Wharf container truck scheduling method and system

Abstract

The invention relates to the technical field of wharf container trucks, in particular to a wharf container truck dispatching method and system. Comprising the following steps: constructing a demand prediction model and a plurality of working sub-regions based on historical monitoring parameters of a wharf; constructing a mapping sub-model of each container truck, and generating a container truck library according to all the mapping sub-models; establishing a plurality of scheduling cycles, and generating a primary scheduling strategy of the current scheduling cycle according to the demand prediction model and the container truck library; generating an operation deviation value of the first-level scheduling strategy according to the preset feedback time node, and judging whether a correction instruction is generated or not according to the operation deviation value; a prediction sub-model of each operation sub-scene is constructed based on a big data analysis technology, so that the container truck demand quantity and time distribution of different working sub-areas in a single scheduling period are estimated, a scheduling strategy of the container trucks is dynamically adjusted according to an estimation result, the empty driving rate is reduced, and the overall operation efficiency of the container terminal is improved.

Description

Technical Field

[0001] The present application relates to the technical field of terminal container truck scheduling, and in particular to a terminal container truck scheduling method and system. Background Art

[0002] In the operation of container terminals, the dispatching efficiency of container trucks directly affects the overall operating efficiency and logistics costs of the terminal.

[0003] Traditional container truck scheduling often relies on experience or simple rules, lacking the full utilization of real-time data and intelligent analysis. As terminal business volume grows, this scheduling method has gradually exposed problems such as insufficient flexibility and slow response. This results in long waiting times for container trucks and high idle rates, seriously affecting terminal operational efficiency and economic benefits. Summary of the Invention

[0004] The purpose of this application is: to solve the above technical problems, this application provides a terminal container truck scheduling method and system, aiming to improve the scheduling efficiency of container trucks and improve the operating efficiency and economic benefits of the terminal.

[0005] In some embodiments of the present application, a method for dispatching terminal trucks is provided, comprising: Build demand forecasting models and multiple work sub-areas based on historical monitoring parameters of the terminal; Construct mapping sub-models for each truck collection unit, and generate a truck collection unit library based on all mapping sub-models; Establish multiple scheduling cycles and generate the first-level scheduling strategy for the current scheduling cycle based on the demand forecast model and the truck depot; Generate an operation deviation value of the first-level scheduling strategy according to the preset feedback time node, and determine whether to generate a correction instruction based on the operation deviation value; Among them, when generating a truck collection warehouse, it includes: Establish a truck sequence A, A=(a1,a2…a i …a n ), where a i is the i-th collection truck; n is the number of collection trucks.

[0006] In some embodiments of the present application, when building a demand forecasting model, the following steps are included: Generate multiple scene characteristic indicators based on historical monitoring parameters; Establish identification sub-models and multiple operation sub-scenarios based on all scenario characteristic indicators; Create a sequence of sub-scenario operations B, B = (b1, b2…b i …b m ), where b i is the i-th operation sub-scenario; m is the number of operation sub-scenarios; According to the sequence B of the operation sub-scenario, set bi as the target sub-scenario in sequence; Generate a training data package for the target sub-scenario based on historical monitoring data, and build a prediction sub-model for the target sub-scenario based on the training data package; Generate prediction sub-models for each operation sub-scenario in sequence; A demand forecasting model is established based on the identification sub-model and all prediction sub-models.

[0007] In some embodiments of the present application, when constructing the mapping sub-model of each truck collection unit, the following steps are included: According to the truck sequence A, set a i Set trucks for the target; Construct a virtual sub-model based on the equipment parameters of the target set trucks; Establishing a status update timeline for target set trucks, wherein the status update timeline includes update time nodes; Obtain monitoring data packets of target trucks according to the update time node; Generate a mapping sub-model of the target set trucks according to the virtual sub-model and the state update timeline; Generate the mapping sub-models of each truck set in turn.

[0008] In some embodiments of the present application, generating a first-level scheduling policy for a current scheduling period includes: Obtain a feature data packet of the current scheduling period, and generate a first-level operation scenario of the current scheduling period based on the feature data packet; Generate similarity evaluation values ​​for the first-level operation scenario and all operation sub-scenarios, and set the first-level prediction model based on all similarity evaluation values; Generate expected demand parameters for each work sub-area based on the first-level forecast model; Set multiple time intervals within the current scheduling cycle; Establish a time interval sequence T, T=(t1,t2…t i …t r ), where t i is the i-th time interval in the current scheduling cycle; r is the number of time intervals in the current adjustment cycle; According to the time interval sequence T, ti is set as the target time interval in sequence; Set the scheduling sub-strategy within the target time interval; Generate scheduling sub-strategies in each time interval in sequence; Generate the first-level scheduling strategy for the current scheduling cycle based on all scheduling sub-strategies.

