sorting the scheduling adaptation evaluation values of the paths in the candidate path set from high to low, selecting the path with the highest scheduling adaptation evaluation value as the execution path of the current transport task, and realizing the optimal path allocation; if there are multiple paths with the same scheduling adaptation evaluation value in the candidate path set, based on the joint path sorting strategy that integrates the Markov decision process and the queuing network model, calling the node waiting time and scheduling waiting number history sequence corresponding to the path, constructing the sorting input state, outputting the current state transition probability and expected delay structure of each path for secondary sorting selection; if the candidate path set for the current task is empty, then the task is marked as an unscheduled task, and its scheduling waiting number is increased by one, and retained until the next scheduling cycle to retry path selection; generating a scheduling instruction for the transport task that has completed path allocation to drive subsequent transport execution; and writing the task number, selected path number, start and target processing point number, path delay evaluation value, scheduling time, and scheduling adaptation evaluation value into the transport task execution log.

[0015] Furthermore, data from the transport task execution log is extracted to establish a path execution feedback dataset. The specific steps for quantifying the degree of deviation of the path during actual execution are as follows: after the baggage transport task is completed, the actual transport time of the task and the node waiting time of each transport processing point in the path are collected to obtain the actual transport time and the average node waiting time; data from the transport task execution log is extracted and used to establish a path execution feedback dataset with the actual transport time and the average node waiting time; based on the path execution feedback dataset, the degree of deviation between the scheduling prediction and the execution result of the path is quantified: the actual transport time minus the path delay evaluation value is divided by the path delay evaluation value plus the minimum term, and the absolute value of the ratio is taken to obtain the scheduling deviation factor; the node waiting time, the number of baggage on the current path, and the number of baggage arriving at the node are added to obtain the path comprehensive load item; the path comprehensive load item is divided by the maximum baggage handling capacity of the node plus the minimum term to obtain the load normalization factor; the scheduling deviation factor is added to the load normalization factor to obtain the path execution deviation evaluation value.

[0016] Furthermore, based on the quantification results, the runtime evaluation value of the pass path and the processing capacity configuration of the processing point are adjusted, and the adjustment results are used as the basis for updating path optimization and task scheduling. The specific steps for constructing a closed-loop scheduling mechanism are as follows: the path execution deviation evaluation value is compared with the execution deviation threshold in real time. When the path execution deviation evaluation value is greater than or equal to the execution deviation threshold, the path delay evaluation value of the path is corrected and improved in the next evaluation cycle to prevent the path from being frequently selected; when the path execution deviation evaluation value is less than the execution deviation threshold, the original path delay evaluation value of the path is maintained; when the node waiting time of a processing point in the path exceeds the node waiting time threshold for three consecutive scheduling cycles, the processing point is marked as an overloaded node, and the node maximum baggage handling capacity of the processing point is reduced in subsequent scheduling cycles; after each scheduling cycle, the path delay evaluation value is updated based on the updated node maximum baggage handling capacity and real-time airport baggage transportation data, and the updated path delay evaluation value is used as the input basis for calculating the scheduling adaptation evaluation value and path sorting in the next cycle, forming a closed-loop mechanism of scheduling-execution-feedback-update.

[0017] Beneficial effects The present invention has the following beneficial effects: (1) This airport transport path optimization and scheduling management system based on big data dynamically integrates path efficiency and task timeliness requirements by building a path operation status evaluation mechanism. It can not only comprehensively reflect the accessibility and processing capacity of the current path, but also identify the effective path range that meets the scheduling conditions in real time under the conditions of flight concentration and node resource imbalance, providing a multi-factor driven path screening basis for subsequent task allocation.

[0018] (2) This airport transport path optimization and scheduling management system based on big data introduces a scheduling adaptation evaluation model that combines the number of scheduling wait times, path delay evaluation values, flight departure time and baggage check-in time, and uniformly maps task urgency, path capacity and scheduling history behavior into scheduling adaptation evaluation values. Based on this, the candidate paths are sorted and optimized, and the two-way adaptation of task status and path status is achieved, thereby enhancing the perception of task time limit and path load during the scheduling process.

[0019] (3) This airport transport path optimization and scheduling management system based on big data builds a path execution feedback data set by establishing a feedback data link between scheduling instructions and transport task execution logs. It characterizes the degree of deviation between the predicted delay and the actual time of the path by calculating the path execution deviation evaluation value, and accordingly corrects key parameters such as the path delay evaluation value and the maximum baggage handling capacity of the node, thereby realizing the continuous updating of the path evaluation model.

