Cold-chain logistics cargo transportation path planning method based on AI processing

Abstract

The invention discloses a cold-chain logistics cargo transportation path planning method based on AI processing, and belongs to the technical field of path planning, and the method comprises the following steps: S1, when the temperature in a box of a cold-chain transportation vehicle is abnormal, determining a standby area according to the current position and the final position of the cold-chain transportation vehicle; s2, screening a plannable grid in the standby area; s3, generating a current traffic flow depth and a relative traffic flow depth according to the vehicle condition of the plannable grid, and constructing a discriminator; and S4, using a discriminator to extract the plannable grid with the smooth driving state from the time when the temperature in the box is abnormal to the predicted arrival time, and generating a transportation path. According to the method, the transportation efficiency and the cargo safety of cold-chain logistics can be remarkably improved, the operation cost can be effectively reduced, and the overall service level and competitiveness of the industry are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of path planning, and specifically relates to a cold chain logistics cargo transportation path planning method based on AI processing. Background Art

[0002] With the rapid development of the cold chain logistics industry, ensuring the quality and safety of goods during transportation has become a core concern in the industry. Cold chain logistics specifically refers to a series of logistics activities that transport, store and distribute fresh goods that require a specific temperature environment under low temperature conditions. Such goods are extremely sensitive to temperature conditions, and any temperature fluctuations may cause the goods to deteriorate or be damaged, which in turn affects their market value and consumer experience. Therefore, the cargo transportation route planning in cold chain logistics must not only consider efficiency and economy, but also take temperature control as the primary factor into consideration.

[0003] In traditional cold chain logistics route planning, once a cold chain transport vehicle encounters abnormal temperature in the box (such as too high or too low temperature) during the journey, there is often a lack of effective and immediate response measures, which may cause damage to the goods, increase operating costs, and affect customer satisfaction. Traditional solutions may include returning directly to the starting point for adjustments, but these methods often ignore route optimization and timeliness, increasing unnecessary transportation time and costs. Summary of the invention

[0004] In order to solve the above problems, the present invention proposes a cold chain logistics cargo transportation path planning method based on AI processing.

[0005] The technical solution of the present invention is: a cold chain logistics cargo transportation path planning method based on AI processing comprises the following steps:

[0006] S1. When the temperature inside the box of the cold chain transport vehicle is abnormal, determine the backup area according to the current position and the terminal position of the cold chain transport vehicle;

[0007] S2, selecting the plannable grids in the spare area;

[0008] S3, according to the vehicle situation of the plannable grid, generate the current traffic flow depth and the relative traffic flow depth, and build a discriminator;

[0009] S4. Use the discriminator to extract the plannable grids whose driving state is smooth from the time when the temperature in the box is abnormal to the expected arrival time, and generate a transportation path.

[0010] The present invention is more suitable for situations where the temperature inside the box is abnormal and the cold chain transport vehicle must reach the destination in the shortest time to avoid affecting the subsequent cold chain logistics process. In S1-S2, the spare area can be as large as possible and can be set manually, or the center of the line connecting the current position and the terminal position can be directly used as the center of the circle to construct a circle as the spare area. The spare area is divided into several grids, and the grids with roads are used as plannable grids. The discriminator can calculate the corresponding value based on the two traffic depths of the input plannable grid, compare the corresponding value with the threshold, and obtain the driving status.

[0011] Furthermore, S3 includes the following sub-steps:

[0012] S31. Collect the number of vehicles in the plannable grid at each transportation time between the start of transportation of the cold chain transportation vehicle and the time when the temperature inside the box is abnormal;

[0013] S32, calculating the current traffic depth of the plannable grid according to the vehicle conditions of the plannable grid at each transportation moment;

[0014] S33, calculating the relative traffic depth of the plannable grid according to the vehicle conditions of the plannable grid in the historical days;

[0015] S34: construct a discriminator according to the current traffic flow depth and relative traffic flow depth of each plannable grid.

[0016] The beneficial effect of the above further scheme is: in the present invention, the degree of grid congestion is accurately quantified to avoid the problem of insufficient sensitivity of the traditional average method to emergencies. By analyzing the historical traffic gradients of the previous N days, the periodic congestion patterns of the grid (such as morning and evening peaks, holiday effects) can be identified. At the same time, the adaptability of the discriminator to historical laws can be enhanced, and misjudgments caused by data fluctuations can be reduced. The introduction of historical traffic gradients and random numbers can simulate the superimposed impact of historical volatility on the current state, and enhance the generalization ability of the discriminator in complex scenarios. The present invention weightedly fuses the current traffic depth with the historical relative depth, while considering the real-time state and long-term laws, and can balance the impact of the current traffic situation and the historical traffic situation, avoid the one-sidedness of a single indicator (such as instantaneous traffic), and improve the comprehensive accuracy of congestion discrimination.