[0009] In some embodiments of the present application, setting a scheduling sub-strategy within a target time interval includes: Establish a sequence of working sub-areas D, D=(d1, d2…d i …d s ), where d i is the i-th working sub-area; s is the number of working sub-areas; Set d in sequence according to the number of working sub-areas D i is the target sub-region; Obtain the expected demand parameters of the target sub-area and generate the expected truck demand w for the target sub-area in the target time interval; Update the truck depot based on historical scheduling sub-strategies; Generate the evaluation value of the fit between each truck and the target sub-area in the target time interval based on the update result; Establish a matching evaluation value sequence C, C=(c1, c2…c i …c n ), where c i is the fit evaluation value of the target sub-area of ​​the i-th truck collection in the target time interval; n is the number of truck collection; Generate the initial sub-strategy for the target sub-region based on the fit evaluation sequence C and the expected demand for trucks w; Generate initial sub-strategies for each working sub-area in the target time interval in sequence; Generate a scheduling sub-strategy for the target time interval based on all initial sub-strategies.

[0010] In some embodiments of the present application, establishing the sequence C of fit evaluation values ​​includes: According to the truck sequence A, set a i Set trucks for evaluation; Generate a fit evaluation value c between the target sub-region and the trucks in the evaluation set; c=e1×Q1×[ β i ×j i ]+e2×Q2×[ η i ×v i ]; Wherein, e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; is the number of associated indicators; β i is the impact factor of the i-th correlation index; j i is the reference value of the i-th correlation index between the trucks in the evaluation set and the target sub-area; 2 is the number of equipment evaluation indicators; η i is the influencing factor of the i-th equipment evaluation index generated based on the update results of the truck depot; v iis the reference value of the i-th equipment evaluation index in the truck to be evaluated; Generate the matching evaluation values of each container truck and the target sub-region in sequence.

[0011] In some embodiments of the present application, when determining whether to generate a correction instruction according to the operation deviation value, it includes: Set the end time node of each time interval as the feedback time node according to the time interval sequence T; Obtain the feedback data packet of the current feedback node, and generate the operation deviation value f of the current feedback time node; Preset an operation deviation value threshold F1 and an operation deviation value threshold F2, and F1 < F2; If f < F1, no correction instruction is generated at the current feedback time node; If F1 < f < F2, a first-level correction instruction is generated at the current feedback time node; If f > F2, a second-level correction instruction is generated at the current feedback time node.

[0012] In some embodiments of the present application, when generating the operation deviation value f of the current feedback time node, it includes: f = e3×Q3× g i ×h i +e4×Q4× ɑ i ×k i ; Where, e3 is a preset third weight coefficient; e4 is a preset fourth weight coefficient; Q3 is a preset third fixed coefficient; Q4 is a preset fourth fixed coefficient; s is the number of working sub-regions; g i is the influence factor of the i-th working sub-region; h i is the demand deviation value of the i-th working sub-region generated based on the feedback data packet; θ3 is the number of operation evaluation indexes; ɑ i is the influence factor of the i-th operation evaluation index; k i is the reference value of the i-th operation evaluation index generated based on the feedback data packet.

[0013] In some embodiments of the present application, a container truck scheduling system for a terminal is provided, including: A central control unit, configured to construct a demand prediction model and multiple working sub-regions based on the historical monitoring parameters of the terminal; An equipment unit, configured to construct a mapping sub-model for each container truck, and generate a container truck library according to all the mapping sub-models; The equipment unit is further configured to establish a container truck sequence A, A=(a1, a2…a i …a n ), where, ai is the i-th truck; n is the number of trucks; Monitoring unit, used to collect monitoring data of each truck; The central control unit includes: The first processing module is used to establish a demand forecasting model; The first scheduling module is used to establish multiple scheduling cycles and generate a first-level scheduling strategy for the current scheduling cycle based on the demand forecast model and the truck depot; The second scheduling module is used to generate an operation deviation value of the first-level scheduling strategy according to the preset feedback time node, and determine whether to generate a correction instruction according to the operation deviation value; The establishment of the demand forecasting model includes: Generate multiple scene characteristic indicators based on historical monitoring parameters; Establish identification sub-models and multiple operation sub-scenarios based on all scenario characteristic indicators; Create a sequence of sub-scenario operations B, B = (b1, b2…b i …b m ), where b i is the i-th operation sub-scenario; m is the number of operation sub-scenarios; According to the sequence B of the operation sub-scenario, set bi as the target sub-scenario in sequence; Generate a training data package for the target sub-scenario based on historical monitoring data, and build a prediction sub-model for the target sub-scenario based on the training data package; Generate prediction sub-models for each operation sub-scenario in sequence; A demand forecasting model is established based on the identification sub-model and all prediction sub-models.

