E-commerce logistics monitoring method and system based on big data

Abstract

The invention discloses an e-commerce logistics monitoring method and system based on big data, and relates to the technical field of logistics management, and the method comprises the following steps: S1, constructing a multi-dimensional data model; s2, acquiring and processing real-time data; s3, performing anomaly detection; s4, decision support of logistics; s5, transparent tracking of logistics is carried out; and S6, dynamically optimizing the logistics strategy. The system comprises function modules corresponding to the operation of each step in the e-commerce logistics monitoring method based on big data. According to the big data-based e-commerce logistics monitoring method and system, by constructing the multi-dimensional data model and the logistics decision support model, efficient processing and analysis of real-time data streams can be realized, abnormal conditions can be identified in time, optimization suggestions can be provided, logistics efficiency and customer satisfaction can be effectively improved, and meanwhile, the method and the system have good application prospects. Through integration of multiple data sources, information islands are eliminated, the accuracy and flexibility of prediction are enhanced, and the overall efficiency of e-commerce logistics management is significantly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics management, and in particular to an e-commerce logistics monitoring method and system based on big data. Background Art

[0002] With the rapid development of the e-commerce industry, logistics, as its core component, faces increasingly complex challenges. Traditional logistics monitoring methods often rely on manual data collection and static analysis, which cannot reflect the logistics status in real time, resulting in delayed response to abnormal situations. In addition, existing technologies are usually difficult to integrate data from multiple sources, such as order information, transportation status, external environment, etc., resulting in information islands and affecting the accuracy and timeliness of decision-making. Current anomaly detection algorithms mostly rely on simple statistical methods, which are difficult to adapt to the dynamically changing logistics environment, thus affecting the ability to predict potential risks. Therefore, there is an urgent need for a new method and system to improve the intelligence and real-time performance of logistics monitoring. Summary of the invention

[0003] In view of the deficiencies in the prior art, the present invention provides an e-commerce logistics monitoring method and system based on big data to solve the problems raised in the above background technology.

[0004] To achieve the above object, the present invention provides the following technical solution: a method for monitoring e-commerce logistics based on big data, comprising the following steps:

[0005] S1. Build a multidimensional data model;

[0006] Integrate various heterogeneous data in e-commerce logistics into a structured model, establish a multidimensional data model, and integrate all relevant data in the logistics process into a unified platform;

[0007] S2, real-time data acquisition and processing;

[0008] Based on IoT devices, GPS positioning, and sensor technology, real-time data acquisition is performed. Data processing uses streaming processing technology to analyze data at the moment it is generated, achieving efficient processing and analysis of real-time data streams, ensuring that the system responds to abnormal situations in a timely manner and makes corresponding adjustments;

[0009] S3, perform anomaly detection;

[0010] Based on anomaly detection algorithms, real-time data streams are analyzed to identify potential abnormal patterns, and abnormal situations in the logistics process are discovered and responded to in a timely manner;

[0011] S4. Logistics decision support;

[0012] After anomaly detection, a data-driven logistics decision support model is provided to analyze the current logistics status, predict future logistics trends, and generate optimization suggestions based on real-time and historical data;

[0013] S5. Conduct transparent tracking of logistics;

[0014] By generating blocks at each stage and storing relevant information on the blockchain, users can track logistics status in real time and audit the transportation process;

[0015] S6. Dynamically optimize logistics strategies;

[0016] By using optimization algorithms, transportation routes and distribution strategies are adjusted in real time. The algorithms are based on various constraints, such as time, cost, and vehicle load, and are continuously iterated through a self-learning mechanism to find the optimal solution in different environments and ensure the feasibility of the optimization solution.

[0017] To further optimize the technical solution, in step S1, the multidimensional data model established includes not only basic data such as order information, cargo type, and transportation method, but also data on external factors such as weather, traffic conditions, and geographic information.

