Cold-chain food distribution analysis method and system based on big data

Abstract

The invention relates to a cold-chain food distribution analysis method and system based on big data, and the method comprises the steps: carrying out the dynamic feature extraction of multi-source monitoring data of a cold-chain food distribution vehicle, and obtaining and generating transportation state parameters; carrying out correlation matching on the transportation state parameters and real-time environment data of the cold-chain food distribution vehicle to obtain an environment influence evaluation result; carrying out distribution efficiency analysis on the vehicle path transportation data of the cold-chain food distribution vehicle based on the environmental influence evaluation result to form initial distribution efficiency data; road condition diagnosis is carried out on the route transportation data, route pre-judgment is carried out in combination with the environmental influence evaluation result, and a route prediction result is generated; and performing cross analysis on the initial delivery efficiency data and the path prediction result, and formulating a global delivery scheme of the cold-chain food delivery vehicle. According to the invention, logistics resource configuration can be optimized, and the overall operation efficiency of cold-chain logistics is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cold-chain food distribution, in particular to a cold-chain food distribution analysis method and system based on big data. BACKGROUND

[0002] Traditional distribution modes are difficult to effectively integrate multi-dimensional data such as vehicle operating state, environmental temperature and humidity, etc., resulting in difficulty in timely early warning and dynamic adjustment of abnormal situations in the transportation process. The existing analysis method often considers path planning, energy consumption management and temperature control requirements separately, and lacks the ability of collaborative optimization of the whole distribution process. In addition, due to the lack of precise modeling of the dynamic correlation between real-time road conditions, environmental changes and cold-chain equipment performance, the existing distribution scheme is prone to time delay or temperature control failure in actual execution, which not only increases the logistics cost, but also causes food safety hazards. SUMMARY

[0003] The main purpose of the present application is to provide a cold-chain food distribution analysis method and system based on big data, which can optimize the allocation of logistics resources and improve the overall operation efficiency of cold-chain logistics.

[0004] To achieve the above purpose, the present application provides a cold-chain food distribution analysis method based on big data, comprising: Dynamic feature extraction is performed on the multi-source monitoring data of the cold-chain food distribution vehicle to obtain generated transportation state parameters; The transportation state parameters are associated and matched with the real-time environmental data of the cold-chain food distribution vehicle to obtain environmental influence evaluation results; Based on the environmental influence evaluation results, the vehicle path transportation data of the cold-chain food distribution vehicle is executed for distribution efficiency analysis to form initial distribution efficiency data; The path transportation data is diagnosed for road conditions, and the path prediction result is generated by combining the environmental influence evaluation results for path prediction; The initial distribution efficiency data and the path prediction result are cross-analyzed to formulate a global distribution scheme for the cold-chain food distribution vehicle.

[0005] Further, the dynamic feature extraction on the multi-source monitoring data of the cold-chain food distribution vehicle to obtain the generated transportation state parameters comprises: Obtain the vehicle temperature and humidity of the cold-chain food distribution vehicle, perform differential fluctuation detection, and obtain vehicle temperature change data; Obtain the vehicle temperature and humidity of the cold-chain food distribution vehicle, perform differential fluctuation detection, and obtain vehicle temperature change data; Obtain vehicle vibration data and combine the vehicle temperature change data to identify abnormal accumulation amount, and obtain vehicle degradation index; Obtain vehicle load distribution data and perform morphing analysis to obtain cargo morphing distribution data; Obtain vehicle-mounted device energy consumption data and perform power consumption attenuation calculation to obtain device energy efficiency attenuation rate; Feature fusion is performed on the vehicle-mounted temperature change data, the vehicle-mounted degradation index, the cargo morphing distribution data and the device energy efficiency attenuation rate to obtain the transportation state parameter.

[0006] Further, the transportation state parameter is associated with the real-time environmental data of the cold-chain food distribution vehicle to obtain an environmental impact assessment result, including: Obtain vehicle perception data and real-time weather data of the cold-chain food distribution vehicle, and perform environmental association to obtain the real-time environmental data; Perform multi-source heterogeneous protocol conversion and outlier cleaning on the real-time environmental data to obtain baseline environmental data; According to the baseline environmental data and the transportation state parameter, multi-dimensional feature fusion is performed to obtain an environmental state association matrix; Perform influence factor clustering on the environmental state association matrix to obtain an environmental impact factor classification result; According to the environmental impact factor classification result, the transportation state parameter is quantitatively evaluated to obtain the environmental impact assessment result.

[0007] Further, based on the environmental impact assessment result, the vehicle path transportation data of the cold-chain food distribution vehicle is executed for distribution efficiency analysis to form initial distribution efficiency data, including: Obtain the distribution target address and map data of the cold-chain food distribution vehicle, and perform path planning to obtain the path transportation data; Based on the cold-chain food distribution vehicle, the path transportation data is matched for driving to obtain road section travel time; According to the environmental impact assessment result, the road section travel time is analyzed for temperature control stability to obtain temperature control compliance rate data; Based on the transportation state parameter, the temperature control compliance rate data is matched for device operation to obtain device operation efficiency index; According to the road section travel time and the device operation efficiency index, timeliness evaluation is performed to obtain the initial distribution efficiency data.

[0008] Further, the path transportation data is diagnosed for road conditions, and the environmental impact assessment result is combined for path prediction to generate a path prediction result, including: According to the preset traffic database, traffic congestion flow analysis is performed on the path transportation data to obtain periodic congestion characteristics; According to the periodic congestion characteristics, dynamic road analysis is performed on the path transportation data to obtain a dynamic road network topology; According to the dynamic road network topology, road section environment correlation is performed on the environmental impact assessment result to obtain an environmental sensitive path set; The environmental sensitive path set is subjected to candidate path prediction, and traffic event matching is performed with the traffic database to obtain a predicted path avoidance strategy; Based on the predicted path avoidance strategy, path planning prediction is performed on the dynamic road network topology to obtain the path prediction result.

[0009] Further, the environmental impact assessment result is subjected to road section environment correlation according to the dynamic road network topology to obtain an environmental sensitive path set, including The dynamic road network topology is subjected to road section segmentation to obtain a basic road network unit set; According to the environmental impact assessment result, road network environment analysis is performed on the basic road network unit set to obtain a road section environment scoring matrix; The road section environment scoring matrix is subjected to road section identification to obtain a target road section identification set; According to the periodic congestion characteristics, the target road section identification set is subjected to hotspot area detection to obtain an environmental traffic hotspot area; Based on the environmental traffic hotspot area, path reorganization is performed on the dynamic road network topology to obtain the environmental sensitive path set.

[0010] Further, the initial distribution efficiency data and the path prediction result are subjected to cross analysis to formulate a global distribution scheme of the cold chain food distribution vehicle, including: The initial distribution efficiency data is subjected to segmented calculation to obtain transportation efficiency data and cargo damage data; According to the path prediction result, deviation detection is performed on the transportation efficiency data to obtain path matching degree; The cargo damage data and the environmental impact assessment result are subjected to joint calculation to obtain damage correction data; According to the path matching degree, the damage correction data and the path prediction result, global distribution construction is performed to obtain the global distribution scheme.

[0011] Further, the cargo damage data and the environmental impact assessment result are subjected to joint calculation to obtain damage correction data, including: The cargo damage data is subjected to transportation period division to obtain a damage rate curve; The environmental impact assessment result is subjected to feature extraction to obtain a temperature and humidity change curve and vibration influence data; According to the alignment analysis of the damage rate curve and the temperature and humidity change curve, a temperature and humidity factor is obtained; According to the vibration response analysis of the damage rate curve and the vibration influence data, a vibration factor is obtained. Based on the temperature and humidity factor and the vibration factor, the cargo damage data is compensated and corrected in multiple factors to obtain the damage correction data.

[0012] The application also provides a cold-chain food distribution analysis system based on big data, which is applied to the cold-chain food distribution analysis method based on big data. The acquisition module is used for dynamic feature extraction on the multi-source monitoring data of the cold-chain food distribution vehicle to obtain generated transportation state parameters. The analysis module is used for associating and matching the transportation state parameters with real-time environmental data of the cold-chain food distribution vehicle to obtain environmental influence evaluation results. The association module is used for performing distribution efficiency analysis on the vehicle path transportation data of the cold-chain food distribution vehicle based on the environmental influence evaluation results to form initial distribution efficiency data. The processing module is used for road condition diagnosis on the path transportation data and path prediction by combining the environmental influence evaluation results to generate path prediction results. The control module is used for cross analysis on the initial distribution efficiency data and the path prediction results to formulate a global distribution scheme of the cold-chain food distribution vehicle.

[0013] The cold-chain food distribution analysis method and system based on big data have the following beneficial effects: The transportation state parameters are obtained by dynamic feature extraction of multi-source monitoring data, real-time and accurate monitoring of the running state of the cold chain distribution vehicle is realized, the problem of slow response to dynamic changes in the transportation process in the traditional method is effectively solved, and the early warning capability of abnormal conditions is significantly improved. By associating and matching the real-time environmental data with the transportation state parameters, a dynamic correlation model of environmental factors and cold chain efficiency is established, which overcomes the problem of unsystematic assessment of the influence of environmental factors in existing methods, and provides a scientific basis for optimizing the temperature control strategy. Based on the environmental impact assessment result of the path transportation data, the distribution efficiency is analyzed, the cooperative optimization of path planning and cold chain energy consumption is realized, and the problem that timeliness and temperature control requirements are difficult to balance in the traditional distribution scheme is solved. By predicting the road condition detection result and the environmental impact assessment, a dynamic intelligent path planning mechanism is constructed, which effectively avoids the problems of distribution delay and temperature control failure caused by road condition changes, and improves the distribution reliability. By integrating the initial distribution efficiency data and the path prediction result to generate a global distribution scheme, intelligent management of the whole process of cold chain distribution is realized, which not only guarantees the food quality and safety, but also optimizes the allocation of logistics resources, and significantly improves the overall operation efficiency of cold chain logistics. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 is a cold chain food distribution analysis method flowchart based on big data provided by the present application; Figure 2 is a cold chain food distribution analysis system structure diagram based on big data provided by the present application.

