Port dispatching and anti-congestion management method based on AI algorithm

Abstract

The present invention discloses a port dispatching and anti-congestion management method based on an AI algorithm, which relates to the technical field of intelligent dispatching of logistics vehicles, and comprises the following steps: counting vehicles to be inspected at a port through intelligent sensors to obtain different vehicle types; analyzing the estimated inspection time of different vehicle types based on historical inspection data; predicting congestion events based on the estimated time and the number of vehicles to be inspected in the current port; and dispatching vehicles to be inspected at the port when a congestion event is predicted. The present invention is used to solve the problem that the existing port dispatching technology still has the problem that the dispatching method is not intelligent enough and the allocation of lanes is not intelligent enough, resulting in insufficient traffic efficiency at the port.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent scheduling of logistics vehicles, and specifically to a port scheduling and anti-congestion management method based on AI algorithm. Background Art

[0002] Port dispatch technology refers to the technical means of planning, organizing and controlling the operation of ports through scientific management and optimization methods. Its purpose is to improve the operating efficiency of ports, reduce congestion and shorten waiting time, and provide smoother and more efficient port clearance services.

[0003] Existing intelligent scheduling technologies are usually only used to correct vehicle routes. For example, in a patent application with application publication number CN117151576A, a logistics vehicle scheduling management system and method are disclosed. The intelligent scheduling method can automatically correct the imported vehicle routes, which saves manpower and is not prone to errors. Existing port scheduling technologies usually guide different types of vehicles by pre-setting guide lanes. The pre-setting method will cause the port's throughput rate to decrease when there are fewer vehicles. If the number of vehicles is lower than a certain base number, the pre-set guide lanes will limit the vehicle's traffic efficiency. Existing port scheduling technologies usually have fixed lane allocations. When the number of vehicles is the same, different number ratios and the time required for vehicle inspection will become the key to affecting traffic efficiency. Therefore, it is necessary to dynamically balance the allocation of lanes to ensure the port's traffic efficiency. Existing port scheduling technologies also have problems such as insufficient intelligence in scheduling methods and insufficient intelligence in lane allocation, resulting in insufficient traffic efficiency at the port. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems in the prior art to a certain extent. By using intelligent sensors to count vehicles to be inspected at the port, different vehicle types are obtained. Then, based on historical inspection data, the estimated inspection time for different vehicle types is analyzed. Congestion events are predicted by the estimated time and the number of vehicles to be inspected in the current port. When congestion events are predicted, the vehicles to be inspected at the port are dispatched. By calculating the vehicle model priorities of different vehicle types, more lanes are allocated to vehicle types with high vehicle model priorities, so as to solve the problem that the existing port dispatching technology still has the scheduling method that is not intelligent enough and the lane allocation is not intelligent enough, resulting in insufficient traffic efficiency at the port.

[0005] To achieve the above objectives, in the first aspect, the present application provides a port dispatching anti-congestion management method based on an AI algorithm, comprising the following steps:

[0006] The intelligent sensors are used to count the vehicles to be inspected at the port and obtain different vehicle types;

[0007] Estimated inspection times for different vehicle types based on historical inspection data;

[0008] Predict congestion events based on the estimated duration and the number of vehicles currently waiting for inspection at the port;

[0009] When congestion is predicted, vehicles awaiting inspection at the port will be dispatched.

[0010] Furthermore, the vehicle types include small trucks, medium trucks, large trucks and non-trucks; and the smart sensor is a high-definition camera.

[0011] Furthermore, the vehicles to be inspected at the port are counted by intelligent sensors, and the vehicles to be inspected are classified and counted by intelligent image recognition to obtain the number of different vehicle types, including small vehicles, medium vehicles, large vehicles and non-freight vehicles.

[0012] Further, analyzing the estimated inspection time of different vehicle types based on historical inspection data includes the following sub-steps:

[0013] The historical inspection data is grouped based on vehicle type and historical inspection duration to obtain inspection duration groups, and the number of different historical inspection durations in the inspection duration groups is counted to obtain group counts;

[0014] Set up an estimated duration analysis plan, further analyze the inspection duration groups and group counts, and obtain the estimated duration corresponding to the vehicle type;

[0015] Different vehicle types are analyzed through the estimated duration analysis solution to obtain the estimated duration of different vehicle types.