[0018] To further optimize the technical solution, the multidimensional data model includes the following processes when being constructed:

[0019] Divide the data into the following dimensions, including:

[0020] Order dimension O: includes information on order ID, customer ID, order time, and total amount;

[0021] Cargo dimension G: includes cargo type, weight, volume, and value;

[0022] Transportation dimension T: covers transportation methods, such as land transportation, sea transportation, air transportation, estimated transportation time, actual transportation time;

[0023] Geographic dimension L: information about the geographic location, traffic flow, and road conditions of the starting and ending points;

[0024] External factors dimension E: including weather conditions, such as temperature, rainfall, and holiday impacts;

[0025] The above dimensions are expressed by the following formula:

[0026] M = f(O, G, T, L, E);

[0027] Among them, M represents a multidimensional data model, and f is a comprehensive function that integrates data from all relevant dimensions;

[0028] The dimensions in the multidimensional data model are quantified and the transportation efficiency E in the logistics process is evaluated. The evaluation formula is as follows:

[0029]

[0030] in,

[0031] O amount The order amount represents the economic value of the order and affects the shipping priority;

[0032] G value The value of the goods reflects the importance of the goods;

[0033] T efficiency The average time efficiency of land transportation is 1.5, and that of air transportation is 0.8. The lower the value, the higher the efficiency.

[0034] L distance The longer the distance, the lower the transportation efficiency;

[0035] E impact A weighting coefficient is used to account for the influence of external factors. The larger the value, the more significant the impact of external conditions on transportation. For example, the impact of weather is 1.2 and the impact of holidays is 1.5.

[0036] To further optimize the technical solution, in step S2, a dynamic response model is constructed to perform real-time monitoring, anomaly detection and decision support on the data stream. The model is shown in the following formula:

[0037] D = f(R, S, A);

[0038] in,

[0039] D is the final decision result;

[0040] R is real-time data input, including order status, transportation status, and external environment;

[0041] S is the status assessment, which reflects the effectiveness of the current logistics environment;

[0042] A is the anomaly detection analysis result, identifying potential problems or risks.

[0043] To further optimize the technical solution, the dynamic response model receives data from IoT devices, GPS and sensors in real time, integrates the data to form R, and then calculates the state evaluation S according to the value of R, and comprehensively scores the data of different dimensions through the set weight coefficient to determine the current logistics state;

[0044] If the result of S is low or R exceeds the set threshold, the system will call the anomaly detection analysis A to identify potential problems, such as transportation delays and cargo damage, and generate corresponding decisions D based on the analysis results;

[0045] If the system detects a delay on a transportation route, the model recommends switching to an alternative transportation route or temporarily adding transportation resources to ensure on-time delivery.

[0046] To further optimize the technical solution, in the dynamic response model, real-time data input, state evaluation and anomaly detection analysis results are designed according to the following formula:

[0047] Real-time data input R:

[0048]

[0049] Among them, O i , G i , T i , L i , E i They represent real-time data of different dimensions respectively, and n represents the number of data points in the data stream;

[0050] Status Assessment S:

[0051]

[0052] in,

[0053] W order , W cargo , W transport is the weight coefficient of each dimension, reflecting its importance;

[0054] O status , G condition , T performance The scores are respectively for order status, cargo condition and transportation performance, such as 1-10 points;

[0055] Anomaly detection analysis results A:

[0056] A=Detect(R)if R>T;

[0057] Among them, T is the preset threshold. When this value is exceeded, anomaly detection is triggered. Through machine learning algorithms such as random forest or support vector machine, the model analyzes the real-time data stream to determine whether there is an anomaly.

[0058] To further optimize the technical solution, in step S3, the anomaly detection algorithm includes the following specific process:

[0059] Data preprocessing: clean the real-time data stream to remove noise and missing values; standardize the data so that different features have the same dimension for effective comparison;

[0060] Feature extraction: Extract important features from the raw data, such as order delay time, transportation distance, and cargo status; use principal component analysis (PCA) to reduce the dimension and retain important features;

[0061] Cluster analysis: Use K-means clustering or density-based spatial clustering algorithm (DBSCAN) to cluster the data and find the normal operation mode; by selecting appropriate clustering parameters such as K value or ε value, the data can be divided into normal and abnormal categories;

[0062] Outlier detection: Based on the clustering results, outlier detection is performed using the local outlier factor (LOF) algorithm; the LOF algorithm calculates the local outlier factor of each data point and identifies data points that are more outliers than their neighboring data points;

[0063] Threshold setting and feedback adjustment: Set reasonable thresholds based on historical data to decide when to mark data points as abnormal. After each detection, use the feedback mechanism to collect actual conditions, compare them with the predicted results, and continuously adjust the thresholds and algorithm parameters.

