Order dispatching optimization method and device based on automobile service supply chain, and storage medium

Abstract

The invention relates to the technical field of supply chain order dispatching optimization, in particular to an order dispatching optimization method and device based on an automobile service supply chain and a storage medium. The method comprises the following steps: obtaining multi-source automobile service supply chain information and user real-time order data based on a supply chain platform; performing supply chain service manufacturer identification and dynamic node space distribution fitting on the multi-source automobile service supply chain information, and constructing a supply chain service node distribution diagram; performing user order service demand analysis on the user real-time order data, and performing real-time service demand load analysis to obtain supply chain real-time service demand load characteristics; performing space-time demand distribution evolution on the supply chain real-time service demand load characteristics to generate dynamic service demand prediction data; and performing optimal demand supply matching on the supply chain service node distribution map based on the dynamic service demand prediction data, and constructing a dynamic service demand distribution strategy. According to the invention, efficient and accurate supply chain dispatching optimization is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of supply chain dispatch optimization, and in particular to a dispatch optimization method, device and storage medium based on an automobile service supply chain. Background Art

[0002] In the modern automotive service industry, with the diversification of consumers' demands for automobile repair and maintenance and the improvement of service quality, the management of the automotive service supply chain has become increasingly complex. The automotive service supply chain usually includes multiple links, such as automobile parts supply, repair services, logistics distribution, customer appointments, etc. Each link may be affected by different factors, such as order volume fluctuations, service personnel's skill level, changes in parts inventory, traffic conditions, etc. These factors have an important impact on the timeliness and quality of automotive services.

[0003] In the traditional automotive service supply chain, the dispatch system usually relies on manual scheduling and manual matching. The dispatch process is cumbersome and error-prone, making it difficult to achieve efficient resource utilization and accurate service matching. Especially when faced with a large number of service demands and a dynamically changing market environment, manual dispatch often finds it difficult to quickly respond to changes in customer demand and the allocation of service resources, resulting in low service efficiency and decreased customer satisfaction.

[0004] In addition, with the development of intelligence and digitalization in the automotive industry, more and more automotive service companies have begun to use information management systems, but the traditional dispatching system still has many problems, such as the mismatch between service demand and actual resources, the inability to reflect the timeliness of customer demand in a timely manner, and scheduling delays in the vehicle maintenance service process. These problems not only increase operating costs, but also affect service quality and customer experience. Therefore, with the continuous development of technologies such as big data, artificial intelligence, and the Internet of Things, the automotive service supply chain needs a more intelligent and flexible dispatching optimization method to deal with various emergencies that may arise during the service process and achieve optimized management of all links in the supply chain. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a dispatch optimization method, device and storage medium based on the automobile service supply chain to solve at least one of the above technical problems.

[0006] To achieve the above object, the present invention provides an order dispatch optimization method based on an automobile service supply chain, comprising the following steps: Step S1: obtaining multi-source automobile service supply chain information and user real-time order data based on the supply chain platform; identifying supply chain service vendors and performing dynamic node spatial distribution fitting on the multi-source automobile service supply chain information, and constructing a supply chain service node distribution map; Step S2: Performing user order service demand analysis on user real-time order data and performing real-time service demand load analysis to obtain supply chain real-time service demand load characteristics; Step S3: Performing temporal and spatial demand distribution evolution on the real-time service demand load characteristics of the supply chain, and then performing dynamic service demand forecasting to generate dynamic service demand forecasting data; Step S4: Based on the dynamic service demand forecast data, optimal demand and supply matching is performed on the supply chain service node distribution map to construct a dynamic service demand allocation strategy; Step S5: Adaptively optimize the dispatch of orders on the supply chain platform according to the service demand optimal node matching data, and adjust the supply chain node service delay to obtain a dynamic delay adjustment strategy; Step S6: Based on the user's real-time order data, multi-scenario transfer learning is performed on the dynamic delay adjustment strategy, and intelligent dispatching decision analysis is performed to build a supply chain intelligent dispatching model.

[0007] The present invention also provides an order dispatch optimization device based on an automobile service supply chain, comprising: The node distribution module is used to obtain multi-source automotive service supply chain information and user real-time order data based on the supply chain platform; identify supply chain service vendors and perform dynamic node spatial distribution fitting on multi-source automotive service supply chain information to build a supply chain service node distribution map; The demand load module is used to perform user order service demand analysis on user real-time order data and perform real-time service demand load analysis to obtain the real-time service demand load characteristics of the supply chain; The demand forecasting module is used to perform temporal and spatial demand distribution evolution on the real-time service demand load characteristics of the supply chain, and then perform dynamic service demand forecasting to generate dynamic service demand forecasting data; The optimal demand matching module is used to perform optimal demand-supply matching on the supply chain service node distribution map based on dynamic service demand forecast data and build a dynamic service demand allocation strategy; The delay adjustment module is used to optimize the supply chain platform's adaptive dispatching according to the optimal node matching data of service demand, and adjust the supply chain node service delay to obtain a dynamic delay adjustment strategy; The intelligent dispatch decision module is used to perform multi-scenario transfer learning of dynamic delay adjustment strategies based on users' real-time order data, conduct intelligent dispatch decision analysis, and build a supply chain intelligent dispatch model.

[0008] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above-mentioned methods for optimizing dispatching orders based on the automotive service supply chain are implemented.