[0020] (4) This airport transport path optimization and scheduling management system based on big data forms a scheduling closed-loop mechanism with path delay evaluation value as the evaluation basis, scheduling adaptation evaluation value as the selection basis, and path execution deviation evaluation value as the update drive, realizing a complete data closed-loop from real-time data collection, path status evaluation, task path allocation to execution result feedback, updating the path evaluation parameters in each scheduling cycle, and enhancing the system's dynamic adaptability and stable scheduling capabilities for baggage transport tasks under high-load and high-fluctuation scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a structural diagram of an airport transportation route optimization and scheduling management system based on big data; Figure 2 A histogram of path delay evaluation values; Figure 3 A ranking graph of candidate path scheduling adaptation evaluation values. DETAILED DESCRIPTION

[0022] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0023] See also Figure 1-Figure 3The embodiment of the present invention provides a technical solution: an airport transportation path optimization and scheduling management system based on big data, including: an airport baggage transportation data collection and preprocessing module, which is used to collect multi-source data from airport transportation path maps, sensor networks, check-in systems and flight scheduling systems in real time, extract airport baggage transportation data from the multi-source data, and preprocess the airport baggage transportation data; a path delay evaluation and normal path screening module, which is used to construct a pass path sequence based on the airport baggage transportation path, and evaluate the operating status of each pass path in the pass path sequence based on the preprocessed airport baggage transportation data, and screen out passable paths from the pass path sequence based on the evaluation results. The normal path is selected and the normal path set is output; the transportation task scheduling and path selection module calls the normal path set, filters the paths that overlap with the passable paths, builds the candidate path set, analyzes the scheduling priority of the candidate paths, completes the optimal allocation of the candidate paths, generates scheduling instructions, and records the transportation task execution log; the transportation execution feedback and path status update module is used to extract data from the transportation task execution log, establish a path execution feedback data set, quantify the degree of deviation of the path during actual execution, adjust the running time evaluation value of the passable path and the processing capacity configuration of the processing point based on the quantification result, and use the adjustment result as the update basis for path optimization and task scheduling to build a closed-loop scheduling mechanism.

[0024] Specifically, multi-source data from the airport transportation route map, sensor network, check-in system, and flight scheduling system are collected in real time. The specific steps for extracting airport baggage transportation data from the multi-source data are as follows: by performing structured analysis on the airport transportation route map, the connection relationship and number information of each transportation processing point are extracted, and a mapping structure between path nodes is established; the sensor network is synchronously connected to periodically read the real-time values of the conveyor belt speed and the current number of baggage on the route, and recorded in combination with the route number; the baggage check-in time and the corresponding processing point identifier are obtained from the check-in system, and the historical route segment passing records are extracted to assist in determining the cumulative change in the number of baggage arriving at the node; the mapping relationship between flight departure time and baggage target segment is extracted from the flight scheduling system, and a corresponding index between the flight number and the route terminal node is established; the above-mentioned source data are aligned according to a unified timestamp format, and the airport baggage transportation data including route length, conveyor belt speed, current route baggage number, node arrival baggage number, node maximum baggage handling capacity, flight departure time, and baggage check-in time are extracted to form a basic data set for subsequent route evaluation and scheduling calculation.

[0025] In this implementation plan, by building a multi-source data collection mechanism covering path structure, operating status and task plan, we can achieve complete acquisition of path length, conveyor belt speed, number of luggage on the current path, number of luggage arriving at the node, maximum luggage handling capacity of the node, flight departure time and luggage check-in time, providing high-efficiency and high-precision data support for the accurate calculation of path delay evaluation value and scheduling adaptation evaluation value.

[0026] Specifically, the specific steps for preprocessing airport baggage transportation data are as follows: outliers in path length and conveyor belt speed are identified through outlier detection and correction algorithms, and data drift caused by equipment abnormalities or manual entry errors is eliminated in combination with historical path structure information; airport baggage transportation data is denoised through a sliding window median filter algorithm, and the jump values of path length, conveyor belt speed, current path baggage number and node arrival baggage number are smoothed within a continuous time window to weaken the interference of system jitter and instantaneous repeated recording of sensors; path length, current path baggage number and node maximum baggage handling capacity are statistically normalized through the Z-score normalization method to eliminate the dimensional influence between different data dimensions; flight departure time, baggage check-in time and node arrival baggage number are interval mapped and adjusted through decimal scaling normalization to achieve normalization of airport baggage transportation data, providing a unified data input basis for the calculation of subsequent path delay evaluation value and scheduling adaptation evaluation value.

[0027] In this implementation plan, by constructing a multi-dimensional data preprocessing process covering path length, conveyor belt speed, number of baggage on the current path, number of baggage arriving at a node, maximum baggage handling capacity at a node, flight departure time, and baggage check-in time, the stability, consistency, and computability of airport baggage transportation data at the input stage are effectively improved, laying a high-quality data foundation for the accurate modeling of path delay evaluation values and scheduling adaptation evaluation values.

