Carrier automatic assessment method and system based on big data analysis

Abstract

The invention relates to the technical field of logistics management, in particular to a carrier automatic assessment method and system based on big data analysis, and the method comprises the following steps: S1, data collection and preprocessing: collecting performance data about carriers; s2, data analysis: extracting key performance indicators of the carrier; s3, complex network analysis and dynamic adjustment: key nodes and bottlenecks are identified, and key performance indicators are dynamically adjusted; s4, model establishment: establishing a carrier assessment model, and comprehensively scoring each index of a carrier; s5, automatic assessment: performing real-time assessment on the carrier through the carrier assessment model, and generating an assessment report; and S6, performing feedback and optimization: feeding back an assessment result and an optimization suggestion to the carrier according to the assessment result, and optimizing the assessment model according to the actual performance of the carrier. According to the invention, the change of the external environment can be reflected in real time, and the accuracy and fairness of the assessment result are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics management, and in particular to a method and system for automatically evaluating carriers based on big data analysis. Background Art

[0002] With the rapid development of the logistics industry, the selection and performance evaluation of carriers have become particularly important. Traditional carrier evaluation methods mainly rely on manual review and simple indicator scoring, which cannot fully reflect the actual performance of carriers. In addition, the logistics process involves a large amount of data, including transportation time, service quality, customer evaluation and transportation costs. These data come from a wide range of sources and in different formats, which brings great challenges to data collection and processing.

[0003] In the existing technology, although there are some carrier assessment methods based on data analysis, most of them have the following shortcomings: first, the data collection channel is single and the data is incomplete, which affects the accuracy and representativeness of the assessment results; second, the data analysis methods are limited and cannot fully explore the complex relationship between the data, especially in terms of multi-dimensional data analysis and network structure analysis; third, the dynamic adjustment ability of the assessment model is insufficient and cannot reflect the changes in the external environment in real time, resulting in the assessment results lagging behind the actual situation.

[0004] In response to the above problems, the present invention proposes a method and system for automatic assessment of carriers based on big data analysis, which provides accurate, timely and scientific performance evaluation and improves the service level and operational efficiency of carriers. Summary of the invention

[0005] Based on the above objectives, the present invention provides a method and system for automatic assessment of carriers based on big data analysis.

[0006] The automatic assessment method for carriers based on big data analysis includes the following steps: S1, data collection and preprocessing: collect performance data about carriers through various channels, including transportation time, service quality, customer evaluation and transportation cost, and preprocess the collected performance data; S2, data analysis: conduct multi-dimensional analysis on the pre-processed performance data to extract the carrier's key performance indicators (KPIs), including transportation timeliness, transportation costs, and customer satisfaction; S3, Complex Network Analysis and Dynamic Adjustment: Build a relationship network between carriers, customers, and suppliers, analyze the network structure, identify key nodes and bottlenecks, and dynamically adjust key performance indicators (KPIs), including: S31, Relationship network construction: Through complex network analysis technology, the relationship network between carriers, customers and suppliers is constructed, the connection status of each node and edge is recorded, and a logistics relationship map is formed; S32, Network Structure Analysis: Analyze the structure of the relationship network and the connections between nodes, identify key nodes and potential bottlenecks in the logistics process, and evaluate their impact on the overall logistics performance; S33, dynamic adjustment: dynamically adjust the key performance indicators (KPIs) based on the network structure analysis results and real-time performance data; S4, model establishment: establish a carrier assessment model based on the adjusted key performance indicators (KPIs) and comprehensively score the various indicators of the carriers; S5, automatic assessment: Carriers are assessed in real time through the carrier assessment model and assessment reports are generated; S6, Feedback and Optimization: Based on the assessment results, feedback the assessment results and optimization suggestions to the carriers, and optimize the carrier assessment model based on the actual performance of the carriers.

[0007] Optionally, the data collection and preprocessing in S1 includes: S11, data collection: obtain transportation time from the carrier's transportation management platform through the API interface, collect customer evaluation data on the carrier's service quality through the customer feedback platform, obtain the carrier's transportation cost data through the financial platform, and obtain customer evaluation data on the carrier's overall satisfaction through online questionnaires; S12, data preprocessing: preprocess the collected carrier performance data, including cleaning, sorting, conversion and normalization.