[0064] To further optimize the technical solution, in step S4, the logistics decision support model includes current logistics status evaluation C, future trend prediction F, and optimization suggestion generation A;

[0065] Current logistics situation assessment C:

[0066] C=α1·O+α2·G+α3·T+α4·L+α5·E+∈;

[0067] in,

[0068] O is the order status score, such as delay and completion;

[0069] G is the goods status score, such as damage rate and inventory status;

[0070] T is the transportation efficiency score, such as punctuality rate and delivery speed;

[0071] L is the score for geographical factors, such as traffic conditions and distance;

[0072] E is the external environment score, such as weather and market dynamics;

[0073] α i is the weight coefficient of each factor;

[0074] ∈ is a noise term, reflecting uncontrollable factors;

[0075] Future Trend Forecast F:

[0076]

[0077] in,

[0078] Xj is the historical order data features, such as the number and type of orders in the past 7 days;

[0079] Y j Historical transportation data features, such as transportation delays and route selections in the past 7 days;

[0080] Z j External factors, such as weather and holidays;

[0081] β j , γ j , δ j is the weight coefficient of the corresponding feature;

[0082] θ is the model bias term, reflecting the overall trend;

[0083] Optimization suggestion generation A:

[0084]

[0085] in,

[0086] R k Suggest functions for optimization, such as changing routes and adjusting delivery plans, based on current conditions C and predicted trends F;

[0087] λ k The weight of each recommendation reflects its importance or applicability;

[0088] p is the total number of recommendations.

[0089] To further optimize the technical solution, in step S5, a smart contract is also provided in the blockchain, and the smart contract automatically executes condition-based operations, such as automatically initiating payment when the goods arrive.

[0090] An e-commerce logistics monitoring system based on big data is constructed based on the above-mentioned e-commerce logistics monitoring method based on big data, and the system establishes corresponding functional modules based on the operations of each step in the e-commerce logistics monitoring method based on big data;

[0091] The system also establishes a multi-party collaboration and information sharing platform to closely connect various logistics participants, such as suppliers, transportation companies, and warehouse management;

[0092] By opening the API interface, information can be shared in real time, suppliers can obtain transportation status in real time and make timely production adjustments, and transportation companies can optimize transportation plans based on order requirements.

[0093] Compared with the prior art, the present invention provides an e-commerce logistics monitoring method and system based on big data, which has the following beneficial effects:

[0094] This e-commerce logistics monitoring method and system based on big data can realize efficient processing and analysis of real-time data streams by constructing multidimensional data models and logistics decision support models. It can not only identify abnormal situations in a timely manner, but also provide optimization suggestions, effectively improve logistics efficiency and customer satisfaction. At the same time, by integrating multiple data sources, eliminating information islands, and enhancing the accuracy and flexibility of predictions, the overall efficiency of e-commerce logistics management will be significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] Figure 1 A schematic diagram of a process of an e-commerce logistics monitoring method based on big data proposed by the present invention;

[0096] Figure 2 This is a flow chart of an anomaly detection algorithm in an e-commerce logistics monitoring method based on big data proposed in the present invention. DETAILED DESCRIPTION

[0097] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0098] Embodiment 1:

[0099] See also Figure 1 , an e-commerce logistics monitoring method based on big data, comprising the following steps:

[0100] S1. Build a multidimensional data model

[0101] Integrate various heterogeneous data in e-commerce logistics into a structured model, establish a multidimensional data model, and integrate all relevant data in the logistics process into a unified platform.

[0102] In this embodiment, in step S1, the multidimensional data model established includes not only basic data such as order information, cargo type, and transportation method, but also data on external factors such as weather, traffic conditions, and geographic information.

[0103] Furthermore, when constructing the multidimensional data model, the following process is included:

[0104] Divide the data into the following dimensions, including:

[0105] Order dimension O: includes information on order ID, customer ID, order time, and total amount;

[0106] Cargo dimension G: includes cargo type, weight, volume, and value;

[0107] Transportation dimension T: covers transportation methods, such as land transportation, sea transportation, air transportation, estimated transportation time, actual transportation time;

[0108] Geographic dimension L: information about the geographic location, traffic flow, and road conditions of the starting and ending points;

[0109] External factors dimension E: including weather conditions, such as temperature, rainfall, and holiday impacts;

[0110] The above dimensions are expressed by the following formula:

[0111] M = f(O, G, T, L, E);

[0112] Among them, M represents a multidimensional data model, and f is a comprehensive function that integrates data from all relevant dimensions;

[0113] The dimensions in the multidimensional data model are quantified and the transportation efficiency E in the logistics process is evaluated. The evaluation formula is as follows:

[0114]

[0115] in,

[0116] O amount The order amount represents the economic value of the order and affects the shipping priority;

[0117] G value The value of the goods reflects the importance of the goods;

[0118] T efficiency The average time efficiency of land transportation is 1.5, and that of air transportation is 0.8. The lower the value, the higher the efficiency.

