Digital logistics freight platform architecture operation method and management system

Abstract

This invention discloses an operational method and management system for a digital logistics freight platform architecture. The operational method includes: collecting multi-dimensional data; preprocessing the multi-dimensional data to form a basic scheduling data pool; constructing a multi-dimensional dynamic intelligent scheduling engine based on a preset multi-objective optimization algorithm; generating scheduling schemes based on the scheduling engine and transportation scenarios; combining the scheduling schemes with publicized dispatch rules for scheduling execution control and feeding back control implementation data; loading the control implementation data into a preset dynamic pricing system, enabling data interaction between the dynamic pricing system and a big data platform; and having the big data platform perform fusion analysis based on the control implementation data and multi-dimensional data, outputting standardized freight rate calculation results and dynamic supply and demand control strategies. The standardized freight rate calculation results are used for pricing control within the dynamic pricing system, and the control strategies are used for strategy optimization of the scheduling engine. The technical solution of this application can improve scheduling accuracy and make pricing more reasonable.

Description

Technical Field

[0001] This invention relates to the field of vehicle IoT technology, specifically to a digital logistics freight platform architecture operation method and management system. Background Technology

[0002] In the actual operation of digital logistics freight platforms, inaccurate scheduling and unreasonable pricing are long-standing problems that influence each other and seriously affect the platform's operational quality. Specifically, the scheduling process lacks scientific coordination and matching logic, failing to accurately connect the core needs of cargo and transportation capacity, leading to inaccurate scheduling. This often results in mismatches between cargo and vehicle qualifications, and incompatible transportation routes and scenarios, affecting the timeliness and stability of cargo transportation. The pricing process, on the other hand, lacks standardized, unified, and fair pricing criteria, resulting in chaotic and unreasonable pricing standards. This makes it difficult to reasonably control the logistics costs for cargo owners and ensure reasonable operating income for drivers. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a digital logistics freight platform architecture operation method that can effectively improve scheduling accuracy and make pricing more reasonable.

[0004] This application provides an operational method for a digital logistics freight platform architecture, wherein the digital logistics freight platform architecture includes a shipper end, a driver end, and an environment end, and the operational method includes:

[0005] Collect comprehensive data from cargo owners, drivers, and the environment; preprocess the comprehensive data to form a basic data pool for scheduling.

[0006] Based on the aforementioned basic data pool, a multi-dimensional dynamic intelligent scheduling engine is constructed according to a preset multi-objective optimization algorithm, and a scheduling scheme is generated based on the scheduling engine and the current transportation scenario.

[0007] The aforementioned scheduling scheme is combined with the published order dispatch rules for scheduling execution control, and the control implementation data is fed back.

[0008] The control and management data is loaded into a preset dynamic pricing system, which interacts with the big data platform.

[0009] The big data platform integrates and analyzes the control and management data with the full-dimensional data to output standardized freight rate calculation results and dynamic supply and demand control strategies. The standardized freight rate calculation results are used to control the pricing of the dynamic pricing system, and the control strategies are used to optimize the scheduling engine.

[0010] In one aspect, the preprocessing steps for the full-dimensional data include:

[0011] Eliminate vague, false, and invalid data, and verify the authenticity and completeness of the data;

[0012] Use historical data from the same source on the platform to fill in any missing key data;

[0013] Establish data timeliness tags to update data that exceeds the preset time limit.

[0014] In one aspect, the step of generating a scheduling scheme based on the scheduling engine and the current transportation scenario includes:

[0015] By combining real-time capacity data and historical capacity change patterns from the environmental perspective, the capacity situation in each region and transportation scenario can be predicted in advance based on a pre-set capacity supply and demand prediction model.

[0016] The scheduling engine generates a scheduling plan based on the available capacity and the current transportation scenario.

[0017] While generating the scheduling plan, the system matches return freight sources in the corresponding regions.

[0018] The return trip cargo is synchronized and matched with the scheduling scheme.

[0019] In one aspect, the steps of combining the scheduling scheme with the publicly announced order dispatch rules for scheduling execution control include:

[0020] Based on the supply and demand dynamic adjustment strategy output by the big data platform, the priority weight of order dispatch is optimized in real time to form at least two types of orders: high-level and low-level.

[0021] A real-time driver credit score verification process is embedded in the dispatching process, and a preset threshold for the credit score is set.

[0022] Drivers with credit scores below the preset threshold are automatically blocked from receiving high-priority orders.

[0023] In one aspect, the steps of combining the aforementioned scheduling scheme with publicly available order dispatch rules for scheduling execution control and feeding back control data include:

[0024] The entire process of recording and documenting dispatch plans, order push records, driver order confirmation information, transportation trajectory data, and anomaly handling is documented.

[0025] When providing feedback on the implementation data of management and control, the blockchain-based evidence hash value is embedded into the implementation data.

[0026] The authenticity of the control data is verified by reverse verification based on the hash value.

[0027] In one aspect, the steps for the dynamic pricing system to interact with the big data platform include:

[0028] Based on the dynamic road consumption data and driver operating cost data output by the big data platform, the standardized fare calculation formula is adaptively adjusted in real time.

