AI-Prediction-Based Cold Chain Transportation Route Optimization Method and System

Abstract

This application provides a cold chain transportation route optimization method and system based on AI prediction. First, a sample cold chain transportation order data set containing transportation node sequences, cargo freshness preservation requirement parameters, environmental temperature fluctuation records, and transportation timeliness completion status is obtained. Then, a combined transportation feature set containing time sensitivity, environmental stability, and cargo spoilage risk characteristics is extracted from it, and based on this, a cold chain demand prediction model is trained to obtain the predicted value of cargo freshness preservation demand and the path congestion probability distribution. Then, multiple candidate transportation routes are generated and scored for freshness preservation reliability and timeliness achievement rate. Finally, the current transportation route node sequence is dynamically adjusted according to the weighted score result, and at the same time, the model input data is updated in real time to optimize subsequent path decisions. This method uses multi-dimensional data and AI technology to effectively improve the scientificity and reliability of cold chain transportation route planning, reduce the risk of cargo spoilage, and improve transportation efficiency.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology. Specifically, it relates to a method and system for optimizing cold chain transportation routes based on AI prediction. Background Art

[0002] With the rapid development of the cold chain logistics industry, cold chain transportation plays a crucial role in ensuring the quality of temperature-sensitive goods such as fresh food and medicine. The particularity of cold chain transportation lies in not only ensuring the timely delivery of goods but also maintaining a suitable temperature environment throughout the process to prevent goods from deteriorating. However, existing cold chain transportation route planning methods have exposed many limitations when dealing with complex and changing actual transportation scenarios, and innovative technical means are urgently needed to solve them.

[0003] In the traditional cold chain transportation route planning process, most methods only consider basic factors such as distance and transportation time to determine the optimal route. These methods are often based on simple empirical rules or mathematical models, ignoring the key factor of the freshness preservation requirements of the goods themselves in cold chain transportation. For example, for different types of fresh food or medicine, their required freshness preservation temperature ranges, tolerable temperature fluctuations, and spoilage risk coefficients are all different, but traditional methods do not take these differences into account in route planning, resulting in frequent cases of goods spoilage due to improper temperature control during actual transportation, causing huge economic losses to enterprises. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a method and system for optimizing cold chain transportation routes based on AI prediction.

[0005] According to the first aspect of this application, a method for optimizing cold chain transportation routes based on AI prediction is provided. The method includes:

[0006] Obtain a sample cold chain transportation order data set, where the sample cold chain transportation order data set includes the transportation node sequences, cargo freshness preservation requirement parameters, environmental temperature fluctuation records, and transportation timeliness completion status of multiple sample orders;

[0007] Extract transportation features from the sample cold chain transportation order data set to generate a combined transportation feature set corresponding to each sample order. The combined transportation feature set includes time-sensitive features, environmental stability features, and cargo spoilage risk features;

[0008] Based on the combined transportation feature set, call a cold chain demand prediction model to generate a predicted value of the cargo freshness preservation demand in the target area within a preset time range, and generate a path congestion probability distribution based on the sample traffic flow data and weather event records in the target area;

[0009] Generate multiple candidate transportation routes from the starting point to the ending point according to the predicted value of the goods freshness preservation demand and the path congestion probability distribution, and perform freshness reliability scoring and timeliness achievement rate scoring for each candidate transportation route;

[0010] Based on the weighted results of the freshness reliability score and the timeliness achievement rate score, dynamically adjust the node sequence of the current transportation route, and update the input data of the cold chain demand prediction model in real time to optimize subsequent path decisions.

[0011] According to a second aspect of the present application, there is provided a cold chain transportation service system, the cold chain transportation service system includes a machine-readable storage medium and a processor, the machine-readable storage medium stores machine-executable instructions, and when the processor executes the machine-executable instructions, the cold chain transportation service system implements the foregoing cold chain transportation path optimization method based on AI prediction.

[0012] According to a third aspect of the present application, there is provided a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed, the foregoing cold chain transportation path optimization method based on AI prediction is implemented.

[0013] According to any one of the foregoing aspects, the technical effect of the present application is as follows:

[0014] By obtaining a sample cold chain transportation order data set including a transportation node sequence, goods freshness preservation requirement parameters, environmental temperature fluctuation records, and transportation timeliness completion status, the embodiments of the present application can comprehensively capture key information in the cold chain transportation process, so that subsequent analysis and model training are no longer limited to a single dimension, but are based on real-scene data with multiple dimensions and multiple variables.

[0015] When extracting transportation features from the sample cold chain transportation order data set, a combined transportation feature set including time-sensitive features, environmental stability features, and goods spoilage risk features is generated, breaking the limitation of only focusing on a single factor in the past, and comprehensively and synthetically depicting the complex characteristics of cold chain transportation. The time-sensitive features ensure the guarantee of transportation timeliness, the environmental stability features consider the impact of the external environment on the freshness preservation of goods, and the goods spoilage risk features are directly related to the quality and safety of goods. The combination of the three provides more practical, deeper, and broader feature inputs for subsequent model training, greatly improving the model's understanding and prediction ability of cold chain transportation scenarios.

[0016] Invoking the cold chain demand prediction model based on this combined transportation feature set can generate accurate predicted values of the cargo fresh-keeping demand and the probability distribution of path congestion within a preset time range for the target area. Different from traditional simple estimation or empirical methods, this cold chain demand prediction model is based on a large amount of actual data and in-depth feature analysis, achieving intelligent and accurate prediction of the cargo fresh-keeping demand and path congestion conditions. The predicted value of the cargo fresh-keeping demand provides a scientific basis for the temperature control strategy during transportation, and the probability distribution of path congestion warns in advance of possible traffic conditions, helping transportation planners make preparations in advance and greatly improving the foresight and scientific nature of transportation decisions.

[0017] Generate multiple candidate transportation paths based on the predicted value of the cargo fresh-keeping demand and the probability distribution of path congestion, and perform a fresh-keeping reliability score and a timeliness achievement rate score for each candidate transportation path. Traditional path planning often only focuses on a single goal, such as the shortest path or the fastest arrival time. However, this method takes into account both the cargo fresh-keeping reliability and the timeliness achievement rate, fully considering the particularity of cold chain transportation, ensuring that the goods are delivered to the destination in good fresh-keeping condition within the specified time. This not only improves the transportation quality of the goods, reduces the economic losses caused by deterioration, but also enhances customer satisfaction and the competitiveness of the enterprise in the cold chain transportation market.

[0018] Finally, dynamically adjust the node sequence of the current transportation path based on the weighted results of the fresh-keeping reliability score and the timeliness achievement rate score, and update the input data of the cold chain demand prediction model in real time to optimize subsequent path decisions, enabling the transportation path to continuously self-optimize according to the actual situation and adapt to the changing transportation environment. In the face of sudden weather changes, traffic congestion and other situations, it can quickly respond, adjust the path in a timely manner, and ensure that the transportation process always remains efficient and reliable. At the same time, real-time updating of the model input data enables the cold chain demand prediction model to continuously learn new information, further improving the accuracy of prediction and the adaptability of the model. Brief Description of the Drawings

[0019] To more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present application and should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0020] Figure 1 Shows a schematic flowchart of the cold chain transportation path optimization method based on AI prediction provided by the embodiments of the present application;

[0021] Figure 2The figure shows a schematic diagram of the component structure of a cold chain transportation service system provided by an embodiment of the present application for implementing the above-mentioned cold chain transportation route optimization method based on AI prediction. Detailed implementation manners

[0022] The embodiments of the present application will be described below with reference to the accompanying drawings in the present application. It should be understood that the implementation manners described below with reference to the accompanying drawings are exemplary descriptions for explaining the technical solutions of the embodiments of the present application, and do not constitute limitations on the technical solutions of the embodiments of the present application.

[0023] Those skilled in the art of the present technology can understand that unless specifically stated otherwise, the singular forms "a", "an", "the" and "said" used herein may also include the plural forms. It should be further understood that the terms "comprising" and "including" used in the embodiments of the present application mean that the corresponding features can be implemented as the presented features, information, data, steps, operations, elements and / or components, but do not exclude the implementation of other features, information, data, steps, operations, elements, components and / or their combinations, etc. supported by the art of the present technology. It should be understood that when an element is "connected" or "coupled" to another element, the one element can be directly connected or coupled to the other element, or it can mean that the one element and the other element establish a connection relationship through an intermediate element. In addition, the "connection" or "coupling" used herein may include a wireless connection or a wireless coupling. The term "and / or" used herein indicates at least one of the items defined by the term, for example, "A and / or B" can be implemented as "A", or implemented as "B", or implemented as "A and B".

[0024] To make the purpose, technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings. The technical solutions of the embodiments of the present application and the technical effects produced by the technical solutions of the present application will be described below through the description of several exemplary implementation manners. It should be noted that the following implementation manners can be referred to, learned from or combined with each other. For the same terms, similar features and similar implementation steps, etc. in different implementation manners, they will not be described repeatedly.

[0025] Figure 1 The figure shows a schematic flowchart of a cold chain transportation route optimization method and system based on AI prediction provided by an embodiment of the present application. It should be understood that in other embodiments, the order of some steps of the cold chain transportation route optimization method based on AI prediction in this embodiment can be shared according to actual needs, or some of the steps can also be omitted or maintained. The detailed steps of the cold chain transportation route optimization method based on AI prediction include:

[0026] Step S110, obtain a sample cold-chain transportation order dataset, where the sample cold-chain transportation order dataset includes the transportation node sequences, cargo freshness preservation requirement parameters, environmental temperature fluctuation records, and transportation timeliness completion statuses of multiple sample orders.

[0027] In this embodiment, in the fresh food cold-chain transportation scenario, taking the transportation of fruits (such as strawberries) and beef as examples. For the transportation order of strawberries, the transportation node sequence may include starting from the picking farm in the origin, passing through the local collection center, regional cold-chain warehouse, transfer and distribution center, and finally reaching each retail supermarket in the city. In terms of the cargo freshness preservation requirement parameters, strawberries are a highly perishable fruit and need to be stored in an environment with low temperature and appropriate humidity. The freshness preservation temperature requirement may be set at 0 - 4 degrees Celsius, the humidity is maintained at about 85 - 90%, and the total duration during the entire transportation process cannot exceed 48 hours, otherwise serious spoilage will occur. For the transportation of beef, the transportation node sequence may start from the slaughterhouse, pass through the meat processing factory, large cold-chain logistics hub, then to the meat wholesalers in various places, and finally reach the restaurant or retail point. The freshness preservation requirement for beef, compared with strawberries, requires a temperature below -18 degrees Celsius to ensure the freshness of the meat and prevent bacterial growth. Although the transportation timeliness is relatively longer, it also needs to complete the transportation within the specified 72 hours to ensure the quality of the beef.

[0028] Regarding the environmental temperature fluctuation records, during the transportation of strawberries, if during the transportation from the collection center to the regional cold-chain warehouse, due to certain aging of the refrigeration equipment of the transportation vehicle or too high external environmental temperature (for example, during the high-temperature period in summer, the external temperature reaches above 35 degrees Celsius), resulting in temperature fluctuations in the carriage, from the standard temperature of 0 - 4 degrees Celsius fluctuating to 5 - 6 degrees Celsius, and this fluctuation lasts for 1 - 2 hours. For the transportation of beef, if during the process from the meat processing factory to the cold-chain logistics hub, when the transportation vehicle passes through the mountainous section, due to the change in altitude and a short-term failure of the refrigeration equipment, the temperature fluctuates from -18 degrees Celsius to -10 degrees Celsius and lasts for about 30 minutes.

[0029] The transportation timeliness completion status is whether the transportation is completed within the specified time in the actual transportation process. For example, for a batch of strawberry transportation, due to a temporary road control at a certain transportation node, the transportation time is extended, and finally it fails to reach the retail supermarket within 48 hours, then the transportation timeliness completion status is not completed. For a batch of beef transportation, it passes smoothly through each transportation node and reaches the restaurant within the specified 72 hours, and the transportation timeliness completion status is completed. By collecting the relevant information of multiple such strawberry and beef transportation orders, including the transportation node sequences, cargo freshness preservation requirement parameters, environmental temperature fluctuation records, and transportation timeliness completion statuses, the sample cold-chain transportation order dataset is formed.

[0030] Step S120: Extract transportation features from the sample cold-chain transportation order dataset to generate a combined transportation feature set corresponding to each sample order. The combined transportation feature set includes time-sensitive features, environmental stability features, and cargo spoilage risk features.

[0031] Taking strawberry transportation as an example, extract the residence duration data of each transportation node from the sample cold-chain transportation order dataset. Suppose the planned residence duration for strawberries from the picking farm to the collection center is 2 hours, but due to the low sorting efficiency at the collection center, the actual residence duration reaches 4 hours. Regarding the number of times the environmental temperature deviates from the threshold, during the transportation from the regional cold-chain warehouse to the transfer and distribution center, due to a malfunction of the temperature sensor in the refrigeration equipment, the temperature inside the carriage deviates from the threshold of 0 - 4 degrees Celsius 3 times. In terms of the record of cargo packaging integrity, if some packaging of strawberries is squeezed and deformed during transportation, resulting in a certain degree of damage to the internal strawberries. Determine the time-sensitive feature based on the difference between the residence duration data and the preset cargo freshness duration limit. Since an additional 2 hours are spent at the collection center, which is close to half of the 48-hour freshness duration limit for strawberries, this indicates a relatively high risk level of exceeding the freshness duration limit in the transportation node sequence.