[0009] The beneficial effects of the present invention are specifically as follows: by acquiring multi-source automotive service supply chain information and user real-time order data, a supply chain service node distribution map can be established, which helps to fully understand the supply chain structure and key node locations. Identifying supply chain service vendors and dynamic node spatial distribution fitting can improve the transparency and operability of the supply chain, and provide basic data support for subsequent dispatch optimization. Through user order service demand analysis and real-time service demand load analysis, user demand characteristics and the current load of the supply chain can be deeply understood, providing a basis for subsequent service demand prediction and optimization. Acquiring real-time supply chain service demand load characteristics can help improve service response speed and efficiency, and ensure a balance between service supply and demand. Through the evolution of spatiotemporal demand distribution and dynamic service demand prediction, accurate prediction of future service demand can be achieved, which helps to plan and adjust service supply in advance. Generating dynamic service demand prediction data can help improve the flexibility of the supply chain. The system ensures that timely adjustments can be made when demand changes. Based on dynamic service demand forecast data, the optimal demand-supply matching and the construction of dynamic service demand allocation strategies can achieve optimal resource allocation and efficient provision of services. The construction of dynamic service demand allocation strategies helps to improve the flexibility and efficiency of the supply chain and ensure that services can meet user needs in a timely manner. Through adaptive dispatch optimization and supply chain node service delay adjustment, the dispatch process can be intelligent and the service response time can be optimized. The establishment of a dynamic delay adjustment strategy helps to improve service quality and user satisfaction and ensure that services can be completed on time and efficiently. Based on users' real-time order data, multi-scenario transfer learning and intelligent dispatch decision analysis can be carried out to build an intelligent dispatch model to improve the accuracy and efficiency of dispatch decisions. Building a supply chain intelligent dispatch model helps to optimize the dispatch process, improve the processing capacity of supply chain services, and enhance the dispatch efficiency and accuracy of the supply chain. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic diagram of the steps of a dispatch optimization method based on an automobile service supply chain according to the present invention; Figure 2 Detailed implementation flow chart of step S1; Figure 3 Detailed implementation flow chart of step S2; Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION

[0011] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0012] The present application example provides a dispatch optimization method, device and storage medium based on the automobile service supply chain. The execution subjects of the dispatch optimization method, device and storage medium based on the automobile service supply chain include but are not limited to: mechanical equipment, data processing platform, cloud server node, network upload device, etc. equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of an audio and image management system, an information management system, and a cloud data management system.

[0013] See also Figures 1 to 4 The present invention provides an order dispatch optimization method based on an automobile service supply chain, and the order dispatch optimization method based on an automobile service supply chain comprises the following steps: Step S1: obtaining multi-source automobile service supply chain information and user real-time order data based on the supply chain platform; identifying supply chain service vendors and performing dynamic node spatial distribution fitting on the multi-source automobile service supply chain information, and constructing a supply chain service node distribution map; Step S2: Performing user order service demand analysis on user real-time order data and performing real-time service demand load analysis to obtain supply chain real-time service demand load characteristics; Step S3: Performing temporal and spatial demand distribution evolution on the real-time service demand load characteristics of the supply chain, and then performing dynamic service demand forecasting to generate dynamic service demand forecasting data; Step S4: Based on the dynamic service demand forecast data, optimal demand and supply matching is performed on the supply chain service node distribution map to construct a dynamic service demand allocation strategy; Step S5: Adaptively optimize the dispatch of orders on the supply chain platform according to the service demand optimal node matching data, and adjust the supply chain node service delay to obtain a dynamic delay adjustment strategy; Step S6: Based on the user's real-time order data, multi-scenario transfer learning is performed on the dynamic delay adjustment strategy, and intelligent dispatching decision analysis is performed to build a supply chain intelligent dispatching model.

[0014] By acquiring multi-source automotive service supply chain information and user real-time order data, it is possible to establish a supply chain service node distribution map, which helps to fully understand the supply chain structure and key node locations. Identifying supply chain service vendors and dynamic node spatial distribution fitting can improve the transparency and operability of the supply chain, and provide basic data support for subsequent dispatch optimization. Through user order service demand analysis and real-time service demand load analysis, we can deeply understand user demand characteristics and the current load of the supply chain, and provide a basis for subsequent service demand forecasting and optimization. Acquiring real-time supply chain service demand load characteristics helps to improve service response speed and efficiency, and ensure the balance between service supply and demand. Through the evolution of spatiotemporal demand distribution and dynamic service demand forecasting, accurate forecasts of future service demand can be achieved, which helps to plan and adjust service supply in advance. Generating dynamic service demand forecast data can help improve the flexibility and adaptability of the supply chain. Ensure that timely adjustments can be made when demand changes. Optimal demand-supply matching and building dynamic service demand allocation strategies based on dynamic service demand forecast data can achieve optimal resource allocation and efficient service provision. Building dynamic service demand allocation strategies helps improve the flexibility and efficiency of the supply chain and ensure that services can meet user needs in a timely manner. Through adaptive dispatch optimization and supply chain node service delay adjustment, the dispatch process can be intelligent and the service response time can be optimized. The establishment of a dynamic delay adjustment strategy helps improve service quality and user satisfaction and ensure that services can be completed on time and efficiently. Based on users' real-time order data, multi-scenario transfer learning and intelligent dispatch decision analysis can be carried out to build an intelligent dispatch model to improve the accuracy and efficiency of dispatch decisions. Building a supply chain intelligent dispatch model helps optimize the dispatch process, improve the processing capacity of supply chain services, and enhance the dispatch efficiency and accuracy of the supply chain.

[0015] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a dispatch optimization method based on an automobile service supply chain of the present invention. In this example, the steps of the dispatch optimization method based on an automobile service supply chain include: Step S1: obtaining multi-source automobile service supply chain information and user real-time order data based on the supply chain platform; identifying supply chain service vendors and performing dynamic node spatial distribution fitting on the multi-source automobile service supply chain information, and constructing a supply chain service node distribution map; In this embodiment, it is ensured that a connection is established with the API interface of the supply chain platform, so that multi-source automobile service supply chain information and user order data can be obtained in real time. A request is sent to the supply chain platform using an HTTP request to obtain the required data, which includes service provider information (such as name, address, contact information, service type) and user real-time order data (such as order ID, user ID, service requirements, order time). The obtained data is formatted into a structure suitable for subsequent processing (such as JSON, CSV or DataFrame format). The requests library in Python can be used to make API calls, and the pandas library can be used to process the data. The obtained data is cleaned, missing values, duplicate values ​​and outliers are processed to ensure data accuracy and consistency. The dropna() and drop_duplicates() methods in pandas can be used to integrate the cleaned multi-source automobile service supply chain information with the user real-time order data to form a unified data set for subsequent analysis and processing. An appropriate identification method is selected, such as based on string matching, classification algorithm or cluster analysis, to identify a unique service provider node. The integrated data is analyzed to extract unique service provider information, including name, address and contact information. The groupby() method in das removes duplicate service providers and stores the identified service provider information in a new data table for subsequent analysis. Ensure that each service provider has corresponding geographic location information (such as address, city, longitude and latitude). If longitude and latitude are missing, you can use geocoding services (such as Google Maps API) for conversion. Select appropriate spatial distribution fitting methods, such as K-Means clustering, heat map analysis, or other spatial statistical methods to analyze the distribution of service nodes. Use clustering algorithms to analyze the geographic coordinates of service providers and identify the distribution patterns of service nodes. You can use KMeans in scikit-learn for clustering analysis, store the fitting results in the database, and use visualization tools (such as matplotlib or folium) to draw a service node distribution map to display the spatial distribution of service providers. Organize the geographic coordinates of service nodes and the identified service provider information, prepare data for drawing the distribution map, use Python's folium or matplotlib library to draw a map, mark the location of each service provider on the map, and highlight it. Save the constructed supply chain service node distribution map as an image file or display it in a web application to facilitate subsequent decision support and analysis.