[0015] The implementation of the object, functional characteristics and advantages of the present application will be further described with reference to the accompanying drawings. DETAILED DESCRIPTION

[0016] In order to make the object, technical scheme and advantages of the present application more clear, the present application will be further described in detail below with reference to the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application, and are not used to limit the present application.

[0017] The present application will be further described below in combination with the drawings and specific embodiments.

[0018] Referring to Figure 1 , a cold chain food distribution analysis method based on big data includes: Step S1: dynamic feature extraction is performed on the multi-source monitoring data of the cold chain food distribution vehicle to obtain generated transportation state parameters; Step S2: associating and matching the transportation state parameters with the real-time environmental data of the cold chain food distribution vehicle to obtain environmental impact assessment results; Step S3: Based on the environmental impact assessment result, the vehicle path transportation data of the cold chain food distribution vehicle is analyzed for distribution efficiency, and initial distribution efficiency data is formed; Step S4: The path transportation data is diagnosed for road conditions, and the path prediction result is generated by combining the environmental impact assessment result; Step S5: Cross analysis is performed on the initial distribution efficiency data and the path prediction result, and a global distribution scheme for the cold chain food distribution vehicle is formulated.

[0019] Based on the steps shown in the above, the detailed step process is described as follows: Step S1: Through dynamic feature extraction of multi-source monitoring data of the cold chain food distribution vehicle, transportation state parameters are obtained. In this step, multi-dimensional monitoring data streams including carriage temperature, humidity, refrigeration unit operating parameters, vehicle speed, engine speed, etc. are collected through the vehicle sensor network. The data acquisition module continuously acquires the original signals of the sensors at a preset sampling frequency, and after the noise interference is eliminated by the signal conditioning circuit, it is transmitted to the vehicle data processing unit.

[0020] The dynamic feature extraction process adopts a sliding time window mechanism to analyze the time domain and frequency domain features of continuous time series data, and extracts key state indicators including temperature fluctuation rate, refrigeration efficiency coefficient, and vehicle vibration spectrum features. The feature extraction algorithm is based on the preset cold chain transportation quality standard, and the original monitoring data is normalized and feature weighted to finally generate a transportation state parameter matrix reflecting the real-time running state of the vehicle. The parameter matrix provides basic data support for subsequent environmental impact assessment, and its dynamic updating mechanism ensures timely response to changes in the transportation process.

[0021] Step S2: The transportation state parameters are associated and matched with the real-time environmental data of the cold chain food distribution vehicle to obtain the environmental impact assessment result. The environmental monitoring module triggers regional meteorological database query through GPS positioning information to obtain real-time environmental parameters such as temperature, humidity, wind speed, and solar intensity on the distribution path. At the same time, the local micro-environmental data collected by the vehicle-mounted external sensors are spatiotemporally aligned and data fused with the cloud meteorological data.

[0022] The association matching engine uses a timestamp-based data alignment algorithm to spatiotemporally associate the environmental data stream with the transportation state parameters. By establishing a mapping relationship model between environmental parameters and carriage temperature control efficiency and refrigeration energy consumption, evaluation indicators including environmental temperature variation influence coefficient and climate adaptability score are calculated. The evaluation result reflects the comprehensive influence of external environmental changes on the cold chain transportation quality, and provides a quantitative basis for environmental adaptability for subsequent distribution efficiency analysis. The dynamic association model established in this step can accurately capture the coupling relationship between environmental factors and transportation state.

[0023] Step S3: Based on the environmental impact assessment results, the vehicle path transportation data of the cold chain food distribution vehicle is analyzed for distribution efficiency, and initial distribution efficiency data is formed. The path data collection module integrates historical distribution records, real-time traffic information and electronic map data to build a multi-dimensional path information library containing path distance, elevation change and typical section characteristics.

[0024] The efficiency analysis engine multi-dimensionally associates the environmental impact assessment results with the path feature data, calculates the expected temperature control energy consumption, time cost and quality maintenance degree of each path segment, and establishes a distribution efficiency evaluation matrix. The analysis process considers the response characteristics of the refrigeration system under different environmental conditions, and predicts the temperature fluctuation risk area combined with the path terrain characteristics. Based on the preset cold chain quality threshold and distribution time limit, an initial distribution efficiency data set reflecting the path applicability is generated. The data set contains the comprehensive score of each alternative path, providing a quantitative comparison basis for subsequent path optimization decision-making. The efficiency analysis result and the environmental assessment data form a feedback mechanism to support dynamic adjustment of distribution strategy.

[0025] Step S4: Through the real-time traffic information platform, dynamic road condition data on the distribution path is obtained, including congestion index, accident report, construction section and weather warning information. The road condition detection module combines historical traffic flow data and real-time sensor feedback to build a traffic capacity evaluation model of the current path. At the same time, the climate adaptability score in the environmental impact assessment results is spatiotemporally matched with the road condition data to analyze the influence of different environmental conditions (such as high temperature, heavy rain, snow and ice) on road traffic efficiency.

[0026] The path prediction engine uses a multi-factor weighted algorithm to comprehensively evaluate the timeliness, stability and cold chain quality guarantee capacity of each alternative path, and generates prediction results including the optimal path, backup path and risk warning. The prediction model supports a dynamic update mechanism to continuously correct the path strategy during the distribution process, ensuring the reliability and efficiency of cold chain transportation. The output of this step provides key path decision basis for the formulation of subsequent global distribution scheme.

[0027] Step S5: Based on the path applicability score in the initial distribution efficiency data, combined with the real-time road condition and environmental adaptability in the path prediction result, a multi-objective optimization model is built. The optimization process considers distribution timeliness, energy consumption, cold chain quality maintenance and transportation cost, etc., and uses heuristic algorithm to generate the optimal distribution strategy. The global distribution scheme includes vehicle scheduling plan, path dynamic adjustment strategy, temperature control parameter optimization suggestion and emergency response plan.

[0028] The scheme generation module supports manual intervention adjustment, allowing distribution managers to fine-tune optimization weights according to actual needs. The final output global distribution scheme is presented in a visual manner, including recommended routes, estimated arrival times, key risk point prompts, and real-time monitoring suggestions. This scheme ensures the quality stability of cold chain food during transportation, while maximizing distribution efficiency, achieving intelligent and adaptive efficient cold chain logistics management.

[0029] The cold chain food distribution analysis method based on big data provided by the application realizes real-time and accurate monitoring of the running state of cold chain distribution vehicles by extracting dynamic characteristics from multi-source monitoring data to obtain transportation state parameters, effectively solves the problem of lag response to dynamic changes in the transportation process of traditional methods, and significantly improves the early warning capability of abnormal situations. By correlating and matching real-time environmental data with transportation state parameters, a dynamic correlation model between environmental factors and cold chain efficiency is established, overcoming the problem of unsystematic assessment of the impact of environmental factors in existing methods, and providing a scientific basis for optimization of temperature control strategies. Based on the environmental impact assessment results of the path transportation data, the distribution efficiency is analyzed, realizing the collaborative optimization of path planning and cold chain energy consumption, and solving the problem that timeliness and temperature control requirements are difficult to balance in traditional distribution schemes. By predicting the path based on the road condition detection results and environmental impact assessment, a dynamic intelligent path planning mechanism is constructed, effectively avoiding the problems of distribution delay and temperature control failure caused by road condition changes, and improving the distribution reliability. By integrating the initial distribution efficiency data and the path prediction results to generate a global distribution scheme, the intelligent management of the whole process of cold chain distribution is realized, which not only guarantees food quality and safety, but also optimizes the allocation of logistics resources, and significantly improves the overall operation efficiency of cold chain logistics.

[0030] In one embodiment, by dynamically extracting features from multi-source monitoring data of cold chain food distribution vehicles, transportation state parameters are generated, including: A distributed temperature and humidity sensor array is deployed in the refrigerated vehicle compartment to collect temperature and humidity raw data of each partition inside the compartment at a sampling frequency of not less than 1Hz. The temperature and humidity sensor uses an industrial-grade platinum resistance temperature probe and a capacitive humidity sensor. After the raw signal is processed by the data acquisition unit to resist interference, the temperature and humidity change gradient between adjacent sampling points is calculated by a time series difference algorithm.

[0031] For the collected continuous temperature and humidity data stream, a sliding time window mechanism is used for dynamic analysis, with a window size of 5 minutes and a sliding step of 1 minute. In each analysis window, the root mean square value of temperature change, the maximum fluctuation amplitude, and the cumulative time of continuous deviation from the set threshold are calculated to form a multi-dimensional feature vector reflecting temperature stability. The differential fluctuation detection module pays special attention to temperature sudden change events, and when the temperature change exceeds the preset threshold within a unit time, an abnormality flag is triggered and the event characteristics are recorded. The final vehicle temperature change data generated includes the real-time temperature change rate of each monitoring point, the historical fluctuation pattern, and the abnormal event statistics, providing a basis for subsequent degradation analysis.

[0032] Through the three-axis acceleration sensor installed on the vehicle chassis and the refrigeration unit, vibration frequency spectrum data during vehicle driving is collected. The vibration sensor records the X, Y, Z three-axis acceleration values at a sampling frequency not less than 100Hz. Vibration signal preprocessing includes removing gravity acceleration component, band-pass filtering to eliminate high-frequency noise, etc.