[0032] When calculating the environmental stability feature, based on the number of times the environmental temperature deviates from the threshold and the proportion of the deviation duration. During the above-mentioned transportation from the regional cold-chain warehouse to the transfer and distribution center, the temperature deviates from the threshold 3 times, and each deviation duration is about 30 minutes. The total transportation duration is 4 hours, and the proportion of the deviation duration is (3×30)÷(4×60) = 37.5%. This proportion indicates that the cumulative impact of temperature fluctuations on the quality of strawberries during transportation is relatively large.

[0033] For the cargo spoilage risk feature, combine the cargo packaging integrity record and the spoilage rate parameter corresponding to the cargo type. Strawberries themselves have a relatively fast spoilage rate, and because some packaging is damaged, the strawberries are exposed to a non-ideal environment, which greatly increases the contribution of different nodes in the transportation path to the decline in cargo quality.

[0034] For beef transportation, from the slaughterhouse to the meat processing plant, the planned residence time is 3 hours, and the actual residence time is 5 hours. During the transportation from the large cold chain logistics hub to the meat wholesaler, due to the influence of the external high temperature, the temperature in the carriage deviated from the -18 °C threshold 2 times, and each deviation lasted for 15 minutes. The transportation duration of this section is 5 hours, and the proportion of the deviation duration is (2×15)÷(5×60) = 10%. If there is a scratch on the beef packaging, combined with the deterioration rate of beef (relatively slower than strawberries but also deteriorating in a bad environment), it will also increase the risk characteristics of cargo deterioration. Finally, the time-sensitive characteristics, environmental stability characteristics, and cargo deterioration risk characteristics of the strawberry and beef transportation orders are normalized to generate their respective combined transportation characteristic sets.

[0035] Step S130, based on the combined transportation characteristic set, call the cold chain demand prediction model to generate the predicted value of the cargo freshness preservation demand in the target area within the preset time range, and generate the path congestion probability distribution based on the sample traffic flow data and weather event records in the target area.

[0036] Continue with the scenarios of strawberry and beef transportation. Input the combined transportation characteristic set of strawberry transportation into the temporal convolutional neural network. Assume that the transportation nodes include the picking farm A, the collection center B, the regional cold chain warehouse C, the transfer and distribution center D, and the retail supermarket E. The temporal convolutional neural network analyzes the transportation data between these nodes, such as the transportation situation from A to B and from B to C at different time periods, and extracts the dynamic association patterns between the transportation nodes. For example, it is detected that the transportation from B to C is prone to transportation delays on Monday mornings because the goods in the collection center B are shipped intensively at this time and the vehicle scheduling is tense. After being processed by the temporal convolutional neural network, the initial path demand prediction result is output.

[0037] Then use the long short-term memory network to correct the time dependence of the initial path demand prediction result. Due to the strict freshness preservation time limit of strawberries, the long short-term memory network takes into account that the freshness of strawberries will decline over time and adjusts the initial path demand prediction result. For example, if the initial prediction shows a low demand for freshness preservation resources at a certain node, but considering the cumulative effect of the transportation duration, the demand for freshness preservation resources at this node is increased after correction, generating the corrected predicted value of the cargo freshness preservation demand. This value includes the distribution of the freshness preservation resource demands of different transportation nodes in the transportation path. For example, more refrigeration resources are needed at the transfer and distribution center D to maintain the freshness of strawberries.

[0038] For a target area (such as a certain city and its surrounding areas), a sub-model for predicting path congestion probability is constructed based on the sample traffic flow data and weather event records in this area. If heavy rain often occurs in this city in summer, according to historical data, waterlogging on the roads during heavy rain will cause traffic paralysis on some sections. By analyzing the correlation between weather events and traffic flow changes, for example, it is detected that when it rains, the congestion probability of the road from regional cold-chain warehouse C to transfer and distribution center D will increase by 30%, so as to output the path congestion probability distribution in the future time period.

[0039] For beef transportation, the same process is adopted. The combined transportation feature set is input into the temporal convolutional neural network to analyze the dynamic association patterns among nodes such as slaughterhouses, meat processing plants, and cold-chain logistics hubs, and the initial path demand prediction result is output. The long short-term memory network makes time-dependent corrections according to the freshness preservation characteristics of beef (such as the deterioration rate is relatively slow but still affected by time at -18 degrees Celsius), and obtains the cargo freshness preservation demand prediction value including the distribution of freshness preservation resource demands at each transportation node. When constructing the sub-model for predicting path congestion probability, considering that snowfall in winter may affect the transportation from the meat processing plant to the cold-chain logistics hub, the relationship between snowfall and traffic flow is analyzed to obtain the corresponding path congestion probability distribution. Finally, the corrected cargo freshness preservation demand prediction values and path congestion probability distributions of strawberry and beef transportation are fused to generate the joint output of the cold-chain demand prediction model. This joint output can identify the occupancy rate of freshness preservation resources and congestion delay risks of different candidate transportation paths. For example, for a candidate path for transporting strawberries, the occupancy rate of freshness preservation resources may reach 80%, and the congestion delay risk is 20%. For a candidate path for beef transportation, the occupancy rate of freshness preservation resources is 60%, and the congestion delay risk is 15%.

[0040] Step S140, generate multiple candidate transportation paths from the starting point to the ending point according to the cargo freshness preservation demand prediction value and the path congestion probability distribution, and perform freshness preservation reliability scoring and timeliness achievement rate scoring on each candidate transportation path.

[0041] Taking strawberry transportation as an example, the starting point and the ending point are from the picking farm in the origin to the urban retail supermarket. The expected total transportation duration of each candidate transportation route is calculated based on the node sequence length of the transportation path, the sample average passing speed, and the path congestion probability distribution. Suppose there is a candidate transportation route: picking farm - collection center - transfer and distribution center - retail supermarket, and the node sequence length of this route is 3 nodes. In terms of the sample average passing speed, if the average speed from the picking farm to the collection center is 60 km / h and the distance is 120 km, the average speed from the collection center to the transfer and distribution center is 50 km / h and the distance is 100 km, and the average speed from the transfer and distribution center to the retail supermarket is 40 km / h and the distance is 80 km. Then, combined with the path congestion probability distribution, if the congestion probability from the collection center to the transfer and distribution center is 30%, which will cause the passing speed to decrease, the expected total transportation duration can be calculated through this.

[0042] Compare the expected total transportation duration with the cargo freshness preservation duration limit to determine the timeliness achievement rate score. Since the freshness preservation duration limit of strawberries is 48 hours, if the calculated expected total transportation duration is 40 hours, then the timeliness achievement rate score is relatively high because the negative deviation degree of the expected total transportation duration relative to the freshness preservation duration limit is small. This score can be an inverse proportional function. For example, 1 - (40 - 48) ÷ 48 = 1 - (-1 / 6) = 7 / 6 (this is just an example calculation method).

[0043] Based on the freshness preservation resource demand distribution corresponding to the nodes in the cargo freshness preservation demand prediction value, combined with the current available refrigeration equipment capacity and the energy supply status, calculate the freshness preservation reliability score. For example, at the transfer and distribution center, according to the cargo freshness preservation demand prediction value, 50% of the refrigeration equipment capacity is required to maintain the freshness of strawberries, while the current available refrigeration equipment capacity at the transfer and distribution center is 80% and the energy supply status is good. Then, the resource matching degree in the freshness preservation reliability score is relatively high. At the same time, considering the temperature control ability of the refrigeration equipment, if the temperature control accuracy within the range of 0 - 4 degrees Celsius is relatively high, then the freshness preservation reliability score will also be relatively high. This score is the weighted average of the node resource matching degree and the temperature control ability.

[0044] For beef transportation, the starting and ending points of transportation are from the slaughterhouse to the restaurant. Similarly, calculate the expected total transportation duration of the candidate transportation routes. For example, for the route of slaughterhouse - meat processing plant - cold chain logistics hub - meat wholesaler - restaurant, calculate the expected total transportation duration based on the distances, average speeds, and path congestion probability distributions of each section. Compare it with the 72-hour freshness preservation duration limit of beef to determine the timeliness achievement rate score. According to the distribution of freshness preservation resource requirements at each node (such as the cold chain logistics hub) in the predicted value of the goods' freshness preservation demand, calculate the freshness preservation reliability score in combination with the capacity of the refrigeration equipment and the energy supply status. Finally, perform dynamic weight allocation on the freshness preservation reliability score and the timeliness achievement rate score of each candidate transportation route for strawberry and beef transportation to generate the comprehensive priority ranking of each candidate transportation route. For example, for a certain candidate route for strawberry transportation, the weight of the freshness preservation reliability score is 0.6, and the weight of the timeliness achievement rate score is 0.4. Calculate its priority among all candidate routes comprehensively.

[0045] Step S150, based on the weighted results of the freshness preservation reliability score and the timeliness achievement rate score, dynamically adjust the node sequence of the current transportation route, and update the input data of the cold chain demand prediction model in real time to optimize subsequent route decisions.

[0046] Taking strawberry transportation as an example, monitor the real-time temperature data, traffic status update information, and abnormal events in the operation of refrigeration equipment at each node during transportation. Real-time collect the temperature distribution data inside the carriage through the Internet of Things sensors deployed on the transportation vehicle. If during the transportation from the goods collection center to the transfer and distribution center, the sensor detects that the temperature distribution inside the carriage is uneven and the local temperature reaches 6 degrees Celsius, identify this abnormal temperature fluctuation pattern. Obtain the real-time road condition data interface of the traffic management system and learn that there is a road construction event on the section from the transfer and distribution center to the retail supermarket, which is the traffic status update information. Monitor the compressor operation frequency, refrigerant pressure, and energy consumption rate of the refrigeration equipment to generate an equipment health status assessment report. If it is detected that the refrigerant pressure is lower than the normal level, this is an abnormal event in the operation of the refrigeration equipment.

[0047] When the duration of the abnormal temperature fluctuation corresponding to the abnormal temperature fluctuation pattern inside the carriage exceeds the preset tolerance threshold (such as 15 minutes), adjust the residence duration limit of the subsequent nodes according to the continuous deviation of the real-time temperature data from the preset temperature threshold. For example, the original plan was to stay at the transfer and distribution center for 1 hour. Due to the temperature fluctuation, in order to ensure the freshness of strawberries, shorten the residence duration to 30 minutes and recalculate the freshness preservation reliability score.

[0048] When it is detected that the traffic status update information indicates that the congestion probability of the path from the transfer and distribution center to the retail supermarket exceeds the preset risk threshold (e.g., 50%), the process of regenerating the candidate transportation path is triggered. According to the risk node location (the section from the transfer and distribution center to the retail supermarket) and the influence radius (e.g., within 10 kilometers) in the path congestion probability distribution corresponding to the traffic status update information, the geographical boundary range of the detour area is determined. Alternative nodes with cold storage facility filings are screened within this range. For example, if a small cold chain warehouse is detected within the detour range and has sufficient refrigeration equipment. The compatibility data of the refrigeration equipment type of this alternative node with the current transported strawberries (e.g., whether the temperature control range is suitable for strawberry preservation) is extracted, and combined with the resource demand distribution in the predicted value of the goods preservation demand, alternative nodes with qualified compatibility data are matched to generate emergency path branches, such as detour main paths, temporary transfer paths, etc. According to the strawberry temperature stability index collected in real time, branches that exceed the remaining preservation duration constraint are excluded from the emergency path branches to generate a subset of effective emergency paths. The subset of effective emergency paths is input into the cold chain demand prediction model, and re-ordered according to the dynamic weights of the preservation reliability score and the timeliness achievement rate score, and the target emergency path with the highest priority is output.

[0049] The adjusted node sequence (such as adding new alternative nodes), real-time temperature data, and traffic status update information are fed back to the cold chain demand prediction model to iteratively optimize the accuracy of the predicted value of the goods preservation demand. For example, a difference analysis is performed between the actual transportation results of the target emergency path (including new node and path information) and the predicted preservation reliability score to generate a path decision deviation index. If it is detected that the preservation resource demand at a certain node during actual transportation is higher than the predicted value, the convolution kernel size of the temporal convolutional neural network is adjusted according to the path decision deviation index to enhance the local feature extraction ability for sudden congestion events. Based on the strawberry quality feedback data after the execution of the target emergency path, the forgetting gate threshold of the time-dependent correction in the long short-term memory network is optimized. Within a preset period (e.g., every day), the adjusted node sequence, real-time temperature data, traffic status update information, and cold storage equipment operation abnormal events are integrated into an incremental training dataset according to the time stamp, and the incremental training dataset is injected into the cold chain demand prediction model through an online learning algorithm to update the model weight information of the cold chain demand prediction model and retain the high-value patterns in the historical path decisions.

[0050] The same process applies to beef transportation. Monitor real-time temperature data, traffic status update information, and cold storage equipment operation abnormal events during transportation. When there are abnormal temperature fluctuations or traffic congestion risks, adjustments are made according to the above processes, such as adjusting the residence duration of subsequent nodes, regenerating the candidate transportation path, feeding back data to optimize the cold chain demand prediction model, etc., to ensure that the beef can reach the destination within the specified preservation conditions and transportation timeliness.

[0051] Based on the above steps, the embodiments of the present application can comprehensively capture the key information in the cold chain transportation process by obtaining a sample cold chain transportation order dataset containing the transportation node sequence, cargo freshness requirement parameters, environmental temperature fluctuation records, and transportation timeliness completion status, enabling subsequent analysis and model training to no longer be limited to a single dimension, but rather be based on multi-dimensional and multi-variable real-scene data.

[0052] When extracting transportation features from the sample cold chain transportation order dataset, a combined transportation feature set including time-sensitive features, environmental stability features, and cargo spoilage risk features is generated, breaking the limitation of only focusing on a single factor in the past and comprehensively and synthetically depicting the complex characteristics of cold chain transportation. The time-sensitive features ensure the guarantee of transportation timeliness, the environmental stability features consider the impact of the external environment on cargo freshness, and the cargo spoilage risk features are directly related to cargo quality and safety. The combination of the three provides a more practical, in-depth, and comprehensive feature input for subsequent model training, greatly enhancing the model's understanding and prediction ability of cold chain transportation scenarios.