[0028] Specifically, the specific steps of constructing a pass path sequence based on the airport baggage transportation path and evaluating the operating status of each pass path in the pass path sequence based on the pre-processed airport baggage transportation data are as follows: in the airport baggage transportation path, a number of physical transportation processing points are set, each transportation processing point is used to perform baggage sorting, transfer and handover operations, and a node number is assigned to each transportation processing point to identify the node position and the target binding of the scheduling instruction; multiple consecutive transportation processing points are connected in sequence, and a pass path sequence from the check-in entrance to the flight loading area is constructed in combination with the path graph topology structure to form a candidate path structure set; based on the pre-processed airport baggage transportation data, the operating status of each passable path is evaluated: The path length is divided by the conveyor belt speed and combined with real-time sensor data to obtain the basic transportation time. The number of bags on the current path is added to the number of bags arriving at the node, and then divided by the maximum baggage handling capacity of the node plus the minimum term to reflect the current node resource load level, resulting in the processing load time. The basic transportation time and the processing load time are added to form the total time consumption for transportation and processing, which is used to characterize the overall operating pressure of the current path. The flight departure time is subtracted from the baggage check-in time and then added to the natural constant, and the logarithm is taken. The inverse of this logarithm is added and one is added to construct the urgency adjustment factor based on the degree of time urgency. The total time consumption term is multiplied by the urgency adjustment factor to obtain the path delay assessment value, which is used to measure the feasibility of the current path within the scheduling cycle and the queuing risk level.

[0029] The specific calculation formula of the path delay evaluation value is: ; Where D is the path delay evaluation value, L is the path length, S is the conveyor belt speed, N is the number of bags on the current path, R is the number of bags arriving at the node, C is the maximum baggage handling capacity of the node, and T is the flight departure time. Indicates the baggage check-in time. represents the minimum term, and e represents the base of natural logarithms.

[0030] In this embodiment, Table 1 is a table of path delay assessment values, detailing the path structure parameters and path delay assessment calculation results for five traversable paths during the current scheduling cycle. Each traversable path is evaluated based on its path length, conveyor speed, current path baggage count, node arrival baggage count, node maximum baggage handling capacity, flight departure time, and baggage check-in time, using a predefined evaluation formula to calculate a corresponding path delay assessment value. This value is used to measure the path's delay risk and scheduling feasibility under the current transport task schedule. For traversable path P1, the path length is 45, the conveyor speed is 1.50, the current path baggage count is 10, the node arrival baggage count is 6, the node maximum baggage handling capacity is 28, the flight departure time is the 100th, and the baggage check-in time is the 66th. The calculated path delay assessment value for this path is 39.81. The path length of traversable path P2 is 50, the conveyor speed is 1.60, the number of bags on the current path is 8, the number of bags arriving at the node is 7, the maximum baggage handling capacity of the node is 30, the flight departure time is the 101st, and the baggage check-in time is the 68th. The corresponding path delay evaluation value is 41.38. The path length of traversable path P3 is 55, the conveyor speed is 1.40, the number of bags on the current path is 12, the number of bags arriving at the node is 5, the maximum baggage handling capacity of the node is 26, the flight departure time is the 104th, and the baggage check-in time is the 70th. The calculated path delay evaluation value is 52.00. The path length of traversable path P4 is 48, the conveyor speed is 1.50, the number of bags on the current path is 9, the number of bags arriving at the node is 6, the maximum baggage handling capacity of the node is 27, the flight departure time is the 105th, and the baggage check-in time is the 71st. The corresponding path delay evaluation value is 43.39. The path length of the traversable path P5 is 52, the conveyor speed is 1.45, the number of bags on the current path is 11, the number of bags arriving at the node is 7, the maximum baggage handling capacity of the node is 29, the flight departure time is the 107th, the baggage check-in time is the 73rd, and the calculated path delay evaluation value is 47.50.

[0031] Table 1 Path delay evaluation value data table

[0032] like Figure 2The figure shows a histogram of path delay evaluation values, displaying the distribution of path delay evaluation values for five accessible paths during the current scheduling cycle and classifying the path status according to the set path delay threshold. Green represents normal paths, red represents congested paths, and the blue dashed line represents the reference baseline for the path delay threshold. As can be seen from the figure, the path delay evaluation values of accessible paths P1, P2, and P4 are 39.81, 41.38, and 43.39, respectively, all below the path delay threshold and classified as normal paths. Paths P3 and P5 have path delay evaluation values of 52.00 and 47.50, respectively, exceeding the path delay threshold and marked as congested paths during the current scheduling cycle. This chart helps to intuitively identify path scheduling availability and provides a visual basis for scheduling optimization and path screening.