[0119] L distance The longer the distance, the lower the transportation efficiency;

[0120] E impact A weighting coefficient is used to account for the influence of external factors. The larger the value, the more significant the impact of external conditions on transportation. For example, the impact of weather is 1.2 and the impact of holidays is 1.5.

[0121] Based on the above model, in actual applications, when a new order is received, the system will update the parameters in real time and calculate E according to the order information, cargo characteristics, transportation mode, and relevant geographical and external information. If the value of E is high, the system may recommend giving priority to the order or adjusting the transportation strategy, such as choosing a more efficient transportation mode or optimizing the transportation route.

[0122] S2. Real-time data acquisition and processing

[0123] Real-time data acquisition is carried out based on IoT devices, GPS positioning, and sensor technology. Data processing uses stream processing technology to analyze the data at the moment it is generated, achieving efficient processing and analysis of real-time data streams, ensuring that the system responds to abnormal situations in a timely manner and makes corresponding adjustments.

[0124] In this embodiment, in step S2, a dynamic response model is constructed to perform real-time monitoring, anomaly detection and decision support on the data stream. The model is shown in the following formula:

[0125] D = f(R, S, A);

[0126] in,

[0127] D is the final decision result;

[0128] R is real-time data input, including order status, transportation status, and external environment;

[0129] S is the status assessment, which reflects the effectiveness of the current logistics environment;

[0130] A is the anomaly detection analysis result, identifying potential problems or risks.

[0131] Furthermore, the dynamic response model receives data from IoT devices, GPS and sensors in real time, integrates the data to form R, and then calculates the status evaluation S according to the value of R, and comprehensively scores the data of different dimensions through the set weight coefficient to determine the current logistics status;

[0132] If the result of S is low or R exceeds the set threshold, the system will call the anomaly detection analysis A to identify potential problems, such as transportation delays and cargo damage, and generate corresponding decisions D based on the analysis results;

[0133] If the system detects a delay on a transportation route, the model recommends switching to an alternative transportation route or temporarily adding transportation resources to ensure on-time delivery.

[0134] Furthermore, in the dynamic response model, real-time data input, state evaluation and anomaly detection analysis results are designed according to the following formula:

[0135] Real-time data input R:

[0136]

[0137] Among them, O i , G i , T i , L i , E i They represent real-time data of different dimensions respectively, and n represents the number of data points in the data stream;

[0138] Status Assessment S:

[0139]

[0140] in,

[0141] W order , W cargo , W transport is the weight coefficient of each dimension, reflecting its importance;

[0142] O status , G condition , T performance The scores are respectively for order status, cargo condition and transportation performance, such as 1-10 points;

[0143] Anomaly detection analysis results A:

[0144] A=Detect(R)if R>T;

[0145] Among them, T is the preset threshold. When this value is exceeded, anomaly detection is triggered. Through machine learning algorithms such as random forest or support vector machine, the model analyzes the real-time data stream to determine whether there is an anomaly.

[0146] S3. Perform anomaly detection

[0147] Based on the anomaly detection algorithm, the real-time data stream is analyzed to identify potential abnormal patterns, and abnormal situations in the logistics process can be discovered and responded to in a timely manner. For example, cluster analysis and outlier detection algorithms (such as LOF algorithm) can be used to timely discover and respond to abnormal situations in the logistics process. The accuracy and sensitivity of anomaly detection can be improved by continuously iterating and optimizing the algorithm.