[0053] Based on this combined transportation feature set, calling the cold chain demand prediction model can generate accurate cargo freshness demand prediction values and path congestion probability distributions in the target area within a preset time range. Different from traditional simple estimation or empirical methods, this cold chain demand prediction model is based on a large amount of actual data and in-depth feature analysis, realizing intelligent and accurate prediction of cargo freshness demand and path congestion conditions. The cargo freshness demand prediction value provides a scientific basis for the temperature control strategy during transportation, and the path congestion probability distribution early warns of possible traffic conditions, helping transportation planners make preparations in advance and greatly improving the forward-looking and scientific nature of transportation decisions.

[0054] Multiple candidate transportation paths are generated according to the cargo freshness demand prediction values and path congestion probability distributions, and each candidate transportation path is scored for freshness reliability and timeliness achievement rate. Traditional path planning often only focuses on a single goal, such as the shortest path or the fastest arrival time. However, this method takes into account both cargo freshness reliability and timeliness achievement rate, fully considering the particularity of cold chain transportation, ensuring that the cargo is delivered to the destination in good freshness condition within the specified time. This not only improves the transportation quality of the cargo, reduces economic losses caused by spoilage, but also enhances customer satisfaction and the competitiveness of enterprises in the cold chain transportation market.

[0055] Finally, based on the weighted results of the freshness preservation reliability score and the timeliness achievement rate score, dynamically adjust the node sequence of the current transportation route, and update the input data of the cold chain demand prediction model in real time to optimize subsequent route decisions, so that the transportation route can continuously self-optimize according to the actual situation and adapt to the changing transportation environment. In the face of sudden weather changes, traffic congestion and other situations, it can quickly respond and adjust the route in time to ensure that the transportation process always remains efficient and reliable. At the same time, the real-time update of the model input data enables the cold chain demand prediction model to continuously learn new information, further improving the prediction accuracy and the adaptability of the model.

[0056] In a possible implementation manner, step S120 includes:

[0057] Step S121, extract the residence duration data, the number of times the environmental temperature deviates from the threshold, and the record of the integrity of the goods packaging for each transportation node from the sample cold chain transportation order dataset.

[0058] In this embodiment, for strawberry transportation, for example, at the transportation node from the picking farm to the collection center, the planned residence duration is 2 hours, but the actual residence duration reaches 3.5 hours; from the collection center to the regional cold chain warehouse, the planned residence is 1.5 hours and the actual residence is 2.5 hours. In terms of the number of times the environmental temperature deviates from the threshold, during the transportation from the regional cold chain warehouse to the transfer and distribution center, due to a short-term failure of the refrigeration equipment, the temperature in the carriage deviated from the threshold of 0-4 degrees Celsius 2 times. The record of the integrity of the goods packaging shows that the packaging of some strawberries was squeezed during transportation, and 10% of the packaging was damaged.

[0059] Step S122, determine the time-sensitive feature according to the difference between the residence duration data and the preset freshness preservation duration limit for the goods, where the time-sensitive feature is used to identify the risk level of exceeding the freshness preservation duration limit in the transportation node sequence.

[0060] For example, the freshness preservation duration limit for strawberries is 48 hours. There is an extra 1.5 hours of residence from the picking farm to the collection center and an extra 1 hour of residence from the collection center to the regional cold chain warehouse. This means that more freshness preservation time has been consumed in the early stage of transportation, and the risk level of exceeding the freshness preservation duration limit in the transportation node sequence is at a relatively high level. This relatively high risk level indicates that more stringent time control is required in subsequent transportation links to prevent the strawberries from deteriorating due to excessive time.

[0061] Step S123, calculate the environmental stability feature according to the number of times the environmental temperature deviates from the threshold and the proportion of the deviation duration, where the environmental stability feature is used to characterize the cumulative impact of temperature fluctuations on the quality of the goods during transportation.

[0062] For example, during the transportation from the regional cold-chain warehouse to the transit distribution center, the temperature deviated from the threshold 2 times, and the duration of each deviation was about 20 minutes. The total transportation duration was 3 hours, and the proportion of the deviation duration was (2×20)÷(3×60)≈22.2%. This proportion reflects that the cumulative impact of temperature fluctuations during transportation on the quality of strawberries is relatively obvious, which may accelerate the deterioration process of strawberries because strawberries are very sensitive to temperature, and even short-term temperature fluctuations may affect their freshness and quality.

[0063] Step S124, generate the goods deterioration risk characteristics by combining the record of the integrity of the goods packaging and the deterioration rate parameters corresponding to the goods type. The goods deterioration risk characteristics are used to quantify the contribution degree of different nodes in the transportation path to the decline in the quality of the goods.

[0064] For example, strawberries are perishable fruits with a relatively fast deterioration rate. Coupled with 10% of the packaging being damaged, this increases the possibility of strawberries being exposed to the external environment, further enhancing the contribution degree of different nodes in the transportation path to the decline in the quality of the goods. For example, at the transit distribution center, if the damaged strawberries are not processed in a timely manner, during the subsequent transportation to the retail supermarket, due to the influence of temperature fluctuations and transportation duration, the deterioration speed of these damaged strawberries will significantly accelerate, thus affecting the quality of the entire batch of strawberries.

[0065] For the transportation of beef, from the slaughterhouse to the meat processing factory, the planned stay duration was 3 hours, and the actual stay duration was 4 hours; from the meat processing factory to the cold-chain logistics hub, the planned stay was 2 hours, and the actual stay was 2.5 hours. During the transportation from the cold-chain logistics hub to the meat wholesaler, due to changes in the external environmental temperature and minor malfunctions of the refrigeration equipment, the temperature in the carriage deviated from the threshold of -18 degrees Celsius once, and the duration was 10 minutes. In terms of the integrity of the beef packaging, 5% of the packaging had slight scratches.

[0066] The preservation duration limit of beef is 72 hours. Staying 1 hour longer from the slaughterhouse to the meat processing factory increases the risk level of exceeding the preservation duration limit, but the risk level is relatively lower compared to strawberries. When calculating the environmental stability characteristics, the transportation duration from the cold-chain logistics hub to the meat wholesaler was 4 hours, the temperature deviated from the threshold once, and the duration of each deviation was 10 minutes. The proportion of the deviation duration was (1×10)÷(4×60)≈4.2%, indicating that the cumulative impact of temperature fluctuations on the quality of beef is relatively small. Combining the slower deterioration rate of beef (compared to strawberries) and the 5% packaging damage, although the overall goods deterioration risk is lower than that of strawberries, the packaging damage will still increase to a certain extent the contribution degree of different nodes in the transportation path to the decline in the quality of the goods. Especially during long-term transportation, the beef at the damaged part may be contaminated by microorganisms and deteriorate.

[0067] Step S125: Normalize the time-sensitive feature, environmental stability feature, and goods spoilage risk feature to generate the combined transportation feature set.

[0068] Finally, in this embodiment, the time-sensitive feature, environmental stability feature, and goods spoilage risk feature of strawberry and beef transportation are normalized to generate their respective combined transportation feature sets. This normalization process is to unify feature values of different magnitudes and ranges into a standard range, so as to enable effective data processing and analysis in subsequent operations such as training the cold chain demand prediction model. For example, for the time-sensitive feature value in strawberry transportation, it may be between 0.6 - 0.8 (hypothetical), the environmental stability feature value is between 0.5 - 0.7, and the goods spoilage risk feature value is between 0.7 - 0.9; for beef transportation, the time-sensitive feature value may be between 0.3 - 0.5, the environmental stability feature value is between 0.2 - 0.4, and the goods spoilage risk feature value is between 0.4 - 0.6. These combined transportation feature sets will serve as an important data basis for subsequent operations, accurately reflecting various characteristics and risk factors of different fresh products during transportation.

[0069] In a possible implementation manner, step S130 includes:

[0070] Step S131: Input the combined transportation feature set into a temporal convolutional neural network, extract the dynamic association pattern between transportation nodes, and output the initial path demand prediction result.

[0071] In the fresh cold chain transportation scenario of this embodiment, take the transportation of strawberries and beef as an example for illustration.

[0072] Strawberry transportation: The time-sensitive feature value reflects the risk level of the transportation node exceeding the freshness preservation duration limit. For example, in strawberry transportation, from the picking farm A to the collection center B, if the transportation is usually in the morning, the road is unobstructed and the staff is efficient, and the transportation time is stable, the time-sensitive feature value of this node is low, indicating a small risk of exceeding the freshness preservation duration limit at this stage. However, from the collection center B to the regional cold chain warehouse C, if the shipment volume at the collection center increases significantly, resulting in an extended vehicle loading and scheduling time and transportation delay, the time-sensitive feature value will increase, meaning an increased risk of exceeding the freshness preservation duration limit.

[0073] This time-sensitive eigenvalue directly affects the estimation of transportation duration and the estimation of fresh-keeping resource requirements in the initial path demand prediction results. When analyzing the transportation from B to C in the initial path demand prediction results, due to the increase in the time-sensitive eigenvalue, the prediction will take into account the possible impact of transportation delays on the freshness of strawberries, thus estimating that the transportation duration will increase. And in order to ensure the freshness of strawberries during the extended transportation time, it will be initially estimated that corresponding fresh-keeping resources need to be increased, such as increasing the refrigeration power or the refrigeration time, to slow down the spoilage rate of strawberries.

[0074] Beef transportation: In the transportation of beef, from the slaughterhouse F to the meat processing factory G, due to the relatively fixed working process of the slaughterhouse, the transportation time is relatively stable and the time-sensitive eigenvalue is low. However, during holidays, the shipment volume from the meat processing factory G to the cold chain logistics hub H increases, and the waiting time of transportation vehicles prolongs, resulting in an increase in the time-sensitive eigenvalue.

[0075] For the initial path demand prediction results, when considering the transportation from G to H, the increase in the time-sensitive eigenvalue makes the prediction aware of the possible transportation delay, and then estimates that the transportation duration will increase. At the same time, in order to ensure that the quality of beef is not greatly affected during the extended transportation time, it will be initially estimated that the fresh-keeping resource requirements need to be adjusted, such as perhaps more precise temperature control or an increase in fresh-keeping packaging materials, to maintain the freshness of beef.

[0076] Regarding the environmental stability characteristic, strawberry transportation: The environmental stability eigenvalue is calculated based on the number of times the environmental temperature deviates from the strawberry fresh-keeping suitable temperature range (assumed to be 0 - 5 degrees Celsius) and the proportion of the deviation duration, and is used to characterize the cumulative impact of temperature fluctuations on the quality of strawberries during transportation. If during the transportation from the regional cold chain warehouse C to the transfer and distribution center D, the temperature sensor records that the temperature frequently deviates from the suitable range and the proportion of the deviation duration is large, then the environmental stability eigenvalue is low, indicating that the temperature fluctuations have a great impact on the quality of strawberries.

[0077] The initial path demand prediction results will make corresponding adjustments based on this environmental stability eigenvalue. Since large temperature fluctuations may accelerate the spoilage of strawberries, the prediction will consider increasing the refrigeration resources in this transportation stage or adopting more advanced temperature control equipment to stabilize the transportation environment temperature and ensure the freshness of strawberries. At the same time, in the estimation of transportation duration, the time changes caused by possible additional measures due to temperature problems may also be considered.

[0078] Beef transportation: For beef transportation, the environmental stability eigenvalue is calculated based on the temperature deviation from the beef fresh-keeping suitable temperature range (assumed to be -18 - -15 degrees Celsius). For example, during the transportation from the cold chain logistics hub H to the meat wholesaler I, if the number of times the temperature deviates from this suitable range is small and the proportion of the deviation duration is small, the environmental stability eigenvalue is relatively high, indicating that the temperature fluctuations have a small impact on the quality of beef.

[0079] In the initial path demand prediction results, considering that the environmental stability is relatively good, the preliminary estimate of the demand for preservation resources may be relatively stable and there will be no significant adjustment. However, if the environmental stability eigenvalue is low, that is, the temperature fluctuates greatly, the prediction may increase the demand for temperature regulation equipment, such as equipment with stronger refrigeration capacity or more precise temperature monitoring instruments, to ensure the quality of beef during transportation.

[0080] Regarding the risk characteristics of goods deterioration, for strawberry transportation: The risk eigenvalue of goods deterioration combines the record of strawberry packaging integrity and the relatively high deterioration rate parameter of strawberries themselves, and is used to quantify the contribution of different nodes in the transportation path to the quality decline of strawberries. Suppose at the transfer and distribution center D, it is found that some strawberry packages are squeezed and damaged, and the strawberries themselves have a fast deterioration rate, then the risk eigenvalue of goods deterioration at this node is high.

[0081] In response to this situation, the initial path demand prediction results will focus on considering increasing the protection measures for strawberries and the input of preservation resources at this node and in the subsequent transportation stages. There may be an increased demand for replacing packaging materials, or an increased use of preservation substances such as preservatives to reduce the risk of goods deterioration. At the same time, in the estimation of transportation duration, the time consumption caused by dealing with the risk of goods deterioration may also be considered.

[0082] For beef transportation: In beef transportation, the risk eigenvalue of goods deterioration is also generated by combining the beef packaging integrity and relatively low deterioration rate parameters. For example, at the meat processing plant G, if the beef packaging is damaged and not dealt with in time, and considering the beef deterioration rate, the risk eigenvalue of goods deterioration at this node will increase.