[0033] This implementation dynamically quantifies the estimated path delay by constructing a sequence of travel paths from the check-in entrance to the flight loading area and integrating path length, conveyor speed, baggage count on the current path, baggage arrival count at a node, maximum baggage handling capacity at a node, flight departure time, and baggage check-in time. This effectively improves the accuracy and stability of path status identification, responding to the new requirements of modern aviation logistics for transport efficiency, resource allocation, and refined process management, providing efficient and reliable structured input support for subsequent path screening and task scheduling.

[0034] Specifically, based on the evaluation results, a passable normal path is screened out from the passable path sequence, and the specific steps of outputting a normal path set are as follows: a real-time comparison of the path delay evaluation value and the path delay threshold, where the path delay evaluation value is determined by the path length, conveyor belt speed, the number of luggage on the current path, the number of luggage arriving at the node, the maximum luggage handling capacity of the node, the flight departure time, and the luggage check-in time; if the path delay evaluation value is less than the path delay threshold, it is marked as a normal path, and the corresponding path number, path structure, and current status are recorded, all normal paths are extracted, and a normal path set is constructed as the initial input of the candidate path; if the path delay evaluation value is greater than or equal to the path delay threshold, it is marked as a congested path, and the path delay evaluation value change trend of the path is registered, and the scheduling of the path is suspended; a fixed scheduling cycle is set, and the path delay evaluation value is regularly updated based on the latest collected airport baggage transportation data. During the update process, the node processing status and historical evaluation records of the current path are synchronously called, and the path label is dynamically updated according to the updated path delay evaluation value to achieve periodic correction and identification of the path usage status.

[0035] This implementation achieves dynamic identification of normal and congested routes by constructing a path status labeling mechanism centered around path delay assessments. This mechanism updates path labels within a fixed scheduling cycle. This addresses the higher requirements for path efficiency and scheduling response accuracy in complex operational scenarios in modern aviation logistics, effectively improving the real-time accuracy of path screening and enhancing the automation of path status identification and label maintenance in airport baggage transport scheduling. This provides stable data support for subsequent scheduling adaptation assessment value calculations and candidate path construction.

[0036] Specifically, the normal path set is called, the passable paths of the current task are screened against the repeated paths in the normal path set, and the specific steps for constructing a candidate path set are as follows: for each baggage transport task to be scheduled, the number of times the current task has been delayed in the past cycle is counted, and this number is written into the task scheduling log in combination with the task number and recorded as the number of scheduling wait times; the starting processing point number of the current task and the loading processing point number corresponding to the target flight are determined, and all passable paths between the two are extracted in combination with the path graph structure information, and the path numbers and node sequences are recorded; the normal path set is called, and paths with path delay evaluation values lower than the path delay threshold are retrieved. The passable paths of the current task are compared and screened against the repeated paths in the normal path to construct a candidate path set, and key data such as the path length, conveyor belt speed, number of baggage on the current path, and number of baggage arriving at the node are registered for the candidate paths, providing an input path set for the calculation of the subsequent scheduling adaptation evaluation value.

[0037] In this implementation, a set of normal paths is called and combined with the starting processing point number and the loading processing point number to extract traversable paths. This constructs a set of candidate paths that matches the current transport task. This implementation implements targeted constraints on scheduling path screening, supported by the number of scheduling wait times, path number, path length, conveyor belt speed, number of baggage on the current path, and number of baggage arriving at the node. This provides a closed-structured data entry for the precise calculation of scheduling adaptation evaluation values and optimal path allocation.

[0038] Specifically, the specific steps for analyzing the scheduling priority of candidate paths are as follows: based on real-time airport baggage transportation data and path delay evaluation values, the scheduling priority of candidate paths is analyzed; the scheduling waiting times are increased by one and divided by the path delay evaluation value, and the path number and historical scheduling records are combined to obtain the path scheduling compensation item, which reflects the corrective impact of the current task on the frequency of path usage; the flight departure time is subtracted from the baggage check-in time and the minimum term is added, and the inverse of this value is taken to construct the task urgency factor, which is used to characterize the degree of constraint of the remaining space of the task time on the path selection; the conveyor belt speed is divided by the path length, and the path structure stability label is combined to obtain the unit transportation efficiency factor, which measures the operating efficiency of the path per unit distance; the path scheduling compensation item, the task urgency factor and the unit transportation efficiency factor are multiplied to obtain the scheduling adaptation evaluation value, which is used as the core indicator for candidate path sorting and preferential allocation.

[0039] The specific calculation formula of the scheduling adaptation evaluation value is: ; Where P represents the scheduling adaptation evaluation value, W represents the scheduling waiting times, D represents the path delay evaluation value, T represents the flight departure time, represents the baggage check-in time, S represents the conveyor belt speed, L represents the path length, and ε represents the minimum term.