[0014] In some embodiments of the present application, the device unit further includes: The first equipment module is used to set a in sequence according to the number of trucks A i Set trucks for the target; Construct a virtual sub-model based on the equipment parameters of the target set trucks; Establishing a status update timeline for target set trucks, wherein the status update timeline includes update time nodes; Obtain monitoring data packets of target trucks according to the update time node; Generate a mapping sub-model of the target set trucks according to the virtual sub-model and the state update timeline; Generate the mapping sub-models of each truck set in turn.

[0015] Compared with the prior art, the terminal container truck dispatching method and system of the present application has the following advantages: Multiple operation sub-scenarios are constructed according to different time periods and operation scenarios, and a prediction sub-model is constructed for each operation sub-scenarios based on big data analysis technology. This allows the demand quantity and time distribution of container trucks in different work sub-areas within a single scheduling cycle to be estimated. The dispatching strategy of container trucks is dynamically adjusted based on the estimated results, thereby improving the dispatching efficiency of container trucks, reducing the empty driving rate, and improving the overall operating efficiency of the container terminal.

[0016] By presetting multiple feedback time nodes, the real-time container truck scheduling situation can be evaluated and adjusted in real time to avoid a large deviation between the initial predicted demand and the actual demand, which would reduce the overall operating efficiency of the terminal. Dynamic and flexible scheduling can be achieved, reducing the waiting time of container trucks and improving operating efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of a terminal truck dispatching method in a preferred embodiment of the present application. DETAILED DESCRIPTION

[0018] The following embodiments are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0019] In the description of this application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on this application.

[0020] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature specified as "first" or "second" may explicitly or implicitly include one or more of such features. Throughout this application, unless otherwise specified, "plurality" means two or more.

[0021] In the description of this application, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they can refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal connections between two components. Those skilled in the art will understand the specific meanings of the above terms in this application based on the specific circumstances.

[0022] like Figure 1 As shown, a terminal truck dispatching method according to a preferred embodiment of the present application includes: S101: Build a demand forecast model and multiple work sub-areas based on the terminal’s historical monitoring parameters; S102: Constructing mapping sub-models for each truck collection unit, and generating a truck collection unit library based on all mapping sub-models; S103: Establish multiple scheduling cycles and generate a first-level scheduling strategy for the current scheduling cycle based on the demand forecast model and the truck depot; S104: Generate an operation deviation value of the first-level scheduling strategy according to the preset feedback time node, and determine whether to generate a correction instruction according to the operation deviation value; Among them, when generating a truck collection warehouse, it includes: Establish a truck sequence A, A=(a1,a2…a i …a n ), where a i is the i-th collection truck; n is the number of collection trucks.

[0023] Specifically, multiple working sub-areas are established based on the yard data of the terminal; Specifically, data mining and machine learning algorithms are used to analyze historical monitoring data at the terminal and construct a demand forecasting model. This demand forecasting model can estimate the number and time distribution of container truck demand in different work sub-areas within a single scheduling cycle.

[0024] Specifically, historical monitoring data includes, but is not limited to, personnel dispatch records, container truck driving trajectories, operation times, container loading and unloading information, and other historical data over past time periods. This data covers the container truck dispatch parameters of the terminal in different time periods and different operation scenarios.

[0025] Specifically, when building a demand forecasting model, it includes: Generate multiple scene characteristic indicators based on historical monitoring parameters; Establish identification sub-models and multiple operation sub-scenarios based on all scenario characteristic indicators; Create a sequence of sub-scenario operations B, B = (b1, b2…b i …b m ), where b i is the i-th operation sub-scenario; m is the number of operation sub-scenarios; According to the sequence B of the operation sub-scenario, set bi as the target sub-scenario in sequence; Generate a training data package for the target sub-scenario based on historical monitoring data, and build a prediction sub-model for the target sub-scenario based on the training data package; Generate prediction sub-models for each operation sub-scenario in sequence; A demand forecasting model is established based on the identification sub-model and all prediction sub-models.