[0083] Based on this increased risk eigenvalue of goods deterioration, the initial path demand prediction results will adjust the demand for preservation resources. There may be an increase in the secondary packaging of beef or strengthening of preservation treatment, such as increasing the injection amount of preservation gas. In the estimation of transportation duration, the time spent on operations related to dealing with the risk of goods deterioration will also be considered to ensure that the quality of beef is not greatly affected during subsequent transportation.

[0084] First, look at strawberry transportation. Its transportation nodes include the picking farm A, the consolidation center B, the regional cold-chain warehouse C, the transfer and distribution center D, and the retail supermarket E. The combined transportation feature set covers the time-sensitive feature values, environmental stability feature values, and cargo spoilage risk feature values of each node. In terms of time-sensitive feature values, for example, in the transportation from A to B, if it is in the morning period, due to fewer road vehicles and high staff efficiency, the transportation time is relatively stable, which makes the time-sensitive feature value relatively low, indicating a low risk of exceeding the freshness-keeping duration limit. However, in the transportation from B to C, if the consolidation center has a large shipment volume, the vehicle loading and scheduling time increase, resulting in transportation delays, and the time-sensitive feature value will increase. In terms of environmental stability feature values, for example, in some transportation sections, the number of times the temperature sensor records the temperature deviating from the suitable temperature range for strawberry freshness-keeping (assumed to be 0 - 5 degrees Celsius) and the proportion of the deviation duration are used to calculate the environmental stability feature value. If the number of deviations is large and the proportion of the duration is large, this feature value is low, meaning that the temperature fluctuation has a large cumulative impact on the strawberry quality. The cargo spoilage risk feature value is generated by combining the strawberry packaging integrity record (such as whether there is extrusion damage, etc.) and the spoilage rate parameter of the strawberry itself (because strawberries are perishable and the spoilage rate is relatively high).

[0085] Input the combined transportation feature set of strawberry transportation into the temporal convolutional neural network. For the first convolutional layer of this network, assume that the convolutional kernel size is 7×3. Here, 7 represents the window size on the transportation node sequence (time dimension), 3 matches the dimension of the three-dimensional feature vector (time-sensitive, environmental stability, cargo spoilage risk feature values) of each node, and the stride is set to 1. The convolutional kernel slides on the combined transportation feature set. For example, starting from the first node A, it will perform multiplication and addition operations on the three-dimensional feature vectors of these 7 nodes from A to A + 6 through a pre-set weight matrix to capture local dynamic association patterns such as the impact of the transportation time change from A to B on the environmental stability and cargo spoilage risk of subsequent nodes. After the convolutional operation, a non-linear transformation is performed through the ReLU activation function. Multiple convolutional kernels with different weight matrices work in parallel. Assume that the first convolutional layer has 15 convolutional kernels. After the operation, a feature map with the same number of nodes but a feature dimension of 15 is obtained. Then, it enters the pooling layer. Max pooling is used, with a pooling window size of 3×1 and a stride of 3. The maximum value is taken within each pooling window to reduce the data volume while retaining important features, and a new feature map is obtained. After alternating processing through multiple convolutional layers and pooling layers, the initial path demand prediction result is output.

[0086] This result includes the preliminary estimates of the expected transportation duration for each node, possible delay situations, and fresh-keeping resource requirements. For example, for the transportation from C to D, due to the long time and high risk of temperature fluctuations (reflected by the low environmental stability eigenvalue), it is initially estimated that 10% more refrigeration resources are needed to maintain the freshness of strawberries. This is obtained by comprehensively considering the dynamic associations among the time sensitivity, environmental stability, and cargo spoilage risk eigenvalues at each node during the transportation process.

[0087] Looking at the beef transportation again, the transportation nodes include slaughterhouse F, meat processing plant G, cold-chain logistics hub H, meat wholesaler I, and restaurant J. Its combined transportation feature set is also composed of the time sensitivity, environmental stability, and cargo spoilage risk eigenvalues of each node. In terms of the time sensitivity eigenvalue, the transportation from F to G has a relatively stable time because of the fixed working process of the slaughterhouse, so the eigenvalue is low; however, during holidays, the transportation from G to H has an increased eigenvalue because the increased shipment volume from the processing plant leads to long waiting times for vehicles. The environmental stability eigenvalue is calculated based on the temperature deviation from the suitable temperature range for beef fresh-keeping (assumed to be -18 to -15 degrees Celsius). The cargo spoilage risk eigenvalue is generated by combining the integrity of the beef packaging and the beef spoilage rate parameter (relatively low compared to the strawberry spoilage rate).

[0088] Input the combined transportation feature set of beef transportation into a temporal convolutional neural network, and the temporal convolutional neural network extracts the dynamic association patterns between transportation nodes in a similar way. For example, analyze the relationship that the transportation time from F to G is stable but the transportation from G to H is affected by the shipment volume and vehicle scheduling. Finally, output the initial path demand prediction result for beef transportation, including the estimated transportation duration for each node, possible delay factors, and the preliminary judgment of fresh-keeping resource requirements. For example, for the transportation from H to I, the estimated duration is 5 hours. Due to the small temperature fluctuations and short time (judged from the environmental stability and time sensitivity eigenvalues), it is initially estimated that the fresh-keeping resource requirements are the same as usual.

[0089] Step S132, use a long short-term memory network to perform time-dependence correction on the initial path demand prediction result to generate a corrected cargo fresh-keeping demand prediction value, and the corrected cargo fresh-keeping demand prediction value includes the distribution of fresh-keeping resource requirements for different transportation nodes in the transportation path.

[0090] For strawberry transportation, due to its strict freshness time limit, the freshness drops rapidly with time. The long short-term memory network considers this time dependence to adjust the initial prediction result. For example, in the initial prediction, for the transportation from the transfer and distribution center D to the retail supermarket E, the estimation of fresh-keeping resource requirements is based on a relatively short transportation duration and low temperature fluctuation risk. However, the long short-term memory network considers the cumulative freshness time starting from the picking farm A. If a large amount of freshness time has been consumed when reaching D, even if the transportation duration and temperature fluctuation risk from D to E remain unchanged, more fresh-keeping resources are needed.

[0091] The input gate, forget gate, output gate, and memory cell of the long short-term memory network play their roles. Taking node D as an example, the input gate combines the prediction result of the initial path demand of the current node (such as the estimated fresh-keeping resource demand based on the short duration and low temperature fluctuation risk from D to E) and the hidden state transmitted by the previous node C (including information related to the fresh-keeping resource demand during the transportation process of node C and before, and affected by the feature values at that time), and obtains the input gate vector through weight matrix operation, determining the degree to which new information enters the memory cell. The forget gate also operates with the input information to obtain the forget gate vector, determining the degree to which the old information in the memory cell is retained. For example, if the environment during the previous transportation process was relatively stable and the demand for temperature regulation resources was stable, but the eigenvalue of environmental stability at node D becomes worse, the forget gate may reduce the retention degree of the old information of the corresponding temperature regulation resource demand. The memory cell is updated by combining the results of the input gate and the forget gate, remembering the long-term dependence relationship of the fresh-keeping resource demand during the transportation process, and dynamically adjusting according to the changes of each feature value. The output gate operates with the state of the memory cell to obtain the output gate vector, determining the output hidden state, which integrates the information of each node and the influence of the change of feature values.

[0092] After correction, the generated corrected predicted value of the cargo fresh-keeping demand includes the distribution of fresh-keeping resource demands at different transportation nodes. At the picking farm A, as the starting point of transportation, the fresh-keeping resource demand mainly ensures that strawberries quickly enter a suitable low-temperature environment, and the demand ratio is about 20%; at the collection center B, in addition to maintaining low temperature, sorting and packaging are also required, and the demand ratio increases to 30%; at the regional cold-chain warehouse C, due to the possible long-term storage waiting for transfer, the demand ratio reaches 40%; at the transit distribution center D, considering transportation losses and remaining fresh-keeping time, the demand ratio is raised to 50%; at the retail supermarket E, to ensure the freshness on the shelves, the demand ratio is maintained at about 30%.

[0093] For beef transportation, the long short-term memory network also performs time-dependent correction. Although the spoilage speed of beef is relatively slower than that of strawberries, long-term transportation is also affected by time. For example, in the initial prediction of the transportation from the cold-chain logistics hub H to the meat wholesaler I, the estimated fresh-keeping resource demand may not fully consider the cumulative transportation time starting from the slaughterhouse F.

[0094] The long short-term memory network works according to a similar mechanism. Taking the H node as an example, the input gate, forget gate, memory cell, and output gate work together to adjust the fresh-keeping resource requirements based on the changes in the characteristic values of time sensitivity, environmental stability, and the risk of goods spoilage at each node during the transportation process. The corrected predicted value of the fresh-keeping demand for goods includes the distribution of fresh-keeping resource requirements at each node. At the slaughterhouse F, the fresh-keeping resource requirements mainly ensure that beef quickly enters a low-temperature environment of -18 degrees Celsius, with a demand ratio of approximately 15%; at the meat processing factory G, in addition to maintaining low temperature, processing is also required, and the demand ratio increases to 20%; at the cold chain logistics hub H, due to the possible long-term storage and waiting for transfer, the demand ratio reaches 25%; at the meat wholesaler I, considering transportation losses and the remaining fresh-keeping time, the demand ratio is raised to 30%; at the restaurant J, to ensure freshness before processing, the demand ratio remains at around 20%.

[0095] Step S133, based on the sample traffic flow data and weather event records of the target area, construct a path congestion probability prediction sub-model. The path congestion probability prediction sub-model outputs the path congestion probability distribution for the future time period by analyzing the correlation between weather events and traffic flow changes.

[0096] Then, construct a path congestion probability prediction sub-model based on the sample traffic flow data and weather event records of the target area (such as a certain city and its surrounding areas). For strawberry transportation, the traffic flow data of the target area shows that during the morning and evening rush hours on weekdays, the roads in the city center area will be congested. The weather event records indicate that during heavy rain in summer and snowfall in winter, the traffic capacity of the roads will be greatly reduced. The path congestion probability prediction sub-model outputs the path congestion probability distribution for the future time period by analyzing the correlation between weather events and traffic flow changes. For example, in the transportation path from the regional cold chain warehouse C to the transit distribution center D, if passing through the city center area, during the morning rush hour on weekdays (7 - 9 o'clock), the congestion probability may reach 30%; if encountering heavy rain, this congestion probability may further increase to 50% because heavy rain will cause road flooding, slow down the vehicle driving speed, and even lead to traffic paralysis in some local sections.

[0097] For beef transportation, similarly construct a path congestion probability prediction sub-model. The traffic flow data of the target area shows that during holidays, the roads leading to commercial areas and residential areas will be congested. The weather event records indicate that during foggy weather, the visibility of the roads decreases and the traffic flow will be restricted. In the transportation path from the meat wholesaler I to the restaurant J, if passing through the commercial area, during the afternoon of holidays (14 - 16 o'clock), the congestion probability may reach 20%; if encountering foggy weather, the congestion probability may increase to 30% because in foggy weather, drivers will reduce their speed for safety, resulting in a decrease in the traffic capacity of the roads.

[0098] Step S134: Integrate the corrected predicted value of the goods fresh-keeping demand and the path congestion probability distribution to generate the joint output of the cold-chain demand prediction model, where the joint output is used to identify the fresh-keeping resource occupancy rate and congestion delay risk of different candidate transportation paths.

[0099] Finally, integrate the corrected predicted value of the goods fresh-keeping demand for strawberry and beef transportation and the path congestion probability distribution to generate the joint output of the cold-chain demand prediction model. This joint output is used to identify the fresh-keeping resource occupancy rate and congestion delay risk of different candidate transportation paths. For example, for a candidate transportation path A - B - C - D - E for strawberry transportation, according to the corrected predicted value of the goods fresh-keeping demand, the distribution of fresh-keeping resource requirements at each node determines the fresh-keeping resource occupancy rate of the entire path. If during the transportation from C to D, due to the high demand for fresh-keeping resources and a relatively high road congestion probability (such as 50%), then the congestion delay risk of this path will increase. The fresh-keeping resource occupancy rate may reach 70%, and the congestion delay risk is 30%. For a candidate transportation path F - G - H - I - J for beef transportation, determine the fresh-keeping resource occupancy rate according to the corrected predicted value of the goods fresh-keeping demand, and then combine it with the path congestion probability distribution. If during the transportation from H to I, the road congestion probability is 30%, then the fresh-keeping resource occupancy rate of this path may be 60%, and the congestion delay risk is 20%. Through this joint output, different candidate transportation paths can be comprehensively evaluated, providing an important basis for subsequent selection of the optimal transportation path, ensuring that strawberries and beef can meet the fresh-keeping requirements during transportation and minimizing the impact of congestion delay as much as possible.

[0100] In a possible implementation manner, step S140 includes:

[0101] Step S141: Calculate the expected total transportation duration of each candidate transportation path according to the node sequence length of the transportation path, the sample average passing speed, and the path congestion probability distribution.

[0102] Step S142: Compare the expected total transportation duration with the goods fresh-keeping duration limit to determine the timeliness achievement rate score, where the timeliness achievement rate score is an inverse proportional function of the negative deviation degree of the expected total transportation duration relative to the fresh-keeping duration limit.

[0103] Step S143: Based on the distribution of fresh-keeping resource requirements corresponding to the nodes in the goods fresh-keeping demand prediction value, combined with the current available capacity of refrigeration equipment and the energy supply status, calculate the fresh-keeping reliability score, where the fresh-keeping reliability score is the weighted average of the node resource matching degree and the temperature control ability.

[0104] Step S144: Dynamically assign weights to the freshness preservation reliability score and the timeliness achievement rate score to generate a comprehensive priority ranking for each candidate transportation route.