[0148] like Figure 2 As shown, in this embodiment, in step S3, the anomaly detection algorithm includes the following specific process:

[0149] Data preprocessing: clean the real-time data stream to remove noise and missing values; standardize the data so that different features have the same dimension for effective comparison;

[0150] Feature extraction: Extract important features from the raw data, such as order delay time, transportation distance, and cargo status; use principal component analysis (PCA) to reduce the dimension and retain important features;

[0151] Cluster analysis: Use K-means clustering or density-based spatial clustering algorithm (DBSCAN) to cluster the data and find the normal operation mode; by selecting appropriate clustering parameters such as K value or ε value, the data can be divided into normal and abnormal categories;

[0152] Outlier detection: Based on the clustering results, outlier detection is performed using the local outlier factor (LOF) algorithm; the LOF algorithm calculates the local outlier factor of each data point and identifies data points that are more outliers than their neighboring data points;

[0153] Threshold setting and feedback adjustment: Set reasonable thresholds based on historical data to decide when to mark data points as abnormal. After each detection, use the feedback mechanism to collect actual conditions, compare them with the predicted results, and continuously adjust the thresholds and algorithm parameters.

[0154] S4. Logistics decision support

[0155] After anomaly detection, a data-driven logistics decision support model is provided to analyze the current logistics situation, predict future logistics trends, and generate optimization suggestions based on real-time and historical data. For example, the system can suggest changing the transportation route during peak traffic hours or adjusting the delivery plan during bad weather to reduce losses and improve efficiency.

[0156] In this embodiment, in step S4, the logistics decision support model includes current logistics status evaluation C, future trend prediction F, and optimization suggestion generation A;

[0157] Current logistics situation assessment C:

[0158] C=α1·O+α2·G+α3·T+α4·L+α5·E+∈;

[0159] in,

[0160] O is the order status score, such as delay and completion;

[0161] G is the goods status score, such as damage rate and inventory status;

[0162] T is the transportation efficiency score, such as punctuality rate and delivery speed;

[0163] L is the score for geographical factors, such as traffic conditions and distance;

[0164] E is the external environment score, such as weather and market dynamics;

[0165] αi is the weight coefficient of each factor;

[0166] ∈ is a noise term, reflecting uncontrollable factors;

[0167] Future Trend Forecast F:

[0168]

[0169] in,

[0170] X j Features of historical order data, such as the number and type of orders in the past 7 days;

[0171] Y j Historical transportation data features, such as transportation delays and route selections in the past 7 days;

[0172] Z j External factors, such as weather and holidays;

[0173] β j , γ j , δ j is the weight coefficient of the corresponding feature;

[0174] θ is the model bias term, reflecting the overall trend;

[0175] Optimization suggestion generation A:

[0176]

[0177] in,

[0178] R k Suggest functions for optimization, such as changing routes and adjusting delivery plans, based on current conditions C and predicted trends F;

[0179] λ k The weight of each recommendation reflects its importance or applicability;

[0180] p is the total number of recommendations.

[0181] The logistics status assessment C comprehensively considers the impact of orders, goods, transportation, geography and external environment, and adjusts the importance of different factors through weight coefficients.

[0182] The future trend prediction model F extracts features based on historical data and obtains the future logistics status through weighted summation. The selection and weight of each feature can be optimized through machine learning algorithms.

[0183] Optimization suggestion generation A is based on current conditions and future forecasts to calculate the best suggestions and help managers make scientific decisions at critical moments.

[0184] S5. Conduct transparent tracking of logistics

[0185] By generating blocks at each link and storing relevant information on the blockchain, users can track logistics status in real time and audit the transportation process.

[0186] In this embodiment, in step S5, a smart contract is also provided in the blockchain, and the smart contract automatically executes condition-based operations, such as automatically initiating payment when the goods arrive.

[0187] S6. Dynamically optimize logistics strategy

[0188] By using optimization algorithms, transportation routes and distribution strategies are adjusted in real time. The algorithms are based on various constraints, such as time, cost, and vehicle load, and are continuously iterated through a self-learning mechanism to find the optimal solution in different environments and ensure the feasibility of the optimization solution.

[0189] Embodiment 2:

[0190] An e-commerce logistics monitoring system based on big data is constructed based on the e-commerce logistics monitoring method based on big data described in Example 1. The system establishes corresponding functional modules based on the operations of each step in the e-commerce logistics monitoring method based on big data.

[0191] The system also establishes a multi-party collaboration and information sharing platform to closely connect various logistics participants, such as suppliers, transportation companies, and warehouse management;

[0192] By opening the API interface, information can be shared in real time, suppliers can obtain transportation status in real time and make timely production adjustments, and transportation companies can optimize transportation plans based on order requirements.