[0016] Step S2: Performing user order service demand analysis on user real-time order data and performing real-time service demand load analysis to obtain supply chain real-time service demand load characteristics; In this embodiment, the real-time order data of the user is extracted from the database to ensure that key fields such as order ID, user ID, service type, order time, service location, order status, etc. are included. The extracted data is converted into a format suitable for analysis (such as pandasDataFrame) for subsequent processing. The Python pandas library is used to read and process the data. The service type (such as repair, maintenance, cleaning, etc.) is defined according to business needs to ensure that the service demand of each order is accurately classified. A suitable demand analysis method is selected. Commonly used methods include frequency statistics, trend analysis, and cluster analysis. The groupby() method in pandas is used to group and count the order data according to the service type, and the demand frequency and total demand of each service type are calculated. Through time series analysis, the changing trend of service demand in different time periods is analyzed (for example, summarized by hour, day, and week), and the key indicators of service demand load analysis are determined, such as real-time demand, response time, etc. time, service completion rate, etc., obtain user order status in real time by establishing a data stream pipeline (such as using Apache Kafka or Apache Spark Streaming) to ensure the real-time nature of the data, use statistical analysis methods (such as descriptive statistics) to calculate the service demand load in the current time period, and use pandas for real-time data analysis to generate load indicators and calculate real-time service demand load characteristics, such as: current service demand (for example, the number of orders in the current hour), demand distribution of each service type (such as the number of demands for each service type), service response time statistics (such as average response time, maximum response time, etc.), store the results of real-time service demand load analysis in the database, and use visualization tools (such as Matplotlib or Tableau) to generate visualization charts of load characteristics. The visualization results may include: time series graphs of service demand, pie charts or bar charts of demand for each service type, and distribution graphs of response times.

[0017] Step S3: Performing temporal and spatial demand distribution evolution on the real-time service demand load characteristics of the supply chain, and then performing dynamic service demand forecasting to generate dynamic service demand forecasting data; In this embodiment, real-time service demand load characteristic data of the supply chain is extracted from the database to ensure that it includes information such as timestamp, service type, demand, service node, etc. The extracted data is organized into a format suitable for analysis using Python's pandas library to ensure the structure and integrity of the data. A suitable spatiotemporal analysis model is selected, such as spatiotemporal clustering, heat map analysis, or time series analysis, to identify the spatiotemporal evolution characteristics of demand. A heat map of service demand is generated using the folium or seaborn library to display the distribution of service demand in different time periods and locations. The pivot_table() method in pandas is used to aggregate demand data by time and location to generate grid data of demand. A time series analysis is performed on the service demand. The seasonal_decompose() method in the statsmodels library is used to perform trend, seasonality, and residual analysis on the demand to identify the cyclical changes and trend characteristics of demand. The results of the spatiotemporal demand distribution evolution (such as Heat maps, time series analysis graphs) are stored in the database for subsequent use and reference. An appropriate dynamic service demand forecasting model is selected, such as the ARIMA model, LSTM (long short-term memory network) or XGBoost regression model. Forecasts are made based on historical demand data. According to the analysis results of the spatiotemporal demand distribution, historical demand data is sorted out, and relevant features (such as date, time, service type, etc.) are extracted. The data is preprocessed, including normalization, missing value processing and timestamp conversion. The selected forecasting model is trained and verified using the training set and the test set. The data can be divided into a training set and a test set using train_test_split(). The statsmodels library in Python is used to fit the ARIMA model, or Keras is used to build and train the LSTM model. Based on the trained model, future service demand is predicted to generate dynamic service demand forecasting data. The forecast results should include the demand for each service type in different time periods in the future.

[0018] Step S4: Based on the dynamic service demand forecast data, optimal demand and supply matching is performed on the supply chain service node distribution map to construct a dynamic service demand allocation strategy; In this embodiment, dynamic service demand forecast data is extracted from the database, including service type, predicted demand, time period and other information; service node information is extracted from the supply chain service node distribution map, including node location, service type that can be provided, available resources, etc.; a suitable optimal matching algorithm is selected, and commonly used methods include linear programming, Hungarian algorithm or minimum cost flow algorithm; a demand and supply matrix is ​​created, in which rows represent service demands (based on forecast data), columns represent service nodes, and the elements of the matrix represent the matching cost between each demand and each node (such as distance, resource availability, etc.); this matrix is ​​constructed using numpy or pandas, and the matching cost between each demand and node is calculated; the constructed matching matrix is ​​solved using the selected algorithm to find the optimal supply matching solution, identify which service nodes are most suitable for meeting which service demands, and Record the matching results, determine the goals of the dynamic service demand allocation strategy, such as minimizing response time, maximizing resource utilization, and improving user satisfaction, etc. Select an appropriate strategy model, such as a rule-based allocation model, an optimization-based allocation model, or a machine learning-based prediction model, and set allocation rules based on business needs, such as giving priority to service nodes closest to users, giving priority to nodes with sufficient resources, etc. If an optimization model is used, define the objective function (such as maximizing the service response rate) and implement the optimization solution. Use Python's scipy.optimize or PuLP library for linear programming to determine the best service node allocation plan. Considering changes in demand, design a dynamic adjustment mechanism so that the allocation strategy can be adjusted according to real-time data in actual operations. This can include regularly re-evaluating demand forecasts and service resource status, and updating the allocation strategy accordingly.