[0105] For example, for strawberry transportation, the starting point and the ending point of the transportation are from the picking farm to the retail supermarket. First, calculate the expected total transportation duration of each candidate transportation route based on the node sequence length of the transportation route, the sample average passing speed, and the path congestion probability distribution. Suppose there are the following three candidate transportation routes: Route 1 is picking farm - collection center - regional cold chain warehouse - transfer distribution center - retail supermarket, and the node sequence length of this route is 5 nodes; Route 2 is picking farm - collection center - transfer distribution center - retail supermarket, and the node sequence length is 4 nodes; Route 3 is picking farm - regional cold chain warehouse - retail supermarket, and the node sequence length is 3 nodes.

[0106] In terms of the sample average passing speed, the sample average passing speed from the picking farm to the collection center is 60 km / h, and the distance is 120 km; the average speed from the collection center to the regional cold chain warehouse is 50 km / h, and the distance is 100 km; the average speed from the regional cold chain warehouse to the transfer distribution center is 40 km / h, and the distance is 80 km; the average speed from the transfer distribution center to the retail supermarket is 30 km / h, and the distance is 60 km. Then, combined with the path congestion probability distribution, for example, the congestion probability from the collection center to the regional cold chain warehouse in Route 1 is 30%, which will reduce the actual passing speed to 35 km / h (assuming calculated according to the congestion impact model on speed), and the congestion probability from the regional cold chain warehouse to the transfer distribution center is 20%, and the actual speed becomes 32 km / h.

[0107] Calculate the expected total transportation duration of each route based on the above data. Route 1: (120÷60 + 100÷50 + 80÷35 + 60÷32)≈2 + 2 + 2.29 + 1.88 = 8.17 hours; Route 2: (120÷60 + 100÷35 + 60÷30)≈2 + 2.86 + 2 = 6.86 hours; Route 3: (120÷50 + 60÷30)≈2.4 + 2 = 4.4 hours.

[0108] Compare the expected total transportation duration with the freshness preservation duration limit (48 hours) of strawberries to determine the timeliness achievement rate score. Since the timeliness achievement rate score is an inverse proportional function of the negative deviation degree of the expected total transportation duration relative to the freshness preservation duration limit, let the timeliness achievement rate score be S, the freshness preservation duration limit be T, and the expected total transportation duration be t. Then S = k÷(T - t) (k is a constant, and here k = 100 is assumed for easy calculation and explanation). For Route 1, S1 = 100÷(48 - 8.17) ≈ 2.5; for Route 2, S2 = 100÷(48 - 6.86) ≈ 2.3; for Route 3, S3 = 100÷(48 - 4.4) ≈ 2.2.

[0109] Based on the distribution of freshness preservation resource requirements corresponding to the nodes in the strawberry cargo freshness preservation demand prediction value, combined with the current available capacity of refrigeration equipment and the energy supply status, calculate the freshness preservation reliability score. Assume that in Route 1, the predicted value of freshness preservation resource requirements at the consolidation center is 30%, while the current available capacity of refrigeration equipment at the consolidation center is 50%, and the energy supply status is good, and the temperature control ability can meet the requirements of strawberries at 0 - 4 degrees Celsius. Let the weight of node resource matching degree be 0.6 and the weight of temperature control ability be 0.4. Then the freshness preservation reliability score of the consolidation center is (0.6×(50÷30)+0.4×1)=1.4 (here 1 is assumed to represent fully meeting the temperature control requirements). Calculate the freshness preservation reliability scores of other nodes in the route in the same way, and then obtain the freshness preservation reliability score of Route 1 by weighted average. Calculate the freshness preservation reliability scores of Route 2 and Route 3 in the same way.

[0110] Finally, perform dynamic weight allocation on the freshness preservation reliability score and the timeliness achievement rate score to generate the comprehensive priority ranking of each candidate transportation route. Assume that the weight of the freshness preservation reliability score is 0.6 and the weight of the timeliness achievement rate score is 0.4. For Route 1, the comprehensive priority is 0.6×freshness preservation reliability score + 0.4×2.5; for Route 2, the comprehensive priority is 0.6×freshness preservation reliability score + 0.4×2.3; for Route 3, the comprehensive priority is 0.6×freshness preservation reliability score + 0.4×2.2. Determine the comprehensive priority ranking of the three candidate transportation routes by comparing the values of these comprehensive priorities.

[0111] For beef transportation, the starting point and the ending point are from the slaughterhouse to the restaurant. There are also candidate transportation routes. For example, Route 1 is Slaughterhouse - Meat Processing Plant - Cold Chain Logistics Hub - Meat Wholesaler - Restaurant; Route 2 is Slaughterhouse - Meat Processing Plant - Meat Wholesaler - Restaurant; Route 3 is Slaughterhouse - Cold Chain Logistics Hub - Restaurant. According to the calculation method of strawberry transportation mentioned above, calculate the expected total transportation duration based on the distance of each section, the average passing speed of the sample, and the probability distribution of route congestion. For example, the average speed from the slaughterhouse to the meat processing plant is 50 km / h, and the distance is 100 km; the average speed from the meat processing plant to the cold chain logistics hub is 40 km / h, and the distance is 80 km, etc. Calculate the expected total transportation duration of each route in combination with the congestion probability. Compare with the cargo freshness preservation duration limit of beef (72 hours) to determine the timeliness achievement rate score.

[0112] Then, based on the distribution of freshness preservation resource requirements corresponding to the nodes in the beef cargo freshness preservation demand prediction value, calculate the freshness preservation reliability score in combination with the current available refrigeration equipment capacity and the energy supply status. For example, in Route 1, the predicted value of the freshness preservation resource requirement at the meat processing plant is 20%, the current available refrigeration equipment capacity is 30%, the energy supply status is normal, and the temperature control ability meets the requirement of -18 degrees Celsius. Calculate the freshness preservation reliability score of this node according to the weight, and then obtain the freshness preservation reliability score of the entire route. Finally, perform dynamic weight allocation on the freshness preservation reliability score and the timeliness achievement rate score to determine the comprehensive priority ranking of each candidate transportation route, so as to provide a basis for selecting the optimal transportation route for beef transportation, ensuring that beef can meet the freshness preservation requirements and arrive at the destination on time during transportation.

[0113] In a possible implementation manner, step S150 includes:

[0114] Step S151, monitoring the real-time temperature data, traffic status update information, and abnormal events of refrigeration equipment operation at each node during transportation.

[0115] In this embodiment, for strawberry transportation, it is necessary to strictly monitor the real-time temperature data, traffic status update information, and abnormal operation events of refrigeration equipment at each node during transportation. During the transportation from the picking farm to the consolidation center, the temperature data inside the carriage is collected in real time through temperature sensors installed on the transport vehicle. Assuming that the preset temperature threshold for strawberries is 0-4 degrees Celsius, if it is detected that the temperature in a certain area inside the carriage reaches 5 degrees Celsius for 5 consecutive minutes during transportation, this is the continuous deviation of the real-time temperature data from the preset temperature threshold. Regarding the traffic status update information, the transport vehicle can obtain the real-time road conditions through the connection with the traffic management system. For example, during the transportation from the consolidation center to the regional cold chain warehouse, a notice sent by the traffic management system is received that there is congestion on the road ahead due to a traffic accident, and this is the traffic status update information. Regarding the abnormal operation events of the refrigeration equipment, by monitoring the operating parameters of the refrigeration equipment, such as detecting that the operating frequency of the refrigeration compressor suddenly decreases, this may indicate that there is a malfunction or abnormal operation of the refrigeration equipment.

[0116] Step S152: According to the continuous deviation of the real-time temperature data from the preset temperature threshold, adjust the residence time limit of the subsequent nodes and recalculate the freshness reliability score.

[0117] In the case of the above temperature deviation, since the increase in temperature may accelerate the deterioration of strawberries, in order to reduce the in-transit time, it is necessary to adjust the residence time limit of the subsequent nodes from the consolidation center to the regional cold chain warehouse and from the regional cold chain warehouse to the transfer and distribution center. Originally, it was planned to stay for 2 hours at the consolidation center, but now it is adjusted to 1.5 hours. When recalculating the freshness reliability score, it is necessary to re-evaluate factors such as the resource matching degree of each node. For example, at the consolidation center, due to the shortened residence time, more freshness preservation resources may need to be provided in a short time, and the available capacity of the refrigeration equipment and the energy supply status at the current consolidation center remain unchanged, then the freshness reliability score is recalculated according to the new situation.

[0118] Step S153: When it is detected that the traffic status update information indicates that the path congestion probability exceeds the preset risk threshold, trigger the process of regenerating the candidate transport path and update the path congestion probability distribution based on the latest data.

[0119] Assume that the preset risk threshold for the path congestion probability from the regional cold chain warehouse to the transfer and distribution center is 30%. If the traffic management system notifies that the congestion probability of this section has reached 40% due to a traffic accident, the process of regenerating the candidate transportation path is triggered. When regenerating the candidate transportation path, it is necessary to comprehensively consider the current traffic conditions, the locations of each node, and the distribution of refrigeration equipment. For example, the original transportation path is the picking farm - the consolidation center - the regional cold chain warehouse - the transfer and distribution center - the retail supermarket. Now the regenerated candidate path may be the picking farm - the consolidation center - another regional cold chain warehouse - the transfer and distribution center - the retail supermarket. This other regional cold chain warehouse may be located in an area near the original regional cold chain warehouse with better traffic conditions. At the same time, update the path congestion probability distribution based on the latest traffic data, weather data, etc. For example, the congestion probability from the consolidation center to another regional cold chain warehouse in the new path is expected to be 10% according to the real-time traffic flow and road conditions.

[0120] Step S154, feedback the adjusted node sequence, real-time temperature data, and traffic status update information to the cold chain demand prediction model, and iteratively optimize the accuracy of the predicted value of the goods freshness preservation demand.

[0121] For example, the new transportation path node sequence (such as the picking farm - the consolidation center - another regional cold chain warehouse - the transfer and distribution center - the retail supermarket), real-time temperature data (such as the actual temperature changes at each node during transportation), and traffic status update information (such as the changes in congestion probability of each section) can be fed back into the cold chain demand prediction model. The model re-analyzes the dynamic association pattern between transportation nodes based on these new data and corrects the previous predicted value of the goods freshness preservation demand. For example, previously it was predicted that 30% of the refrigeration resources were needed at the transfer and distribution center to maintain the freshness of strawberries. According to the new data, it may be adjusted to 35% to more accurately reflect the actual transportation situation, thereby improving the accuracy of the predicted value of the goods freshness preservation demand.

[0122] Step S155, generate a path adjustment instruction according to the iterated prediction result. The path adjustment instruction includes operation commands such as adding standby nodes, skipping significant risk nodes, or enabling emergency refrigeration resources.

[0123] For example, if the predicted result after iteration shows an increase in the fresh-keeping resource demand at a certain node, and the current operating condition of the refrigeration equipment is poor, such as a decrease in refrigeration effect, corresponding path adjustment instructions need to be generated. If it is detected at the transfer and distribution center that the refrigeration effect of the refrigeration equipment cannot meet the demand, the path adjustment instructions may include adding a standby node, such as finding a temporary storage point with sufficient refrigeration resources near the transfer and distribution center as a standby node; or skipping significantly risky nodes. If the refrigeration equipment at a certain node fails and cannot be repaired in time, directly skip this node and transport the strawberries to the next node; or enabling emergency refrigeration resources, such as adding temporary refrigeration equipment to the transport vehicle to meet the fresh-keeping requirements.

[0124] For beef transportation, the same process applies. During transportation, real-time temperature data (the preset temperature threshold for beef is below -18 degrees Celsius), traffic status update information (such as congestion caused by road construction, traffic accidents, etc.), and abnormal events in the operation of refrigeration equipment (such as abnormal refrigerant pressure in the refrigeration equipment) are monitored at each node from the slaughterhouse to the meat processing plant, from the meat processing plant to the cold chain logistics hub, etc. The residence time limit of the subsequent nodes is adjusted according to the deviation between the real-time temperature data and the preset temperature threshold, and the fresh-keeping reliability score is recalculated. When the traffic status update information indicates that the path congestion probability exceeds the preset risk threshold, a candidate transportation path is regenerated and the path congestion probability distribution is updated. The adjusted relevant data is fed back to the cold chain demand prediction model for iterative optimization. Finally, appropriate path adjustment instructions are generated based on the predicted result after iteration and abnormal events in the operation of refrigeration equipment. For example, during the transportation from the meat wholesaler to the restaurant, if the refrigeration equipment fails, a standby refrigerated storage point may be added, or this node may be skipped and the beef directly transported to the restaurant, and emergency refrigeration resources are enabled to ensure the freshness of the beef. Through these operations, the fresh-keeping effect and transportation timeliness of strawberries and beef during transportation can be ensured, while continuously optimizing the transportation path decision-making.

[0125] In a possible implementation manner, step S151 may include:

[0126] Step S1511, real-time collect the temperature distribution data inside the carriage through the Internet of Things sensors deployed on the transportation carrier, and identify the temperature abnormal fluctuation pattern of the temperature distribution data inside the carriage.

[0127] In the scenario of fresh food cold chain transportation, taking the transportation of strawberries and beef as examples, during the transportation of strawberries, from the picking farm to the collection center, multiple Internet of Things sensors are deployed on the transport vehicle. These sensors can accurately collect the temperature data at different positions inside the carriage, forming the temperature distribution data inside the carriage. For example, the sensors collect data every 5 minutes, and the temperature data at the front, middle, and rear of the carriage are all recorded. Under normal circumstances, the fresh-keeping temperature requirement for strawberries is 0-4 degrees Celsius. When the data collected by the sensors shows that, within a certain period of time, the temperature in the middle of the carriage gradually rises from 3 degrees Celsius to 5 degrees Celsius and then drops to 4 degrees Celsius, this fluctuation pattern is regarded as an abnormal temperature fluctuation pattern. For the transportation of beef, from the slaughterhouse to the meat processing factory, its preset fresh-keeping temperature is below -18 degrees Celsius. If the sensors detect that the local temperature inside the carriage rises to -10 degrees Celsius within a short period of time, this is also an abnormal temperature fluctuation pattern. The identification of this abnormal temperature fluctuation pattern is crucial for taking timely measures to protect the quality of the goods.