[0008] Optionally, the data analysis in S2 includes: S21, Transportation Timeliness Analysis: By calculating the difference between the actual completion time and the estimated completion time of each transportation, the transportation timeliness index is obtained, which is expressed as: Among them, T delay is the average transport delay, T actual，i is the actual completion time of the i-th transport, T estimated，i is the estimated completion time of the i-th transport, and N is the number of transports; S22, Transportation cost analysis: By calculating the total cost of each transportation, including fuel cost, labor cost and other related expenses, the transportation cost index is obtained, which is expressed as: Among them, C total is the average transportation cost, C fuel，i , C labor，i and C other are the fuel cost, labor cost and other related costs of the i-th transportation, and N is the number of transportations; S23, customer satisfaction analysis: By performing sentiment analysis on customer evaluation data, the customer satisfaction score is extracted and expressed as: Among them, S customer is the average customer satisfaction score, S text，i is the sentiment analysis score of the i-th customer review, and N is the number of reviews; S24, key performance indicator extraction: Based on the analysis results of transportation timeliness, transportation cost and customer satisfaction, the key performance indicators (KPI) of the carrier are extracted and each indicator is standardized.

[0009] Optionally, the relationship network construction in S31 includes: S311, Data Collection: Collect relevant data from the carrier management platform, customer relationship management platform (CRM) and supply chain management platform (SCM), including basic information and interaction records of carriers, customers and suppliers; S312, Node definition: Carriers, customers and suppliers are considered as nodes in the relationship network, each node represents an entity, and the attributes of the node include basic information and performance data of the entity; S313, edge definition and weight calculation: The interaction between carriers and customers, and between carriers and suppliers are regarded as edges. The weight of the edge represents the intensity or frequency of the interaction, which is expressed as: Among them, W ij is the weight of the edge between node i and node j, I ij，t is the number of interactions between node i and node j at time t, F ij，t is the frequency of interaction between node i and node j at time t, and T is the total time period; S314, Relationship network construction: Through complex network analysis technology, all nodes and edges are connected to form a relationship network, which is expressed as: Among them, A ij is the element in the adjacency matrix that represents the relationship between node i and node j, W ij is the weight of the edge.

[0010] Optionally, the network structure analysis in S32 includes: S321, Node degree analysis: Calculate the degree of each node, expressed as: Among them, D i is the degree of node i, A ijis the element of the relationship between node i and node j in the adjacency matrix, and N is the total number of nodes in the network; S322, Betweenness Centrality Analysis: Calculate the betweenness centrality of each node, expressed as: Where CB(v) is the betweenness centrality of node v, σ st is the total number of shortest paths between node s and node t, σ st (v) is the number of shortest paths between node s and node t that passes through node v; S323, bottleneck identification: By calculating the edge load, analyzing the node and edge load, the bottleneck in the logistics network is identified, which is expressed as: Among them, L ij is the load of edge ij, M is the number of all logistics flow paths, f k is the flow rate of path k, δ k (i, j) is an indicator function, indicating whether path k passes through edge ii; S324, Critical Node and Bottleneck Assessment: Based on the analysis results of node degree, betweenness centrality and edge load, the critical nodes and bottlenecks in the logistics network are identified, and the impact of critical nodes and bottlenecks on the overall logistics performance is evaluated, including: Key node evaluation: Comprehensive node degree and betweenness centrality are evaluated, expressed as: K i =α·D i +β·C B (i); Among them, K i is the critical evaluation value of node i, D i is the degree of node i, C B (i) is the betweenness centrality of node i, α and β are weight coefficients; Bottleneck evaluation formula: Evaluate through the edge load, expressed as: B ij =L ij ; Among them, B ij is the bottleneck evaluation value of edge ij, L ij is the load of edge ij.

[0011] Optionally, the dynamic adjustment in S33 includes: S331, Real-time data collection: Continuously collect real-time performance data of carriers; S332, dynamic update of network structure: update the properties of nodes and edges in the relationship network according to the real-time collected performance data; S333, key performance indicator adjustment: based on the network structure analysis results and real-time performance data, dynamically adjust the key performance indicators, expressed as: KPI i ′=KPI i ×(1+β·ΔP i ); Among them, KPl i ' is the adjusted key performance indicator, KPI i is the original key performance indicator, δ is the adjustment coefficient, ΔP i is the performance change rate of the i-th node; Among them, P current，i is the current performance value of the i-th node, P previous，i is the performance value of the i-th node in the previous period.