[0026] Specifically, scenario characteristic indicators include but are not limited to time periods, peak and off-seasons of terminal business, operation scenarios and other parameters. Multiple operation sub-scenarios are constructed by quantifying and randomly combining each scenario characteristic indicator.

[0027] Specifically, based on historical monitoring parameters from different operational scenarios, corresponding prediction sub-models are constructed to accurately predict terminal workload and container flow under different time periods and operational scenarios. This provides data support for subsequent container truck scheduling decisions, improving both container truck scheduling efficiency and the overall operational efficiency of the terminal.

[0028] Specifically, when building the mapping sub-model for each truck, it includes: According to the truck sequence A, set a i Set trucks for the target; Construct a virtual sub-model based on the equipment parameters of the target set trucks; Establish a status update timeline for the target set trucks, which includes update time nodes; Obtain monitoring data packets of target trucks according to the update time node; Generate a mapping sub-model of the target set trucks according to the virtual sub-model and the state update timeline; Generate the mapping sub-models of each truck set in turn.

[0029] Specifically, the mapping sub-model can display the load status and operating trajectory of the container truck in real time. At the same time, it can analyze and judge the operating status of the container truck through the monitoring data packets of each update time node, avoiding the long-term operation and long-term emptying of the container truck, and reducing the overall empty driving rate.

[0030] Specifically, each container and container truck is equipped with an RFID tag, and RFID readers are deployed in various areas of the terminal, such as the yard, loading bridge, and road crossings. RFID technology allows real-time location information of containers and container trucks to be obtained, accurately down to the specific yard location and loading bridge operating area, thereby monitoring the real-time location and load status of each container truck.

[0031] In a preferred embodiment of the present application, when generating the first-level scheduling strategy of the current scheduling period, the following steps are included: Obtain a feature data packet of the current scheduling period, and generate a first-level operation scenario of the current scheduling period based on the feature data packet; Generate similarity evaluation values ​​for the first-level operation scenario and all operation sub-scenarios, and set the first-level prediction model based on all similarity evaluation values; Generate expected demand parameters for each work sub-area based on the first-level forecast model; Set multiple time intervals within the current scheduling cycle; Establish a time interval sequence T, T=(t1,t2…t i …t r ), where t i is the i-th time interval in the current scheduling cycle; r is the number of time intervals in the current adjustment cycle; According to the time interval sequence T, ti is set as the target time interval in sequence; Set the scheduling sub-strategy within the target time interval; Generate scheduling sub-strategies in each time interval in sequence; Generate the first-level scheduling strategy for the current scheduling cycle based on all scheduling sub-strategies.

[0032] Specifically, the duration of the time interval can be set based on historical parameters and the first-level prediction model, preferably 1-3 hours. The specific duration can be adjusted based on the confidence level of the first-level prediction model, thereby improving the dispatch efficiency of container trucks.

[0033] Specifically, the feature data packet of the current scheduling cycle is processed by the identification sub-model to generate similarity evaluation values ​​of the first-level operation scenario and each operation sub-scenario. The larger the similarity evaluation value, the more similar the expected operation status in the current scheduling cycle is to the expected operation status of the corresponding operation sub-scenario.

[0034] Specifically, the prediction sub-model of the operation sub-scenario corresponding to the maximum similarity evaluation value is set as the first-level prediction model.

[0035] Specifically, when setting the scheduling sub-strategy within the target time interval, it includes: Establish a sequence of working sub-areas D, D=(d1, d2…d i …d s ), where d i is the i-th working sub-area; s is the number of working sub-areas; Set d in sequence according to the number of working sub-areas D i is the target sub-region; Obtain the expected demand parameters of the target sub-area and generate the expected truck demand w for the target sub-area in the target time interval; Update the truck depot based on historical scheduling sub-strategies; Generate the evaluation value of the fit between each truck and the target sub-area in the target time interval based on the update result; Establish a matching evaluation value sequence C, C=(c1, c2…c i …c n ), where ci is the fit evaluation value of the target sub-area of ​​the i-th truck collection in the target time interval; n is the number of truck collection; Generate the initial sub-strategy for the target sub-region based on the fit evaluation sequence C and the expected demand for trucks w; Generate initial sub-strategies for each working sub-area in the target time interval in sequence; Generate a scheduling sub-strategy for the target time interval based on all initial sub-strategies.