[0128] Step S1512, obtain the real-time traffic condition data interface of the traffic management system, and obtain road construction events, traffic accident reports, and weather warning signals in the target area as the traffic status update information.

[0129] For example, during the transportation of strawberries, when the transport vehicle travels from the regional cold chain warehouse to the transfer and distribution center, the transportation enterprise obtains the road construction event information in the target area by obtaining the real-time traffic condition data interface of the traffic management system. For example, road construction is underway on a main road in the city where the transfer and distribution center is located, resulting in the closure of some lanes, which will affect the vehicle's passing speed. At the same time, a traffic accident report is also obtained. For example, on the way from the collection center to the regional cold chain warehouse, a two-vehicle rear-end collision occurred on a certain section, causing traffic congestion. In addition, weather warning signals are also obtained. For example, during summer transportation, a high-temperature warning signal is received, which may affect the performance of the refrigeration equipment of the transport vehicle and increase the risk of goods spoilage. For the transportation of beef, during the process from the meat processing factory to the cold chain logistics hub, if a road icing weather warning signal is received, this will affect transportation safety and transportation duration. At the same time, road construction events and past traffic accident reports near the cold chain logistics hub are obtained. All these information are used as traffic status update information for subsequent transportation decision-making adjustments.

[0130] Step S1513, monitor the compressor operation frequency, refrigerant pressure, and energy consumption rate of the refrigeration equipment, generate an equipment health status assessment report, and analyze the abnormal refrigeration equipment operation events existing in the equipment health status assessment report.

[0131] For example, during strawberry transportation, a refrigeration device is installed on the transport vehicle, continuously monitoring the operating frequency of its compressor, refrigerant pressure, and energy consumption rate. For example, under normal circumstances, the operating frequency of the compressor remains at around 3000 revolutions per minute, the refrigerant pressure is within a certain standard range, and the energy consumption rate is relatively stable. If during transportation, it is monitored that the operating frequency of the compressor suddenly drops to 2000 revolutions per minute, the refrigerant pressure is below the lower limit of the standard, and the energy consumption rate increases abnormally, a device health status assessment report will be generated. By analyzing this device health status assessment report, abnormal operating events of the refrigeration device can be detected, which may be problems such as compressor failure or refrigerant leakage. For beef transportation, during the transportation process from the cold chain logistics hub to the meat wholesaler, if the operating frequency of the compressor of the refrigeration device is unstable, the refrigerant pressure fluctuates greatly, and the energy consumption rate is 20% higher than normal, these situations will be recorded in the device health status assessment report, and abnormal operating events of the refrigeration device will be analyzed, which may affect the preservation effect of beef and need to be dealt with in a timely manner.

[0132] Step S1514, when it is detected that the duration of the temperature abnormal fluctuation corresponding to the temperature abnormal fluctuation pattern exceeds the preset tolerance threshold, generate a first-level warning signal and trigger a local path adjustment.

[0133] For example, during strawberry transportation, assume that the preset tolerance threshold is 10 minutes. If during the transportation from the collection center to the regional cold chain warehouse, the temperature abnormal fluctuation pattern in the carriage lasts for 15 minutes, exceeding the preset tolerance threshold. At this time, the system will generate a first-level warning signal. This warning signal will trigger a local path adjustment. For example, originally planned to directly go to the regional cold chain warehouse without stopping on the way, but due to the abnormal temperature fluctuation, it may be temporarily decided to stop at a small refrigeration station on the way to check the cargo status and adjust the refrigeration device to prevent strawberries from spoiling due to temperature fluctuations. For beef transportation, during the process from the slaughterhouse to the meat processing factory, if the abnormal temperature fluctuation lasts for more than 10 minutes, the first-level warning signal will be triggered, and the transportation speed may be adjusted or a short stop may be made at a place with refrigeration facilities nearby to ensure the preservation temperature of the beef.

[0134] Step S1515, when the received weather warning signal includes a red warning for freezing or high temperature, generate a second-level warning signal and force the activation of an alternative transportation path.

[0135] For example, during the transportation of strawberries, if a high-temperature red warning signal is received during transportation in summer, it indicates that the external temperature is extremely high, which may pose a serious threat to the freshness preservation of strawberries. At this time, the system will generate a second-level warning signal and forcibly enable an alternative transportation route. For example, the original transportation route is from the picking farm - the collection center - the regional cold-chain warehouse - the transfer and distribution center - the retail supermarket, and the alternative transportation route may be from the picking farm - another collection center - the alternative regional cold-chain warehouse - the transfer and distribution center - the retail supermarket. This alternative route may pass through areas with relatively lower temperatures or better traffic conditions to ensure the freshness preservation of strawberries in a high-temperature environment. For the transportation of beef, if a freezing red warning signal is received during transportation in winter, since the freezing weather may affect road traffic and the normal operation of refrigeration equipment, the second-level warning signal will be triggered to enable an alternative transportation route, such as from the slaughterhouse - the alternative meat processing factory - the cold-chain logistics hub - the meat wholesaler - the restaurant. The nodes on this alternative route may be more adaptable to the transportation requirements in freezing weather.

[0136] In a possible implementation manner, step S152 may include:

[0137] Step S1521, statistically calculate the cumulative time ratio and the maximum deviation amplitude of the temperature deviation from the threshold value within the current transportation segment.

[0138] For example, during the transportation of strawberries, for the transportation segment from the regional cold-chain warehouse to the transfer and distribution center, the preset temperature threshold is 0 - 4 degrees Celsius. Assume that within this transportation segment, due to some minor malfunctions of the refrigeration equipment, the temperature deviates from the threshold value. Through the analysis of real-time temperature data, it is detected that the cumulative time of temperature deviation from the threshold value is 30 minutes, and the total duration of the transportation segment is 3 hours (180 minutes). Then the cumulative time ratio is 30÷180 = 1 / 6. In terms of the maximum deviation amplitude, the highest temperature reaches 6 degrees Celsius. Compared with the highest threshold value of 4 degrees Celsius, the maximum deviation amplitude is 2 degrees Celsius. For the transportation of beef, for the transportation segment from the cold-chain logistics hub to the meat wholesaler, the preset temperature threshold is below -18 degrees Celsius. If the cumulative time ratio is 1 / 5 and the maximum deviation amplitude is 8 degrees Celsius (rising from -18 degrees Celsius to -10 degrees Celsius), the statistics of these data provide a basis for subsequent calculations.

[0139] Step S1522, calculate the acceleration coefficient of the goods quality decay according to the cumulative time ratio and the maximum deviation amplitude.

[0140] For example, according to a specific calculation formula (assumed to be an empirical formula based on the spoilage characteristics of strawberries and the impact of temperature on strawberries), the acceleration coefficient of cargo quality decay can be calculated using the cumulative time ratio of 1 / 6 and the maximum deviation amplitude of 2 degrees Celsius. For example, this acceleration coefficient of cargo quality decay may be positively correlated with the cumulative time ratio and the maximum deviation amplitude. After calculation, the acceleration coefficient of cargo quality decay is 0.3 (this is just an example value). For beef transportation, according to its corresponding calculation formula, combined with the cumulative time ratio of 1 / 5 and the maximum deviation amplitude of 8 degrees Celsius, the acceleration coefficient of cargo quality decay is calculated to be 0.5 (also an example value). This coefficient reflects the degree of influence of temperature deviation on the rate of cargo quality decline.

[0141] Step S1523: Dynamically shorten the maximum allowable residence time of subsequent transportation nodes based on the acceleration coefficient of cargo quality decay.

[0142] For example, for the subsequent transportation node from the transfer and distribution center to the retail supermarket, the originally planned maximum allowable residence time is 2 hours. Since the previously calculated acceleration coefficient of cargo quality decay is 0.3, according to the pre-set adjustment rule (for example, for every 0.1 increase in the acceleration coefficient of cargo quality decay, the maximum allowable residence time of the subsequent node is shortened by 15%), then the maximum allowable residence time of this node needs to be shortened by 0.3÷0.1×15% = 45%. The adjusted maximum allowable residence time is 2×(1 - 0.45) = 1.1 hours. For beef transportation, for the subsequent transportation node from the meat wholesaler to the restaurant, the originally maximum allowable residence time is 1.5 hours. According to the calculated acceleration coefficient of cargo quality decay of 0.5, in accordance with the corresponding adjustment rule (for example, for every 0.1 increase in the acceleration coefficient of cargo quality decay, the maximum allowable residence time of the subsequent node is shortened by 20%), the adjusted maximum allowable residence time is 1.5×(1 - 0.5÷0.1×0.2) = 0.6 hours.

[0143] Step S1524: Re-evaluate the distribution of fresh-keeping resource requirements for the adjusted node sequence, and update the resource matching degree weight in the fresh-keeping reliability score.

[0144] For example, the adjusted node sequence is picking farm - consolidation center - regional cold chain warehouse - transfer and distribution center - retail supermarket, and the residence time at the transfer and distribution center is shortened. Due to the shortened residence time, in order to ensure the fresh - keeping quality of strawberries, more concentrated fresh - keeping resources need to be provided at the transfer and distribution center. Originally at the transfer and distribution center, the distribution of fresh - keeping resource requirements was 30% for temperature control, 20% for air circulation, etc. Now it may be adjusted to 40% for temperature control, 30% for air circulation, etc. For the resource matching degree weight in the fresh - keeping reliability score, originally the resource matching degree weight was calculated based on the planned resource allocation and available resources. Now, due to the change in the distribution of fresh - keeping resource requirements, it needs to be recalculated. For example, assume that the available refrigeration equipment capacity at the transfer and distribution center is 50%. Previously, the resource matching degree was calculated as 0.6 (30%÷50%) according to the planned resource requirements. Now, the resource matching degree is calculated as 0.8 (40%÷50%) according to the adjusted resource requirements, thereby updating the resource matching degree weight in the fresh - keeping reliability score. For beef transportation, after the node adjustment from meat wholesaler to restaurant, the distribution of fresh - keeping resource requirements is re - evaluated, such as allocating more resources to temperature control to cope with the fresh - keeping pressure brought by the shortened residence time, and at the same time updating the resource matching degree weight in the fresh - keeping reliability score.

[0145] Step S1525, when the goods quality decay acceleration coefficient exceeds the critical value, trigger an emergency unloading instruction and initiate a quality remedy plan at the nearest node.

[0146] For example, assume that the critical value of the goods quality decay acceleration coefficient is 0.5. If within a certain transportation section, due to a serious failure in temperature control, the calculated goods quality decay acceleration coefficient reaches 0.6, exceeding the critical value. At this time, the system will trigger an emergency unloading instruction. For example, at the nearest transfer and distribution center, the vehicle will go there and unload as soon as possible. A quality remedy plan is initiated at the transfer and distribution center, which may include quickly cooling the strawberries, checking the spoilage situation of the strawberries, repackaging the unspoiled strawberries and transferring them to a standby refrigeration equipment for continued transportation. For beef transportation, if the goods quality decay acceleration coefficient exceeds the critical value, an emergency unloading instruction will also be triggered, and a quality remedy plan will be initiated at the nearest node (such as a meat wholesaler), such as inspecting the beef and readjusting the refrigeration environment, etc.

[0147] In a possible implementation manner, step S153 may include:

[0148] Step S1531, based on the risk node positions and influence radii in the path congestion probability distribution corresponding to the traffic status update information, determine the geographical boundary range of the detour area.

[0149] For example, during the transportation of strawberries, on the transportation route from the regional cold chain warehouse to the transfer and distribution center, according to the updated traffic status information, it is detected that the probability of path congestion on a certain section exceeds the preset risk threshold (assumed to be 30%) due to a traffic accident. In the path congestion probability distribution, the location of the risk node is determined to be the accident point, and the influence radius is assumed to be 10 kilometers (determined according to factors such as traffic flow and road type). Then, the geographical boundary range of the detour area is a circular area with the accident point as the center and a radius of 10 kilometers. For the transportation of beef, on the path from the meat processing factory to the cold chain logistics hub, if the congestion probability exceeds the preset risk threshold due to road construction, the geographical boundary range of the detour area is determined according to the location of the risk node (the road construction section) and the influence radius (such as 8 kilometers).

[0150] Step S1532: Based on the geographical boundary range, screen alternative nodes with cold storage facility filings, and extract the compatibility data between the cold storage equipment types of the alternative nodes and the currently transported goods.

[0151] For example, within the determined geographical boundary range of the detour area, by querying the cold storage facility filing database, alternative nodes with cold storage facilities are screened out. For example, several small cold storage warehouses and some logistics sites with cold storage equipment are screened out. Then, the compatibility data between the cold storage equipment types of these alternative nodes and the currently transported strawberries is extracted. If the temperature control range of the cold storage equipment in a certain small cold storage warehouse is -5 to 5 degrees Celsius, and the fresh-keeping temperature requirement for strawberries is 0 to 4 degrees Celsius, the compatibility data needs to evaluate whether this temperature range can meet the fresh-keeping needs of strawberries, and it may be concluded that the compatibility is 80% (according to certain evaluation criteria). For the transportation of beef, after screening out alternative nodes within the detour area, if the temperature of the cold storage equipment of the alternative node is -20 to -10 degrees Celsius, the compatibility data is evaluated according to the fresh-keeping requirement of beef below -18 degrees Celsius.