[0012] Optionally, the carrier assessment model in S4 adopts a factorization machine model, and the factorization machine model includes: S41, feature vectorization: converting the adjusted carrier performance data into a feature vector, expressed as: x=[x 1 , x 2 , ..., x n ]; Among them, x is the feature vector, x 1 , x 2 , ..., x n are the 1st, 2nd, ..., nth features respectively; S42, the model is constructed based on the factorization machine, introducing the interactive characteristics of external factors (such as real-time traffic, weather, etc.), expressed as: Among them, w 0 is the global bias, w i is feature x i The weight of <v i , v j > is the feature vector v i and v j The inner product of ij is the interaction feature x i and z j The weight of z j for external factors (such as real-time traffic data, weather conditions); S43, parameter adjustment: dynamically adjust model parameters according to real-time external factors to reflect the latest performance status, expressed as: Among them, β ij (t) is the interaction feature weight at time t, η is the learning rate, y t is the actual value, is the predicted value; S44, comprehensive score: Comprehensive score of each indicator of the carrier, expressed as: Among them, Score is the comprehensive score, αi is the weight of different indicators, is the predicted value of the ith indicator.

[0013] Optionally, the automatic assessment in S5 includes: S51, real-time data input: continuously obtain the real-time performance data of the carrier and input the real-time performance data into the carrier assessment model; S52, real-time assessment: the comprehensive score of the carrier is calculated in real time through the carrier assessment model; S53, assessment report generation: Generate an assessment report based on the calculated comprehensive score, including the score of each key performance indicator, the total score, and the performance change trend.

[0014] Optionally, the feedback and optimization in S6 includes: S61, Assessment result feedback: Feedback the assessment result to the carrier based on the assessment report generated by automatic assessment; S62, optimization suggestion generation: Generate optimization suggestions based on the assessment results and external environmental factors, expressed as: Optimization_Score = α·KPI current +β·External_Factors; Among them, Optimization_Score is the optimization score, KPl current is the current key performance indicator, External_Factors is the external factor, α and β are the weight coefficients; S63, model optimization: Based on the actual performance and feedback data of the carrier, the parameters of the carrier assessment model are continuously optimized using an adaptive learning algorithm, which is expressed as: Among them, θ new is the updated model parameter, θ old are the old model parameters, η is the learning rate, and L(θ) is the loss function.

[0015] The automatic carrier assessment system based on big data analysis is used to implement the above-mentioned automatic carrier assessment method based on big data analysis, and includes the following modules: Data collection and preprocessing module: collects performance data about carriers through various channels, including transportation time, service quality, customer evaluation and transportation cost, and preprocesses the collected performance data; Data analysis module: conducts multi-dimensional analysis on the pre-processed performance data and extracts the carrier's key performance indicators, including transportation timeliness, transportation costs, and customer satisfaction; Complex network analysis and dynamic adjustment module: builds the relationship network between carriers, customers and suppliers, analyzes the network structure, identifies key nodes and bottlenecks, and dynamically adjusts key performance indicators; Model building module: Based on the adjusted key performance indicators, a carrier assessment model is established to comprehensively score the various indicators of the carrier; Automatic assessment module: conducts real-time assessment of carriers through the carrier assessment model and generates assessment reports; Feedback and optimization module: Based on the assessment results, the assessment results and optimization suggestions are fed back to the carriers, and the carrier assessment model is optimized based on the carriers' actual performance.

[0016] Beneficial effects of the present invention: The present invention realizes comprehensive, accurate and real-time analysis and evaluation of carrier performance data through data collection and preprocessing, multi-dimensional data analysis and complex network analysis and dynamic adjustment. The data collection and preprocessing module collects data through multiple channels, and cleans, organizes and normalizes the data to ensure data quality and consistency. The data analysis module conducts in-depth multi-dimensional analysis of the preprocessed data, extracts key performance indicators, and provides a detailed and accurate performance evaluation basis.

[0017] The present invention can reflect the complexity and real-time changes of the actual logistics network by constructing a relationship network among carriers, customers and suppliers, analyzing the network structure, identifying key nodes and bottlenecks, and dynamically adjusting key performance indicators. The dynamic adjustment mechanism ensures the flexibility and adaptability of the model, can reflect changes in the external environment in real time, and improves the accuracy and fairness of the assessment results.