[0036] Specifically, the higher the fit evaluation value, the higher the corresponding dispatch order. Trucks to be dispatched are selected based on the expected demand for trucks in the target sub-region during the target time interval and the dispatch order. This generates the initial sub-strategy for the target sub-region during the target time interval.

[0037] Specifically, a simulation model is performed based on the scheduling sub-strategy of the previous time interval of the target time interval, thereby updating the expected operating status of each truck in the truck depot in the target time interval.

[0038] Specifically, when establishing the fit evaluation value sequence C, it includes: According to the truck sequence A, set a i Set trucks for evaluation; Generate a fit evaluation value c between the target sub-region and the trucks in the evaluation set; c=e1×Q1×[ β i ×j i ]+e2×Q2×[ η i ×v i ]; Wherein, e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; is the number of associated indicators; β i is the impact factor of the i-th correlation index; j i is the reference value of the i-th correlation index between the trucks in the evaluation set and the target sub-area; 2 is the number of equipment evaluation indicators; η i is the influencing factor of the i-th equipment evaluation index generated based on the update results of the truck depot; v i is the reference value of the evaluation index of the i-th equipment in the truck set to be evaluated; Generate the evaluation value of the fit between each truck and the target sub-area in turn.

[0039] Specifically, the larger the fit evaluation value, the more suitable the trucks in the evaluation set are for performing the loading and unloading tasks in the target sub-area within the target time interval.

[0040] Specifically, the equipment evaluation indicators include, but are not limited to, the duration of the empty container truck, the number of tasks executed, whether there are tasks to be executed within the target time interval, the operating status of the container truck, and other parameters. By quantifying each equipment evaluation indicator, real-time analysis is performed on each container truck. η i ×v i The larger the value of [], the more suitable the current container truck is for executing the current loading and unloading task.

[0041] Specifically, the associated indicators include, but are not limited to, the number of tasks executed by the container truck to be evaluated in the target sub-region, the familiarity with the expected driving trajectory, the expected position of the container truck, and other parameters. By quantifying each associated indicator, the suitability of each container truck for the loading and unloading tasks in the target sub-region is judged, reducing the empty load rate of the container truck and improving the transportation efficiency of the container truck at the same time.

[0042] Specifically, by presetting the first fixed coefficient and the second fixed coefficient, all parameters in the model are normalized, so that each parameter in the model is within the same value range.

[0043] It can be understood that in the above embodiments, by constructing the prediction sub-models of each operation sub-scene, the demand quantity and time distribution of container trucks in different working sub-regions within a single scheduling cycle are estimated, and the scheduling strategy of container trucks is dynamically adjusted according to the estimation results, improving the scheduling efficiency of container trucks, reducing the empty running rate, and enhancing the overall operation efficiency of the container terminal.

[0044] In the preferred embodiment of the embodiment of the present application, when judging whether to generate a correction instruction according to the operation deviation value, it includes: Setting the end time node of each time interval as the feedback time node according to the time interval sequence T; Obtaining the feedback data packet of the current feedback node and generating the operation deviation value f of the current feedback time node; Presetting the operation deviation value threshold F1 and the operation deviation value threshold F2, and F1 < F2; If f < F1, no correction instruction is generated at the current feedback time node; If F1 < f < F2, a first-level correction instruction is generated at the current feedback time node; If f > F2, a second-level correction instruction is generated at the current feedback time node.

[0045] Specifically, the feedback data packet contains all the scheduling data and demand data of the time interval corresponding to the current feedback time node.

[0046] Specifically, the first-level correction instruction indicates that the overall terminal operation efficiency is currently low, and it is necessary to optimize the working parameters of the relevant operating systems of the terminal, such as the loading and unloading bridge control system, the yard management system, etc., to improve the efficiency of the coordinated operation of the entire terminal operation process. At the same time, the operation path of each container truck is optimized to reduce driving time and empty mileage, thereby improving the operation effect of the current first-level scheduling strategy.

[0047] Specifically, the second-level correction instruction is that the original first-level scheduling strategy cannot meet the current actual scheduling needs, and the prediction accuracy of its first-level prediction model is poor. The first-level prediction model needs to be optimized and updated, and the expected demand parameters of each working sub-area need to be re-predicted, and a new first-level scheduling strategy needs to be rebuilt to improve the overall operating efficiency of the terminal.