[0040] In this embodiment, Table 2 is a scheduling adaptation evaluation value data table, which records in detail the path state parameters and scheduling adaptation evaluation value calculation results of the five candidate paths in the current scheduling cycle. Among them, the scheduling waiting number of candidate path H1 is 0, the path delay evaluation value is 14.47, the flight departure time is the 2324th, the baggage check-in time is the 2159th, the conveyor belt speed is 3.48, the path length is 16, and the calculated scheduling adaptation evaluation value is 9.11; the scheduling waiting number of candidate path H2 is 1, the path delay evaluation value is 19.82, the flight departure time is the 2077th, the baggage check-in time is the 1862nd, the conveyor belt speed is 2.47, the path length is 22, and the calculated result is 8.12; the scheduling waiting number of candidate path H3 is 0, the path delay evaluation value is 22.40, the flight departure time is the 2648th, and the baggage check-in time is the 1862nd. The baggage check-in time is the 2357th, the conveyor belt speed is 3.09, the path length is 19, and the corresponding scheduling adaptation evaluation value is 6.54; the path parameters of candidate path H4 include the scheduling wait time of 0, the path delay evaluation value of 14.56, the departure time of 3115th, the check-in time of 2969th, the conveyor belt speed of 3.36, the path length of 21, and the scheduling adaptation evaluation value of 7.53; the parameters of candidate path H5 are: the scheduling wait time of 1, the path delay evaluation value of 22.28, the flight departure time of 2043, the check-in time of 1867, the conveyor belt speed of 3.45, the path length of 25, and the calculated scheduling adaptation evaluation value is 6.73.

[0041] Table 2 Scheduling adaptation evaluation value data table

[0042] like Figure 3 The figure shows a ranking chart of candidate path scheduling adaptation evaluation values, displaying the ranking results of the scheduling adaptation evaluation values of five candidate paths within the current scheduling cycle. The horizontal axis represents the candidate path number, and the vertical axis represents the corresponding scheduling adaptation evaluation value P. A higher value indicates a path's greater adaptability to the current transport task. As can be seen from the figure, candidate path H1 has the highest scheduling adaptation evaluation value of 9.11, making it the preferred choice. H2 and H4 follow with evaluation values of 8.12 and 7.53, respectively. H5 and H3 have relatively low scheduling adaptation evaluation values of 6.73 and 6.54, respectively. During the scheduling task optimization process, the system prioritizes paths with higher scheduling adaptation evaluation values to undertake the current transport task, thereby achieving optimal allocation of path resources. This provides intuitive support for the path ranking mechanism and assists in the implementation of evaluation-driven path scheduling strategies.

[0043] In this implementation plan, by constructing a scheduling adaptation evaluation value calculation mechanism with scheduling waiting times, path delay evaluation value, flight departure time, baggage check-in time, conveyor belt speed and path length as input, a multi-factor comprehensive characterization of candidate path scheduling priorities is achieved, the accuracy of path sorting and task matching are improved, and a quantifiable decision-making basis is provided for subsequent path optimization allocation.

[0044] Specifically, the specific steps to complete the optimal allocation of candidate paths, generate scheduling instructions, and record the transportation task execution log are as follows: sort the scheduling adaptation evaluation values of the paths in the candidate path set from high to low, combine the path number and the historical task allocation frequency, and select the path with the highest scheduling adaptation evaluation value as the execution path of the current transportation task to achieve path optimization allocation; if there are multiple paths with the same scheduling adaptation evaluation value in the candidate path set, based on the joint path sorting strategy that integrates the Markov decision process and the queuing network model, call the node waiting time and scheduling waiting number historical sequence corresponding to the path, construct the sorting input state, and model each path as a queuing system node, and simulate the path through the Markov state transition method. The evolution process of the scheduling state combines the queue delay characteristics and the state value function to output the joint ranking score of the path and perform secondary ranking selection. If the candidate path set for the current task is empty, the task is marked as an unscheduled task, and its scheduling wait time is increased by one. At the same time, the task number and path demand information are retained in the scheduling buffer queue, and the path selection is retried in the next scheduling cycle. The transportation task with completed path allocation generates a scheduling instruction. The scheduling instruction content includes the task number, selected path number, starting processing point number, target processing point number, path delay evaluation value, scheduling time and scheduling adaptation evaluation value, and is written into the transportation task execution log to drive the subsequent transportation execution and support the data backtracking of the execution feedback mechanism.

[0045] In this implementation plan, by introducing a scheduling adaptation evaluation value sorting mechanism and integrating the Markov decision process with the queuing network model to construct a path secondary sorting strategy, the joint modeling of variables such as the scheduling waiting number, node waiting time, and path delay evaluation value is achieved. In the case of scheduling conflicts, the judgment accuracy of path selection is improved, and a dynamic and robust scheduling logic is provided for transportation task allocation. A structured scheduling feedback chain is established by generating scheduling instructions and recording transportation task execution logs.