[0018] The present invention establishes a carrier assessment model based on adjusted key performance indicators, comprehensively scores various indicators of the carrier, and generates an assessment report. The system can conduct assessments in real time and provide detailed assessment results and optimization suggestions to the carrier through a feedback mechanism. The assessment model is continuously optimized according to the actual performance of the carrier to ensure the scientificity and effectiveness of the assessment. This comprehensive, dynamic and real-time assessment method greatly improves the efficiency and quality of carrier performance evaluation, and helps to improve the service level and operational efficiency of the carrier. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 A schematic diagram of the flow chart of the assessment method according to an embodiment of the present invention; Figure 2 Schematic diagram of system function modules according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0022] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0023] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0024] like Figure 1 As shown, the automatic carrier assessment method based on big data analysis includes the following steps: S1, data collection and preprocessing: collect performance data about carriers through various channels, including transportation time, service quality, customer evaluation and transportation cost, and preprocess the collected performance data to remove invalid data and unify the data format; S2, data analysis: conduct multi-dimensional analysis on the pre-processed performance data to extract the carrier's key performance indicators (KPIs), including transportation timeliness, transportation costs, and customer satisfaction; S3, Complex Network Analysis and Dynamic Adjustment: Build a relationship network among carriers, customers, and suppliers, analyze the network structure, identify key nodes and bottlenecks, and dynamically adjust key performance indicators (KPIs) to ensure that the assessment model can reflect the actual situation and provide more accurate assessment results. Specifically, it includes: S31, Relationship network construction: Through complex network analysis technology, the relationship network between carriers, customers and suppliers is constructed, the connection status of each node and edge is recorded, and a logistics relationship map is formed; S32, Network Structure Analysis: Analyze the structure of the relationship network and the connections between nodes, identify key nodes and potential bottlenecks in the logistics process, and evaluate their impact on the overall logistics performance; S33, Dynamic Adjustment: Dynamically adjust key performance indicators (KPIs) based on network structure analysis results and real-time performance data; S4, model establishment: based on the adjusted key performance indicators (KPIs), establish a carrier assessment model and comprehensively score the various indicators of the carriers; S5, automatic assessment: Carriers are assessed in real time through the carrier assessment model and assessment reports are generated; S6, Feedback and Optimization: Based on the assessment results, feedback the assessment results and optimization suggestions to the carriers, and optimize the carrier assessment model based on the actual performance of the carriers; Through the above steps, the performance of the carrier can be comprehensively and accurately evaluated, and the key performance indicators can be dynamically adjusted to ensure that the assessment model can reflect the actual situation in real time, thereby providing more accurate assessment results, improving the fairness and efficiency of the assessment, and continuously improving the carrier's service quality and operational performance through real-time feedback and optimization suggestions.

[0025] Data collection and preprocessing in S1 include: S11, data collection: obtain transportation time from the carrier's transportation management platform through the API interface, collect customer evaluation data on the carrier's service quality through the customer feedback platform, obtain the carrier's transportation cost data through the financial platform, and obtain customer evaluation data on the carrier's overall satisfaction through online questionnaires; S12, data preprocessing: preprocessing the collected carrier performance data, including cleaning, sorting, conversion and normalization; Data cleaning uses mean filling method to handle missing values ​​and delete obviously erroneous data points; Data collation unifies the time format and integrates data from different sources into a unified format; Data conversion converts non-numeric data into numeric data, for example, converting text data of customer reviews into ratings; Data normalization uses the Min-Max normalization method to convert data of different dimensions into the same dimension; Through the above steps, the integrity and consistency of the collected carrier performance data can be ensured, and the accuracy and reliability of subsequent data analysis and assessment models can be improved.

[0026] The data analysis in S2 includes: S21, Transportation Timeliness Analysis: By calculating the difference between the actual completion time and the estimated completion time of each transportation, the transportation timeliness index is obtained, which is expressed as: Among them, T delay is the average transport delay, T actual，i is the actual completion time of the i-th transport, T estimated，i is the estimated completion time of the i-th transport, and N is the number of transports; S22, Transportation cost analysis calculates the total cost of each transportation, including fuel cost, labor cost and other related expenses, to obtain the transportation cost index, which is expressed as: Among them, C total is the average transportation cost, C fuel，i , C labor，i and C other，i are the fuel cost, labor cost and other related costs of the i-th transportation, and N is the number of transportations; S23, customer satisfaction analysis: By performing sentiment analysis on customer evaluation data, the customer satisfaction score is extracted and expressed as: Among them, S customer is the average customer satisfaction score, S text，i is the sentiment analysis score of the i-th customer review, and N is the number of reviews; S24, extraction of key performance indicators: Based on the analysis results of transportation timeliness, transportation cost and customer satisfaction, the key performance indicators (KPIs) of the carriers are extracted, and each indicator is standardized to ensure that different indicators are comparable, expressed as: Among them, KPI′ is the standardized key performance indicator, KPI is the original indicator value, and KPI min and KPIs max are the minimum and maximum values ​​of the indicator respectively; Through the above content, we ensure the comparability and consistency of various indicators, provide a more reliable and scientific basis for assessment, help improve the accuracy, fairness and practicality of the assessment, and promote carriers to continuously optimize service quality and operational efficiency.