[0033] The feature extraction process divides the vibration frequency spectrum into 0-20Hz low frequency band and 20-100Hz high frequency band, and calculates the energy proportion, main frequency position and harmonic component of each frequency band respectively. The abnormality detection algorithm aligns the vibration features with the vehicle temperature change data in time and space, establishes a vibration-temperature change correlation model. When the vibration energy exceeds the threshold and is accompanied by abnormal temperature fluctuation, it is determined as a potential degradation event. The degradation index calculation module calculates the abnormal event density, duration and energy accumulation per unit mileage, and establishes the degradation trend curve combined with historical data. The final generated vehicle degradation index includes mechanical vibration features, temperature change correlation degradation degree and equipment health status score, providing equipment state basis for subsequent energy efficiency analysis.

[0034] Through the pressure sensor array installed on the side wall of the cargo compartment and the top laser ranging device, the cargo loading state is monitored in real time. The pressure sensor array measures the lateral pressure distribution of the cargo on the cabin wall with a spatial resolution of 10cm x 10cm, and the laser range finder scans the cargo surface height change at a frequency of 1Hz. The data fusion module reconstructs the pressure distribution and height change data in three dimensions to establish a spatial model of cargo deformation. The deformation analysis algorithm combines vehicle vibration data to calculate the displacement response and pressure redistribution characteristics of the cargo under different vibration intensities.

[0035] Special attention is paid to the resonance effect of the vibration main frequency and the inherent frequency of the cargo, and when the resonance phenomenon is detected, its frequency characteristics and spatial distribution are recorded. The dynamic deformation model evaluates the changing trend of the physical state of the cargo during transportation by analyzing parameters such as cargo center of gravity shift, local compression degree, etc. The final generated cargo deformation distribution data includes real-time deformation of each cargo location, pressure gradient distribution, and resonance risk area marking, providing cargo physical state features for cold chain transportation quality evaluation.

[0036] Through the real-time monitoring module of the refrigerated truck energy management system, the operating energy consumption data of key equipment such as refrigeration units, ventilation equipment, and on-board power sources are collected. The energy consumption monitoring unit records electrical parameters such as current, voltage, and power factor of each device at a sampling frequency of not less than 1 Hz, and combines with the device operating state (such as compressor start-stop, fan speed) to extract energy consumption characteristics. The power consumption decay analysis module synchronously matches the energy consumption data with the on-board degradation indicators in time, establishing a correlation model between device performance degradation and energy consumption changes. For refrigeration units, the focus is on analyzing the energy consumption change trend under unit refrigeration capacity; for on-board power systems, the decay of charging and discharging efficiency is monitored.

[0037] Energy efficiency decay calculation is based on historical operation data, comparing the current energy consumption level with the baseline value under the initial state or standard working condition of the device, and calculating the relative energy efficiency loss percentage. The final generated device energy efficiency decay rate includes the refrigeration system energy efficiency ratio decline, battery capacity decay rate, and overall energy consumption degradation trend, providing quantitative basis for transportation economy evaluation.

[0038] Through the multi-source data fusion engine, the feature data extracted in the foregoing steps are spatio-temporally aligned and normalized. The feature fusion module uses a weighted aggregation strategy, dynamically assigning weights according to the influence of different monitoring indicators on transportation quality. Temperature change data and cargo deformation distribution data are combined to evaluate the impact of cold chain environment stability on the physical state of goods; on-board degradation indicators are associated with device energy efficiency decay rate to reflect the potential risks of device health status on transportation reliability.

[0039] The fusion algorithm generates a multi-dimensional transportation state vector, including environmental control stability score, device operation health degree, cargo preservation quality index, and comprehensive transportation efficiency evaluation value. Transportation state parameters are graded through a dynamic threshold mechanism, and different levels of alarm signals are triggered when key indicators exceed the preset safety range. The final output of the transportation state parameters provides comprehensive and structured data support for real-time monitoring, abnormal diagnosis, and optimization decision-making in the cold chain distribution process.

[0040] This embodiment realizes comprehensive perception and accurate evaluation of cold chain food transportation state through dynamic feature extraction of multi-source monitoring data. Using differential fluctuation detection technology to process temperature and humidity data can accurately capture subtle changes in the cold chain environment and effectively identify temperature abnormal fluctuations, providing reliable basis for cold chain quality control. Combined with vibration data and temperature change data, abnormal accumulation quantity recognition can timely detect device degradation trends and prevent transportation risks caused by cold chain equipment failure. Through deformation analysis of cargo distribution and vibration data, the physical state changes of goods during transportation can be objectively evaluated to ensure the integrity of cold chain food. The correlation calculation of device energy consumption and degradation indicators accurately reflects the device energy efficiency decay, providing data support for transportation cost optimization.

[0041] In one embodiment, the transportation status parameters are matched with real-time environmental data of the cold-chain food delivery vehicle to obtain an environmental impact assessment result, including: The vehicle perception data of the cold-chain food delivery vehicle is collected by on-board environmental sensors, including but not limited to infrared temperature sensors, humidity sensors, gas composition detection modules, and optical imaging devices. The infrared temperature sensor monitors the external environmental temperature of the vehicle, the humidity sensor detects the relative humidity of the air, the gas composition detection module analyzes the components such as carbon dioxide and volatile organic compounds in the surrounding air that can affect the quality of cold-chain transportation, and the optical imaging device captures the road surface conditions and weather visibility. Real-time weather data is derived from the API interface published by the meteorological department or the meteorological monitoring station deployed on the delivery route, covering macro weather indicators such as wind speed, precipitation probability, and ultraviolet intensity.

[0042] The environmental correlation process adopts a spatio-temporal alignment technique to match the vehicle perception data with real-time weather data in the time stamp and geographic location dimensions. The on-board GPS module provides real-time coordinates of the vehicle, and the weather data is spatially aligned through coordinate interpolation or nearest site matching; the time stamp is unified in UTC standard to ensure data synchronization. The correlated real-time environmental data forms a structured record, including the temperature, humidity, gas composition, road surface conditions, and macro weather parameters of the external environment where the vehicle is located, providing complete input for subsequent analysis.

[0043] Multi-source heterogeneous protocol conversion is carried out for the communication protocol differences of different sensors and weather data sources. On-board sensors can use CAN bus, Modbus, or MQTT protocol, and weather data is transmitted in XML format. The protocol conversion layer converts the data into a standardized key-value pair structure through pre-defined mapping rules, such as converting the hexadecimal temperature value in the CAN bus to a floating-point number, and renaming the wind speed field in the weather data to a unified identifier. The converted data is stored in a time series database, and the field naming and units comply with international standards (such as temperature unit in Celsius and humidity in percentage).

[0044] Anomaly value cleaning is based on statistical rules and domain knowledge. Statistical rules use box plots to identify outliers, such as readings outside the range of -30°C to 50°C for vehicle external temperature sensors being considered abnormal; domain knowledge rules exclude values that do not conform to physical laws, such as humidity exceeding 100% or carbon dioxide concentration being negative. The cleaning process marks abnormal data and fills in reasonable values, using forward propagation or linear interpolation to ensure the continuity of the time series. The final output of the baseline environmental data is a high-quality data set that has been standardized and verified for effectiveness.

[0045] Transportation state parameters include interior temperature, cargo load, vehicle speed, refrigeration unit power, and other operating indicators. Multi-dimensional feature fusion aligns benchmark environmental data with transportation state parameters by time window. The window size is set according to business needs (e.g., 5 minutes or 1 kilometer of travel). The aligned data extracts joint features, such as the difference between external temperature and interior temperature, the vector composition of wind speed and vehicle speed, and the weighted score of precipitation probability and road surface wetness.

[0046] The environmental state correlation matrix is organized in a two-dimensional table format, with rows representing time points or road segments and columns containing fused features. Cross-features, such as the product of external humidity and refrigeration unit power, are introduced during matrix construction to reflect the impact of humidity on refrigeration efficiency. The dimensions of the matrix can be reduced through principal component analysis, retaining features with a variance contribution rate exceeding a threshold. This matrix serves as input for subsequent clustering analysis, implicitly revealing the dynamic correlation between environmental factors and transportation states.

[0047] The environmental state correlation matrix contains multi-dimensional fused features, and the influence factor clustering aims to identify potential impact patterns of key environmental variables on cold-chain transportation states. The clustering process uses an unsupervised learning method, with standardized feature vectors in the matrix as input data. Feature importance evaluation is performed before clustering to eliminate low-variance or redundant features, ensuring that the clustering results focus on key influence factors.

[0048] The clustering algorithm divides data points based on feature similarity, forming several clusters. Each cluster represents a typical environmental impact combination, such as high external temperature and low wind speed clusters corresponding to high-load operation of refrigeration units, and high humidity and strong precipitation clusters associated with the risk of cargo compartment sealing. The clustering results are verified by silhouette coefficients or elbow rules to ensure compactness within clusters and separation between clusters. The final output of environmental impact factor classification results is labeled in the original data in the form of tags, clearly indicating the differentiated impact of different environmental conditions on transportation states.

[0049] Quantitative evaluation is based on the mapping relationship between classification results and transportation state parameters. For each class of environmental impact factors, the transportation state parameter changes within the corresponding time window are calculated, such as the average power increase percentage of refrigeration units under high external temperature clusters and the standard deviation of cargo temperature fluctuations under strong precipitation clusters. The evaluation uses a comparative analysis method to compare the mean value of the current environmental category parameters with the historical benchmark value (e.g., ideal transportation conditions on the same route), and calculates the deviation degree.

[0050] The environmental impact assessment results are output in a structured report, including key indicators: environmental risk level (e.g. low / medium / high risk), main impact dimensions (e.g. temperature stability, energy efficiency), and recommended measures (e.g. adjusting refrigeration settings or optimizing routes). The assessment process introduces a weighting mechanism, giving higher weights to parameters with high sensitivity to goods, such as temperature tolerance for pharmaceutical transportation. The final results support dynamic decision-making, such as real-time alerts or subsequent delivery strategy optimization.