[0046] Specifically, the data in the transport task execution log are extracted to establish a path execution feedback data set. The specific steps to quantify the degree of deviation of the path during actual execution are as follows: after the baggage transport task is completed, the actual transport time of the task and the node waiting time of each transport processing point in the path are collected, and the actual execution status of the current task within the scheduling cycle is recorded in combination with the task number. The actual transport time and the average waiting time of the node are obtained as direct observations of the path operation status; the data in the transport task execution log are extracted, including the selected path number, path delay evaluation value, scheduling time and scheduling adaptation evaluation value, and the data are fused with the actual transport time and the average waiting time of the node to establish a path execution feedback data set; based on the path execution feedback data set, the path is quantified. The degree of deviation between the scheduling prediction and the execution result is measured by taking the absolute value of the ratio of the actual transportation time minus the path delay assessment value and dividing it by the path delay assessment value plus the minimum term. The scheduling deviation factor is constructed to measure the error margin of the scheduling prediction in terms of timeliness. The node waiting time, the number of bags on the current path, and the number of bags arriving at the node are added together to form the path comprehensive load term, which reflects the degree of concentrated use of path resources. The path comprehensive load term is divided by the maximum baggage handling capacity of the node plus the minimum term to obtain the load normalization factor, which is used to eliminate the impact of differences in service capabilities of different paths. The scheduling deviation factor is added to the load normalization factor to obtain the path execution deviation assessment value, which is used as a key evaluation indicator in the feedback phase to support subsequent path status updates and parameter corrections.

[0047] The specific calculation formula for the path execution deviation evaluation value is: ; Where E represents the path execution deviation evaluation value, represents the actual transportation time, D represents the path delay evaluation value, represents the average waiting time of the node, N represents the number of bags on the current path, R represents the number of bags arriving at the node, C represents the maximum baggage handling capacity of the node, and ε represents the minimum term.

[0048] In this implementation plan, by constructing a path execution feedback dataset with actual transportation time, average waiting time at nodes, path delay evaluation value, number of luggage on the current path, number of luggage arriving at nodes, and maximum luggage handling capacity at nodes as core variables, quantitative modeling of the degree of deviation between scheduling predictions and execution results is achieved. The path execution deviation evaluation value is constructed as a feedback indicator, forming a quantitative feedback basis for the path status after task execution, providing closed-loop support for subsequent path delay evaluation value correction and node parameter update.

[0049] Specifically, based on the quantification results, the running time evaluation value of the pass path and the processing capacity configuration of the processing point are adjusted, and the adjustment results are used as the update basis for path optimization and task scheduling. The specific steps of building a closed-loop scheduling mechanism are as follows: compare the path execution deviation evaluation value and the execution deviation threshold in real time, and combine the current path number and the corresponding task execution log to determine the degree of deviation of the path scheduling performance; when the path execution deviation evaluation value is greater than or equal to the execution deviation threshold, the path delay evaluation value of the path is corrected and improved in the next evaluation cycle, and the number of corrections and historical fluctuation range of the path are recorded to suppress the path from being frequently selected; when the path execution deviation evaluation value is less than the execution deviation threshold, the original path delay evaluation value of the path is maintained, and its historical evaluation record is retained for trend monitoring When the node waiting time of a processing point in the path exceeds the node waiting time threshold for three consecutive scheduling cycles, the processing point is marked as an overloaded node, and the processing status label of the node is updated. In the subsequent scheduling cycle, the node maximum baggage handling capacity of the processing point is reduced. After each scheduling cycle, the path delay evaluation value is updated based on the updated node maximum baggage handling capacity and the real-time collected airport baggage transportation data such as path length, conveyor belt speed, number of baggage on the current path, and number of baggage arriving at the node. During the update process, the path structure and processing point number are synchronously retained to ensure the consistency of scheduling logic. The updated path delay evaluation value is used as the input basis for calculating the scheduling adaptation evaluation value and path sorting in the next cycle, forming a closed-loop mechanism of scheduling-execution-feedback-update.

[0050] This implementation establishes a path delay assessment correction mechanism triggered by path execution deviation assessments, combined with a dynamic adjustment strategy based on node wait times and maximum baggage handling capacity. This allows for periodic correction of path scheduling performance and node operational status, enhancing the automation of path state modeling and input correction. To meet the complex demands of modern aviation logistics for scheduling accuracy and operational robustness, the path delay assessment is updated after the scheduling cycle and serves as the basis for calculating the scheduling adaptation assessment. This creates a closed-loop structure of scheduling-execution-feedback-update, improving the real-time nature of scheduling inputs and the adaptability of path status.