[0019] Step S5: Adaptively optimize the dispatch of orders on the supply chain platform according to the service demand optimal node matching data, and adjust the supply chain node service delay to obtain a dynamic delay adjustment strategy; In this embodiment, a suitable adaptive dispatch optimization model is selected. Common models include rule-based dispatch systems, optimization algorithms (such as genetic algorithms, particle swarm optimization, etc.) or machine learning models. The objective function of the optimization model is designed, such as minimizing the average response time, maximizing resource utilization, etc. If an optimization algorithm is used, the constraints of the problem (such as service type consistency, resource availability, etc.) are set. Python's scipy.optimize or PuLP library is used to input the optimal node matching data, solve the optimization model, obtain an adaptive dispatch plan, generate the best service node allocation plan for each service demand, and design a dynamic delay adjustment strategy based on the current service delay data. The goal is to reduce the load of high-delay service nodes and reasonably allocate service needs. Request, use statistical analysis methods (such as descriptive statistics, box plot analysis) to analyze service delay data, identify high-latency nodes, and adjust their dispatch strategies for identified high-latency nodes, for example: reduce the allocation ratio of their new service demands, and give priority to allocating new demands to nodes with lower delays. Design rules, such as threshold-based adjustment rules: if the average response time of a node exceeds the set threshold, reduce the number of dispatches for the node, adjust the actual dispatch behavior according to the designed dynamic delay adjustment strategy, implement automated dispatching in the system, ensure that new demands can be reasonably allocated according to the latest delay data, monitor the response time and service quality of service nodes in real time, ensure that the dynamic adjustment strategy can run effectively, collect data after implementation, and analyze the actual effect of the adjustment strategy.

[0020] Step S6: Based on the user's real-time order data, multi-scenario transfer learning is performed on the dynamic delay adjustment strategy, and intelligent dispatching decision analysis is performed to build a supply chain intelligent dispatching model.

[0021] In this embodiment, multiple scenarios are determined, such as holidays, weekdays, promotional activities, weather changes, etc., the impact of these scenarios on user service needs and response time are identified, and user order data in different scenarios are integrated to ensure that each scenario has corresponding characteristics (such as time, location, user characteristics, etc.). Python's pandas is used for data cleaning, missing values, standardized features, etc. are processed, and a suitable transfer learning framework is selected, such as transfer learning based on deep learning (such as convolutional neural network CNN or recurrent neural network RNN). Pre-trained models are used for transfer learning to adapt to service demand characteristics in different scenarios. Deep learning frameworks such as Keras or PyTorch are used to load pre-trained models and perform fine-tuning to ensure that the model can adapt to new scenario data. When training the model, user demand characteristics in different scenarios are input, and model parameters are optimized to improve the model's predictive ability in new scenarios. The goals of the intelligent dispatch decision model are designed, such as minimizing response time, maximizing service quality, etc., and a suitable Decision analysis methods such as reinforcement learning, decision trees, and random forests are used. Based on the user's real-time order data and the implementation effect of the dynamic delay adjustment strategy, relevant features (such as user historical behavior, service node performance, service demand, etc.) are extracted. The feature selection methods in the sklearn library (such as random forest feature importance) are used to identify key features. The extracted features are used to train the intelligent dispatch decision model, such as using random forests for classification or regression analysis to predict the optimal dispatch plan, divide the training set and the test set, use train_test_split() to split the data, evaluate the model performance (such as accuracy, recall, etc.), integrate the transfer learning model with the intelligent dispatch decision model to form a comprehensive intelligent dispatch system, improve the accuracy and stability of the model through integrated learning methods (such as voting method and weighted average method), design a dynamic adjustment mechanism to ensure that the intelligent dispatch model can adapt to the real-time changing user needs and service environment, add a real-time feedback mechanism to the model, and regularly update the model parameters and decision rules to cope with changes in new data.

[0022] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Step S11: acquiring multi-source automobile service supply chain information and user real-time order data based on the supply chain platform; Step S12: Identify supply chain service vendors for multi-source automobile service supply chain information and extract multiple supply chain service nodes; Step S13: Analyze the supply chain service types of multiple supply chain service nodes to obtain the service type of each node; Step S14: Calculate the geographic spatial position of multiple supply chain service nodes to obtain the geographic spatial coordinates of each service node; Step S15: performing node space distribution analysis on the geographic space coordinates of each service node, thereby obtaining service node distribution data; Step S16: Perform dynamic node spatial distribution fitting on the service node distribution data based on the service type of each node to construct a supply chain service node distribution map.

[0023] In this embodiment, after obtaining authorization from the platform and the user, it is ensured that the interface (API) with the supply chain platform has been connected, and relevant automobile service supply chain information and user order data can be obtained in real time. The multi-source automobile service supply chain information, including service providers, service types, prices, availability and other information, is obtained by using API calls. At the same time, the user's real-time order data, including order ID, customer information, service requirements, etc., is obtained. The obtained data is integrated into a unified database, and a data framework (such as Pandas) is used to clean and preprocess the data to ensure the consistency and integrity of the data. An appropriate data analysis method (such as cluster analysis or classification algorithm) is selected to identify the supply chain service providers, and the integrated Analyze the subsequent supply chain information and extract the unique supply chain service vendor node, including the name, type and contact information of the service provider. Use Python's pandas library for data processing and deduplication. Store the identified multiple supply chain service nodes in the database for subsequent analysis. Develop classification standards for service types (such as repair, maintenance, cleaning, etc.) to ensure the accuracy and consistency of classification. Perform service type analysis on each supply chain service node, identify and mark the service type of each node. Use pandas for data mapping to ensure that each node is assigned the correct service type. Store the service type information of each node in the database to form a detailed record of the service type. Ensure that each service node has relevant geographic information (such as address, city, or postal code), use geocoding services (such as Google Maps API, Geocodio, or OpenStreetMap) to convert the address to geographic coordinates (latitude and longitude), geocode the addresses of all service nodes, obtain the geospatial coordinates of each service node, record the geographic coordinates of each node, and store them in the database. Select a suitable spatial distribution analysis method, such as K-Means clustering, heat map analysis, or spatial statistics, analyze the geospatial coordinates of each service node, generate spatial distribution data of service nodes, and use Python's scikit-learn. arn library for cluster analysis, or use folium library to generate heat map, store the service node distribution data in the database for subsequent visualization and analysis, select appropriate fitting methods (such as polynomial regression, local weighted regression or time series analysis) to adapt to the changes in node distribution in different time periods, perform dynamic node spatial distribution fitting based on the service type and spatial distribution data of each node, build a supply chain service node distribution map, use visualization tools such as matplotlib or seaborn to display the fitting results, store the generated service node distribution map in the database, and display the distribution map through a visualization platform (such as Tableau or PowerBI) to facilitate analysis and decision-making.