[0051] This embodiment realizes dynamic monitoring and accurate assessment of the cold chain transportation environment by obtaining real-time environmental data of cold chain food delivery vehicles and correlating them with transportation state parameters. The use of multi-source heterogeneous protocol conversion and outlier cleaning technology effectively improves the accuracy and reliability of environmental data, providing a high-quality data foundation for subsequent analysis. The environmental state correlation matrix constructed through multi-dimensional feature fusion can comprehensively reflect the complex correlation between environmental factors and transportation state. The environmental impact factor classification method based on clustering analysis can automatically identify the influence mode of key environmental variables on transportation quality, significantly improving the analysis efficiency. The quantitative evaluation results not only accurately reflect the actual impact of environmental factors on cold chain transportation, but also provide data support for transportation strategy optimization, effectively ensuring the transportation quality and safety of cold chain food.

[0052] In one embodiment, based on the environmental impact assessment results, the vehicle path transportation data of the cold chain food delivery vehicle is analyzed for delivery efficiency, forming initial delivery efficiency data, including: The delivery target address of the cold chain food delivery vehicle is the core input of the delivery task. After the target address is determined, path planning is completed in combination with map data. The delivery target address includes specific geographic location information of the consignee, such as latitude and longitude coordinates or detailed street address. Map data covers road network, traffic rules, restricted areas, real-time or historical traffic flow information, and road classification.

[0053] Path planning is based on the target address and map data, and uses a preset path algorithm to generate multiple feasible routes. The length of the route, the estimated travel time, the probability of road congestion, and the special needs of cold chain transportation, such as avoiding bumpy sections to reduce goods vibration, are comprehensively evaluated. Path transportation data includes detailed node information of the planned route, distance of each road segment, estimated speed, and key landmarks along the way. The generation of path transportation data needs to ensure the feasibility of the route to avoid delivery interruption due to road closure or height and weight restrictions.

[0054] During route planning, the accuracy of map data directly impacts the reliability of route transportation data. High-precision map data can identify narrow roads, steep slopes, tunnels, and other road sections that may hinder the passage of cold chain vehicles. The accuracy of the delivery destination address ensures the correct destination of route planning, avoiding detours or delays due to ambiguous addresses. Route transportation data not only contains static route information but also integrates dynamic factors, such as the impact of weather changes on road traffic. Route transportation data provides the foundation for subsequent steps, and the calculation of road section travel time relies on accurate route division.

[0055] Driving matching for cold chain food delivery vehicles is the process of combining route transportation data with the vehicle's actual driving characteristics. Vehicle characteristics include vehicle model, load capacity, power performance, refrigeration equipment power consumption, and driver operating habits. The distance, slope, and number of curves of each road section in the route transportation data are matched with vehicle characteristics to estimate the vehicle's actual driving speed on different sections. For example, uphill sections can cause vehicles to slow down, while flat, high-speed sections allow vehicles to travel at higher speeds. The section travel time is the sum of the travel time of each section, including the time spent on starting, accelerating, maintaining a constant speed, decelerating, and stopping.

[0056] Driving matching needs to take into account the particularities of cold chain transportation. The operation of refrigeration equipment can affect vehicle power distribution, especially in urban roads with frequent starts and stops, where vehicle speeds fluctuate greatly. The calculation of section travel time integrates historical driving data or statistical results of similar vehicles to improve the accuracy of time estimation. For example, the travel time of the same delivery vehicle through the same section during peak hours and off-peak hours is different. As intermediate data, section travel time directly affects the input of temperature control stability analysis. The longer the travel time of the section, the greater the challenge faced by the temperature control equipment, which may lead to increased energy consumption or temperature fluctuations due to continuous operation.

[0057] The results of the environmental impact assessment include the impact of the external environment on cold chain transportation, such as temperature, humidity, sunlight intensity, and the distribution of heat sources around the road section. The temperature control stability analysis evaluates the temperature control capabilities of cold chain vehicles on each road section. The road section travel time is combined with environmental factors to determine whether the temperature control equipment can maintain the temperature of the goods within a safe range within a specific period of time. For example, long-term driving in a high-temperature environment can lead to an increase in the refrigeration load, while low-temperature environments at night can reduce the demand for refrigeration. The temperature control compliance rate data reflects the degree of compliance of the temperature control on each road section, and is expressed as a percentage of the proportion of time that meets the temperature control requirements.

[0058] The temperature control stability analysis relies on the matching of vehicle refrigeration performance parameters and environmental data. If the road section has a long travel time and the environmental temperature is extremely high, the temperature control equipment needs to run at high load continuously, and there may be a risk of temporary overheating. The temperature control compliance rate data quantifies such risks and identifies potential problem road sections. The analysis process can introduce historical data from vehicle sensors to verify the actual performance of the temperature control equipment under similar environments. The temperature control compliance rate data provides a basis for subsequent equipment operation efficiency indicators, and low compliance rate road sections indicate that the equipment is under greater operating pressure, prompting further optimization of delivery routes or adjustment of equipment parameters.

[0059] Transportation state parameters cover the dynamic operation data of cold chain delivery vehicles during transportation, including the working mode of refrigeration equipment, energy consumption level, compressor start-stop frequency, temperature fluctuation range in the compartment, and load status of the battery or fuel system. Temperature control compliance rate data reflects the temperature control effect of each road section, while equipment operation matching aims to associate the compliance rate with the actual operation parameters of the vehicle to quantify the operation efficiency of the equipment. For example, if the temperature control compliance rate of a road section is 95%, but the refrigeration equipment runs continuously at high power, it indicates that the equipment can maintain the temperature but has low energy efficiency; on the contrary, if the compliance rate is 98% and the equipment runs intermittently, it indicates that the energy efficiency is better.

[0060] The equipment operation efficiency indicator is generated by analyzing the balance between the running time of the refrigeration equipment, energy consumption, and temperature control effect. This indicator not only focuses on whether the temperature meets the standard, but also evaluates the economic efficiency of the equipment under specific environmental conditions. For example, on a high-temperature road section, the equipment needs to run at full load for a long time to maintain low temperature, and at this time the efficiency indicator will reflect high energy consumption and low efficiency; while on a road section with low temperature control demand, the equipment can maintain stability at low power, and the efficiency indicator is better. The calculation of the equipment operation efficiency indicator needs to combine the vehicle model, refrigeration technology type, and historical operation data to ensure that the indicator is consistent with the actual performance. This indicator provides key input for subsequent timeliness evaluation, and high-efficiency equipment can shorten the temperature control adjustment time, thereby improving delivery efficiency.

[0061] The timeliness evaluation combines road section travel time and equipment operation efficiency indicators to quantify the overall efficiency of cold chain delivery. Road section travel time reflects the movement efficiency of the vehicle on the path, while the equipment operation efficiency indicator reflects the stability and economy of the temperature control system. After combining the two, it is determined whether the delivery process is completed within a reasonable time while meeting the temperature control requirements. For example, if the total travel time of a delivery route is short, but the energy consumption increases due to frequent start-stop of the equipment, it can affect the overall efficiency; on the contrary, if the travel time is slightly longer but the equipment runs smoothly, the overall efficiency can be higher.

[0062] The initial delivery efficiency data is presented in the form of a comprehensive score or quantitative value, covering time efficiency (such as total delivery time, road segment delay) and temperature control efficiency (such as temperature fluctuation range, equipment energy consumption ratio). This data can be used to compare different delivery schemes horizontally, or to optimize existing routes to improve the overall performance of cold chain logistics. The initial delivery efficiency data serves as the basis for subsequent optimization analysis, such as adjusting path planning strategies, optimizing vehicle scheduling, or upgrading temperature control device configurations, ultimately achieving more efficient and stable cold chain food delivery.

[0063] This embodiment realizes accurate analysis of delivery efficiency by obtaining path transportation data of cold chain food delivery vehicles and combining environmental impact assessment results, effectively improving the overall operation efficiency of cold chain logistics. Based on the road segment travel time obtained through path planning and driving matching, the actual time consumption in the transportation process can be accurately reflected, providing a reliable basis for subsequent analysis. The temperature control compliance rate data obtained through temperature control stability analysis can directly evaluate the temperature control quality in the cold chain transportation process, ensuring food quality and safety. The establishment of equipment operation efficiency indicators combines temperature control effect and equipment operation state, providing quantitative basis for optimizing equipment configuration and operation strategy. The initial delivery efficiency data obtained through timeliness evaluation can comprehensively reflect the time efficiency and temperature control efficiency of the delivery process.

[0064] In one embodiment, path transportation data is subjected to road condition diagnosis, combined with environmental impact assessment results for path prediction, resulting in path prediction results, including: Path transportation data includes vehicle driving trajectory, historical speed record, timestamp information, and road identification. The pre-set traffic database stores multi-source traffic information, including real-time traffic flow, historical congestion records, road types, and traffic signal distribution. The goal of traffic congestion flow analysis is to extract congestion patterns with time regularity from path transportation data.

[0065] By matching path transportation data with historical congestion records in the traffic database, repeated congestion segments in a specific time period are identified. For example, the speed decrease phenomenon of urban trunk roads during morning and evening peak hours is marked as a fixed congestion point. Through time series analysis, the average travel speed of the same period on different dates is calculated, and congestion segments with significant periodicity are selected. Congestion characteristics not only include repeatability in the time dimension, but also involve congestion duration, congestion intensity, and influencing factors such as weather or holidays.

[0066] The extraction of periodic congestion features relies on data aggregation and pattern recognition. The route transportation data is divided into time windows, and the average travel time deviation of each road segment in the same time period is calculated. Road segments with deviation values exceeding the threshold are classified as periodic congestion road segments. Further analysis of the correlation of congestion road segments, such as whether adjacent road segments are congested at the same time, forms the congestion propagation rule. The final output of the periodic congestion features includes the geographic location, time period, duration, and congestion level of the congestion road segment.

[0067] The core of dynamic road analysis is to combine static road network with real-time traffic state, and to construct a road network model that reflects the actual traffic conditions. Periodic congestion features reveal the time-varying characteristics of the road network, and the dynamic road network topology needs to integrate these features to optimize path planning.