[0051] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0052] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. An airport transportation route optimization and scheduling management system based on big data, characterized in that: include: The airport baggage transportation data collection and preprocessing module is used to collect multi-source data from airport transportation route maps, sensor networks, consignment systems, and flight scheduling systems in real time, extract airport baggage transportation data from these multi-source data, and preprocess the airport baggage transportation data. A path delay assessment and normal path screening module is used to construct a pass path sequence based on the airport baggage transportation path, evaluate the operating status of each pass path in the pass path sequence based on the pre-processed airport baggage transportation data, screen out passable normal paths from the pass path sequence based on the evaluation results, and output a normal path set; The transport task scheduling and path selection module calls the normal path set, filters the passable paths of the current task and the duplicate paths in the normal path set, builds a candidate path set, analyzes the scheduling priority of the candidate paths, completes the optimal allocation of candidate paths, generates scheduling instructions, and records the transport task execution log; The transport execution feedback and path status update module is used to extract data from the transport task execution log, establish a path execution feedback dataset, quantify the degree of deviation of the path during actual execution, adjust the running time evaluation value of the pass path and the processing capacity configuration of the processing point based on the quantification results, and use the adjustment results as the update basis for path optimization and task scheduling, thereby building a closed-loop scheduling mechanism.

2. The airport transportation route optimization and scheduling management system based on big data according to claim 1 is characterized by: The specific steps of real-time collection of multi-source data from the airport transportation route map, sensor network, consignment system, and flight scheduling system, and extracting airport baggage transportation data from the multi-source data are as follows: By performing a structured analysis of the airport transport route map, the connection relationship and numbering information of each transport processing point are extracted; synchronous access is made to the sensor network to periodically read the real-time values of the conveyor belt speed and the number of bags on the current route; the baggage check-in time and the corresponding processing point identifier are obtained from the check-in system; and the mapping relationship between flight departure time and baggage target segment is extracted from the flight scheduling system. The data from the above sources are aligned according to a unified timestamp format to extract airport baggage transport data including route length, conveyor belt speed, number of bags on the current route, number of bags arriving at the node, maximum baggage handling capacity at the node, flight departure time and baggage check-in time.

3. The airport transportation route optimization and scheduling management system based on big data according to claim 1 is characterized by: The specific steps of pre-processing the airport baggage transportation data are as follows: An outlier detection and correction algorithm is used to identify outliers in path length and conveyor belt speed, eliminating data drift caused by erroneous entries. A sliding window median filter algorithm is used to denoise airport baggage transportation data, smoothing out data jumps caused by system jitter and instantaneous repeated recordings by sensors. The Z-score normalization method is used to statistically normalize the airport baggage transportation data. Decimal scaling normalization is used to uniformly adjust the range of the airport baggage transportation data to achieve normalization of the airport baggage transportation data.

4. The airport transportation route optimization and scheduling management system based on big data according to claim 1 is characterized by: The specific steps of constructing a pass path sequence based on the airport baggage transportation path and evaluating the operating status of each pass path in the pass path sequence based on the pre-processed airport baggage transportation data are as follows: In the airport baggage transportation route, several physical transportation processing points are set up. Each transportation processing point is used to perform baggage sorting, transfer and handover operations, and a node number is assigned to each transportation processing point. Multiple consecutive transportation processing points are connected in sequence to construct a passage path sequence from the check-in entrance to the flight loading area. Based on the pre-processed airport baggage transportation data, the operating status of each passable path is evaluated: the path length is divided by the conveyor belt speed to obtain the basic transportation time; The total time for transportation and processing is calculated by adding the baggage count on the current path to the baggage count arriving at the node, then dividing the sum by the node's maximum baggage handling capacity plus the minimum term. The total time for transportation and processing is calculated by adding the basic transportation time to the handling load time. The logarithm of the flight departure time minus the baggage check-in time plus the natural constant is then taken. The inverse of this logarithm is then added to obtain the urgency adjustment factor. The total time is multiplied by the urgency adjustment factor to obtain the path delay assessment value.

5. The airport transportation route optimization and scheduling management system based on big data according to claim 1 is characterized by: The specific steps of screening out passable normal paths from the passable path sequence based on the evaluation result and outputting the normal path set are as follows: Compare the path delay evaluation value with the path delay threshold in real time. If the path delay evaluation value is less than the path delay threshold, mark it as a normal path, extract all normal paths, and build a normal path set. If the path delay evaluation value is greater than or equal to the path delay threshold, mark it as a congested path and suspend scheduling of the path. A fixed scheduling cycle is set to regularly update the path delay evaluation value based on the latest collected airport baggage transportation data, and the path label is dynamically updated according to the updated path delay evaluation value.