[0152] Step S1533: Combining the resource demand distribution in the predicted value of the goods fresh-keeping demand, match alternative nodes with qualified compatibility data to generate emergency path branches, and the emergency path branches include the main detour path, the temporary transfer path, and the multimodal transport connection path.

[0153] For example, according to the resource demand distribution in the predicted value of the goods fresh-keeping demand, 30% of the refrigeration resources may be required in the transfer and distribution center. From the alternative nodes screened out previously, select those nodes with qualified compatibility data. For example, there is a small refrigerated warehouse, and its temperature control range and capacity of the refrigeration equipment meet the fresh-keeping requirements of strawberries, so it can be included in the emergency path branch. The emergency path branch includes bypassing the main path, such as bypassing the congested section from the regional cold chain warehouse, passing through this small refrigerated warehouse and then arriving at the transfer and distribution center; the temporary transfer path, if the goods need to be temporarily transferred in this small refrigerated warehouse, such as repackaging, adjusting the refrigeration equipment, etc.; the multimodal transport connection path, if the transport mode needs to be converted in this area, such as converting from road transport to rail transport, corresponding connection arrangements can also be made. For beef transportation, similarly, combined with the resource demand distribution in the predicted value of the goods fresh-keeping demand, match the alternative nodes to generate the emergency path branch to ensure that the beef can be transported smoothly and kept fresh when encountering traffic congestion.

[0154] Step S1534, according to the goods temperature stability index collected in real time, exclude the branches that exceed the remaining fresh-keeping duration constraint from the emergency path branches, and generate an effective emergency path subset.

[0155] For example, the goods temperature stability index collected in real time shows the current temperature state and the remaining fresh-keeping duration of strawberries. If in a certain emergency path branch, due to too long a detour distance or too many transfer links, the expected arrival time at the retail supermarket will exceed the remaining fresh-keeping duration of strawberries, then this branch will be excluded. For example, there is an emergency path branch that can bypass the congested section, but due to passing through multiple temporary transfer points, the total transportation time is expected to exceed the remaining fresh-keeping duration of strawberries, so it is excluded. After such screening, an effective emergency path subset is generated. For beef transportation, according to the temperature stability index and the remaining fresh-keeping duration of beef, exclude the branches that do not meet the requirements from the emergency path branches to obtain an effective emergency path subset.

[0156] Step S1535, input the effective emergency path subset into the cold chain demand prediction model, reorder according to the dynamic weights of the fresh-keeping reliability score and the timeliness achievement rate score, and output the target emergency path with the highest priority.

[0157] For example, the subset of effective emergency paths is input into the cold chain demand prediction model. The cold chain demand prediction model re-ranks these paths according to the dynamic weights of the previously calculated freshness preservation reliability scores (based on the matching degree of refrigeration equipment resources and temperature control capabilities of each node) and timeliness achievement rate scores (according to the expected transportation duration and the freshness preservation duration limit of strawberries). For example, one emergency path has a relatively high freshness preservation reliability score but a slightly lower timeliness achievement rate score, and another path has a high timeliness achievement rate score but a slightly lower freshness preservation reliability score. The comprehensive score is calculated according to the set dynamic weights (such as the weight of the freshness preservation reliability score is 0.6 and the weight of the timeliness achievement rate score is 0.4). After comparison, the target emergency path with the highest priority is output, and this target emergency path will be used for actual transportation adjustment. For beef transportation, similarly, the subset of effective emergency paths is input into the cold chain demand prediction model, re-ranked according to the dynamic weights of the freshness preservation reliability score and the timeliness achievement rate score, and the target emergency path is determined.

[0158] In a possible implementation manner, step S154 may include:

[0159] Step S1541, perform a difference analysis on the actual transportation result of the target emergency path and the predicted freshness preservation reliability score to generate a path decision deviation index.

[0160] For example, in strawberry transportation, assume the target emergency path is from the regional cold chain warehouse - small refrigerated warehouse - transfer distribution center - retail supermarket. During the actual transportation process, there are differences between the arrival times at each node, the actual freshness preservation resource usage at each node, etc. and the data on which the predicted freshness preservation reliability score is based. For example, it is predicted that 30% of the refrigeration resources are required in the small refrigerated warehouse to maintain the freshness of strawberries, but actually 35% is used. Through the comparison between the actual transportation result and the prediction, the deviation value of each node is calculated, and then the path decision deviation index is comprehensively generated. For beef transportation, during the actual transportation of the target emergency path, compare the actual freshness preservation resource usage, transportation duration, etc. with the predicted situation to generate a path decision deviation index.

[0161] Step S1542, adjust the convolutional kernel size of the temporal convolutional neural network according to the path decision deviation index to enhance the local feature extraction ability for sudden congestion events.

[0162] For example, if the path decision deviation index shows that the model has insufficient ability to extract local features when dealing with sudden congestion events (such as congestion caused by traffic accidents). According to this deviation index, adjust the convolution kernel size of the temporal convolutional neural network. For example, if the deviation index indicates that the node temperature control and transportation duration prediction near the congested section are inaccurate, the convolution kernel size may be increased to better capture local features such as temperature changes and residence duration of these nodes under sudden congestion events. For beef transportation, also adjust the convolution kernel size of the temporal convolutional neural network according to the path decision deviation index to improve the ability to respond to sudden congestion events.

[0163] Step S1543, optimize the forgetting gate threshold of the time-dependency correction in the long short-term memory network based on the cargo quality feedback data after the execution of the target emergency path.

[0164] For example, after the execution of the target emergency path, according to the cargo quality feedback data when arriving at the retail supermarket, such as the freshness and spoilage rate of strawberries. If it is detected that the spoilage rate of strawberries is higher than expected, it indicates that the correction of time-dependency may not be accurate enough during transportation. At this time, by analyzing the cargo quality feedback data, optimize the forgetting gate threshold of the time-dependency correction in the long short-term memory network. For example, if the high spoilage rate of strawberries is due to insufficient emphasis on time in the later stage of transportation, the forgetting gate threshold may be lowered to make the long short-term memory network more strongly emphasize the impact of time factors on cargo quality in subsequent predictions. For beef transportation, optimize the forgetting gate threshold in the long short-term memory network according to the cargo quality feedback data when arriving at the restaurant.

[0165] Step S1544, within a preset period, integrate the adjusted node sequence, the real-time temperature data, the traffic state update information, and the abnormal events of the refrigeration equipment operation into an incremental training dataset according to timestamps.

[0166] For example, in strawberry transportation, assume that the preset cycle is daily. During a day of transportation, multiple adjusted node sequences will be generated. For example, the original transportation path node sequence when leaving the picking farm in the morning is picking farm - collection center - regional cold chain warehouse - transfer and distribution center - retail supermarket. During transportation, due to temperature fluctuations and traffic congestion, etc., the adjusted node sequence may become picking farm - collection center - backup regional cold chain warehouse - transfer and distribution center - retail supermarket. At the same time, the real-time temperature data collected at each transportation node, such as the temperature at the collection center is 2 degrees Celsius, the temperature at the backup regional cold chain warehouse is 3 degrees Celsius, etc.; traffic status update information, including road construction, traffic accidents, and weather warnings, etc.; abnormal events in the operation of refrigeration equipment, such as the compressor operation frequency of the refrigeration equipment decreases during transportation, etc., are all integrated according to the time stamp. The time stamp is accurate to the minute, so that the sequence of events can be clearly reflected. Integrating these data forms an incremental training data set. For beef transportation, with the daily preset cycle as well, relevant data during transportation are integrated to form an incremental training data set, including relevant information of each node from the slaughterhouse to the meat processing plant, cold chain logistics hub, etc.

[0167] Step S1545, inject the incremental training data set into the cold chain demand prediction model through an online learning algorithm, update the model weight information of the cold chain demand prediction model, and retain the high-value patterns in the historical path decisions.

[0168] For example, the online learning algorithm will adjust the parameters of the model according to the new incremental training data set. For example, for the temporal convolutional neural network part, according to the new node sequence and temperature data, adjust the connection weights between neurons to better extract the dynamic association patterns between transportation nodes. For the long short-term memory network part, according to the new time series data (including the time series of traffic status update information and abnormal events in the operation of refrigeration equipment), adjust the weights of the forget gate, input gate, and output gate to optimize the time-dependent correction. While updating the model weight information, the high-value patterns in the historical path decisions will also be retained. For example, a transportation path decision from the picking farm to the retail supermarket in a certain season before has achieved good fresh-keeping effects and timeliness achievement rates in multiple transports, and this pattern will be retained. For beef transportation, inject the incremental training data set into the cold chain demand prediction model through the online learning algorithm, similarly update the model weight information, and retain the valuable historical path decision patterns, so as to improve the accuracy of the model's prediction of the fresh-keeping demand for beef transportation and path decision-making.

[0169] Among them, step S155 includes:

[0170] Step S1551: Analyze the refrigeration resource occupancy rate and energy supply time window of each alternative node in the target emergency path.

[0171] For example, in strawberry transportation, the target emergency path is from the regional cold chain warehouse - small refrigerated warehouse - transfer distribution center - retail supermarket. For the small refrigerated warehouse as an alternative node, analyze its refrigeration resource occupancy rate. Suppose the total refrigeration capacity of the small refrigerated warehouse is 100 cubic meters and 30 cubic meters are currently occupied, then the refrigeration resource occupancy rate is 30%. In terms of the energy supply time window, the energy supply of the small refrigerated warehouse is through power supply, and there is a fixed power maintenance time every day from 2 to 3 am, and this time period is the energy supply time window. For beef transportation, for the alternative nodes in the target emergency path, also analyze the refrigeration resource occupancy rate and energy supply time window. For example, in a certain alternative meat refrigerated warehouse, the refrigeration resource occupancy rate is 40% and the energy supply time window is from 10 pm to 11 pm.

[0172] Step S1552: Calculate the feasibility probability of the carrier reaching each alternative node based on the remaining cruising range of the transportation carrier and the real-time power consumption of the refrigeration equipment.

[0173] For example, the transportation carrier is a refrigerated truck with a remaining cruising range of 200 kilometers and a real-time power consumption of the refrigeration equipment of 0.5 liters of diesel per kilometer (hypothetical). The distance from the current location to the small refrigerated warehouse is 150 kilometers. At the normal driving speed, it takes 3 hours to reach the small refrigerated warehouse, and the power consumption of the refrigeration equipment within these 3 hours is 1.5 liters of diesel. Considering possible fuel consumption fluctuations and other factors during the journey, the calculated feasibility probability of the carrier reaching the small refrigerated warehouse is 80% (according to a pre-set calculation model, comprehensively considering factors such as the remaining cruising range, real-time power consumption, and distance). For beef transportation, factors such as the remaining cruising range of the transportation carrier, the real-time power consumption of the refrigeration equipment, and the distance to the alternative node are used to calculate the feasibility probability of reaching each alternative node. For example, from the current location to a certain alternative cold chain logistics hub, the feasibility probability is calculated to be 70% based on relevant data.

[0174] Step S1553: Generate an instruction sequence matching the feasibility probability, where the instruction sequence includes detour turning coordinates, refrigeration power adjustment gradient, and the upper limit of the stay duration at the transfer node.

[0175] For example, if the feasibility probability of the carrier arriving at the small cold storage warehouse is 80%, the generated instruction sequence is as follows. The detour turning coordinates are to turn left at the intersection 50 kilometers away from the current location (determined according to the map coordinates and route planning). The refrigeration power adjustment gradient is to reduce the refrigeration power by 10% within a range of 10 kilometers close to the small cold storage warehouse (because the refrigeration environment of the small cold storage warehouse is better). The upper limit of the residence time at the transfer node is 1 hour (determined according to the cargo freshness preservation requirements and the operation efficiency of the small cold storage warehouse). For beef transportation, the corresponding instruction sequence is generated according to the calculated feasibility probability. For example, when approaching the alternative cold chain logistics hub, the detour turning coordinates, the refrigeration power adjustment gradient, and the upper limit of the residence time at the transfer node are determined to ensure the smooth transportation and freshness preservation of beef.

[0176] Step S1554, after synchronizing the instruction sequence to the path execution terminal of the transport carrier and the refrigeration equipment controller, real-time monitor the execution status data of the instruction sequence. When it is detected that the refrigeration power adjustment gradient does not meet the standard or the turning coordinate deviation exceeds the limit, trigger an artificial intervention instruction and transmit the abnormal data back to the cold chain demand prediction model.

[0177] For example, the path execution terminal navigates according to the detour turning coordinates, and the refrigeration equipment controller adjusts the refrigeration power according to the refrigeration power adjustment gradient. During the transportation process, the execution status data of the instruction sequence is real-time monitored through the sensors installed on the transport carrier. If it is detected that the refrigeration power adjustment gradient does not meet the standard, for example, the refrigeration power should be reduced by 10% but is actually only reduced by 5%, or the turning coordinate deviation exceeds the allowable range, such as deviating from the predetermined turning coordinate by 100 meters, an artificial intervention instruction will be triggered. The transport personnel will receive a notice for manual adjustment. At the same time, the abnormal data (such as the unqualified refrigeration power, the deviated turning coordinates, etc.) will be transmitted back to the cold chain demand prediction model for the model to further optimize and adjust the subsequent predictions and decisions. For beef transportation, the instruction sequence is also synchronized to the relevant equipment of the transport carrier, the execution status data is real-time monitored, and an artificial intervention instruction is triggered and the abnormal data is transmitted back to the cold chain demand prediction model when an abnormality occurs, continuously improving the freshness preservation and transportation efficiency of beef transportation.

[0178] Figure 2 Fig. 100 shows a cold chain transportation service system 100 provided in an embodiment of the present application, including a processor 1001, a memory 1003, and program code stored on the memory 1003. The processor 1001 executes the above program code to implement the steps of the cold chain transportation path optimization method based on AI prediction.