[0016] Furthermore, grouping the historical inspection data based on vehicle type and historical inspection duration includes the following sub-steps:

[0017] Connecting to the port's historical traffic database to obtain historical inspection data, including historical inspection duration and historical vehicle types;

[0018] For any vehicle type, obtain the maximum and minimum values ​​of the historical inspection duration, which are marked as the maximum inspection duration and the minimum inspection duration, respectively, and are represented by Tmax and Tmin;

[0019] Calculate the maximum inspection time minus the minimum inspection time, mark the calculation result as the time span, set the first grouping number, divide the time span by the first grouping number, calculate the time gradient, and represent it by the symbol t;

[0020] Set the inspection time of the first group number, by symbol S nRepresents, where 1≤n≤the first grouping number, and n is a positive integer;

[0021] S n The corresponding inspection duration range is [Tmin+(n-1)×t,Tmin+n×t], and the statistics are in S n The number of historical inspection durations in the group is marked as group count. If the historical inspection duration is exactly at the boundary of two inspection duration groups, it is recorded once in both inspection duration groups.

[0022] Furthermore, the estimated duration analysis scheme includes the following sub-steps:

[0023] A histogram is constructed with inspection duration groups as the X-axis and group counts as the Y-axis. The histogram is named duration count histogram. The inspection duration groups and corresponding group counts are entered into the duration count histogram.

[0024] Narrow the inspection duration range of the inspection duration grouping, double the number of inspection duration groups, i.e. double the number of base bars in the duration count bar chart;

[0025] Continue to reduce the inspection duration range until the top of the benchmark bar in the duration count bar chart can smoothly transition to form a curve, then stop reducing it and name the curve the duration count line;

[0026] Get the peak in the duration count line and name it as the curve peak;

[0027] The inspection time range corresponding to the peak of the curve is named as the range to be calculated, and the upper limit and lower limit of the range to be calculated are named as the upper limit of the range and the lower limit of the range respectively;

[0028] The upper limit of the range is summed and averaged to obtain the average upper limit; the lower limit of the range is summed and averaged to obtain the average lower limit;

[0029] Add the average upper limit to the average lower limit and take the average to calculate the estimated duration.

[0030] Furthermore, analyzing different vehicle types through the estimated duration analysis scheme includes the following sub-steps:

[0031] Analyze small trucks through the estimated duration analysis solution to obtain the estimated duration of small trucks;

[0032] Analyze medium-sized trucks through the estimated duration analysis solution to obtain the estimated duration of medium-sized trucks;

[0033] Analyze large trucks through the estimated duration analysis solution to obtain the estimated duration of large trucks;

[0034] Non-trucks are analyzed through the estimated duration analysis solution to obtain the non-truck estimated duration.

[0035] Furthermore, predicting the congestion event based on the estimated duration and the number of vehicles to be inspected at the current port includes the following sub-steps:

[0036] Get the number of small vehicles, medium vehicles, large vehicles, and non-freight vehicles;

[0037] Calculate the number of small vehicles and multiply it by the estimated time for small cargo to get the first estimated time; calculate the number of medium vehicles and multiply it by the estimated time for medium cargo to get the second estimated time; calculate the number of large vehicles and multiply it by the estimated time for large cargo to get the third estimated time; calculate the number of non-cargo vehicles and multiply it by the estimated time for non-cargo to get the fourth estimated time;

[0038] Add the first estimated duration, the second estimated duration, the third estimated duration, and the fourth estimated duration to obtain an estimated total duration;

[0039] Obtain the standard traffic capacity of the port, which is the number of vehicles that can pass through the port per hour;

[0040] Calculate the standard transit time at the port through the standard transit time calculation formula;

[0041] The standard travel time is compared with the estimated total time. If the standard travel time is less than the estimated total time, a congestion risk signal is output; if the standard travel time is greater than or equal to the estimated total time, a congestion risk-free signal is output.

[0042] Furthermore, when a congestion event is predicted, the scheduling of vehicles to be inspected at the port includes the following sub-steps:

[0043] When a congestion risk signal is output, the vehicle to be inspected is guided by lane guidance lights;

[0044] Obtain the number of port lanes at the port, obtain the number of small vehicles, medium vehicles, large vehicles and non-freight vehicles and collectively refer to them as the number of vehicle types;

[0045] The number of vehicle models is compared with a first number threshold, and if the number of vehicle models is less than or equal to the first number threshold, a signal indicating that the number of vehicles is small is output; if the number of vehicle models is greater than the first number threshold, a signal indicating that the number of vehicles is large is output;

[0046] If the output signal is for fewer vehicle types, there is no need to guide the corresponding vehicle type; if the output signal is for more vehicle types, the number of lanes required for different vehicle types is analyzed based on the number of lanes at the port and the number of vehicle types.