[0068] The basic road network data uses the city electronic map, which contains road grade, lane number, and speed limit information. Periodic congestion features are mapped to the road network, marking the congestion road segments and their time attributes. For example, a road segment is marked as "peak congestion" from 7:00 to 9:00 on weekdays, with a 50% decrease in traffic capacity. The dynamic road network topology is achieved through weight adjustment, and the travel time weight of the congestion road segment changes dynamically with the time period.

[0069] Analyze the alternative paths of congestion road segments and identify the adjacent roads that can be bypassed. If a major road is congested during peak hours, the traffic pressure on the surrounding branches increases, and the dynamic road network updates the weights of the branches simultaneously. In addition, traffic control or construction information is extracted from the traffic database, and the connectivity of the affected road segments is temporarily adjusted. The output of the dynamic road network topology includes the time-varying attributes of nodes (intersections) and edges (road segments), such as the travel speed, congestion probability, and bypass recommendation of a road segment at different times.

[0070] The environmental impact assessment results include high-risk areas that need to be avoided during cold chain transportation, such as high-temperature exposure areas, frequently stopped road segments, or pollution-prone areas. The goal of road segment environmental correlation is to integrate environmental factors into the dynamic road network and identify paths that are sensitive to temperature or susceptible to external influences.

[0071] The road segments in the dynamic road network topology are matched with the environmental assessment data. For example, a road segment is marked as a "high-temperature risk road segment" due to strong midday sunlight, with road surface temperature significantly higher than the surrounding area. Another road segment is adjacent to an industrial area, with poor air quality affecting the heat dissipation of refrigeration equipment. The basis for determining environmentally sensitive paths includes historical temperature and humidity data, weather records, and geographic features.

[0072] The generation of the environmentally sensitive path set employs multi-condition screening. The environmental score of each road segment in the dynamic road network is calculated based on temperature fluctuations, pollution indexes, and sunshade conditions. Road segments with scores exceeding a threshold are included in the set and labeled with specific risk types. For example, road segments in urban business districts are classified as "start-stop sensitive paths" due to frequent parking and loading / unloading, causing temperature fluctuations in refrigerated truck compartments. The final output of the environmentally sensitive path set includes road segment locations, risk types, and severity, providing constraint conditions for subsequent path prediction.

[0073] The goal of candidate path prediction is to generate feasible paths that meet the requirements of cold chain transportation based on the dynamic road network topology and the environmentally sensitive path set. The environmentally sensitive path set defines high-risk road segments that need to be avoided, while the dynamic road network topology provides road capacity under real-time traffic conditions.

[0074] A path search algorithm is used to calculate multiple feasible paths from the starting point to the ending point in the dynamic road network, excluding high-risk road segments in the environmentally sensitive path set. For example, if a main road is marked as an environmentally sensitive path due to high temperatures during the noon, alternative routes with good shading or ventilation are preferred. The generation of candidate paths needs to comprehensively evaluate path length, travel time, and environmental stability to minimize temperature fluctuations of cold chain food during transportation.

[0075] Traffic event matching further optimizes the reliability of candidate paths. Real-time event data in the traffic database, such as traffic accidents, temporary regulations, or severe weather warnings, are compared with candidate paths. If a candidate path passes through a sudden congestion area, the path planning is adjusted to avoid temporary traffic obstacles. The matching process uses a rule engine, such as selecting paths without real-time events or choosing the option with the least impact from events among multiple feasible paths.

[0076] The final output of the prediction path avoidance strategy includes the recommended path and its avoidance basis. For example, a path is preferentially recommended due to avoiding high-temperature road segments and real-time accident points, with alternative paths as backup options. The strategy is not only dependent on current data but also combines historical event occurrence probabilities to predict potential risks. This strategy provides decision support for final path planning, ensuring the timeliness and safety of cold chain transportation.

[0077] In the path planning prediction stage, the dynamic road network, environmental constraints, and traffic event avoidance strategies are integrated to generate the optimal transportation path. The prediction path avoidance strategy has clearly defined road segments to be avoided and alternative solutions, while the dynamic road network topology provides real-time weight adjustments to ensure the accuracy of path calculation.

[0078] The graph optimization method is used to calculate the optimal path in the dynamic road network, which combines multiple objective optimizations such as time, distance, and environmental stability. For example, a cold chain distribution task requires reaching the destination in the shortest time while ensuring that the temperature fluctuation in the vehicle does not exceed the set threshold. The path planning algorithm selects the route with the shortest total travel time while meeting the environmental sensitive path constraints. If multiple paths meet the requirements, further comparison is made on the historical congestion probability, average vehicle speed, and frequency of unexpected events to select the optimal solution.

[0079] The output of the path prediction result includes detailed navigation information such as the route, estimated travel time, and environmental risk warning. For example, if the prediction result indicates that a certain road section has a high temperature risk at a specific time period, it suggests adjusting the departure time or taking additional temperature protection measures. In addition, the result supports dynamic updating, and if new traffic events or environmental changes are monitored during transportation, the path is re-planned in real-time. The final output of the path prediction result not only meets the distribution efficiency but also guarantees the quality and safety of cold chain food, realizing intelligent logistics decision-making driven by big data.

[0080] This embodiment can accurately identify the road congestion rules at different time periods through periodic congestion feature analysis, providing time-dimension path planning basis for cold chain transportation and significantly improving the timeliness of distribution. The construction of dynamic road network topology combines static road network with real-time traffic status, making the path planning more in line with the actual road traffic conditions and effectively avoiding distribution delays caused by traffic congestion. The establishment of the environmental sensitive path set fully considers environmental factors such as temperature-sensitive areas to ensure the temperature stability of cold chain food during transportation and guarantee food quality and safety. The prediction path avoidance strategy integrates historical data and real-time traffic events to achieve intelligent avoidance of risky road sections, significantly reducing the impact of unexpected situations during transportation. The final path prediction result considers multiple factors such as timeliness, environmental stability, and unexpected event avoidance to provide the optimal distribution scheme for cold chain logistics and achieve dual protection of distribution efficiency and food quality.

[0081] In one embodiment, the environmental impact evaluation result is associated with the road section according to the dynamic road network topology structure to obtain an environmental sensitive path set, including The dynamic road network topology structure is the core data carrier for cold chain distribution path analysis, and its road segment segmentation directly affects the accuracy of subsequent environmental evaluation. Road segment segmentation uses a discretization method based on road network characteristics to divide continuous road networks into the smallest calculable units with independent attributes. During implementation, segmentation nodes are determined based on road grades, intersection distributions, and traffic control areas to ensure that each basic road network unit has complete traffic attributes and environmental characteristics.

[0082] The urban expressway is divided by the on-ramp and off-ramp, the main road is divided by the intersection, and the branch road is divided by the actual length. The divided basic road network unit contains standardized coding, geographic coordinate range, road type, and basic traffic capacity metadata. The unit boundary setting needs to consider both the calculation efficiency and the data granularity. Over-fine division will lead to a sharp increase in calculation, and over-coarse division will affect the accuracy of the evaluation. The final generated basic road network unit set constitutes the minimum evaluation unit for subsequent environmental analysis, providing a structured data basis for road network environment scoring.

[0083] The road network environment analysis maps the environmental constraints of cold chain transportation to the basic road network unit and establishes a quantitative evaluation system. The environmental impact evaluation results include temperature field distribution, air pollution index, and sunshine duration, etc. These data are spatially overlaid with the basic road network unit. During implementation, the grid matching algorithm is used to align the environmental monitoring data with the geographic coordinates of the road network unit, and the relevant environmental parameters are extracted for each unit. When building the scoring model, different weights are set for different types of cold chain commodities. For example, frozen meat focuses on temperature stability, and fresh agricultural products pay extra attention to air humidity indicators.

[0084] The scoring dimensions include: temperature fluctuation coefficient (based on historical temperature change data), environmental risk index (integrating pollution and weather data), and sunshade coverage (identified through street view images). After normalization, the scores of each dimension are weighted and summed to generate a comprehensive environmental score for each road network unit. The final road segment environmental score matrix is stored in a two-dimensional table structure, with rows representing road network unit numbers and columns corresponding to scoring values for different time slices, supporting time-sensitive analysis.

[0085] Target road section identification filters out path sections that pose significant environmental risks to cold chain transportation from massive scoring data. The identification process uses a dynamic threshold judgment method, with the threshold set referring to the environmental tolerance standards allowed for cold chain commodities. When implemented, a time sequence sliding window analysis is performed on the scoring matrix, and units with scores exceeding the limit for multiple consecutive time slices are marked. For example, if the temperature score of a road section exceeds the threshold for frozen food in three consecutive time slices during the afternoon, it is determined to be a high-risk unit. The identification algorithm introduces spatial continuity verification to merge discrete high-scoring units into complete risk road sections based on road connectivity.

[0086] For long-distance risk areas that span multiple units, the connected component detection algorithm in graph theory is used for aggregation processing. The identification results are manually verified to exclude false positives, such as abnormal data caused by temporary construction. The final output target road section identification set contains three types of data: red road sections that must be avoided (environmental risk is unacceptable), yellow road sections that should be bypassed (risk is controllable but needs optimization), and blue road sections that need to be monitored (potential risk). The identification set and the dynamic road network topology establish a bidirectional index, supporting fast geographic positioning and attribute querying.

[0087] The environmental traffic hotspot area detection combines the environmental risk and the spatiotemporal distribution characteristics of traffic congestion of the target road section to identify the key area that has a combined impact on cold chain distribution. In the implementation process, the historical traffic flow data is used to construct a periodic congestion mode library, and the distribution of congestion road sections in typical time periods such as weekdays, holidays, morning and evening peak hours is extracted. The target road section identification set and the congestion data are subjected to spatiotemporal correlation analysis to screen out the superimposed area that meets the high environmental risk and high frequency congestion at the same time.