[0027] The relationship network construction in S31 includes: S311, Data Collection: Collect relevant data from the carrier management platform, customer relationship management platform (CRM) and supply chain management platform (SCM), including basic information and interaction records of carriers, customers and suppliers; S312, Node definition: Carriers, customers and suppliers are considered as nodes in the relationship network, each node represents an entity, and the attributes of the node include basic information and performance data of the entity; S313, edge definition and weight calculation: The interaction between carriers and customers, and between carriers and suppliers are regarded as edges. The weight of the edge represents the intensity or frequency of the interaction, which is expressed as: Among them, W ij is the weight of the edge between node i and node j, I ij，t is the number of interactions between node i and node j at time t, F ij，t is the frequency of interaction between node i and node j at time t, and T is the total time period; S314, Relationship network construction: Through complex network analysis technology, all nodes and edges are connected to form a relationship network, which is expressed as: Among them, A ij is the element in the adjacency matrix that represents the relationship between node i and node j, W ij is the weight of the edge; Through the above steps, a relationship network among carriers, customers and suppliers is constructed, and the connection status of each node and edge is recorded to form a logistics relationship map, which can intuitively display the interactive relationships and key nodes in the logistics network, providing an important basis for subsequent data analysis and performance evaluation.

[0028] The network structure analysis in S32 includes: S321, Node degree analysis: Calculate the degree of each node, that is, the number of edges directly connected to the node, expressed as: Among them, D i is the degree of node i, A ij is the element of the relationship between node i and node j in the adjacency matrix, N is the total number of nodes in the network, and nodes with higher degree values ​​indicate that they have stronger connectivity in the network and are potential key nodes; S322, Betweenness Centrality Analysis: Calculate the betweenness centrality of each node, that is, the number of shortest paths passing through the node, expressed as: Among them, C B (v) is the betweenness centrality of node v, σ st is the total number of shortest paths between node s and node t, σ st (v) is the number of shortest paths between node s and node t that passes through node v; S323, bottleneck identification: By calculating the edge load, analyzing the node and edge load, the bottleneck in the logistics network is identified, which is expressed as: Among them, L ij is the load of edge ij, M is the number of all logistics flow paths, f k is the flow rate of path k, δ k (i, j) is an indicator function, indicating whether path k passes through edge ii. If the load of the edge is too high, it means that the edge may become a bottleneck in the logistics process; S324, key node and bottleneck assessment: Based on the analysis results of node degree, betweenness centrality and edge load, key nodes and bottlenecks in the logistics network are identified, and the impact of key nodes and bottlenecks on the overall logistics performance is evaluated. Optimization suggestions are put forward to improve the overall efficiency and reliability of the logistics network, including: Key node evaluation: Comprehensive node degree and betweenness centrality are evaluated, expressed as: K i =α·D i +β·C B (i); Among them, K i is the critical evaluation value of node i, D i is the degree of node i, C B (i) is the betweenness centrality of node i, α and β are weight coefficients used to adjust the relative importance of degree and betweenness centrality; Bottleneck evaluation formula: Evaluate through the edge load, expressed as: B ij =L ij ; Among them, B ij is the bottleneck evaluation value of edge ij, L ij is the load of edge ij; Through the above steps, we can comprehensively analyze the structure of the relationship network and the connections between nodes, identify key nodes and potential bottlenecks in the logistics process, evaluate their impact on the overall logistics performance, and provide data support and decision-making basis for optimizing the logistics network.

[0029] Dynamic adjustments in S33 include: S331, Real-time data collection: Continuously collect real-time performance data of carriers; S332, dynamic update of network structure: based on the real-time collected performance data, update the attributes of nodes and edges in the relationship network to ensure that the network structure reflects the latest logistics status; S333, Key Performance Indicator Adjustment: Based on the network structure analysis results and real-time performance data, the key performance indicators are dynamically adjusted, expressed as: KPI i ′=KPI i ×(1+γ·ΔP i ); Among them, KPl i ' is the adjusted key performance indicator, KPI i is the original key performance indicator, γ is the adjustment coefficient, ΔP i is the performance change rate of the i-th node; Among them, P current，i is the current performance value of the i-th node, P previous，i is the performance value of the previous period of the ith node; through the above steps, according to the network structure analysis results and real-time performance data, the key performance indicators are dynamically adjusted to ensure that the assessment model can reflect the actual situation in real time and improve the accuracy and fairness of the assessment.