[0024] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Step S21: Performing user order service demand analysis on user real-time order data to obtain multiple order service demand data; Step S22: performing demand type frequency calculation on a plurality of order service demand data, thereby obtaining the demand frequency of each service type; Step S23: quantify the time-series demand change of the demand frequency of each service type, and generate user service type demand change data; Step S24: Calculate the demand peak value of the user service type demand change data to obtain the user service demand peak value; Step S25: Perform real-time service demand load analysis based on the user service demand peak to obtain the real-time service demand load characteristics of the supply chain.

[0025] In this embodiment, the real-time order data of the user is extracted from the database to ensure the integrity and consistency of the data, including fields such as order ID, user ID, service type, order time, etc. The user order data is analyzed to extract the service demand information of each order to form multiple order service demand data sets. The Python pandas library is used to filter and process the data to extract the service demand field, and the extracted multiple order service demand data are stored in a new data table or database to facilitate subsequent analysis, determine the type classification standard of service demand, such as repair, maintenance, cleaning, etc., to ensure the consistency of classification, use statistical analysis methods (such as counting and proportion calculation) to calculate the demand frequency of each service type, perform demand type frequency calculation on multiple order service demand data, count the number of times each service type appears, use the value_counts() function in pandas to implement frequency calculation, store the demand frequency of each service type in the database for subsequent analysis, convert the demand frequency data into a time series format to facilitate the analysis of demand change trends, select an appropriate change quantification method, such as the difference method, percentage change method or sliding average method, and perform time series demand change on the demand frequency of each service type. Quantify and generate user service type demand change data, use the diff() function in pandas to calculate the change, or use pct_change() to calculate the percentage change, store the generated user service type demand change data in the database for subsequent analysis, select a suitable peak identification method, such as local maximum detection, sliding window method or Z-score method, calculate the demand peak of user service type demand change data to identify the peak period of demand, you can use the find_peaks() function in the scipy.signal library for peak detection, store the calculated user service demand peak in the database for subsequent analysis, select a suitable load analysis model, such as a queuing theory model, a load prediction model or a regression model based on machine learning, analyze the real-time service demand load based on the user service demand peak, calculate the load characteristics of the service provider during the peak period (such as processing capacity, response time, etc.), use statistical analysis or machine learning methods to predict the service load under different demand conditions, store the analysis results in the database, and use data visualization tools (such as Matplotlib or Tableau) to display the service demand load characteristics for decision support.

[0026] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: Step S31: extracting the service demand location of each user based on the user's real-time order data; Step S32: dividing the service demand location of each user into demand location areas, thereby obtaining multiple service demand areas; Step S33: performing spatiotemporal demand distribution evolution on the real-time service demand load characteristics of the supply chain based on multiple service demand areas to obtain spatiotemporal demand distribution characteristics; Step S34: mining the dynamic demand fluctuation law based on the spatiotemporal demand distribution characteristics to generate the user dynamic demand fluctuation law; Step S35: Perform dynamic service demand prediction based on the user dynamic demand fluctuation pattern to generate dynamic service demand prediction data.

[0027] In this embodiment, service demand related fields are extracted from the user's real-time order data to ensure that the user ID, service type, service location and other information are included. A data processing tool (such as Python's pandas) is used to filter and clean the order data, extract the service demand location of each user, ensure that the extracted location information is accurate, and process missing or erroneous data. The extracted service demand location of each user is stored in a new data table for subsequent analysis. The standard for regional division of service demand locations is determined, such as by city, administrative area, or cluster analysis based on geographic coordinates. A suitable regional division method is selected, such as K-Means clustering, DBSCAN, etc., to group the service demand locations, perform regional division on the extracted service demand locations, generate multiple service demand areas, use Python's scikit-learn library to implement cluster analysis, and assign each location to a corresponding area. The service demand area is combined with time information to construct a spatiotemporal demand data set, ensure that the timestamp, service area and demand fields are included. Statistical analysis methods, such as frequency statistics, average value calculation, etc., are used to analyze the demand situation of each service demand area in different time periods. Time series analysis methods (such as ARI) are used MA model) or spatiotemporal data visualization tools (such as heat maps) to display the evolution characteristics of demand distribution, select appropriate analysis methods, such as sliding average method, Fourier transform or wavelet transform, to identify demand fluctuation patterns, analyze the dynamic demand fluctuation patterns of spatiotemporal demand distribution characteristics, identify the patterns and trends of demand fluctuations, use the statsmodels library in Python for time series analysis, extract fluctuation patterns, store the identified user dynamic demand fluctuation patterns in the database for subsequent demand forecasting, select suitable dynamic demand forecasting models, such as linear regression, decision tree, random forest or LSTM (long short-term memory network), use historical demand data and the identified dynamic demand fluctuation patterns to train the forecasting model, and verify it to ensure the accuracy of the model, use the trained model to forecast dynamic service demand based on the user dynamic demand fluctuation patterns, generate forecast data, use Python's scikit-learn or TensorFlow to train and forecast the model, store the generated dynamic service demand forecast data in the database, and use visualization tools (such as Matplotlib or Tableau) to display the forecast results to support decision-making.

[0028] In this embodiment, step S4 includes the following steps: Step S41: Analyze the current idle service nodes on the supply chain service node distribution map and extract the idle service nodes; Step S42: performing node resource configuration analysis on the idle service nodes and extracting resource configuration data of the idle nodes; Step S43: performing optimal demand-supply matching on resource configuration data of idle nodes based on dynamic service demand prediction data to generate service demand optimal node matching data; Step S44: Make dynamic demand allocation decisions on the service demand optimal node matching data and construct a dynamic service demand allocation strategy.