[0088] The detection algorithm introduces a sliding time window matching mechanism, for example, a road section has both high temperature exposure risk and regular congestion during 10:00-12:00 every day, which is determined as an environmental traffic hotspot area. The spatial clustering algorithm further aggregates discrete high-risk congestion road sections into continuous hotspot areas to ensure the integrity of the detection results. The final output of the environmental traffic hotspot area includes structured information such as geographical boundaries, risk level, high-risk period, and recommended detour strategy, which provides decision basis for subsequent path reconstruction.

[0089] The path reconstruction takes the environmental traffic hotspot area as the constraint condition to optimize the cold chain distribution path in the dynamic road network, ensuring the environmental stability of the transportation process. A multi-objective path planning strategy is adopted, which superimposes environmental score weight on the basis of traditional shortest path algorithm, so that the reconstructed path meets the dual requirements of optimal distance and lowest environmental risk. The reconstruction process is executed in stages: in the initial stage, all red road sections are excluded to generate a basic feasible path set; in the optimization stage, the dynamic adjustment mechanism of yellow road sections is introduced, and the detour scheme is calculated combined with real-time traffic data; in the final stage, the historical environmental data is integrated to predict future risks, and the alternative path is generated.

[0090] The path evaluation indicators include: environmental fluctuation index (whole journey temperature stability), risk exposure time (cumulative time passing through high-risk areas), and traffic efficiency (estimated travel time). The reconstruction result is output in the form of an environmental sensitive path set, each path is accompanied by an environmental fitness score and an applicable scenario description (such as suitable for frozen food, fresh fruits and vegetables, etc. different categories). The set supports a dynamic updating mechanism, which iteratively optimizes periodically with the changes of road network environment.

[0091] The embodiment realizes fine environmental risk assessment of cold chain transportation path through segment division and environment association processing of dynamic road network topology structure, and effectively improves the accuracy of distribution path planning. The road segment environment scoring matrix and target road segment identification mechanism can accurately capture sensitive road segments with environmental risks, providing reliable risk avoidance basis for cold chain transportation. Combined with the detection of hot spot areas with periodic congestion characteristics, the environmental risk and traffic conditions are analyzed cooperatively to ensure that the distribution path meets the requirements of environmental stability and traffic efficiency. Through dynamic recombination of environmental sensitive paths, an optimized path set considering transportation timeliness and temperature control requirements is constructed, which significantly improves the reliability and economy of cold chain logistics. The method supports real-time data-driven path dynamic adjustment, which can adapt to complex and variable road network environment changes, and provides an intelligent solution for safe distribution of cold chain food.

[0092] In one embodiment, the initial distribution efficiency data and the path prediction results are cross-analyzed to develop a global distribution scheme for the cold chain food distribution vehicles, including: In the process of distribution analysis of the cold chain food distribution vehicles according to the initial distribution efficiency data and the path prediction results, the initial distribution efficiency data as the original input reflects the basic running state of the vehicle in the historical distribution task. The initial distribution efficiency data usually contains multi-dimensional information such as time stamp, vehicle speed, energy consumption, temperature and humidity record, loading and unloading time, etc. The purpose of segmented calculation is to decompose the complex original data into two key indicators: transportation efficiency data and goods damage data.

[0093] The transportation efficiency data is generated by extracting the fields in the initial distribution efficiency data that are directly related to the vehicle operation, including average driving speed, idling time, route deviation times, and cold machine energy consumption proportion. These fields are normalized to form quantitative indicators reflecting the transportation efficiency of the vehicle. For example, the cold machine energy consumption proportion indirectly reflects the running load of the temperature control system by calculating the proportion of the refrigeration equipment in the total energy consumption. The idling time measures the invalid waiting of the vehicle outside the distribution node from the time dimension.

[0094] The goods damage data is generated by analyzing the fields in the initial distribution efficiency data related to the quality of the goods, including temperature and humidity fluctuation frequency, over-limit temperature difference duration, and vibration amplitude peak value. The temperature and humidity fluctuation frequency counts the number of times that the temperature and humidity exceed the preset threshold within a unit time, and the over-limit temperature difference duration records the cumulative duration of the temperature deviation from the safety interval in the cargo hold. The vibration amplitude peak value reflects the mechanical damage that the goods can suffer during transportation from the physical impact angle. When segmented calculation, abnormal values such as instantaneous temperature and humidity jump caused by sensor failure need to be excluded.

[0095] Transport efficiency data and goods damage data serve path optimization and goods safety guarantee respectively in subsequent analysis. The generation of the two types of data relies on the same set of initial distribution efficiency data, but the calculation logic and target dimensions are completely different. The accuracy of the segmented calculation directly affects the reliability of subsequent deviation detection and joint calculation.

[0096] Path prediction results are theoretical values of the route to be traveled based on historical traffic conditions, traffic rules, and weather conditions, including estimated travel time, distance, and sequence of nodes passed through. Transport efficiency data reflects the actual operating state, and there is a certain difference between it and the path prediction results. The core of deviation detection is to quantify this difference and form a comparable path matching degree index.

[0097] The calculation of path matching degree focuses on the comparison of key parameters. The travel time deviation rate reflects the degree of route congestion or detour by comparing the percentage difference between actual travel time and predicted travel time. The distance deviation degree measures the closeness of path execution by taking the absolute difference between actual travel distance and predicted travel distance as a proportion of predicted travel distance. The node coincidence rate counts the proportion of actual nodes passed through and predicted node sequence, reflecting the integrity of route execution.

[0098] Deviation detection needs to distinguish between systematic deviation and random deviation. Systematic deviation shows persistent deviation in a specific period or area, such as fixed detour due to construction. Random deviation is caused by sudden conditions, such as temporary traffic control. Path matching degree needs to mark the type of deviation to provide correction basis for subsequent global distribution construction. During the detection process, abnormal data caused by human factors such as drivers changing routes without reporting need to be excluded.

[0099] Path matching degree is a bridge connecting path prediction and actual transportation. High matching degree indicates that the prediction model is accurate, and existing strategies can be directly used; low matching degree triggers the need for dynamic adjustment of the path prediction model. This index provides data screening conditions for subsequent joint calculation and global construction.

[0100] Goods damage data has extracted indicators such as temperature, humidity, and vibration that are directly related to goods quality through segmented calculation, while environmental impact assessment results include external factor data such as temperature change curve along the way, ultraviolet intensity, and rainfall probability. The purpose of joint calculation is to quantify the additive effect of environmental factors on goods damage and generate more accurate damage correction data.

[0101] The temperature and humidity related correction is based on the temperature and humidity fluctuation in the cargo damage data, combined with the external temperature and humidity changes in the environmental impact assessment results for compensation calculation. For example, when the external temperature rises suddenly, the increased cooling load can cause a temporary fluctuation in the cabin temperature. Such fluctuations need to be distinguished from abnormal fluctuations caused by equipment failure. The vibration damage correction introduces road flatness rating and weather-induced visibility reduction data to attribute mechanical impact to specific environmental conditions.

[0102] The joint calculation adopts a weight distribution mechanism. High-frequency but low-intensity environmental disturbances (such as short-term gusts) are given lower weights, while low-frequency but high-intensity disturbances (such as sustained heavy rain) are given higher weights. The corrected data needs to be labeled with environmental impact factor labels, such as "temperature difference correction" and "rain and fog correction", to facilitate traceability analysis. The calculation process needs to avoid repeated correction, such as environmental impact that has been offset by the temperature control system self-adjustment and is no longer repeated.

[0103] The damage correction data is the basis for decision-making of cargo safety protection strategy. The corrected indicators more truly reflect the state of the goods under environmental forces majeures, providing a basis for path selection and temperature control parameter adjustment in global distribution scheme. This step converts static cargo damage data into dynamic risk prediction indicators, which together with path matching degree form the dual input of global distribution construction.

[0104] Path matching degree, damage correction data, and path prediction results are the outputs of the previous steps, which together form the core input of global distribution construction. The goal of global distribution construction is to generate a comprehensive distribution scheme that balances transportation efficiency and cargo safety. The core is to balance the contradiction between path optimization and risk avoidance.

[0105] Path matching degree directly affects the credibility of path prediction results. For road segments with a matching degree higher than the preset threshold, the planned route in the path prediction results is directly used. For road segments with a matching degree lower than the threshold, a dynamic path adjustment mechanism is triggered. The adjustment basis includes real-time traffic data, historical deviation analysis conclusions, and high-risk area labeling in damage correction data. For example, if a road segment has a high cargo damage rate due to frequent temperature and humidity fluctuations, it should be avoided even if the path matching degree is acceptable.

[0106] Damage correction data plays a screening role in route selection. For products that are easily affected by the environment (such as fresh dairy products), routes with small temperature and humidity fluctuations and low vibration intensity are preferred, even if the route is longer or takes more time. For cargo with strong damage resistance (such as frozen meat), environmental restrictions are appropriately relaxed to prioritize transportation efficiency. The global distribution scheme needs to prioritize different types of cargo and match different path strategies.

[0107] The spatiotemporal dimensions of the path prediction results are reorganized. In the time dimension, the time period risk distribution in the damage correction data is used to avoid high-temperature periods or traffic peaks. In the spatial dimension, the path matching degree is combined to screen alternative routes, forming a multi-level backup path library. The final global distribution scheme includes structured data such as the main path, backup path, expected damage rate of each road segment, cold machine preset parameters, and environmental risk warning markers.

[0108] The global distribution scheme is not a static output, but has a dynamic updating mechanism. When the deviation between real-time monitoring data and scheme preset parameters exceeds the tolerance, local path re-planning is automatically triggered. The executability of the scheme is verified through simulation, ensuring a balance between theoretical optimality and practical operability. The output of this step directly guides the real-time scheduling and path execution of cold chain vehicles, forming a closed loop of analysis-decision-execution.