[0179] Figure 2The cold chain transportation service system 100 shown includes: a processor 1001 and a memory 1003. Among them, the processor 1001 and the memory 1003 are connected, such as connected through a bus 1002. Optionally, the cold chain transportation service system 100 may further include a transceiver 1004, and the transceiver 1004 can be used for data interaction between this cold chain transportation service system and other cold chain transportation service systems, such as data sending and / or data receiving, etc. It should be noted that in actual scheduling, the transceiver 1004 is not limited to one, and the structure of this cold chain transportation service system 100 does not constitute a limitation on the embodiments of the present application.

[0180] The processor 1001 can be a CPU (Central Processing Unit, central processor), a general-purpose processor, a DSP (Digital Signal Processor, data signal processor), an ASIC (Application Specific Integrated Circuit, application-specific integrated circuit), an FPGA (Field Programmable Gate Array, field programmable gate array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It can implement or execute various exemplary logic blocks, modules and circuits described in combination with the disclosure of the present application. The processor 1001 can also be a combination that implements computing functions, such as a combination including one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0181] The bus 1002 may include a path for transmitting information between the above components. The bus 1002 can be a PCI (Peripheral Component Interconnect, peripheral component interconnect template) bus or an EISA (Extended Industry Standard Architecture, extended industry template structure) bus, etc. The bus 1002 can be divided into an address bus, a data bus, a control bus, etc.

[0182] The memory 1003 can be a ROM (Read Only Memory), or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory), or other types of dynamic storage devices that can store information and instructions. It can also be an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, other magnetic storage devices, or any other medium that can be used to carry or store program code and can be read by a computer, which is not limited herein.

[0183] The memory 1003 is used to store the program code for implementing the embodiments of this application and is controlled by the processor 1001 for execution. The processor 1001 is used to execute the program code stored in the memory 1003 to implement the steps shown in the foregoing method embodiments.

[0184] The embodiments of this application provide a computer-readable storage medium, on which program code is stored. When the program code is executed by a processor, the steps and corresponding content of the foregoing method embodiments can be implemented.

[0185] It should be understood that although the flowchart of the embodiments of this application indicates each operation step by arrows, the execution order of these steps is not limited to the order indicated by the arrows. Unless otherwise clearly stated in this application, in some implementation scenarios of the embodiments of this application, the implementation steps in each flowchart can be executed in other orders based on requirements. In addition, some or all of the steps in each flowchart may include multiple sub-steps or multiple stages according to the actual implementation scenarios. Some or all of these sub-steps or stages can be executed at the same time, and each sub-step or stage of these sub-steps or stages can also be executed at different times respectively. In the scenario where the execution times are different, the execution order of these sub-steps or stages can be flexibly configured according to requirements, and the embodiments of this application do not limit this.

[0186] The above are only optional implementation manners of some implementation scenarios of this application. It should be noted that for those of ordinary skill in the art, without departing from the technical concept of the solution of this application, using other similar implementation means based on the technical idea of this application also belongs to the protection scope of the embodiments of this application.

Claims

1. A cold chain transportation route optimization method based on AI prediction, characterized in that: The method comprises: Acquire a sample cold chain transport order data set, wherein the sample cold chain transport order data set includes a transport node sequence, cargo preservation requirement parameters, ambient temperature fluctuation records, and transport timeliness completion status of multiple sample orders; Extracting transport features from the sample cold chain transport order data set to generate a combined transport feature set corresponding to each sample order, wherein the combined transport feature set includes a time-sensitive feature, an environmental stability feature, and a cargo deterioration risk feature; Based on the combined transport feature set, a cold chain demand forecasting model is called to generate a forecast value of the cargo preservation demand in the target area within a preset time range, and based on sample traffic flow data and weather event records in the target area, a path congestion probability distribution is generated; According to the predicted value of the cargo preservation demand and the probability distribution of route congestion, multiple candidate transportation routes from the starting point to the end point are generated, and each candidate transportation route is scored for preservation reliability and timeliness achievement rate; Based on the weighted results of the freshness-keeping reliability score and the timeliness achievement rate score, dynamically adjust the node sequence of the current transportation path, and update the input data of the cold chain demand prediction model in real time to optimize subsequent path decisions; The extracting of transport features from the sample cold chain transport order data set to generate a combined transport feature set corresponding to each sample order includes: Extracting the dwell time data, the number of times the ambient temperature deviates from the threshold, and the cargo packaging integrity records of each transport node from the sample cold chain transport order data set; Determine the time-sensitive feature according to the difference between the dwelling time data and the preset cargo freshness-keeping time limit, wherein the time-sensitive feature is used to identify the risk level of exceeding the freshness-keeping time limit in the transport node sequence; Calculating the environmental stability feature according to the number of times the ambient temperature deviates from the threshold and the proportion of the deviation duration, wherein the environmental stability feature is used to characterize the cumulative impact of temperature fluctuations on the quality of the goods during transportation; Combining the cargo packaging integrity record and the deterioration rate parameter corresponding to the cargo type, generating the cargo deterioration risk feature, wherein the cargo deterioration risk feature is used to quantify the contribution of different nodes in the transportation path to the deterioration of cargo quality; The time-sensitive features, environmental stability features and cargo deterioration risk features are normalized to generate the combined transport feature set.

2. The cold chain transportation route optimization method based on AI prediction according to claim 1 is characterized in that: The cold chain demand prediction model is called based on the combined transport feature set to generate a forecast value of the cargo preservation demand in the target area within a preset time range, and a path congestion probability distribution is generated based on sample traffic flow data and weather event records in the target area, including: Inputting the combined transport feature set into a time series convolutional neural network, extracting dynamic correlation patterns between transport nodes, and outputting initial path demand prediction results; Using a long short-term memory network to perform time-dependency correction on the initial path demand forecast result, and generate a revised cargo preservation demand forecast value, wherein the revised cargo preservation demand forecast value includes the distribution of preservation resource demand of different transportation nodes in the transportation path; Based on the sample traffic flow data and weather event records of the target area, a path congestion probability prediction sub-model is constructed, wherein the path congestion probability prediction sub-model outputs the path congestion probability distribution in the future time period by analyzing the correlation between weather events and traffic flow changes; The corrected cargo preservation demand prediction value and the path congestion probability distribution are integrated to generate a joint output of the cold chain demand prediction model, where the joint output is used to identify the preservation resource occupancy rate and congestion delay risk of different candidate transportation paths.

3. The cold chain transportation route optimization method based on AI prediction according to claim 2 is characterized in that: According to the predicted value of the cargo preservation demand and the probability distribution of path congestion, a plurality of candidate transportation paths from the starting point to the end point are generated, and each candidate transportation path is scored for preservation reliability and timeliness achievement rate, including: Calculate the expected total transportation time of each candidate transportation route based on the node sequence length of the transportation route, the sample average travel speed and the congestion probability distribution of the route; Compare the expected total transportation time with the cargo preservation time limit to determine the timeliness achievement rate score, where the timeliness achievement rate score is an inversely proportional function of the negative deviation degree of the expected total transportation time relative to the cargo preservation time limit; Based on the distribution of fresh-keeping resource requirements of the corresponding node in the predicted value of the goods fresh-keeping demand, combined with the current available refrigeration equipment capacity and energy supply status, the fresh-keeping reliability score is calculated, and the fresh-keeping reliability score is a weighted average of the node resource matching degree and the temperature control capability; Dynamically weight the preservation reliability score and the timeliness achievement rate score to generate a comprehensive priority ranking for each candidate transportation route.

4. The cold chain transportation route optimization method based on AI prediction according to claim 3 is characterized in that: The method dynamically adjusts the node sequence of the current transportation path based on the weighted result of the freshness-keeping reliability score and the timeliness achievement rate score, and updates the input data of the cold chain demand prediction model in real time to optimize subsequent path decisions, including: Monitor real-time temperature data, traffic status updates and abnormal operation of refrigeration equipment at each node during transportation; According to the continuous deviation of the real-time temperature data from the preset temperature threshold, adjusting the stay time limit of the subsequent nodes and recalculating the freshness preservation reliability score; When it is detected that the traffic status update information indicates that the path congestion probability exceeds a preset risk threshold, a regeneration process of the candidate transportation path is triggered, and the path congestion probability distribution is updated based on the latest data; Feeding back the adjusted node sequence, real-time temperature data and traffic status update information to the cold chain demand forecasting model, iteratively optimizing the accuracy of the cargo preservation demand forecast value; A path adjustment instruction is generated according to the prediction result after the iteration, and the path adjustment instruction includes an operation command of adding a spare node, skipping a significant risk node, or activating an emergency cold storage resource.

5. The cold chain transportation route optimization method based on AI prediction according to claim 4 is characterized in that: The monitoring of real-time temperature data, traffic status update information and abnormal operation events of refrigeration equipment at each node during transportation includes: The IoT sensors deployed on the transport carrier collect the temperature distribution data inside the carriage in real time, and identify the abnormal temperature fluctuation pattern of the temperature distribution data inside the carriage; Obtaining a real-time traffic data interface of a traffic management system, obtaining road construction events, traffic accident reports and weather warning signals in a target area as the traffic status update information; Monitor the compressor operating frequency, refrigerant pressure and energy consumption rate of the refrigeration equipment, generate an equipment health status assessment report, and analyze abnormal operation events of the refrigeration equipment in the equipment health status assessment report; When it is detected that the duration of abnormal temperature fluctuation corresponding to the abnormal temperature fluctuation pattern exceeds the preset tolerance threshold, a first-level warning signal is generated and a local path adjustment is triggered; When the received weather warning signal contains a red warning for freezing or high temperature, a second-level warning signal is generated and the alternative transportation route is forcibly activated.

6. The cold chain transportation route optimization method based on AI prediction according to claim 5 is characterized in that: The step of adjusting the stay time limit of subsequent nodes and recalculating the freshness-keeping reliability score according to the continuous deviation of the real-time temperature data from the preset temperature threshold comprises: Count the cumulative time percentage and maximum deviation range of the temperature deviation threshold in the current transport section; Calculate the cargo mass decay acceleration factor based on the cumulative time proportion and the maximum deviation amplitude; Based on the cargo mass decay acceleration coefficient, dynamically shorten the maximum allowed stay time at the subsequent transportation node; Re-evaluate the distribution of fresh-keeping resource requirements of the adjusted node sequence, and update the resource matching weight in the fresh-keeping reliability score; When the cargo quality decay acceleration coefficient exceeds a critical value, an emergency unloading instruction is triggered and a quality remediation plan is started at the nearest node.

7. The cold chain transportation route optimization method based on AI prediction according to claim 6 is characterized in that: When it is detected that the traffic status update information indicates that the path congestion probability exceeds a preset risk threshold, a process of regenerating a candidate transportation path is triggered, including: Determine the geographical boundary range of the detour area according to the risk node position and impact radius in the path congestion probability distribution corresponding to the traffic status update information; Screening alternative nodes with refrigerated facilities on file based on the geographic boundary range, and extracting compatibility data between the refrigerated equipment type of the alternative nodes and the currently transported goods; Combined with the resource demand distribution in the cargo preservation demand forecast value, matching the alternative nodes with qualified compatibility data to generate emergency path branches, the emergency path branches include a detour main path, a temporary transit path and a multimodal transport connection path; According to the cargo temperature stability index collected in real time, branches exceeding the remaining fresh-keeping time constraint are excluded from the emergency path branches to generate a valid emergency path subset; The effective emergency path subset is input into the cold chain demand prediction model, and is reordered according to the dynamic weights of the freshness-keeping reliability score and the timeliness achievement rate score, and the target emergency path with the highest priority is output.

8. The cold chain transportation route optimization method based on AI prediction according to claim 7 is characterized in that: Feeding back the adjusted node sequence, real-time temperature data and traffic status update information to the cold chain demand prediction model, iteratively optimizing the accuracy of the cargo preservation demand prediction value, includes: Performing a difference analysis between the actual transportation result of the target emergency route and the predicted freshness-keeping reliability score to generate a route decision deviation index; Adjusting the convolution kernel size of the temporal convolutional neural network according to the path decision deviation index to enhance the local feature extraction capability of sudden congestion events; Based on the cargo quality feedback data after the execution of the target emergency path, optimizing the forget gate threshold of the time-dependent correction in the long short-term memory network; Integrate the adjusted node sequence, the real-time temperature data, the traffic status update information and the abnormal operation events of the refrigeration equipment into an incremental training data set according to timestamps within a preset period; Injecting the incremental training data set into the cold chain demand prediction model through an online learning algorithm, updating the model weight information of the cold chain demand prediction model and retaining high-value patterns in historical path decisions; The step of generating a path adjustment instruction according to the prediction result after iteration includes: Analyze the refrigeration resource occupancy rate and energy replenishment time window of each alternative node in the target emergency path; Calculate the feasibility probability of the carrier arriving at each alternative node based on the remaining range of the transport carrier and the real-time power consumption of the refrigeration equipment; Generate an instruction sequence matching the feasibility probability, the instruction sequence including a detour turning coordinate, a cooling power adjustment gradient, and an upper limit of a stay time at a transfer node; After synchronizing the instruction sequence to the path execution terminal of the transport carrier and the refrigeration equipment controller, the execution status data of the instruction sequence is monitored in real time. When it is detected that the refrigeration power adjustment gradient does not meet the standard or the steering coordinate deviation exceeds the limit, the manual intervention instruction is triggered and the abnormal data is fed back to the cold chain demand prediction model.

9. A cold chain transportation service system, characterized in that: It includes a processor and a computer-readable storage medium, wherein the computer-readable storage medium stores machine-executable instructions, and when the machine-executable instructions are executed by the processor, the cold chain transportation path optimization method based on AI prediction described in any one of claims 1 to 8 is implemented.

Images