[0017] Furthermore, in S32, the current traffic depth H of the grid may be planned. now The calculation formula is:

[0018] Where, Y t represents the number of vehicles in the plannable grid at the t-th transportation time on the transportation day, T represents the total number of times of cold chain transportation vehicles from the start of transportation to the time when the temperature inside the box is abnormal, exp(·) represents exponential calculation, and log2(·) represents logarithmic calculation.

[0019] Further, S33 includes the following sub-steps:

[0020] S331. The estimated transportation time period for the cold chain transportation vehicle is from the start of transportation to the estimated arrival time;

[0021] S332, extracting the number of vehicles in the plannable grid at each historical moment in the same expected transportation time period within the previous N days, where N represents a preset number of days;

[0022] S333, calculating the traffic flow gradient of the plannable grid on the nth day according to the number of vehicles at each historical moment on the nth day of the plannable grid, where n represents the day number;

[0023] S334. Calculate the relative traffic flow depth according to the traffic flow gradient of the plannable grid within N days.

[0024] N is a manual setting and can be determined according to actual conditions.

[0025] Furthermore, in S333, the traffic flow gradient Δσ of the grid on the nth day can be planned. n The calculation formula is:

[0026] ; In the formula, represents the number of vehicles in the plannable grid at k historical moments on day n, It represents the number of vehicles in the plannable grid at the time of abnormal temperature on the nth day, and K represents the total number of historical moments.

[0027] Furthermore, in S334, the relative traffic depth H of the grid can be planned. dis The calculation formula is:

[0028] ; Where N represents the preset number of days, Δσ n represents the traffic flow gradient of the plannable grid on day n, represents the number of vehicles in the plannable grid at the time of abnormal temperature on the nth day, exp(·) represents exponential calculation, log2(·) represents logarithmic calculation, and rand(0,1) represents the generation of a random number between 0 and 1.

[0029] Furthermore, in S34, the expression of the loss function Loss of the discriminator is:

[0030] Where H now_j represents the current traffic depth of the j-th plannable grid, H dis_j represents the relative traffic depth of the j-th plannable grid, γ j represents the position weight of the j-th plannable grid.

[0031] Further, S4 includes the following sub-steps:

[0032] S41, using a discriminator to extract a plannable grid whose driving state is smooth from the time when the temperature in the box is abnormal to the estimated arrival time;

[0033] S42. All plannable grids with unblocked driving status are used as gene sequences, and a transportation path is obtained using a genetic algorithm.

[0034] The unimpeded grid number is used as the gene sequence, and the path is represented as a gene arrangement and combination, such as grid 1→grid 3→end point.

[0035] The beneficial effects of the present invention are as follows: the present invention discloses an innovative cold chain logistics cargo transportation route planning method, which automatically searches for and evaluates potential backup areas when temperature anomalies occur, screens out plannable grid paths, and constructs a discriminator to determine that the traffic state is a smooth plannable grid, so as to ensure that the cold chain transportation vehicles are guided to the destination for adjustment in the shortest time, while ensuring the smooth progress of subsequent transportation; the present invention can not only significantly improve the transportation efficiency and cargo safety of cold chain logistics, but also effectively reduce operating costs, and improve the overall service level and competitiveness of the industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a flow chart of the cold chain logistics cargo transportation route planning method based on AI processing. DETAILED DESCRIPTION

[0037] The embodiments of the present invention will be further described below in conjunction with the accompanying drawings.

[0038] like Figure 1 As shown, the present invention provides a cold chain logistics cargo transportation path planning method based on AI processing, comprising the following steps:

[0039] S1. When the temperature inside the box of the cold chain transport vehicle is abnormal, determine the backup area according to the current position and the terminal position of the cold chain transport vehicle;

[0040] S2, selecting the plannable grids in the spare area;

[0041] S3, according to the vehicle situation of the plannable grid, generate the current traffic flow depth and the relative traffic flow depth, and build a discriminator;

[0042] S4. Use the discriminator to extract the plannable grids whose driving state is smooth from the time when the temperature in the box is abnormal to the expected arrival time, and generate a transportation path.