[0030] The carrier assessment model in S4 adopts the factorization machine model, which includes: S41, feature vectorization: The adjusted carrier performance data is converted into feature vectors, expressed as: x=[x 1 , x 2 , ..., x n ]; Among them, x is the feature vector, x 1 , x 2 , ..., x n are the 1st, 2nd, ..., nth features respectively; S42, the model is constructed based on the factorization machine, introducing the interactive characteristics of external factors (such as real-time traffic, weather, etc.) to enhance the adaptability and accuracy of the model, which is expressed as: Among them, w 0 is the global bias, w i is feature x i The weight of <v i , vj > is the feature vector v i and v j The inner product of ij is the interaction feature x i and z j The weight of z j for external factors (such as real-time traffic data, weather conditions); S43, parameter adjustment: dynamically adjust model parameters according to real-time external factors to reflect the latest performance status, expressed as: Among them, β ij (t) is the interaction feature weight at time t, η is the learning rate, y t is the actual value, is the predicted value; S44, comprehensive score: Comprehensive score of each indicator of the carrier, expressed as: Among them, Score is the comprehensive score, αi is the weight of different indicators, is the predicted value of the i-th indicator; Through the above, not only can the internal interaction of carrier performance data be captured, but it can also be dynamically adjusted to reflect the latest changes in the external environment. The result is a more comprehensive and accurate assessment of carriers, providing a more reliable performance evaluation and improving the service quality and operational efficiency of carriers.

[0031] The automatic assessment in S5 includes: S51, real-time data input: continuously obtain the real-time performance data of the carrier and input the real-time performance data into the carrier assessment model; S52, real-time assessment: the comprehensive score of the carrier is calculated in real time through the carrier assessment model; S53, assessment report generation: based on the calculated comprehensive score, an assessment report is generated, including the score of each key performance indicator, the total score, and the performance change trend; Through the above steps, the automatic assessment system can conduct real-time assessment of carriers through the carrier assessment model and generate assessment reports to provide managers with timely and accurate performance evaluation and decision-making support.

[0032] Feedback and optimization in S6 include: S61, Assessment result feedback: Feedback the assessment result to the carrier based on the assessment report generated by automatic assessment; S62, optimization suggestion generation: Combine the assessment results and external environmental factors to generate optimization suggestions to help carriers identify problems and develop improvement plans, expressed as: Optimization_Score = α·KPI current +β·External_Factors; Among them, Optimization_Score is the optimization score, KPI current is the current key performance indicator, External_Factors is the external factor, α and β are the weight coefficients; S63, model optimization: Based on the actual performance and feedback data of the carrier, the adaptive learning algorithm is used to continuously optimize the parameters of the carrier assessment model to ensure that the model can reflect the latest performance status and external environment changes in real time, expressed as: Among them, θ new is the updated model parameter, θ old is the old model parameter, η is the learning rate, and L(θ) is the loss function. Through the above steps, detailed assessment results and optimization suggestions can be fed back to the carrier based on the assessment results, and the assessment model can be continuously optimized according to the actual performance of the carrier to ensure the accuracy and practicality of the assessment.

[0033] like Figure 2 As shown, the automatic carrier assessment system based on big data analysis is used to implement the above-mentioned automatic carrier assessment method based on big data analysis, and includes the following modules: Data collection and preprocessing module: collects performance data about carriers through various channels, including transportation time, service quality, customer evaluation and transportation cost, and preprocesses the collected performance data; Data analysis module: conducts multi-dimensional analysis on the pre-processed performance data and extracts the carrier's key performance indicators, including transportation timeliness, transportation costs, and customer satisfaction; Complex network analysis and dynamic adjustment module: builds the relationship network between carriers, customers and suppliers, analyzes the network structure, identifies key nodes and bottlenecks, and dynamically adjusts key performance indicators; Model building module: Based on the adjusted key performance indicators, a carrier assessment model is established to comprehensively score the various indicators of the carrier; Automatic assessment module: conducts real-time assessment of carriers through the carrier assessment model and generates assessment reports; Feedback and optimization module: Based on the assessment results, the assessment results and optimization suggestions are fed back to the carriers, and the carrier assessment model is optimized based on the carriers' actual performance.

[0034] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0035] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. The automatic assessment method for carriers based on big data analysis is characterized by: The following steps are involved: S1, data collection and preprocessing: collect performance data about carriers through various channels, including transportation time, service quality, customer evaluation and transportation cost, and preprocess the collected performance data; S2, data analysis: conduct multi-dimensional analysis on the pre-processed performance data to extract the key performance indicators of the carrier, including transportation timeliness, transportation cost and customer satisfaction; S3, Complex Network Analysis and Dynamic Adjustment: Build a relationship network between carriers, customers, and suppliers, analyze the network structure, identify key nodes and bottlenecks, and dynamically adjust key performance indicators, including: S31, Relationship network construction: Through complex network analysis technology, the relationship network between carriers, customers and suppliers is constructed, the connection status of each node and edge is recorded, and a logistics relationship map is formed; S32, Network Structure Analysis: Analyze the structure of the relationship network and the connections between nodes, identify key nodes and potential bottlenecks in the logistics process, and evaluate their impact on overall logistics performance; S33, Dynamic Adjustment: Dynamically adjust key performance indicators based on network structure analysis results and real-time performance data; S4, model establishment: based on the adjusted key performance indicators, establish a carrier assessment model and comprehensively score the various indicators of the carrier; S5, automatic assessment: Carriers are assessed in real time through the carrier assessment model and assessment reports are generated; S6, Feedback and Optimization: Based on the assessment results, feedback the assessment results and optimization suggestions to the carriers, and optimize the carrier assessment model based on the actual performance of the carriers.