[0193] The beneficial effects of the present invention are:

[0194] This e-commerce logistics monitoring method and system based on big data can realize efficient processing and analysis of real-time data streams by constructing multidimensional data models and logistics decision support models. It can not only identify abnormal situations in a timely manner, but also provide optimization suggestions, effectively improve logistics efficiency and customer satisfaction. At the same time, by integrating multiple data sources, eliminating information islands, and enhancing the accuracy and flexibility of predictions, the overall efficiency of e-commerce logistics management will be significantly improved.

[0195] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0196] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for monitoring e-commerce logistics based on big data, characterized in that: The following steps are involved: S1. Build a multidimensional data model; Integrate various heterogeneous data in e-commerce logistics into a structured model, establish a multidimensional data model, and integrate all relevant data in the logistics process into a unified platform; S2, real-time data acquisition and processing; Based on IoT devices, GPS positioning, and sensor technology, real-time data acquisition is performed. Data processing uses streaming processing technology to analyze data at the moment it is generated, achieving efficient processing and analysis of real-time data streams, ensuring that the system responds to abnormal situations in a timely manner and makes corresponding adjustments; S3, perform anomaly detection; Based on anomaly detection algorithms, real-time data streams are analyzed to identify potential abnormal patterns, and abnormal situations in the logistics process are discovered and responded to in a timely manner; S4. Logistics decision support; After anomaly detection, a data-driven logistics decision support model is provided to analyze the current logistics status, predict future logistics trends, and generate optimization suggestions based on real-time and historical data; S5. Conduct transparent tracking of logistics; By generating blocks at each stage and storing relevant information on the blockchain, users can track logistics status in real time and audit the transportation process; S6. Dynamically optimize logistics strategies; By using optimization algorithms, transportation routes and distribution strategies are adjusted in real time. The algorithms are based on various constraints, such as time, cost, and vehicle load, and are continuously iterated through a self-learning mechanism to find the optimal solution in different environments and ensure the feasibility of the optimization solution.

2. The e-commerce logistics monitoring method based on big data according to claim 1 is characterized in that: In step S1, the multidimensional data model established includes not only basic data such as order information, cargo type, and transportation method, but also data on external factors such as weather, traffic conditions, and geographic information.

3. The e-commerce logistics monitoring method based on big data according to claim 2 is characterized in that: The multidimensional data model includes the following processes when it is constructed: Divide the data into the following dimensions, including: Order dimension O: includes information on order ID, customer ID, order time, and total amount; Cargo dimension G: includes cargo type, weight, volume, and value; Transportation dimension T: covers transportation methods, such as land transportation, sea transportation, air transportation, estimated transportation time, actual transportation time; Geographic dimension L: information about the geographical location, traffic flow, and road conditions of the starting and ending points; External factors dimension E: including weather conditions, such as temperature, rainfall, and holiday impacts; The above dimensions are expressed by the following formula: M = f(O, G, T, L, E); Among them, M represents a multidimensional data model, and f is a comprehensive function that integrates data from all relevant dimensions; The dimensions in the multidimensional data model are quantified and the transportation efficiency E in the logistics process is evaluated. The evaluation formula is as follows: in, O amount The order amount represents the economic value of the order and affects the shipping priority; G value The value of the goods reflects the importance of the goods; T efficiency The average time efficiency of land transportation is 1.5, and that of air transportation is 0.

8. The lower the value, the higher the efficiency. L distance The longer the distance, the lower the transportation efficiency; E impact A weighting coefficient is used to account for the influence of external factors. The larger the value, the more significant the impact of external conditions on transportation. For example, the impact of weather is 1.2 and the impact of holidays is 1.

5.

4. The e-commerce logistics monitoring method based on big data according to claim 1 is characterized in that: In step S2, a dynamic response model is constructed to perform real-time monitoring, anomaly detection and decision support on the data stream. The model is shown in the following formula: D = f(R, S, A); in, D is the final decision result; R is real-time data input, including order status, transportation status, and external environment; S is the status assessment, which reflects the effectiveness of the current logistics environment; A is the anomaly detection analysis result, identifying potential problems or risks.