[0048] Specifically, when generating the running deviation value f of the current feedback time node, it includes: f=e3×Q3×[ g i ×h i ]+e4×Q4×[ ɑ i ×k i ]; Among them, e3 is the preset third weight coefficient; e4 is the preset fourth weight coefficient; Q3 is the preset third fixed coefficient; Q4 is the preset fourth fixed coefficient; s is the number of working sub-areas; g i is the impact factor of the i-th working sub-area; h i is the demand deviation value of the i-th working sub-area generated based on the feedback data packet; θ3 is the number of operation evaluation indicators; ɑ i is the influencing factor of the i-th operation evaluation index; k i Generates the reference value of the i-th operation evaluation index based on the feedback data packet.

[0049] Specifically, the demand deviation value can be set based on the difference between the expected demand parameters and the actual demand parameters within the current time interval. The larger the difference, the lower the prediction accuracy of the current first-level forecast model. The impact factor of each working sub-area can be set based on historical monitoring data. The larger the impact factor, the greater the actual demand in the corresponding working sub-area.

[0050] Specifically, operational evaluation metrics include, but are not limited to, average truck mileage, empty load ratio, and the difference between actual and expected transport times for each loading and unloading task. By quantifying these operational metrics, we can analyze and evaluate the overall terminal operating efficiency within the target timeframe.

[0051] Specifically, the larger the operating deviation value, the greater the deviation between the current terminal's overall operating efficiency and the expected efficiency, and the original first-level scheduling strategy needs to be corrected in a timely manner.

[0052] Specifically, all parameters in the model are normalized by presetting the third fixed coefficient and the fourth fixed coefficient, so that each parameter in the model is in the same value range.

[0053] It can be understood that in the above embodiment, by presetting multiple feedback time nodes, the real-time container truck scheduling situation is evaluated and adjusted in real time, avoiding the large deviation between the initial predicted demand and the actual demand, thereby reducing the overall operating efficiency of the terminal, realizing dynamic and flexible scheduling, reducing the waiting time of container trucks, and improving operating efficiency.

[0054] Based on another preferred embodiment of a terminal truck dispatching method in any of the above preferred embodiments, this preferred embodiment provides a terminal truck dispatching method, comprising: The central control unit is used to build a demand forecast model and multiple work sub-areas based on the terminal's historical monitoring parameters; The equipment unit is used to construct mapping sub-models for each truck and generate a truck library based on all mapping sub-models; The equipment unit is also used to establish a truck array A, A=(a1,a2…a i …a n ), where a i is the i-th truck; n is the number of trucks; Monitoring unit, used to collect monitoring data of each truck; The central control unit includes: The first processing module is used to establish a demand forecasting model; The first scheduling module is used to establish multiple scheduling cycles and generate a first-level scheduling strategy for the current scheduling cycle based on the demand forecast model and the truck depot; The second scheduling module is used to generate an operation deviation value of the first-level scheduling strategy according to the preset feedback time node, and determine whether to generate a correction instruction according to the operation deviation value; When building a demand forecasting model, include: Generate multiple scene characteristic indicators based on historical monitoring parameters; Establish identification sub-models and multiple operation sub-scenarios based on all scenario characteristic indicators; Create a sequence of sub-scenario operations B, B = (b1, b2…b i …b m ), where b i is the i-th operation sub-scenario; m is the number of operation sub-scenarios; According to the sequence B of the operation sub-scenario, set bi as the target sub-scenario in sequence; Generate a training data package for the target sub-scenario based on historical monitoring data, and build a prediction sub-model for the target sub-scenario based on the training data package; Generate prediction sub-models for each operation sub-scenario in sequence; A demand forecasting model is established based on the identification sub-model and all prediction sub-models.

[0055] Specifically, the equipment unit also includes: The first equipment module is used to set a in sequence according to the number of trucks A i Set trucks for the target; Construct a virtual sub-model based on the equipment parameters of the target set trucks; Establishing a status update timeline for target set trucks, wherein the status update timeline includes update time nodes; Obtain monitoring data packets of target trucks according to the update time node; Generate a mapping sub-model of the target set trucks according to the virtual sub-model and the state update timeline; Generate the mapping sub-models of each truck set in turn.

[0056] According to the first concept of this application, multiple operation sub-scenarios are constructed according to different time periods and operation scenarios, and a prediction sub-model of each operation sub-scenario is constructed based on big data analysis technology, so as to estimate the demand quantity and time distribution of container trucks in different work sub-areas within a single scheduling cycle, and dynamically adjust the scheduling strategy of container trucks according to the estimation results, thereby improving the scheduling efficiency of container trucks, reducing the empty driving rate, and improving the overall operating efficiency of the container terminal.