[0047] Furthermore, analyzing the number of lanes required for different vehicle types based on the number of port lanes and the number of vehicle types includes the following sub-steps:

[0048] Small trucks and non-trucks are collectively referred to as small vehicles; the number of signals with more vehicle types is counted and marked as the number to be guided, and the corresponding vehicle type is named the vehicle type to be guided;

[0049] The vehicle type priority of the vehicle type with more vehicle type signals is calculated by the vehicle type priority calculation formula;

[0050] If the number of vehicles to be guided is greater than the number of lanes at the port, small vehicles will be assigned to one lane, and medium trucks and large trucks will each be assigned to one lane;

[0051] If the number of vehicles to be guided is equal to the number of lanes at the port, a lane is allocated for each type of vehicle to be guided;

[0052] If the number of vehicles to be guided is less than the number of lanes at the port, a lane will be allocated to each vehicle type to be guided in descending order of vehicle type priority, and the remaining lanes will be allocated again in descending order of vehicle type priority until all lanes are allocated.

[0053] Beneficial effects of the present invention: The present invention uses intelligent sensors to count vehicles to be inspected at ports, obtains different vehicle types, and then analyzes the estimated inspection time of different vehicle types based on historical inspection data. The advantage is that different vehicle types require different inspection times. Generally, larger vehicles require longer inspection times. Therefore, vehicles are divided into different vehicle types, and then their estimated inspection times are analyzed based on vehicle types. This can obtain the estimated inspection times of different vehicles, providing a reliable data basis for subsequent port congestion predictions and improving the reliability and rationality of port anti-congestion scheduling.

[0054] The present invention predicts congestion events by estimating the duration and the number of vehicles to be inspected at the current port, and dispatches the vehicles to be inspected at the port when a congestion event is predicted. The advantage is that when the traffic volume at the port is not large, allowing vehicles to freely choose lanes can improve vehicle traffic efficiency, while when the traffic volume is large, free choice will reduce traffic efficiency, so it is necessary to regulate vehicles when the traffic volume is large, and different types of vehicles have different effects on traffic efficiency, so they need to be evaluated by their estimated duration, thereby improving the effectiveness and rationality of port anti-congestion dispatching;

[0055] The present invention calculates the model priorities of different vehicle types and allocates more lanes to vehicle types with higher model priorities. The advantage is that, generally, the more vehicle types a vehicle has, the more lanes it requires, and the longer the inspection time a vehicle takes, the more lanes it requires. A dynamic balance is required between them, so that lanes can be allocated to vehicles of different numbers and types, further improving the effectiveness and rationality of port anti-congestion scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is a flow chart of the steps of the method of the present invention;

[0057] Figure 2 is a schematic diagram of a duration counting bar graph of the present invention;

[0058] Figure 3 is a schematic diagram of a duration counting curve of the present invention;

[0059] Figure 4 Schematic diagram of a lane guide light of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0061] Example 1, please refer to Figure 1 As shown, in the first aspect, the present application provides a port dispatching anti-congestion management method based on an AI algorithm, comprising the following steps:

[0062] Step S1, counting the vehicles to be inspected at the port through the intelligent sensor to obtain different vehicle types; the vehicle types include small trucks, medium trucks, large trucks and non-trucks; the intelligent sensor is a high-definition camera; counting the vehicles to be inspected at the port through the intelligent sensor is to classify and count the vehicles to be inspected through intelligent image recognition to obtain the number of different vehicle types, including the number of small vehicles, the number of medium vehicles, the number of large vehicles and the number of non-freight vehicles;

[0063] In the specific implementation, the evaluation standards for small trucks, medium trucks and large trucks are the same as the existing standards; the high-definition camera uses the existing high-definition camera, and the intelligent image recognition uses the existing image recognition technology to classify and count the vehicles to be inspected.