2. The method for automatically assessing carriers based on big data analysis according to claim 1, characterized in that: The data collection and preprocessing in S1 include: S11, data collection: obtain transportation time from the carrier's transportation management platform through the API interface, collect customer evaluation data on the carrier's service quality through the customer feedback platform, obtain the carrier's transportation cost data through the financial platform, and obtain customer evaluation data on the carrier's overall satisfaction through online questionnaires; S12, data preprocessing: preprocess the collected carrier performance data, including cleaning, sorting, conversion and normalization.

3. The method for automatic carrier assessment based on big data analysis according to claim 1 is characterized in that: The data analysis in S2 includes: S21, Transportation Timeliness Analysis: By calculating the difference between the actual completion time and the estimated completion time of each transportation, the transportation timeliness index is obtained, which is expressed as: Among them, T delay is the average transport delay, T actual，i is the actual completion time of the i-th transport, T estimated，i is the estimated completion time of the i-th transport, and N is the number of transports; S22, Transportation cost analysis: By calculating the total cost of each transportation, including fuel cost, labor cost and other related expenses, the transportation cost index is obtained, which is expressed as: Among them, C total is the average transportation cost, C fuel，i , C labor，i and C other，i are the fuel cost, labor cost and other related costs of the i-th transportation, and N is the number of transportations; S23, customer satisfaction analysis: By performing sentiment analysis on customer evaluation data, the customer satisfaction score is extracted and expressed as: Among them, S customer is the average customer satisfaction score, S text，i is the sentiment analysis score of the i-th customer review, and N is the number of reviews; S24, key performance indicator extraction: Based on the analysis results of transportation timeliness, transportation cost and customer satisfaction, the key performance indicators of the carrier are extracted and each indicator is standardized.

4. The method for automatic carrier assessment based on big data analysis according to claim 1, characterized in that: The relationship network construction in S31 includes: S311, data collection: collect relevant data from the carrier management platform, customer relationship management platform and supply chain management platform, including basic information and interaction records of carriers, customers and suppliers; S312, Node definition: Carriers, customers and suppliers are considered as nodes in the relationship network, each node represents an entity, and the attributes of the node include basic information and performance data of the entity; S313, edge definition and weight calculation: The interaction between carriers and customers, and between carriers and suppliers are regarded as edges. The weight of the edge represents the intensity or frequency of the interaction, which is expressed as: Among them, W ij is the weight of the edge between node i and node j, I ij，t is the number of interactions between node i and node j at time t, F ij，t is the frequency of interaction between node i and node j at time t, and T is the total time period; S314, Relationship network construction: Through complex network analysis technology, all nodes and edges are connected to form a relationship network, which is expressed as: Among them, A ij is the element in the adjacency matrix that represents the relationship between node i and node j, W ij is the weight of the edge.

5. The method for automatic carrier assessment based on big data analysis according to claim 4 is characterized in that: The network structure analysis in S32 includes: S321, Node degree analysis: Calculate the degree of each node, expressed as: Among them, D i is the degree of node i, A ij is the element of the relationship between node i and node j in the adjacency matrix, and N is the total number of nodes in the network; S322, Betweenness Centrality Analysis: Calculate the betweenness centrality of each node, expressed as: Among them, C B (v) is the betweenness centrality of node v, σ st is the total number of shortest paths between node s and node t, σ st (v) is the number of shortest paths between node s and node t that passes through node v; S323, bottleneck identification: By calculating the edge load, analyzing the node and edge load, the bottleneck in the logistics network is identified, which is expressed as: Among them, L ij is the load of edge ij, M is the number of all logistics flow paths, f k is the flow rate of path k, δ k (i, j) is an indicator function, indicating whether path k passes through edge ij; S324, Critical Node and Bottleneck Assessment: Based on the analysis results of node degree, betweenness centrality and edge load, the critical nodes and bottlenecks in the logistics network are identified, and the impact of critical nodes and bottlenecks on the overall logistics performance is evaluated, including: Key node evaluation: Comprehensive node degree and betweenness centrality are evaluated, expressed as: K i =a·D i +β·C B (i); Among them, K i is the critical evaluation value of node i, D i is the degree of node i, C B (i) is the betweenness centrality of node i, α and β are weight coefficients; Bottleneck evaluation formula: Evaluate through the edge load, expressed as: B ij =L ij ; Among them, B ij is the bottleneck evaluation value of edge ij, L ij is the load of edge ij.