5. The e-commerce logistics monitoring method based on big data according to claim 4 is characterized in that: The dynamic response model receives data from IoT devices, GPS and sensors in real time, integrates the data to form R, and then calculates the status evaluation S according to the value of R, and comprehensively scores the data of different dimensions through the set weight coefficient to determine the current logistics status; If the result of S is low or R exceeds the set threshold, the system will call the anomaly detection analysis A to identify potential problems, such as transportation delays and cargo damage, and generate corresponding decisions D based on the analysis results; If the system detects a delay on a transportation route, the model recommends switching to an alternative transportation route or temporarily adding transportation resources to ensure on-time delivery.

6. The e-commerce logistics monitoring method based on big data according to claim 4 is characterized in that: In the dynamic response model, real-time data input, state evaluation, and anomaly detection analysis results are designed using the following formula: Real-time data input R: Among them, O i , G i , T i , L i , E i They represent real-time data of different dimensions respectively, and n represents the number of data points in the data stream; Status Assessment S: in, W order , W cargo , W trensport is the weight coefficient of each dimension, reflecting its importance; O status , G condition , T performence The scores are respectively for order status, cargo condition and transportation performance, such as 1-10 points; Anomaly detection analysis results A: A=Detect(R) if R>T; Among them, T is the preset threshold. When this value is exceeded, anomaly detection is triggered. Through machine learning algorithms such as random forest or support vector machine, the model analyzes the real-time data stream to determine whether there is an anomaly.

7. The e-commerce logistics monitoring method based on big data according to claim 1 is characterized in that: In step S3, the anomaly detection algorithm includes the following specific process: Data preprocessing: clean the real-time data stream to remove noise and missing values; standardize the data so that different features have the same dimension for effective comparison; Feature extraction: Extract important features from the raw data, such as order delay time, transportation distance, and cargo status; use principal component analysis (PCA) to reduce the dimension and retain important features; Cluster analysis: Use K-means clustering or density-based spatial clustering algorithm (DBSCAN) to cluster the data and find the normal operating mode; by selecting appropriate clustering parameters such as K value or ε value, the data can be divided into two categories: normal and abnormal; Outlier detection: Based on the clustering results, outlier detection is performed using the local outlier factor (LOF) algorithm; the LOF algorithm calculates the local outlier factor of each data point and identifies data points that are more outliers than their neighboring data points; Threshold setting and feedback adjustment: Set reasonable thresholds based on historical data to decide when to mark data points as abnormal. After each detection, use the feedback mechanism to collect actual conditions, compare them with the predicted results, and continuously adjust the thresholds and algorithm parameters.

8. The e-commerce logistics monitoring method based on big data according to claim 1 is characterized in that: In step S4, the logistics decision support model includes current logistics status evaluation C, future trend prediction F, and optimization suggestion generation A; Current logistics situation assessment C: C=α1·O+α2·C+α3·T+α4·L+α5·E+∈; in, O is the order status score, such as delay and completion; G is the goods status score, such as damage rate and inventory status; T is the transportation efficiency score, such as punctuality rate and delivery speed; L is the score for geographical factors, such as traffic conditions and distance; E is the external environment score, such as weather and market dynamics; α i is the weight coefficient of each factor; ∈ is a noise term, reflecting uncontrollable factors; Future Trend Forecast F: in, X j Features of historical order data, such as the number and type of orders in the past 7 days; Y j Historical transportation data features, such as transportation delays and route selections in the past 7 days; Z j External factors, such as weather and holidays; β j , γ j , δ j is the weight coefficient of the corresponding feature; θ is the model bias term, reflecting the overall trend; Optimization suggestion generation A: in, R k Suggest functions for optimization, such as changing routes and adjusting delivery plans, based on current conditions C and predicted trends F; λ k The weight of each recommendation reflects its importance or applicability; p is the total number of recommendations.

9. The e-commerce logistics monitoring method based on big data according to claim 1 is characterized in that: In step S5, a smart contract is also set up in the blockchain, and the smart contract automatically executes conditional operations, such as automatically initiating payment when the goods arrive.

10. An e-commerce logistics monitoring system based on big data, constructed based on the e-commerce logistics monitoring method based on big data according to any one of claims 1 to 9, characterized in that: The system establishes corresponding functional modules based on the operations of each step in the e-commerce logistics monitoring method based on big data; The system also establishes a multi-party collaboration and information sharing platform to closely connect various logistics participants, such as suppliers, transportation companies, and warehouse management; By opening the API interface, information can be shared in real time, suppliers can obtain transportation status in real time and make timely production adjustments, and transportation companies can optimize transportation plans based on order requirements.

Images