[0109] This embodiment separates the transportation efficiency and damage risk by segmenting the initial distribution efficiency data, accurately quantifying key indicators in the cold chain distribution process, and providing a reliable data foundation for subsequent analysis. Deviation detection on transportation efficiency data based on path prediction results can dynamically identify the matching degree of actual transportation paths and predicted paths, and timely discover route execution deviations, providing objective basis for path optimization. By integrating path matching degree, damage correction data, and path prediction results for global distribution construction, the transportation efficiency and product safety are optimized, and the generated distribution scheme ensures timeliness and reduces damage risk. The global distribution scheme has a dynamic updating mechanism, which can automatically adjust the path planning according to real-time monitoring data, significantly improving the adaptability and reliability of cold chain distribution.

[0110] In one embodiment, the damage data and environmental impact assessment results are jointly calculated to obtain damage correction data, including: As a core indicator reflecting the quality change of goods during cold chain transportation, the time continuity of damage data determines that it must be divided into time periods to reveal the damage pattern. Transportation period division is based on fixed time windows, combined with the business characteristics of cold chain distribution, and divides the entire transportation cycle into several time period units with business significance. Typical division methods include hourly division, transportation stage division, and temperature control interval division.

[0111] Hourly division divides each period by a fixed time interval (e.g., 30 minutes) and calculates the average damage data for each period. This division method is suitable for short-distance transportation or scenarios with frequent damage changes. Transportation stage division divides the transportation process into loading stage, trunk transportation stage, and urban distribution stage, each stage using different statistical strategies. Temperature control interval division divides the time period according to the temperature fluctuation range of the cargo compartment, focusing on damage performance under temperature critical conditions.

[0112] The generation of the damage rate curve relies on the data aggregation after time period division. Each time period unit calculates the cargo damage rate indicators, including the duration proportion of temperature and humidity exceeding the standard, the number of times of vibration exceeding the limit, etc. These indicators are connected in time sequence to form a curve, which intuitively shows the trend of the damage rate changing with time. The curve is smoothed to eliminate accidental fluctuations and retain statistically significant peaks and valleys. The horizontal axis of the damage rate curve is the time dimension, and the vertical axis is the normalized damage degree score.

[0113] The granularity of time period division directly affects the accuracy of the damage rate curve. Too coarse division may mask key damage events, and too fine division may introduce noise interference. In practical applications, a dynamic adjustment mechanism is used to automatically increase the division during periods of severe damage fluctuations and relax it during stable periods. The generated damage rate curve provides a time alignment benchmark for subsequent environmental impact analysis.

[0114] The environmental impact assessment results serve as an external factor data set, containing multi-dimensional environmental monitoring information. The purpose of feature extraction is to separate the key environmental factors directly related to cargo damage, namely temperature and humidity changes and vibration effects, two core data. The extraction process follows the "de-redundancy, correlation preservation" principle, eliminating environmental parameters unrelated to cold chain transportation.

[0115] The extraction of temperature and humidity change curves is based on meteorological station data along the route and vehicle-mounted environmental sensor records. External temperature and humidity data are processed through coordinate mapping and time synchronization to ensure spatio-temporal matching with the transportation route. When generating the curve, the effects of different altitudes on temperature and humidity are compensated to eliminate measurement deviations caused by topographic factors. Key feature parameters such as diurnal temperature difference amplitude and humidity sudden change points are extracted, and key periods that can cause condensation are labeled.

[0116] The extraction of vibration impact data integrates road roughness detection records, vehicle vibration sensor data, and historical jolt event reports. The vibration frequency spectrum is decomposed by road section to distinguish between persistent mild vibration and sudden severe vibration. Data standardization is performed according to the shock absorption performance of different vehicle types to eliminate the influence of vehicle characteristics on vibration perception. The extraction results include structured indicators such as vibration intensity classification of each road section and typical vibration frequency distribution.

[0117] During the feature extraction process, a mapping relationship between environmental data and transportation routes is established. The temperature and humidity change curve is segmented and labeled according to geographic coordinates, and the vibration impact data is bound to specific road numbers. This spatial correlation ensures that subsequent analysis can accurately locate the position of environmental factors. The extracted environmental feature data maintains the same time reference as the damage rate curve, laying the foundation for joint analysis.

[0118] The alignment of damage rate curve and temperature-humidity curve aims to reveal the quantitative relationship between environmental temperature-humidity fluctuation and cargo damage. The alignment process includes three key steps: time axis calibration, data interpolation, and correlation detection. Time axis calibration addresses the inconsistency in the collection frequency of the two types of curves, ensuring that the data points correspond to the same physical time through timestamp matching.

[0119] Data interpolation processing fills in missing data caused by sensor failure. The temperature-humidity curve uses cubic spline interpolation to maintain smoothness, and the damage rate curve uses nearest neighbor interpolation to avoid introducing false fluctuations. The interpolated double curve performs sliding window correlation calculation, and the window size is dynamically adjusted according to the total transportation time, usually set to a duration that covers a typical temperature-humidity change cycle.

[0120] The extraction of temperature-humidity factors is based on correlation analysis results. High correlation periods are identified as sensitive intervals, and the contribution weight of temperature-humidity change to damage rate is calculated within this interval. The factor is represented as a multi-dimensional vector, including temperature influence coefficient, humidity influence coefficient, and interaction coefficient. The calculation of the coefficient excludes the interference of other environmental factors, for example, when vibration impact data indicates that there is a severe jolt at a certain time, the data at that time is not used in the calculation of temperature-humidity factors.

[0121] Temperature-humidity factors are accompanied by effectiveness annotations. Continuous and stable correlation produces high-confidence factors, while occasional correlation is marked as pending verification factors. Independent factor libraries are established for different types of goods, such as dairy products and frozen meat, and their temperature-humidity factors are calculated and stored separately. The obtained temperature-humidity factors will serve as the basis for compensation correction, and their accuracy directly affects the reliability of the final damage correction data.

[0122] Vibration response analysis focuses on the quantitative impact of mechanical vibration on cargo damage, and its core is to establish the mapping relationship between vibration characteristics and damage rate fluctuations. The matching of damage rate curve and vibration impact data uses an event-driven mode, rather than simple time alignment. Vibration events are defined as acceleration changes exceeding a certain threshold, including instantaneous impact vibration and continuous resonance vibration.

[0123] The extraction of vibration factors is based on the spatio-temporal coupling analysis of vibration events and damage rate surges. For each vibration event, an affected time window is drawn on the damage rate curve, and the correlation coefficient between vibration parameters and damage change is calculated. Vibration parameters include peak acceleration, vibration duration, and main frequency component, while damage change reflects the damage rate increment or quality degradation amplitude. The correlation coefficient is calculated through non-parametric statistical methods to avoid bias caused by data distribution assumptions.

[0124] The expression of the vibration factor adopts a hierarchical quantization strategy. Low-frequency vibration (<10 Hz) mainly affects the integrity of the packaging structure, and its factor is expressed as a cumulative damage coefficient; high-frequency vibration (>50 Hz) is prone to cause damage to the microstructure of the contents, and the corresponding transient impact factor. Each frequency band vibration factor is accompanied by a direction weight, and the vibration in the vertical direction usually has a higher damage contribution than that in the horizontal direction.

[0125] Vibration response analysis needs to exclude pseudo-correlation interference. When the temperature and humidity change curve shows that the environmental parameters change dramatically at the same time, the partial correlation analysis is used to separate the independent influence of the vibration factor. The vibration factor library is established according to the vibration resistance characteristics of the goods, for example, different factor calculation models are used for glass bottle liquid and carton packaged food. The final output vibration factor contains three-dimensional parameters of frequency domain characteristics, time domain intensity and direction sensitivity.

[0126] Multi-factor compensation correction is a key step to eliminate environmental interference and restore the true damage state. The correction process follows the principle of "decomposition first and then synthesis", and the original damage data is decoupled into environmental damage components and inherent damage components. The temperature and humidity factor is used to compensate for the accelerated degradation caused by the hot and humid environment, and the vibration factor is used to offset the additional damage caused by mechanical impact.

[0127] Temperature and humidity compensation uses a reverse deduction method. According to the factor coefficient of each time period in the temperature and humidity change curve, the theoretical environmental damage is calculated, and the component is deducted from the measured damage data. The compensation amount calculation introduces a hysteresis effect correction, considering the continuity characteristics of the impact of temperature changes on goods. For example, the quality decline caused by temperature exceeding the standard in a certain period can continue to affect the next 2-3 periods, and an attenuation model needs to be used for cross-period compensation.

[0128] Vibration compensation implements an impact event backtracking mechanism. For each identified vibration event, the expected damage increment is calculated according to the vibration factor parameters, and the data in the corresponding time window is corrected. The compensation algorithm distinguishes between reversible damage and irreversible damage: reversible damage (such as vegetable and fruit skin indentation) is partially backfilled according to the elastic recovery model; irreversible damage (such as packaging rupture) is directly marked as inherent damage.

[0129] The synthesis of the corrected data is completed by weighted fusion. After the environmental factor compensation, the damage data of each period is re-aggregated by assigning different weight coefficients according to the type of goods. The weight allocation considers the sensitivity difference of goods to environmental factors, for example, the weight of temperature and humidity of frozen seafood is higher than that of vibration, while the weight of vibration of precision instruments dominates. The generated damage correction data eliminates the environmental interference component and can more accurately reflect the impact of transportation operations on the quality of goods. The corrected data is stored with the mapping relationship between the original data and the environmental factors, supporting traceability analysis.