6. The method for automatic carrier assessment based on big data analysis according to claim 5 is characterized in that: The dynamic adjustment in S33 includes: S331, Real-time data collection: Continuously collect real-time performance data of carriers; S332, dynamic update of network structure: updating the attributes of nodes and edges in the relationship network based on the performance data collected in real time; S333, Key Performance Indicator Adjustment: Based on the network structure analysis results and real-time performance data, the key performance indicators are dynamically adjusted, expressed as: KPI i ′=KPI i ×(1+γ·ΔP i ); Among them, KPI i ' is the adjusted key performance indicator, KPI i is the original key performance indicator, γ is the adjustment coefficient, ΔP i is the performance change rate of the i-th node; Among them, P current，i is the current performance value of the i-th node, P previous，i is the performance value of the i-th node in the previous period.

7. The method for automatic carrier assessment based on big data analysis according to claim 6, characterized in that: The carrier assessment model in S4 adopts a factorization machine model, and the factorization machine model includes: S41, feature vectorization: The adjusted carrier performance data is converted into feature vectors, expressed as: x=[x1,x2,...,x n ]; Where x is the eigenvector, x1, x2, ..., x n are the 1st, 2nd, ..., nth features respectively; S42, model construction: Based on the basis of factor decomposition machine, the interactive characteristics of external factors are introduced, which is expressed as: Among them, w0 is the global bias, w i is feature x i The weight of <v i , v j > is the feature vector v i and v j The inner product of ij is the interaction feature x i and z j The weight of z j For external factors; S43, parameter adjustment: dynamically adjust model parameters according to real-time external factors to reflect the latest performance status, expressed as: Among them, β ij (t) is the interaction feature weight at time t, η is the learning rate, y t is the actual value, is the predicted value; S44, comprehensive score: Comprehensive score of each indicator of the carrier, expressed as: Among them, Score is the comprehensive score, α i is the weight of different indicators, is the predicted value of the ith indicator.

8. The method for automatic carrier assessment based on big data analysis according to claim 7 is characterized in that: The automatic assessment in S5 includes: S51, real-time data input: continuously obtain the real-time performance data of the carrier and input the real-time performance data into the carrier assessment model; S52, real-time assessment: the comprehensive score of the carrier is calculated in real time through the carrier assessment model; S53, assessment report generation: Generate an assessment report based on the calculated comprehensive score, including the score of each key performance indicator, the total score, and the performance change trend.

9. The method for automatic carrier assessment based on big data analysis according to claim 8, characterized in that: The feedback and optimization in S6 include: S61, Assessment result feedback: Feedback the assessment result to the carrier based on the assessment report generated by automatic assessment; S62, optimization suggestion generation: Combine the assessment results and external environmental factors to generate optimization suggestions, expressed as: Optimization_Score=α·KPI current +β·External_Factors; Among them, Optimization_Score is the optimization score, KPI current is the current key performance indicator, External_Factors is the external factor, α and β are the weight coefficients; S63, model optimization: Based on the actual performance and feedback data of the carrier, the parameters of the carrier assessment model are continuously optimized using an adaptive learning algorithm, which is expressed as: Among them, θ new is the updated model parameter, θ old are the old model parameters, η is the learning rate, and L(θ) is the loss function.

10. A carrier automatic assessment system based on big data analysis, used to implement the carrier automatic assessment method based on big data analysis as claimed in any one of claims 1 to 9, characterized in that: Includes the following modules: Data collection and preprocessing module: collects performance data about carriers through various channels, including transportation time, service quality, customer evaluation and transportation cost, and preprocesses the collected performance data; Data analysis module: conducts multi-dimensional analysis on the pre-processed performance data and extracts the key performance indicators of the carrier, including transportation timeliness, transportation cost and customer satisfaction; Complex network analysis and dynamic adjustment module: builds the relationship network between carriers, customers and suppliers, analyzes the network structure, identifies key nodes and bottlenecks, and dynamically adjusts key performance indicators; Model building module: Based on the adjusted key performance indicators, a carrier assessment model is established to comprehensively score the various indicators of the carrier; Automatic assessment module: conducts real-time assessment of carriers through the carrier assessment model and generates assessment reports; feedback and optimization module: provides assessment results and optimization suggestions to carriers based on the assessment results, and optimizes the carrier assessment model based on the carrier's actual performance.

Images