[0029] In this embodiment, the status information of all service nodes is extracted from the service node distribution map to ensure that the current status of each node (such as busy or idle) is included. The service node status data is filtered using a data processing tool (such as Python's pandas) to extract the currently idle service nodes. For example, a conditional filtering method can be used to obtain a list of idle nodes, such as: free_nodes=nodes[nodes['status']=='free'], and resource configuration data of idle service nodes is collected, including available personnel, equipment, time and other information. An appropriate analysis method is selected, and descriptive statistics (such as mean, standard deviation) or cluster analysis is used to evaluate the resource configuration of idle nodes. The resource configuration data of idle service nodes is analyzed, and the resource configuration characteristics of each idle node are extracted. Pandas is used for data analysis to obtain the availability and configuration of resources. The resource configuration data of the idle nodes obtained by the analysis is stored in a database to facilitate subsequent matching and decision-making, and dynamic Dynamic service demand forecast data, ensure that the data includes the predicted service demand type, quantity and time, select a suitable matching algorithm, such as linear programming, Hungarian algorithm or optimization algorithm based on machine learning, to ensure that demand and resources can be effectively matched, based on dynamic service demand forecast data, optimal demand and supply matching is performed on the resource configuration data of idle nodes, create a matching model, associate the predicted service demand with the available service node resources, generate the optimal node matching data for service demand, use Python's scipy.optimize library or PuLP library to implement the matching algorithm, determine the decision criteria for dynamic service demand allocation, such as response time, service quality, resource utilization, etc., select a suitable decision model, such as multi-objective optimization model, hierarchical analysis method (AHP) or dynamic decision-making method based on reinforcement learning, make dynamic demand allocation decisions based on the optimal node matching data for service demand, build a dynamic service demand allocation strategy, evaluate the effectiveness of the allocation strategy by simulating different scenarios, and ensure that it can adapt to demand changes in real time.

[0030] In this embodiment, the specific steps of step S43 are: Calculate the spatial location of idle service nodes; Highlighting nodes on the supply chain service node distribution map based on the spatial position to obtain an idle service node distribution map; Calculate the user's current location based on the dynamic service demand prediction data to obtain the user's current location coordinates; Based on the user's current location coordinates, the minimum distance matching calculation is performed on the idle service node distribution map to obtain the minimum distance of the user's service demand; Based on the dynamic service demand prediction data, the resource configuration data of the idle nodes is matched with service type resources to obtain resource matching data of the user demand type; Performing a minimum service response analysis on the resource matching data of the minimum distance of user service demand and the type of user demand to generate user minimum service response data; Based on the user's minimized service response data, the idle service node distribution map is optimally matched with the supply and demand to generate the service demand optimal node matching data.

[0031] In this embodiment, the geographic coordinate information (such as latitude and longitude) of all idle service nodes is extracted from the database, and the extracted coordinate information is organized into a format suitable for visualization, usually a coordinate list. The geopandas library or matplotlib in Python is used to visualize the node position, and the coordinates of the idle service nodes are highlighted on the supply chain service node distribution map. The distribution map is drawn using the matplotlib library, and the scatter() method is called to mark the idle nodes with different colors or shapes, and the idle service node distribution map is generated, and it is saved as an image file or displayed in the user interface. GPS or other positioning services (such as Google Maps API) are used to obtain the user's current location coordinates (longitude, latitude), ensuring that the obtained location information is real-time and accurate, and the user's current location coordinates are stored in the database for subsequent use. The record format can be JSON or directly stored in the database table. A suitable distance calculation method is selected, usually using the Euclidean distance or Haversine distance formula (applicable to geographic coordinates), and the distance between the coordinates of each idle service node and the user's current location is calculated, and a list containing the distance between each node and the user is generated, and the cdist() function in the scipy.spatial.distance module of Python is used for calculation. Calculate, identify the minimum distance and its corresponding idle service node from the calculated distance list, determine the service type resource matching criteria, such as service type, number of available resources, etc., extract the resource configuration data of idle service nodes from the database, including the available service type and number of each node, match the resource configuration of idle nodes based on the user's dynamic service demand prediction data, use the conditional screening method to determine whether each idle node meets the user's service needs, determine the factors that affect the service response, such as distance, service type matching degree, etc., collect the minimum distance and resource matching data of user service needs, form the basic data for response analysis, use the multi-objective optimization method to analyze the minimum The impact of distance and resource matching data on service response, generating minimized service response data, can be calculated and analyzed using Python's numpy library, and appropriate matching algorithms, such as linear programming, multi-objective optimization, or genetic algorithms, can be selected to achieve optimal demand-supply matching. Based on user minimized service response data and resource allocation of idle service nodes, optimal demand-supply matching is performed. Using Python's scipy.optimize library, a matching model is constructed to solve the optimal configuration, and the service demand optimal node matching data is stored in the database for subsequent decision support and dynamic scheduling. The optimal node matching results are visualized for easy viewing and analysis by decision makers.

[0032] In this embodiment, the specific steps of step S5 are: Step S51: Adaptively optimize the dispatch of orders on the supply chain platform according to the service demand optimal node matching data, thereby generating an adaptive dispatch optimization strategy; Step S52: performing real-time dispatch tracking on the adaptive dispatch optimization strategy, and extracting real-time dispatch tracking response data; Step S53: Calculate the supply chain node service delay for the real-time dispatch tracking response data to obtain the supply chain node service delay data; Step S54: dynamically adjust the delay of the adaptive dispatch optimization strategy based on the supply chain node service delay data to obtain a dynamic delay adjustment strategy.