6. The airport transportation route optimization and scheduling management system based on big data according to claim 1 is characterized by: The specific steps of calling the normal path set, screening the passable paths of the current task and the repeated paths in the normal path set, and constructing the candidate path set are as follows: For each baggage transport task to be scheduled, the number of times the current task has been delayed in the past cycle is counted and recorded as the scheduling wait number; the starting processing point number of the current task and the loading processing point number corresponding to the target flight are determined, and all possible paths between the two are extracted; the normal path set is called, and the possible paths of the current task and the repeated paths in the normal path set are screened to construct a candidate path set.

7. The airport transportation route optimization and scheduling management system based on big data according to claim 1 is characterized by: The specific steps of analyzing the scheduling priority of the candidate paths are as follows: Based on real-time airport baggage transportation data and path delay evaluation values, the scheduling priority of candidate paths is analyzed: the path scheduling compensation term is obtained by adding one to the scheduling wait times and dividing it by the path delay evaluation value; the task urgency factor is obtained by subtracting the baggage check-in time from the flight departure time and adding the minimum term, and taking the inverse of this value; the unit transportation efficiency factor is obtained by dividing the conveyor belt speed by the path length; and the path scheduling compensation term, the task urgency factor, and the unit transportation efficiency factor are multiplied together to obtain the scheduling adaptation evaluation value.

8. The airport transportation route optimization and scheduling management system based on big data according to claim 1 is characterized by: The specific steps for completing the optimal allocation of candidate routes, generating scheduling instructions, and recording the transport task execution log are as follows: The scheduling adaptation evaluation values of the paths in the candidate path set are sorted from high to low, and the path with the highest scheduling adaptation evaluation value is selected as the execution path of the current transportation task to achieve path optimization allocation; if there are multiple paths with the same scheduling adaptation evaluation value in the candidate path set, based on the joint path sorting strategy that integrates the Markov decision process and the queuing network model, the node waiting time and scheduling waiting number history sequence corresponding to the path is called to construct the sorting input state, and the current state transition probability and expected delay structure of each path are output for secondary sorting selection; if the candidate path set for the current task is empty, the task is marked as an unscheduled task, and its scheduling waiting number is increased by one, and it is retained until the next scheduling cycle to retry path selection; Generate scheduling instructions for the transport tasks that have completed path allocation to drive subsequent transport execution; and write the task number, selected path number, start and target processing point numbers, path delay evaluation value, scheduling time and scheduling adaptation evaluation value into the transport task execution log.

9. The airport transportation route optimization and scheduling management system based on big data according to claim 1 is characterized by: The specific steps for extracting data from the transport task execution log, establishing a route execution feedback dataset, and quantifying the degree of deviation of the route during actual execution are as follows: After the baggage transport task is completed, the actual transport time of the task and the node waiting time of each transport processing point in the route are collected to obtain the actual transport time and the average node waiting time; Extract data from the transport task execution log and build a path execution feedback dataset based on the actual transport time and average node waiting time. Based on the path execution feedback dataset, the degree of deviation between the scheduling prediction and the execution result of the path is quantified: the actual transportation time minus the path delay evaluation value is divided by the path delay evaluation value plus the minimum term, and the absolute value of this ratio is taken to obtain the scheduling deviation factor. The node waiting time, the number of bags on the current path, and the number of bags arriving at the node are added together to obtain the path comprehensive load term. The path comprehensive load term is divided by the maximum baggage handling capacity of the node plus the minimum term to obtain the load normalization factor. The scheduling deviation factor is added to the load normalization factor to obtain the path execution deviation evaluation value.

10. The airport transportation route optimization and scheduling management system based on big data according to claim 1, characterized in that: The specific steps of constructing a closed-loop scheduling mechanism by adjusting the runtime evaluation value of the pass path and the processing capacity configuration of the processing point based on the quantification results and using the adjustment results as the basis for updating path optimization and task scheduling are as follows: Compare the path execution deviation evaluation value with the execution deviation threshold in real time. When the path execution deviation evaluation value is greater than or equal to the execution deviation threshold, the path delay evaluation value of the path is corrected and improved in the next evaluation cycle to prevent the path from being frequently selected. When the path execution deviation evaluation value is less than the execution deviation threshold, the original path delay evaluation value of the path is maintained. When the node waiting time of a processing point in the path exceeds the node waiting time threshold in three consecutive scheduling cycles, the processing point is marked as an overloaded node, and the maximum baggage handling capacity of the processing point is reduced in subsequent scheduling cycles; After each scheduling cycle, the path delay evaluation value is updated based on the updated node's maximum baggage handling capacity and real-time airport baggage transportation data. The updated path delay evaluation value is used as the input basis for the scheduling adaptation evaluation value calculation and path sorting in the next cycle, forming a closed-loop mechanism of scheduling-execution-feedback-update.

Images