[0057] According to the second concept of this application, by presetting multiple feedback time nodes, the real-time container truck scheduling situation is evaluated and adjusted in real time to avoid a large deviation between the initial predicted demand and the actual demand, thereby reducing the overall operating efficiency of the terminal, achieving dynamic and flexible scheduling, reducing the waiting time of container trucks, and improving operating efficiency.

[0058] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and replacements can be made without departing from the technical principles of the present application. These improvements and replacements should also be regarded as the scope of protection of the present application.

Claims

1. A method for dispatching terminal trucks, characterized in that: include: Build demand forecasting models and multiple work sub-areas based on historical monitoring parameters of the terminal; Construct mapping sub-models for each truck collection unit, and generate a truck collection unit library based on all mapping sub-models; Establish multiple scheduling cycles and generate the first-level scheduling strategy for the current scheduling cycle based on the demand forecast model and truck depot; Generate an operation deviation value of the first-level scheduling strategy according to the preset feedback time node, and determine whether to generate a correction instruction according to the operation deviation value; Among them, when generating a truck collection warehouse, it includes: Establish a truck sequence A, A=(a1,a2…a i …a n ), where a i is the i-th truck collection; n is the number of truck collections.

2. The method for dispatching terminal container trucks according to claim 1, characterized in that: When building a demand forecasting model, include: Generate multiple scene characteristic indicators based on historical monitoring parameters; Establish identification sub-models and multiple operation sub-scenarios based on all scenario feature indicators; Establish a sequence of operation sub-scenario numbers B, B=(b1, b2…b i …b m ), where b i is the i-th operation sub-scenario; m is the number of operation sub-scenarios; According to the operation sub-scenario sequence B, set bi as the target sub-scenario in sequence; Generate a training data packet of the target sub-scenario based on the historical monitoring data, and build a prediction sub-model of the target sub-scenario based on the training data packet; Generate prediction sub-models for each operation sub-scenario in turn; A demand forecasting model is established based on the identification sub-model and all forecasting sub-models.

3. The method for dispatching terminal trucks as claimed in claim 2, characterized in that: When building the mapping sub-model for each truck set, include: According to the truck sequence A, set a i Set trucks for the target; Construct a virtual sub-model based on the equipment parameters of the target set trucks; Establishing a status update timeline of the target set of trucks, wherein the status update timeline includes update time nodes; Obtain monitoring data packets of target set trucks according to update time nodes; Generate a mapping sub-model of the target set trucks according to the virtual sub-model and the state update timeline; Generate mapping sub-models for each truck set in turn.

4. The method for dispatching terminal container trucks according to claim 3, characterized in that: When generating the first-level scheduling policy for the current scheduling period, it includes: Obtain a feature data packet of the current scheduling period, and generate a first-level operation scenario of the current scheduling period according to the feature data packet; Generate similarity evaluation values ​​for the primary operation scenario and all operation sub-scenarios, and set the primary prediction model based on all similarity evaluation values; Generate expected demand parameters for each work sub-area based on the first-level forecast model; Set multiple time intervals within the current scheduling cycle; Establish a time interval sequence T, T=(t1, t2…t i …t r ), where t i is the i-th time interval in the current scheduling cycle; r is the number of time intervals in the current adjustment cycle; According to the time interval sequence T, ti is set as the target time interval in sequence; Set the scheduling sub-strategy within the target time interval; Generate scheduling sub-strategies in each time interval in sequence; Generate the first-level scheduling strategy for the current scheduling cycle based on all scheduling sub-strategies.

5. The method for dispatching terminal container trucks according to claim 4, characterized in that: When setting the scheduling sub-strategy within the target time interval, include: Establish a sequence of working sub-areas D, D = (d1, d2…d i …d s ), where d i is the i-th working sub-area; s is the number of working sub-areas; Set d in sequence according to the number of working sub-areas D i is the target sub-region; Obtain the expected demand parameters of the target sub-area and generate the expected demand volume w of the target sub-area for trucks in the target time interval; Update the truck depot based on the historical scheduling sub-strategy; Generate a matching evaluation value between each truck and the target sub-area in the target time interval according to the update result; Establish a matching evaluation value sequence C, C=(c1, c2…c i …c n ), where c i is the fit evaluation value of the target sub-area of ​​the i-th truck collection in the target time interval; n is the number of truck collection; Generate the initial sub-strategy of the target sub-area according to the fit evaluation sequence C and the expected demand of trucks w; Generate initial sub-strategies for each working sub-area in the target time interval in sequence; Generate a scheduling sub-strategy for the target time interval based on all initial sub-strategies.