[0033] In this embodiment, the optimal node matching data of service demand is extracted from the database, including user demand, service node information and its resource configuration status, and a suitable adaptive dispatch optimization algorithm is selected, such as genetic algorithm, particle swarm optimization or linear programming, to ensure the efficiency and flexibility of the dispatch strategy. Based on the optimal node matching data of service demand, an adaptive dispatch optimization strategy is generated by using the selected optimization algorithm, and the algorithm is implemented using Python's scipy.optimize or DEAP library. The objective function (such as minimizing response time, maximizing resource utilization, etc.) is set, and a real-time dispatch tracking system is constructed to ensure that the dispatch status and service node response information can be obtained in real time. Real-time dispatch response data is obtained from the supply chain platform, including dispatch time, completion time, service node status, etc. A data stream processing framework (such as Apache Kafka or Apache Spark) is used to perform real-time data stream processing, extract real-time dispatch tracking response data, and store the real-time data in the memory data. The system stores the data in a database (such as Redis) for quick access. It selects a suitable delay calculation method, such as response time calculation, average delay calculation, or percentile delay analysis. It analyzes the real-time dispatch tracking response data and calculates the service delay of each supply chain node (i.e., the time difference from dispatch to response). It uses Python's pandas library to calculate the response time and generate a list of delay data. It determines the standards and goals for dynamic delay adjustment, such as the maximum allowable delay and service level agreement (SLA). It analyzes the distribution characteristics and changing trends of delays based on the service delay data of supply chain nodes, identifies delay peaks and anomalies, and adjusts the adaptive dispatch optimization strategy based on the delay data. It determines the response strategy (such as reallocating tasks, increasing resources, etc.) when the delay exceeds the preset value. It uses machine learning models (such as regression analysis and decision trees) to predict delays and dynamically adjust the dispatch strategy. It stores the generated dynamic delay adjustment strategy in the database and feeds it back to the dispatch system to ensure the improvement of real-time response capabilities.

[0034] In this embodiment, the specific steps of step S6 are: Step S61: Analyze the evolution of user personalized demand on user real-time order data to generate user personalized demand habits; Step S62: Perform multi-scenario transfer learning on the user's personalized demand habits, so as to obtain multi-scenario user demand habit characteristics; Step S63: Based on the characteristics of user demand habits in multiple scenarios, intelligent dispatching decision analysis is performed on the dynamic delay adjustment strategy to build a supply chain intelligent dispatching model.

[0035] In this embodiment, the real-time order data of the user is extracted from the database, including information such as user ID, order time, service type, and service location. An appropriate analysis method is selected, such as cluster analysis, time series analysis, or trend analysis, to identify the evolution pattern of user demand. The user's order data is processed to identify the user's personalized demand habits. For example, the user's needs are divided into different categories through cluster analysis, or the seasonal changes in demand are identified through time series analysis. The scikit-learn library in Python is used for cluster analysis, or statsmodels is used for time series analysis. Multiple scenarios that need to be analyzed are determined, such as different time periods, seasons, holidays, or special events (such as promotional activities). An appropriate transfer learning framework and model are selected, such as a transfer learning method based on deep learning (such as convolutional neural network CNN, recurrent neural network RNN). Data integration and preprocessing are performed on user demand habits in different scenarios to ensure data consistency and availability. , use pandas for data cleaning and feature engineering, extract relevant features, use transfer learning models to train and analyze user demand habits in different scenarios, and apply the model to other scenarios after training it in one scenario to identify user demand characteristics. Use Python's TensorFlow or PyTorch to build and train the model, determine the goals of the intelligent dispatch model, such as minimizing response time, improving user satisfaction, or optimizing resource utilization, and select appropriate intelligent decision-making analysis methods, such as reinforcement learning, decision trees, random forests, or hybrid models. Based on the characteristics of user demand habits in multiple scenarios, build an intelligent dispatch decision model, input user demand characteristics and the current service node status, and output the best dispatch strategy. Use Python's scikit-learn library or deep learning framework to train and evaluate the model, store the constructed supply chain intelligent dispatch model and the corresponding decision results in the database, and develop a feedback mechanism to continuously optimize the performance of the model.

[0036] In this embodiment, a dispatch optimization device based on an automobile service supply chain is also provided, including: The node distribution module is used to obtain multi-source automotive service supply chain information and user real-time order data based on the supply chain platform; identify supply chain service vendors and perform dynamic node spatial distribution fitting on multi-source automotive service supply chain information to build a supply chain service node distribution map; The demand load module is used to perform user order service demand analysis on user real-time order data and perform real-time service demand load analysis to obtain the real-time service demand load characteristics of the supply chain; The demand forecasting module is used to perform temporal and spatial demand distribution evolution on the real-time service demand load characteristics of the supply chain, and then perform dynamic service demand forecasting to generate dynamic service demand forecasting data; The optimal demand matching module is used to perform optimal demand-supply matching on the supply chain service node distribution map based on dynamic service demand forecast data and build a dynamic service demand allocation strategy; The delay adjustment module is used to optimize the supply chain platform's adaptive dispatching according to the optimal node matching data of service demand, and adjust the supply chain node service delay to obtain a dynamic delay adjustment strategy; The intelligent dispatch decision module is used to perform multi-scenario transfer learning of dynamic delay adjustment strategies based on users' real-time order data, conduct intelligent dispatch decision analysis, and build a supply chain intelligent dispatch model.

[0037] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above-mentioned methods for optimizing dispatching orders based on the automotive service supply chain are implemented.

[0038] Those skilled in the art clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0039] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it is stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or partly or all or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media for storing program codes such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks or optical disks.

[0040] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0041] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A dispatch optimization method based on automobile service supply chain, characterized in that: The following steps are involved: Step S1: obtaining multi-source automobile service supply chain information and user real-time order data based on the supply chain platform; identifying supply chain service vendors and performing dynamic node spatial distribution fitting on the multi-source automobile service supply chain information, and constructing a supply chain service node distribution map; Step S2: Performing user order service demand analysis on user real-time order data and performing real-time service demand load analysis to obtain supply chain real-time service demand load characteristics; Step S3: Performing temporal and spatial demand distribution evolution on the real-time service demand load characteristics of the supply chain, and then performing dynamic service demand forecasting to generate dynamic service demand forecasting data; Step S4: Based on the dynamic service demand forecast data, optimal demand and supply matching is performed on the supply chain service node distribution map to construct a dynamic service demand allocation strategy; Step S5: Adaptively optimize the dispatch of orders on the supply chain platform according to the service demand optimal node matching data, and adjust the supply chain node service delay to obtain a dynamic delay adjustment strategy; Step S6: Based on the user's real-time order data, multi-scenario transfer learning is performed on the dynamic delay adjustment strategy, and intelligent dispatching decision analysis is performed to build a supply chain intelligent dispatching model.