[0064] Step S2, analyzing the estimated duration of inspections for different vehicle types based on historical inspection data; Step S2 includes the following sub-steps:

[0065] Step S201, grouping historical inspection data based on vehicle type and historical inspection duration to obtain inspection duration groups, and counting the number of different historical inspection durations in the inspection duration groups to obtain group counts;

[0066] Step S201 includes the following sub-steps:

[0067] Step S2011, connect with the port historical traffic database to obtain historical inspection data, which includes historical inspection time and historical vehicle type;

[0068] In specific implementation, the port historical passage database is provided by the port administrator; some of the historical inspection data are shown in Table 1 below:

[0069] Table 1 Part of the historical inspection data

[0070]

[0071] The inspection date, entry time and exit time in Table 1 are necessary data for historical inspection data and are irrelevant to this embodiment, which only considers the historical inspection duration and vehicle type;

[0072] Step S2012, for any vehicle type, obtain the maximum value and the minimum value in the historical inspection time, which are marked as the maximum inspection time and the minimum inspection time, respectively, and are represented by Tmax and Tmin;

[0073] Step S2013, calculating the maximum inspection duration minus the minimum inspection duration, marking the calculation result as the duration span, setting the first grouping number, dividing the duration span by the first grouping number, and calculating the duration gradient, which is represented by the symbol t;

[0074] Step S2014, set the inspection time length group of the first group number, by symbol S n Represents, where 1≤n≤the first grouping number, and n is a positive integer;

[0075] Step S2015, S n The corresponding inspection duration range is [Tmin+(n-1)×t,Tmin+n×t], and the statistics are in S n The number of historical inspection durations in the group is marked as group count. If the historical inspection duration is exactly at the boundary of two inspection duration groups, it is recorded once in both inspection duration groups.

[0076] In the specific implementation, this embodiment analyzes a small truck to illustrate the specific process of the analysis of the vehicle type. The analysis process of other vehicle types is the same as that of the small truck, so it is omitted in this embodiment; Tmax and Tmin are obtained as 302s and 186s respectively, and the time span is calculated to be 116s. The first group number has no practical meaning. It is only for setting the initial state of the time count bar graph. The value is arbitrary. In this embodiment, the first group number is set to 10, and the time map t=11.6s is calculated; 10 inspection time groups S are set 1 To S 10 , Sn The corresponding inspection duration range is [186+(n-1)×11.6,186+n×11.6], and the statistics show that S 1 To S 10 The group counts are 896, 732, 963, 1028, 1259, 1433, 1368, 1197, 1034, and 875;

[0077] Step S202, setting an estimated duration analysis scheme, further analyzing the inspection duration groups and group counts, and obtaining an estimated duration corresponding to the vehicle type;

[0078] Step S202 includes the following sub-steps:

[0079] See also Figure 2 As shown, in step S2021, a bar chart is constructed with the inspection duration group as the X-axis and the group count as the Y-axis, named as the duration count bar chart, and the inspection duration group and the corresponding group count are entered into the duration count bar chart;

[0080] Step S2022, narrowing the inspection duration range of the inspection duration group, doubling the number of inspection duration groups, that is, doubling the number of reference bars in the duration counting bar chart;

[0081] See also Figure 3 As shown, step S2023, continuously narrowing the inspection duration range until the top of the reference column in the duration counting bar graph can smoothly transition to form a curve and then stop narrowing, and the curve is named the duration counting line;

[0082] Step S2024, obtaining the peak in the duration count line and naming it as the curve peak;

[0083] Step S2025, naming the inspection time range corresponding to the peak of the curve as the to-be-calculated range, and naming the upper limit and the lower limit of the to-be-calculated range as the range upper limit and the range lower limit respectively;

[0084] Step S2026, summing up the upper limit of the range and taking the average value to calculate the average upper limit; summing up the lower limit of the range and taking the average value to calculate the average lower limit;

[0085] Step S2027, adding the average upper limit and the average lower limit and taking the average value to calculate the estimated duration;

[0086] In the specific implementation, the duration count bar chart is constructed as follows Figure 2 As shown, the processing time count line is as follows Figure 3As shown; the calculated ranges corresponding to the peaks of the curve include [192.96, 194.12], [212.68, 213.84], [238.2, 239.36], [249.8, 250.96], [262.56, 263.72] and [283.44, 284.6]. The average upper limit is 241.1 and the average lower limit is 239.94. The estimated duration is 241 after further calculation. The calculation result retains integers.

[0087] Step S203, analyzing different vehicle types through an estimated duration analysis scheme to obtain estimated durations for different vehicle types;

[0088] Step S203 includes the following sub-steps:

[0089] Step S2031, analyzing the small truck using the estimated time analysis solution to obtain the estimated time of the small truck;

[0090] Step S2032, analyzing the medium-sized truck using the estimated duration analysis solution to obtain the estimated duration of the medium-sized truck;

[0091] Step S2033, analyzing the large truck using the estimated time analysis solution to obtain the estimated time for the large truck;

[0092] Step S2034, analyzing the non-truck using the estimated duration analysis solution to obtain the non-truck estimated duration;

[0093] In the specific implementation, the analysis shows that the estimated time for small goods is 241s, the estimated time for medium goods is 263s, the estimated time for large goods is 332s, and the estimated time for non-goods is 48s.