6. The method for dispatching terminal container trucks according to claim 5, characterized in that: When establishing the matching evaluation value sequence C, it includes: According to the truck sequence A, set a i Set trucks for evaluation; Generate a fit evaluation value c between the target sub-area and the trucks in the evaluation set; c=e1×Q1×[ b i ×j i ]+e2×Q2×[ or i ×v i ]; Among them, e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; is the number of associated indicators; β i is the impact factor of the i-th correlation index; j i is the reference value of the i-th correlation index between the trucks in the evaluation set and the target sub-area; 2 is the number of equipment evaluation indicators; η i is the influencing factor of the i-th equipment evaluation index generated based on the update result of the truck depot; v i is the reference value of the evaluation index of the i-th equipment in the truck set to be evaluated; Generate the fit evaluation value of each truck collection unit and the target sub-area in turn.

7. The method for dispatching terminal container trucks according to claim 5, characterized in that: The step of determining whether to generate a correction instruction according to the running deviation value includes: According to the time interval sequence T, the end time node of each time interval is set as the feedback time node; Obtain the feedback data packet of the current feedback node, and generate the operation deviation value f of the current feedback time node; Preset the operation deviation value threshold F1 and the operation deviation value threshold F2, and F1 < F2; If f < F1, no correction instruction is generated at the current feedback time node; If F1 < f < F2, a first-level correction instruction is generated at the current feedback time node; If f > F2, a second-level correction instruction is generated at the current feedback time node.

8. The method for dispatching terminal trucks as claimed in claim 7, characterized in that: When generating the operation deviation value f of the current feedback time node, it includes: f=e3×Q3×[ g i ×h i ]+e4×Q4×[ ɑ i ×k i ]; Wherein, e3 is the preset third weight coefficient; e4 is the preset fourth weight coefficient; Q3 is the preset third fixed coefficient; Q4 is the preset fourth fixed coefficient; s is the number of working sub-areas; g i is the impact factor of the i-th working sub-area; h i is the demand deviation value of the i-th working sub-area generated based on the feedback data packet; θ3 is the number of operation evaluation indicators; ɑ i is the influencing factor of the i-th operation evaluation index; k i Generates a reference value for the i-th operation evaluation index based on the feedback data packet.

9. A terminal container truck dispatching system, using the terminal container truck dispatching method according to any one of claims 1 to 8, characterized in that: Including: The central control unit is used to construct a demand prediction model and multiple working sub-areas based on the historical monitoring parameters of the terminal; The equipment unit is used to construct a mapping sub-model for each truck, and generate a truck library according to all the mapping sub-models; The equipment unit is also used to establish a truck collection sequence A, A=(a1, a2…a i …a n ), where a i is the i-th truck; n is the number of trucks; The monitoring unit is used to collect the monitoring data of each truck; The central control unit includes: The first processing module is used to establish a demand prediction model; The first scheduling module is used to establish multiple scheduling cycles, and generate the first-level scheduling strategy of the current scheduling cycle according to the demand prediction model and the truck library; The second scheduling module is used to generate the operation deviation value of the first-level scheduling strategy according to the preset feedback time node, and judge whether to generate a correction instruction according to the operation deviation value; When establishing the demand prediction model, it includes: Generate multiple scenario feature indicators according to the historical monitoring parameters; Establish an identification sub-model and multiple operation sub-scenarios according to all the scenario feature indicators; Establish a sequence of operation sub-scenario numbers B, B=(b1, b2…b i …b m ), where b i is the i-th operation sub-scenario; m is the number of operation sub-scenarios; Set bi as the target sub-scenario in turn according to the operation sub-scenario sequence B; Generate the training data packet of the target sub-scenario according to the historical monitoring data, and construct the prediction sub-model of the target sub-scenario according to the training data packet; Generate the prediction sub-models of each operation sub-scenario in turn; Establish a demand prediction model according to the identification sub-model and all the prediction sub-models.

10. The terminal container truck dispatching system according to claim 9, characterized in that: The equipment unit further includes: The first equipment module is used to set a according to the number of trucks A in sequence i Set trucks for the target; Construct a virtual sub-model according to the equipment parameters of the target truck; Establish the status update time axis of the target truck, and the status update time axis includes update time nodes; Obtain the monitoring data packet of the target truck according to the update time node; Generate the mapping sub-model of the target truck according to the virtual sub-model and the status update time axis; Generate the mapping sub-models of each truck in turn.

Images