2. The dispatch optimization method based on the automobile service supply chain according to claim 1 is characterized in that: The specific steps of step S1 are: Step S11: acquiring multi-source automobile service supply chain information and user real-time order data based on the supply chain platform; Step S12: Identify supply chain service vendors for multi-source automobile service supply chain information and extract multiple supply chain service nodes; Step S13: Analyze the supply chain service types of multiple supply chain service nodes to obtain the service type of each node; Step S14: Calculate the geographic spatial position of multiple supply chain service nodes to obtain the geographic spatial coordinates of each service node; Step S15: performing node space distribution analysis on the geographic space coordinates of each service node, thereby obtaining service node distribution data; Step S16: Perform dynamic node spatial distribution fitting on the service node distribution data based on the service type of each node to construct a supply chain service node distribution map.

3. The dispatch optimization method based on the automobile service supply chain according to claim 1 is characterized in that: The specific steps of step S2 are: Step S21: Performing user order service demand analysis on user real-time order data to obtain multiple order service demand data; Step S22: performing demand type frequency calculation on a plurality of order service demand data, thereby obtaining the demand frequency of each service type; Step S23: quantify the time-series demand change of the demand frequency of each service type, and generate user service type demand change data; Step S24: Calculate the demand peak value of the user service type demand change data to obtain the user service demand peak value; Step S25: Perform real-time service demand load analysis based on the user service demand peak to obtain the real-time service demand load characteristics of the supply chain.

4. The dispatch optimization method based on the automobile service supply chain according to claim 1 is characterized in that: The specific steps of step S3 are: Step S31: extracting the service demand location of each user based on the user's real-time order data; Step S32: dividing the service demand location of each user into demand location areas, thereby obtaining multiple service demand areas; Step S33: performing spatiotemporal demand distribution evolution on the real-time service demand load characteristics of the supply chain based on multiple service demand areas to obtain spatiotemporal demand distribution characteristics; Step S34: mining the dynamic demand fluctuation law based on the spatiotemporal demand distribution characteristics to generate the user dynamic demand fluctuation law; Step S35: Perform dynamic service demand prediction based on the user dynamic demand fluctuation pattern to generate dynamic service demand prediction data.

5. The dispatch optimization method based on the automobile service supply chain according to claim 1 is characterized in that: The specific steps of step S4 are: Step S41: Analyze the current idle service nodes on the supply chain service node distribution map and extract the idle service nodes; Step S42: performing node resource configuration analysis on the idle service nodes and extracting resource configuration data of the idle nodes; Step S43: performing optimal demand-supply matching on resource configuration data of idle nodes based on dynamic service demand prediction data to generate service demand optimal node matching data; Step S44: Make dynamic demand allocation decisions on the service demand optimal node matching data and construct a dynamic service demand allocation strategy.

6. The dispatch optimization method based on the automobile service supply chain according to claim 5 is characterized in that: The specific steps of step S43 are: Calculate the spatial location of idle service nodes; Highlighting nodes on the supply chain service node distribution map based on the spatial position to obtain an idle service node distribution map; Calculate the user's current location based on the dynamic service demand prediction data to obtain the user's current location coordinates; Based on the user's current location coordinates, the minimum distance matching calculation is performed on the idle service node distribution map to obtain the minimum distance of the user's service demand; Based on the dynamic service demand prediction data, the resource configuration data of the idle nodes is matched with service type resources to obtain resource matching data of the user demand type; Performing a minimum service response analysis on the resource matching data of the minimum distance of user service demand and the type of user demand to generate user minimum service response data; Based on the user's minimized service response data, the idle service node distribution map is optimally matched with the supply and demand to generate the service demand optimal node matching data.

7. The dispatch optimization method based on the automobile service supply chain according to claim 1 is characterized in that: The specific steps of step S5 are: Step S51: Adaptively optimize the dispatch of orders on the supply chain platform according to the service demand optimal node matching data, thereby generating an adaptive dispatch optimization strategy; Step S52: performing real-time dispatch tracking on the adaptive dispatch optimization strategy, and extracting real-time dispatch tracking response data; Step S53: Calculate the supply chain node service delay for the real-time dispatch tracking response data to obtain the supply chain node service delay data; Step S54: dynamically adjust the delay of the adaptive dispatch optimization strategy based on the supply chain node service delay data to obtain a dynamic delay adjustment strategy.

8. The dispatch optimization method based on the automobile service supply chain according to claim 1 is characterized in that: The specific steps of step S6 are: Step S61: Analyze the evolution of user personalized demand on user real-time order data to generate user personalized demand habits; Step S62: Perform multi-scenario transfer learning on the user's personalized demand habits, so as to obtain multi-scenario user demand habit characteristics; Step S63: Based on the characteristics of user demand habits in multiple scenarios, intelligent dispatching decision analysis is performed on the dynamic delay adjustment strategy to build a supply chain intelligent dispatching model.

9. A dispatch optimization device based on an automobile service supply chain, characterized in that: The method for optimizing dispatching orders based on the automobile service supply chain according to claim 1 comprises: The node distribution module is used to obtain multi-source automotive service supply chain information and user real-time order data based on the supply chain platform; identify supply chain service vendors and perform dynamic node spatial distribution fitting on multi-source automotive service supply chain information to build a supply chain service node distribution map; The demand load module is used to perform user order service demand analysis on user real-time order data and perform real-time service demand load analysis to obtain the real-time service demand load characteristics of the supply chain; The demand forecasting module is used to perform temporal and spatial demand distribution evolution on the real-time service demand load characteristics of the supply chain, and then perform dynamic service demand forecasting to generate dynamic service demand forecasting data; The optimal demand matching module is used to perform optimal demand-supply matching on the supply chain service node distribution map based on dynamic service demand forecast data and build a dynamic service demand allocation strategy; The delay adjustment module is used to optimize the supply chain platform's adaptive dispatching according to the optimal node matching data of service demand, and adjust the supply chain node service delay to obtain a dynamic delay adjustment strategy; The intelligent dispatch decision module is used to perform multi-scenario transfer learning of dynamic delay adjustment strategies based on users' real-time order data, conduct intelligent dispatch decision analysis, and build a supply chain intelligent dispatch model.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the dispatch optimization method based on the automobile service supply chain described in any one of claims 1 to 8 are implemented.

Images