[0043] The present invention is more suitable for situations where the temperature inside the box is abnormal and the cold chain transport vehicle must reach the destination in the shortest time to avoid affecting the subsequent cold chain logistics process. In S1-S2, the spare area can be as large as possible and can be set manually, or the center of the line connecting the current position and the terminal position can be directly used as the center of the circle to construct a circle as the spare area. The spare area is divided into several grids, and the grids with roads are used as plannable grids. The discriminator can calculate the corresponding value based on the two traffic depths of the input plannable grid, compare the corresponding value with the threshold, and obtain the driving status.

[0044] In this embodiment of the present invention, S3 includes the following sub-steps:

[0045] S31. Collect the number of vehicles in the plannable grid at each transportation time between the start of transportation of the cold chain transportation vehicle and the time when the temperature inside the box is abnormal;

[0046] S32, calculating the current traffic depth of the plannable grid according to the vehicle conditions of the plannable grid at each transportation moment;

[0047] S33, calculating the relative traffic depth of the plannable grid according to the vehicle conditions of the plannable grid in the historical days;

[0048] S34: construct a discriminator according to the current traffic flow depth and relative traffic flow depth of each plannable grid.

[0049] The beneficial effect of the above further scheme is: in the present invention, the degree of grid congestion is accurately quantified to avoid the problem of insufficient sensitivity of the traditional average method to emergencies. By analyzing the historical traffic gradients of the previous N days, the periodic congestion patterns of the grid (such as morning and evening peaks, holiday effects) can be identified. At the same time, the adaptability of the discriminator to historical laws can be enhanced, and misjudgments caused by data fluctuations can be reduced. The introduction of historical traffic gradients and random numbers can simulate the superimposed impact of historical volatility on the current state, and enhance the generalization ability of the discriminator in complex scenarios. The present invention weightedly fuses the current traffic depth with the historical relative depth, while considering the real-time state and long-term laws, and can balance the impact of the current traffic situation and the historical traffic situation, avoid the one-sidedness of a single indicator (such as instantaneous traffic), and improve the comprehensive accuracy of congestion discrimination.

[0050] In the embodiment of the present invention, in S32, the current traffic depth H of the grid may be planned. now The calculation formula is:

[0051] Where, Y trepresents the number of vehicles in the plannable grid at the t-th transportation time on the transportation day, T represents the total number of times of cold chain transportation vehicles from the start of transportation to the time when the temperature inside the box is abnormal, exp(·) represents exponential calculation, and log2(·) represents logarithmic calculation.

[0052] In this embodiment of the present invention, S33 includes the following sub-steps:

[0053] S331. The estimated transportation time period for the cold chain transportation vehicle is from the start of transportation to the estimated arrival time;

[0054] S332, extracting the number of vehicles in the plannable grid at each historical moment in the same expected transportation time period within the previous N days, where N represents a preset number of days;

[0055] S333, calculating the traffic flow gradient of the plannable grid on the nth day according to the number of vehicles at each historical moment on the nth day of the plannable grid, where n represents the day number;

[0056] S334. Calculate the relative traffic flow depth according to the traffic flow gradient of the plannable grid within N days.

[0057] N is a manual setting and can be determined according to actual conditions.

[0058] In the embodiment of the present invention, in S333, the traffic flow gradient Δσ of the grid on the nth day can be planned. n The calculation formula is:

[0059] ; In the formula, represents the number of vehicles in the plannable grid at k historical moments on day n, It represents the number of vehicles in the plannable grid at the time of abnormal temperature on the nth day, and K represents the total number of historical moments.

[0060] In the embodiment of the present invention, in S334, the relative traffic depth H of the grid may be planned. dis The calculation formula is:

[0061] ; Where N represents the preset number of days, Δσ n represents the traffic flow gradient of the plannable grid on day n, represents the number of vehicles in the plannable grid at the time of abnormal temperature on the nth day, exp(·) represents exponential calculation, log2(·) represents logarithmic calculation, and rand(0,1) represents the generation of a random number between 0 and 1.

[0062] In the embodiment of the present invention, in S34, the expression of the loss function Loss of the discriminator is:

[0063] Where Hnow_j represents the current traffic depth of the j-th plannable grid, H dis_j represents the relative traffic depth of the j-th plannable grid, γ j represents the position weight of the j-th plannable grid.