[0094] Step S3, predicting the congestion event based on the estimated duration and the number of vehicles to be inspected in the current port; Step S3 includes the following sub-steps:

[0095] Step S301, obtaining the number of small vehicles, medium vehicles, large vehicles and non-freight vehicles;

[0096] Step S302, calculate the number of small vehicles multiplied by the estimated time for small cargo, to obtain a first estimated time; calculate the number of medium vehicles multiplied by the estimated time for medium cargo, to obtain a second estimated time; calculate the number of large vehicles multiplied by the estimated time for large cargo, to obtain a third estimated time; calculate the number of non-cargo vehicles multiplied by the estimated time for non-cargo, to obtain a fourth estimated time;

[0097] Step S303, adding the first estimated duration, the second estimated duration, the third estimated duration and the fourth estimated duration to obtain an estimated total duration;

[0098] In the specific implementation, the number of small vehicles is 7, the number of medium vehicles is 4, the number of large vehicles is 6, and the number of non-freight vehicles is 4. The first estimated duration is calculated to be 1678s, the second estimated duration is 1052, the third estimated duration is 1992, and the fourth estimated duration is 192. The total estimated duration is 4919s.

[0099] Step S304, obtaining the standard traffic capacity of the port, the standard traffic capacity is the number of vehicles that can pass through the port per hour, and obtaining the number of port lanes of the port;

[0100] Step S305, calculating the standard transit time of the port by using the standard transit time calculation formula;

[0101] Step S306, comparing the standard travel time with the estimated total time, if the standard travel time is less than the estimated total time, outputting a congestion risk signal; if the standard travel time is greater than or equal to the estimated total time, outputting a no congestion risk signal;

[0102] In the specific implementation, the standard traffic capacity is 60, the number of port lanes is 4, and the standard traffic time calculation formula is configured as: Among them, STT is the standard travel time, NOL is the number of port lanes, STC is the standard travel capacity, SV is the number of small vehicles, MSV is the number of medium-sized vehicles, LV is the number of large vehicles, FV is the number of non-freight vehicles, and the constant 3600 is the value of one hour converted to seconds; substituting it into the calculation, the standard travel time is 5040. Through comparison, it is found that the standard travel time is greater than the estimated total time, and a signal of no congestion risk is output.

[0103] Step S4, dispatching vehicles to be inspected at the port when a congestion event is predicted; Step S4 includes the following sub-steps:

[0104] Step S401, when a congestion risk signal is output, the vehicle to be inspected is guided by lane guide lights;

[0105] Step S402, obtaining the number of port lanes at the port, obtaining the number of small vehicles, the number of medium vehicles, the number of large vehicles and the number of non-freight vehicles and collectively referring to the number of vehicle types;

[0106] Step S403, comparing the number of vehicle models with a first number threshold, if the number of vehicle models is less than or equal to the first number threshold, outputting a signal indicating that the number of vehicles is small; if the number of vehicle models is greater than the first number threshold, outputting a signal indicating that the number of vehicles is large;

[0107] Step S404: if the output signal is that there are fewer vehicle types, there is no need to guide the corresponding vehicle type; if the output signal is that there are more vehicle types, the number of lanes required for different vehicle types is analyzed based on the number of port lanes and the number of vehicle types;

[0108] Step S404 includes the following sub-steps:

[0109] Step S4041, small trucks and non-trucks are collectively referred to as small vehicles; the number of signals with more vehicle types is counted and marked as the number to be guided, and the corresponding vehicle type is named the vehicle type to be guided;

[0110] Step S4042, calculating the vehicle type priority of the vehicle type that outputs more vehicle type signals by using the vehicle type priority calculation formula, and combining the small truck and non-truck for calculation;

[0111] Step S4043: If the number of vehicles to be guided is greater than the number of lanes at the port, a small vehicle is assigned to one lane, and a medium truck and a large truck are each assigned to one lane;

[0112] Step S4044, if the number of vehicles to be guided is equal to the number of lanes at the port, a lane is allocated to each type of vehicle to be guided;

[0113] Step S4045: if the number of vehicles to be guided is less than the number of lanes at the port, a lane is allocated to each vehicle type to be guided in descending order of vehicle type priority, and the remaining lanes are allocated again in descending order of vehicle type priority until all lanes are allocated;

[0114] In the specific implementation, since no congestion risk signal is output in this embodiment, step S4 is not executed. For the specific explanation of step S4, please refer to embodiment 2.