[0130] The embodiment realizes accurate quantitative analysis of cargo damage in the cold chain transportation process by joint calculation of cargo damage data and environmental impact evaluation results. Dividing the transportation process into multiple time periods and generating a damage rate curve can accurately capture the damage characteristic changes of different transportation stages and provide a reliable time benchmark for subsequent analysis. The extraction of temperature and humidity change curves and vibration influence data effectively separates key environmental impact factors, avoiding the interference of redundant data. Through the alignment analysis of the damage rate curve and the environmental parameters, the temperature and humidity factors and the vibration factors can objectively reflect the actual influence degree of environmental factors on cargo damage. Based on the multi-factor compensation correction method, the influence of environmental interference on damage evaluation is significantly reduced, and the corrected damage data more truly reflects the influence of transportation operation itself on cargo quality. The technical scheme not only improves the accuracy of cold chain transportation quality evaluation, but also provides reliable data support for optimizing the transportation scheme, and has important practical application value.

[0131] Referring to Figure 2 The application also provides a cold chain food distribution analysis system based on big data, which is applied to the cold chain food distribution analysis method based on big data. The acquisition module is used for dynamic feature extraction on the multi-source monitoring data of the cold chain food distribution vehicle to obtain generated transportation state parameters. The analysis module is used for associating and matching the transportation state parameters with the real-time environmental data of the cold chain food distribution vehicle to obtain environmental impact evaluation results. The association module is used for performing distribution efficiency analysis on the vehicle path transportation data of the cold chain food distribution vehicle based on the environmental impact evaluation results to form initial distribution efficiency data. The processing module is used for road condition diagnosis on the path transportation data, path prediction in combination with the environmental impact evaluation results, and path prediction results. The control module is used for cross analysis on the initial distribution efficiency data and the path prediction results to formulate a global distribution scheme of the cold chain food distribution vehicle.

[0132] The present invention provides a cold chain food distribution analysis system based on big data, which obtains transportation status parameters through dynamic feature extraction of multi-source monitoring data, realizes real-time and accurate monitoring of the operating status of cold chain distribution vehicles, effectively solves the problem of delayed response of traditional methods to dynamic changes in the transportation process, and significantly improves the early warning capability of abnormal situations. By correlating and matching real-time environmental data with transportation status parameters, a dynamic correlation model of environmental factors and cold chain efficiency is established, which overcomes the problem of unsystematic assessment of the impact of environmental factors in existing methods and provides a scientific basis for the optimization of temperature control strategies. Based on the path transportation data, the distribution efficiency analysis of the environmental impact assessment results is carried out to achieve the coordinated optimization of path planning and cold chain energy consumption, and solves the difficult problem of balancing timeliness and temperature control requirements in traditional distribution solutions. By performing path prediction based on road condition detection results and environmental impact assessment, a dynamic intelligent path planning mechanism is constructed, which effectively avoids the problems of distribution delays and temperature control failures caused by changes in road conditions and improves distribution reliability. By integrating initial delivery efficiency data and route prediction results to generate a global delivery plan, intelligent management of the entire cold chain delivery process is achieved, which not only ensures food quality and safety, but also optimizes logistics resource allocation and significantly improves the overall operational efficiency of cold chain logistics.

[0133] It should be noted that, those skilled in the art will clearly understand that, for the sake of convenience and brevity of description, the specific working processes of the above-described system and each module can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0134] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A cold chain food distribution analysis method based on big data, characterized in that: include: By extracting dynamic features from multi-source monitoring data of cold chain food distribution vehicles, transportation status parameters are generated. Correlating and matching the transport status parameters with the real-time environmental data of the cold chain food delivery vehicle to obtain an environmental impact assessment result; Performing distribution efficiency analysis on the vehicle route transportation data of the cold chain food distribution vehicle based on the environmental impact assessment result to form initial distribution efficiency data; Performing road condition diagnosis on the route transportation data, performing route prediction based on the environmental impact assessment results, and generating a route prediction result; The initial delivery efficiency data and the path prediction results are cross-analyzed to formulate a global delivery plan for the cold chain food delivery vehicles.

2. The cold chain food distribution analysis method based on big data according to claim 1 is characterized in that: The method generates transport status parameters by dynamically extracting features from multi-source monitoring data of cold chain food delivery vehicles, including: Obtaining the vehicle temperature and humidity of the cold chain food delivery vehicle, performing differential fluctuation detection, and obtaining vehicle-mounted temperature change data; Acquire vehicle vibration data, and identify abnormal accumulation in combination with the vehicle temperature change data to obtain a vehicle degradation index; Obtain vehicle cargo distribution data and perform deformation analysis to obtain cargo deformation distribution data; Obtain the energy consumption data of the vehicle-mounted equipment and perform power consumption attenuation calculation to obtain the equipment energy efficiency attenuation rate; The vehicle-mounted temperature change data, the vehicle-mounted degradation index, the cargo deformation distribution data, and the equipment energy efficiency attenuation rate are feature-fused to obtain the transportation status parameter.

3. The cold chain food distribution analysis method based on big data according to claim 1 is characterized in that: The associating and matching the transport status parameters with the real-time environmental data of the cold chain food delivery vehicle to obtain an environmental impact assessment result includes: Acquire vehicle perception data and real-time meteorological data of the cold chain food delivery vehicle, perform environmental correlation, and obtain the real-time environmental data; Performing multi-source heterogeneous protocol conversion and outlier cleaning on the real-time environmental data to obtain benchmark environmental data; Performing multi-dimensional feature fusion on the reference environmental data and the transport state parameters to obtain an environmental state correlation matrix; Clustering the influencing factors of the environmental state association matrix to obtain a classification result of the environmental influencing factors; The transport status parameters are quantitatively evaluated according to the classification results of the environmental impact factors to obtain the environmental impact assessment results.

4. The cold chain food distribution analysis method based on big data according to claim 1 is characterized in that: The performing of distribution efficiency analysis on the vehicle route transportation data of the cold chain food distribution vehicle based on the environmental impact assessment result to form initial distribution efficiency data includes: Obtaining the delivery destination address and map data of the cold chain food delivery vehicle, and performing route planning to obtain the route transportation data; Performing driving matching on the route transportation data based on the cold chain food delivery vehicle to obtain the road section travel time; Conducting a temperature control stability analysis on the travel time of the road section according to the environmental impact assessment results to obtain temperature control compliance rate data; Performing equipment operation matching on the temperature control compliance rate data based on the transportation status parameters to obtain an equipment operation efficiency index; A timeliness evaluation is performed based on the road section travel time and the equipment operation efficiency index to obtain the initial delivery efficiency data.

5. The cold chain food distribution analysis method based on big data according to claim 1 is characterized in that: The performing of road condition diagnosis on the route transportation data, performing route prediction in combination with the environmental impact assessment result, and generating a route prediction result includes: Performing traffic congestion flow analysis on the route transportation data based on a preset traffic database to obtain periodic congestion characteristics; Performing dynamic road analysis on the path transportation data according to the periodic congestion characteristics to obtain a dynamic road network topology structure; Performing road segment environment association on the environmental impact assessment results according to the dynamic road network topology structure to obtain an environmentally sensitive path set; Predicting candidate paths for the environment-sensitive path set and matching traffic events with the traffic database to obtain a predicted path avoidance strategy; Based on the predicted path avoidance strategy, path planning prediction is performed on the dynamic road network topology structure to obtain the path prediction result.

6. The cold chain food distribution analysis method based on big data according to claim 5 is characterized in that: The environmental impact assessment results are associated with road sections according to the dynamic road network topology to obtain an environmentally sensitive path set, including Segmenting the dynamic road network topology structure to obtain a basic road network unit set; Performing a road network environmental analysis on the basic road network unit set according to the environmental impact assessment results to obtain a road section environmental scoring matrix; Performing road segment identification on the road segment environment score matrix to obtain a target road segment identification set; Performing hotspot detection on the target road segment identifier set according to the periodic congestion characteristics to obtain environmental traffic hotspots; The dynamic road network topology structure is reorganized based on the environmental traffic hotspot area to obtain the environmentally sensitive path set.

7. The cold chain food distribution analysis method based on big data according to claim 1 is characterized in that: The cross-analysis of the initial delivery efficiency data and the route prediction results to formulate a global delivery plan for the cold chain food delivery vehicle includes: Performing segmented calculations on the initial delivery efficiency data to obtain transportation efficiency data and cargo damage data; Performing deviation detection on the transportation efficiency data according to the path prediction result to obtain a path matching degree; Performing a combined calculation of the cargo damage data and the environmental impact assessment results to obtain damage correction data; A global distribution plan is obtained by constructing a global distribution plan based on the path matching degree, the damage correction data and the path prediction result.

8. The cold chain food distribution analysis method based on big data according to claim 7 is characterized in that: The combined calculation of the cargo damage data and the environmental impact assessment results to obtain damage correction data includes: The cargo damage data is divided into transport periods to obtain a damage rate curve; Performing feature extraction on the environmental impact assessment results to obtain temperature and humidity change curves and vibration impact data; Performing alignment analysis on the damage rate curve and the temperature and humidity change curve to obtain a temperature and humidity factor; Performing vibration response analysis based on the damage rate curve and the vibration impact data to obtain a vibration factor; The cargo damage data is corrected by performing multi-factor compensation based on the temperature and humidity factors and the vibration factors to obtain the damage correction data.

9. A cold chain food distribution analysis system based on big data, characterized in that: The cold chain food distribution analysis method based on big data applied to any one of claims 1 to 8 above comprises: An acquisition module is used to generate transport status parameters by dynamically extracting features from multi-source monitoring data of cold chain food delivery vehicles; An analysis module, configured to correlate and match the transport status parameters with the real-time environmental data of the cold chain food delivery vehicle to obtain an environmental impact assessment result; an association module configured to perform distribution efficiency analysis on the vehicle path transportation data of the cold chain food distribution vehicle based on the environmental impact assessment result to form initial distribution efficiency data; a processing module, the processing module being used to perform road condition diagnosis on the route transportation data, perform route prediction in combination with the environmental impact assessment result, and generate a route prediction result; A control module is used to cross-analyze the initial delivery efficiency data and the path prediction results to formulate a global delivery plan for the cold chain food delivery vehicle.

Images