[0064] In this embodiment of the present invention, S4 includes the following sub-steps:

[0065] S41, using a discriminator to extract a plannable grid whose driving state is smooth from the time when the temperature in the box is abnormal to the estimated arrival time;

[0066] S42. All plannable grids with unblocked driving status are used as gene sequences, and a transportation path is obtained using a genetic algorithm.

[0067] The unimpeded grid number is used as the gene sequence, and the path is represented as a gene arrangement and combination, such as grid 1→grid 3→end point.

[0068] In the embodiment of the present invention, all grids can participate in path planning, and smooth plannable grids are preferred for path planning.

[0069] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A cold chain logistics cargo transportation route planning method based on AI processing, characterized in that: The following steps are involved: S1. When the temperature inside the box of the cold chain transport vehicle is abnormal, determine the backup area according to the current position and the terminal position of the cold chain transport vehicle; S2, selecting the plannable grids in the spare area; S3, according to the vehicle situation of the plannable grid, generate the current traffic flow depth and the relative traffic flow depth, and build a discriminator; S4. Use the discriminator to extract the plannable grids whose driving state is smooth from the time when the temperature in the box is abnormal to the expected arrival time, and generate a transportation path.

2. The cold chain logistics cargo transportation path planning method based on AI processing according to claim 1 is characterized in that: The S3 comprises the following sub-steps: S31. Collect the number of vehicles in the plannable grid at each transportation time between the start of transportation of the cold chain transportation vehicle and the time when the temperature inside the box is abnormal; S32, calculating the current traffic depth of the plannable grid according to the vehicle conditions of the plannable grid at each transportation moment; S33, calculating the relative traffic depth of the plannable grid according to the vehicle conditions of the plannable grid in the historical days; S34: construct a discriminator according to the current traffic flow depth and relative traffic flow depth of each plannable grid.

3. The cold chain logistics cargo transportation path planning method based on AI processing according to claim 2 is characterized in that: In S32, the current traffic depth H of the grid can be planned. now The calculation formula is: Where, Y t represents the number of vehicles in the plannable grid at the t-th transportation time on the transportation day, T represents the total number of times of cold chain transportation vehicles from the start of transportation to the time when the temperature inside the box is abnormal, exp(·) represents exponential calculation, and log2(·) represents logarithmic calculation.

4. The cold chain logistics cargo transportation path planning method based on AI processing according to claim 2 is characterized in that: The S33 comprises the following sub-steps: S331. The estimated transportation time period for the cold chain transportation vehicle is from the start of transportation to the estimated arrival time; S332, extracting the number of vehicles in the plannable grid at each historical moment in the same expected transportation time period within the previous N days, where N represents a preset number of days; S333, calculating the traffic flow gradient of the plannable grid on the nth day according to the number of vehicles at each historical moment on the nth day of the plannable grid, where n represents the day number; S334. Calculate the relative traffic flow depth according to the traffic flow gradient of the plannable grid within N days.

5. The cold chain logistics cargo transportation path planning method based on AI processing according to claim 4 is characterized in that: In S333, the traffic flow gradient Δσ of the grid on the nth day can be planned. n The calculation formula is: ; In the formula, represents the number of vehicles in the plannable grid at k historical moments on day n, It represents the number of vehicles in the plannable grid at the time of abnormal temperature on the nth day, and K represents the total number of historical moments.

6. The cold chain logistics cargo transportation path planning method based on AI processing according to claim 4 is characterized in that: In S334, the relative traffic depth H of the grid may be planned. dis The calculation formula is: ; Where N represents the preset number of days, Δσ n represents the traffic flow gradient of the plannable grid on day n, It represents the number of vehicles in the plannable grid at the time of abnormal temperature on the nth day, exp(·) represents exponential calculation, log2(·) represents logarithmic calculation, and rand(0,1) represents the generation of a random number between 0 and 1.

7. The cold chain logistics cargo transportation path planning method based on AI processing according to claim 2 is characterized in that: In S34, the expression of the loss function Loss of the discriminator is: Where H now_j represents the current traffic depth of the j-th plannable grid, H dis_j represents the relative traffic depth of the j-th plannable grid, γ j Represents the position weight of the j-th plannable grid.

8. The cold chain logistics cargo transportation path planning method based on AI processing according to claim 1 is characterized in that: The S4 comprises the following sub-steps: S41, using a discriminator to extract a plannable grid whose driving state is smooth from the time when the temperature in the box is abnormal to the estimated arrival time; S42. All plannable grids with unblocked driving status are used as gene sequences, and a transportation path is obtained using a genetic algorithm.

Images