[0115] Example 2: This example differs from Example 1 in that a no-congestion risk signal is output in Example 1, so step S4 is no longer executed. This example is a supplement to the specific execution process of step S4. In this example, the number of small vehicles is 12, the number of medium-sized vehicles is 24, the number of large vehicles is 20, and the number of non-freight vehicles is 8. The calculated estimated total duration is 16228, and the standard travel duration is 15360. By comparison, the standard travel duration is less than the estimated total duration, and a congestion risk signal is output;

[0116] Step S4, dispatching vehicles to be inspected at the port when a congestion event is predicted; Step S4 includes the following sub-steps:

[0117] See also Figure 4 As shown, in step S401, when a congestion risk signal is output, the vehicle to be inspected is guided by a lane guide light;

[0118] Step S402, obtaining the number of port lanes at the port, obtaining the number of small vehicles, medium vehicles, large vehicles and non-freight vehicles and collectively referring to the number of vehicle types;

[0119] Step S403, comparing the number of vehicle models with a first number threshold, if the number of vehicle models is less than or equal to the first number threshold, outputting a signal indicating that the number of vehicles is small; if the number of vehicle models is greater than the first number threshold, outputting a signal indicating that the number of vehicles is large;

[0120] In a specific implementation, the lane guide lights are as follows: Figure 4 As shown, above the arrow in the figure is an LED screen, and the content therein can be changed at any time; the number of small vehicles is 12, the number of medium-sized vehicles is 24, the number of large vehicles is 20, and the number of non-freight vehicles is 8; the first quantity threshold is set to 5. When the number of vehicle types is less than the first quantity threshold, the impact on the passage of the port is small, so there is no need to guide them. During the calculation, since the non-freight vehicles have a short passing time and basically have no impact, and there are fewer non-freight vehicles passing through the port, non-freight vehicles are combined with small trucks for reference calculation; by comparison, it is found that the number of small trucks, medium trucks and large trucks is greater than the first quantity threshold, and a signal of a large number of vehicles is output;

[0121] Step S404: if the output signal is that there are fewer vehicle types, there is no need to guide the corresponding vehicle type; if the output signal is that there are more vehicle types, the number of lanes required for different vehicle types is analyzed based on the number of port lanes and the number of vehicle types;

[0122] Step S404 includes the following sub-steps:

[0123] Step S4041, small trucks and non-trucks are collectively referred to as small vehicles; the number of signals with more vehicle types is counted and marked as the number to be guided, and the corresponding vehicle type is named the vehicle type to be guided;

[0124] Step S4042, calculating the vehicle type priority of the vehicle type that outputs more vehicle type signals by using the vehicle type priority calculation formula;

[0125] In the specific implementation, the number of vehicles to be guided is 3, and the vehicles to be guided include small vehicles to be guided, medium vehicles to be guided, and large vehicles to be guided, which correspond to small cars, medium trucks, and large trucks respectively; the vehicle model priority calculation formula is configured as: Among them, MP is the vehicle model priority, MN is the number of vehicle models, ED is the estimated duration, and α is a constant of 100. The setting of α is to make the calculation result more intuitive and has no practical significance. The small priority is 4.98, the medium priority is 9.13, and the large priority is 6.02. The calculation result is rounded to two decimal places.

[0126] Step S4043: If the number of vehicles to be guided is greater than the number of lanes at the port, a small vehicle is assigned to one lane, and a medium truck and a large truck are each assigned to one lane;

[0127] Step S4044, if the number of vehicles to be guided is equal to the number of lanes at the port, a lane is allocated to each type of vehicle to be guided;

[0128] Step S4045: if the number of vehicles to be guided is less than the number of lanes at the port, a lane is allocated to each vehicle type to be guided in descending order of vehicle type priority, and the remaining lanes are allocated again in descending order of vehicle type priority until all lanes are allocated;

[0129] In the specific implementation, the number of lanes at the port is 4, and the number of vehicles to be guided is 3. Through comparison, it is found that the number of vehicles to be guided is less than the number of lanes at the port. Therefore, a lane is allocated to each type of vehicle to be guided in order of vehicle type priority from large to small, that is, one lane is allocated to medium-sized trucks, one lane is allocated to large trucks, one lane is allocated to small cars, and then one lane is allocated to medium trucks. At this time, the lane allocation is completed, and the vehicles to be inspected are guided by lane guide lights. When guiding, large trucks are guided to the right, medium trucks are guided to the middle, and small cars are guided to the left.

[0130] Embodiment 3, in the third aspect, the present application provides an electronic device, including a processor and a memory, the memory storing computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in the above method are executed. Through the above technical solution, the processor and the memory are interconnected and communicate with each other through a communication bus and / or other forms of connection mechanisms, and the memory stores a computer program executable by the processor. When the electronic device is running, the processor executes the computer program to execute the method in any optional implementation of the above embodiment to achieve the following functions: counting the vehicles to be inspected at the port through intelligent sensors to obtain different vehicle types; analyzing the estimated time for inspecting different vehicle types based on historical inspection data; predicting congestion events based on the estimated time and the number of vehicles to be inspected in the current port; and scheduling the vehicles to be inspected at the port when a congestion event is predicted.

[0131] Embodiment 4, in a fourth aspect, the present application provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method are executed. Through the above technical solution, when the computer program is executed by the processor, the method in any optional implementation of the above embodiment is executed to achieve the following functions: counting the vehicles to be inspected at the port through intelligent sensors to obtain different vehicle types; analyzing the estimated inspection time of different vehicle types based on historical inspection data; predicting congestion events based on the estimated time and the number of vehicles to be inspected in the current port; and scheduling the vehicles to be inspected at the port when a congestion event is predicted.

[0132] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0133] It should be understood by those skilled in the art that the embodiments of the present invention can be provided as methods, systems or computer program products. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program codes. Among them, the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Read-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, disk or optical disk. These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0134] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

Claims

1. The port dispatching anti-congestion management method based on AI algorithm is characterized by: The steps include: The intelligent sensors are used to count the vehicles to be inspected at the port and obtain different vehicle types; Estimated inspection times for different vehicle types based on historical inspection data; Predict congestion events based on the estimated duration and the number of vehicles currently waiting for inspection at the port; Dispatch vehicles waiting for inspection at the port when congestion is predicted; The estimated inspection time for different vehicle types based on historical inspection data includes the following sub-steps: The historical inspection data is grouped based on vehicle type and historical inspection duration to obtain inspection duration groups, and the number of different historical inspection durations in the inspection duration groups is counted to obtain group counts; Set up an estimated duration analysis plan, further analyze the inspection duration groups and group counts, and obtain the estimated duration corresponding to the vehicle type; Analyze different vehicle types through the estimated duration analysis solution to obtain the estimated duration of different vehicle types; Grouping historical inspection data based on vehicle type and historical inspection duration includes the following sub-steps: Connecting to the port's historical traffic database to obtain historical inspection data, including historical inspection duration and historical vehicle types; For any vehicle type, obtain the maximum and minimum values ​​of the historical inspection duration, which are marked as the maximum inspection duration and the minimum inspection duration, respectively, and are represented by Tmax and Tmin; Calculate the maximum inspection time minus the minimum inspection time, mark the calculation result as the time span, set the first grouping number, divide the time span by the first grouping number, calculate the time gradient, and represent it by the symbol t; Set the inspection time of the first group number, by symbol S n Represents, where 1≤n≤the first grouping number, and n is a positive integer; S n The corresponding inspection duration range is [Tmin+(n-1)×t,Tmin+n×t], and the statistics are in S n The number of historical inspection durations in the group is marked as group count. If the historical inspection duration is exactly at the boundary of two inspection duration groups, it is recorded once in both inspection duration groups. The estimated duration analysis solution includes the following sub-steps: A histogram is constructed with inspection duration groups as the X-axis and group counts as the Y-axis. The histogram is named duration count histogram. The inspection duration groups and corresponding group counts are entered into the duration count histogram. Narrow the inspection duration range of the inspection duration grouping, double the number of inspection duration groups, i.e. double the number of base bars in the duration count bar chart; Continue to reduce the inspection duration range until the top of the benchmark column in the duration count bar chart can smoothly transition to form a curve, then stop reducing it and name the curve the duration count line; Get the peak in the duration count line and name it as the curve peak; The inspection time range corresponding to the peak of the curve is named as the range to be calculated, and the upper limit and lower limit of the range to be calculated are named as the upper limit of the range and the lower limit of the range respectively; The upper limit of the range is summed and averaged to obtain the average upper limit; the lower limit of the range is summed and averaged to obtain the average lower limit; Add the average upper limit to the average lower limit and take the average to calculate the estimated duration.

2. The port dispatching anti-congestion management method based on AI algorithm according to claim 1 is characterized in that: The vehicle types include small trucks, medium trucks, large trucks and non-trucks; the smart sensor is a high-definition camera.

3. The port dispatching anti-congestion management method based on AI algorithm according to claim 2 is characterized in that: The vehicles to be inspected at the port are counted by intelligent sensors. The vehicles to be inspected are classified and counted through intelligent image recognition to obtain the number of different vehicle types, including small vehicles, medium vehicles, large vehicles and non-freight vehicles.

4. The port dispatching anti-congestion management method based on AI algorithm according to claim 3 is characterized in that: Analyzing different vehicle types through the estimated duration analysis solution includes the following sub-steps: Analyze small trucks through the estimated duration analysis solution to obtain the estimated duration of small trucks; Analyze medium-sized trucks through the estimated duration analysis solution to obtain the estimated duration of medium-sized trucks; Analyze large trucks through the estimated duration analysis solution to obtain the estimated duration of large trucks; Non-trucks are analyzed through the estimated duration analysis solution to obtain the non-truck estimated duration.

5. The port dispatching anti-congestion management method based on AI algorithm according to claim 4 is characterized in that: Predicting congestion events based on the estimated duration and the number of vehicles currently waiting for inspection at the port includes the following sub-steps: Get the number of small vehicles, medium vehicles, large vehicles, and non-freight vehicles; Calculate the number of small vehicles and the estimated time for small trucks to get the first estimated time; Calculate the number of medium-sized vehicles and multiply the estimated duration of the medium-sized truck to get the second estimated duration; Calculate the number of large vehicles and the estimated duration of large cargo to get the third estimated duration; Calculate the number of non-freight vehicles and multiply the non-freight estimated duration to obtain the fourth estimated duration; Add the first estimated duration, the second estimated duration, the third estimated duration, and the fourth estimated duration to obtain an estimated total duration; Obtain the standard traffic capacity of the port, which is the number of vehicles that can pass through the port per hour; Calculate the standard transit time at the port through the standard transit time calculation formula; The standard travel time is compared with the estimated total time. If the standard travel time is less than the estimated total time, a congestion risk signal is output; if the standard travel time is greater than or equal to the estimated total time, a congestion risk-free signal is output.

6. The port dispatching anti-congestion management method based on AI algorithm according to claim 5 is characterized in that: When a congestion event is predicted, the dispatch of vehicles to be inspected at the port includes the following sub-steps: When a congestion risk signal is output, the vehicle to be inspected is guided by lane guidance lights; Obtain the number of port lanes at the port, obtain the number of small vehicles, medium vehicles, large vehicles and non-freight vehicles and collectively refer to them as the number of vehicle types; The number of vehicle models is compared with a first number threshold, and if the number of vehicle models is less than or equal to the first number threshold, a signal indicating that the number of vehicles is small is output; if the number of vehicle models is greater than the first number threshold, a signal indicating that the number of vehicles is large is output; If the output vehicle type has fewer signals, there is no need to guide the corresponding vehicle type; If the output signals have more vehicle types, the number of lanes required for different vehicle types is analyzed based on the number of lanes at the port and the number of vehicle types.

7. The port dispatching anti-congestion management method based on AI algorithm according to claim 6 is characterized in that: Analyzing the number of lanes required for different vehicle types based on the number of lanes at the port and the number of vehicle types includes the following sub-steps: Small trucks and non-trucks are collectively referred to as small vehicles; the number of signals with more vehicle types is counted and marked as the number to be guided, and the corresponding vehicle type is named the vehicle type to be guided; The vehicle type priority of the vehicle type with more vehicle type signals is calculated by the vehicle type priority calculation formula; If the number of vehicles to be guided is greater than the number of lanes at the port, small vehicles will be assigned to one lane, and medium trucks and large trucks will each be assigned to one lane; If the number of vehicles to be guided is equal to the number of lanes at the port, a lane is allocated for each type of vehicle to be guided; If the number of vehicles to be guided is less than the number of lanes at the port, a lane will be allocated to each vehicle type to be guided in descending order of vehicle type priority, and the remaining lanes will be allocated again in descending order of vehicle type priority until all lanes are allocated.

Images