[0029] Establish a data interaction anomaly early warning mechanism and set anomaly judgment criteria;

[0030] When data transmission is delayed or interrupted, a temporary call to the local cached data is automatically triggered.

[0031] In one aspect, the big data platform performs fusion analysis based on the control and management implementation data and the full-dimensional data, and outputs standardized freight rate calculation results and dynamic supply and demand adjustment strategies, including:

[0032] A big data fusion analysis framework is built by adopting both real-time analysis mode and offline review mode;

[0033] Through the real-time analysis mode, the control and management implementation data and the full-dimensional data are processed to output standardized freight rate calculation results and supply and demand regulation strategies;

[0034] Through the offline review mode, a review analysis is conducted based on historical management and control implementation data and the full-dimensional data, and the parameters of the multi-objective optimization algorithm are optimized based on the review results.

[0035] In one aspect, the steps of optimizing the scheduling engine's strategy include:

[0036] Extract relevant data on empty run rate and order fulfillment rate from the aforementioned control and management implementation data;

[0037] Based on the relevant data, targeted subsidies will be implemented for drivers of different levels and for different types of cargo.

[0038] The subsidy standards are linked to the scheduling engine, and the capacity is guided to flow to the gap areas based on the subsidies;

[0039] Subsidies are used as incentives to guide drivers to improve their performance and optimize the dispatching strategy of the scheduling engine.

[0040] One aspect involves the steps of collecting comprehensive data from the shipper, driver, and environmental perspectives, including:

[0041] Establish separate data collection channels for cargo owners, drivers, and the environment, and determine the scope of data collection.

[0042] Collect data from cargo owners regarding cargo attributes, shipping and receiving addresses, timeliness requirements, and value-added service needs;

[0043] Collect data from the driver's end, including vehicle parameters, real-time operating status, order preferences, and historical fulfillment records;

[0044] Collect real-time road conditions, regional transportation capacity supply and demand, and road consumption data from the environmental perspective;

[0045] The collected data are aggregated to form an initial full-dimensional data set.

[0046] Furthermore, to address the aforementioned issues, this application also provides a digital logistics freight platform architecture management system. The digital logistics freight platform architecture includes a shipper's end, a driver's end, and an environment end. The management system includes:

[0047] The data acquisition module is used to collect full-dimensional data from the cargo owner, driver, and environment, and preprocess the full-dimensional data to form a basic data pool for scheduling.

[0048] The generation module is used to construct a multi-dimensional dynamic intelligent scheduling engine based on the basic data pool and according to a preset multi-objective optimization algorithm, and to generate a scheduling scheme based on the scheduling engine and the current transportation scenario.

[0049] The module is used to combine the scheduling scheme with the published order dispatch rules for scheduling execution control and to feed back the control implementation data;

[0050] The interaction module is used to load the control and management data into a preset dynamic pricing system, which interacts with the big data platform.

[0051] The fusion analysis module is used by the big data platform to perform fusion analysis on the control and management landing data and the full-dimensional data, and output standardized freight rate calculation results and supply and demand dynamic adjustment strategies. The standardized freight rate calculation results are used to adjust the pricing of the dynamic pricing system, and the adjustment strategies are used to optimize the scheduling engine.

[0052] The beneficial effects of this invention are as follows: by collecting and preprocessing comprehensive data from cargo owners, drivers, and the environment, a precise and effective basic data pool for scheduling is formed, reducing scheduling deviations caused by incomplete or inaccurate data; based on the basic data pool, a multi-dimensional dynamic intelligent scheduling engine is constructed through a multi-objective optimization algorithm, and scheduling schemes are generated in conjunction with the current transportation scenario, achieving precise matching of cargo sources, transportation capacity, and transportation scenarios, reducing scheduling inaccuracies; at the same time, the landing data fed back from scheduling execution control provides optimization basis for the scheduling engine, forming a scheduling optimization closed loop and continuously improving scheduling accuracy. Regarding pricing rationality, the data from dispatch execution and control is loaded into the dynamic pricing system and interacts with the big data platform. This allows the big data platform to integrate multi-dimensional data with control data for comprehensive analysis, outputting standardized freight rate calculation results. This ensures that freight rate calculation is systematic and based on evidence, reducing the problem of chaotic pricing standards. At the same time, the dynamic supply and demand control strategies output by the big data platform enable pricing to adapt to real-time supply and demand of transportation capacity, balancing the logistics cost control of shippers with reasonable operating income for drivers. This achieves fair and reasonable pricing, further reducing the problem of unreasonable pricing, realizing the synergistic optimization of platform dispatch and pricing, improving the platform's operational quality, and meeting the core needs of the platform, shippers, and drivers. Attached Figure Description

[0053] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0054] Figure 1 This is a flowchart illustrating the operational steps of the digital logistics freight platform architecture in this application.

[0055] Figure 2 This is a schematic diagram of the functional modules of the digital logistics freight platform architecture management system of this application. Detailed Implementation

[0056] The embodiments of the technical solution of the present invention will now be described in detail with reference to the accompanying drawings. These embodiments are merely illustrative of the technical solution of the present invention and are therefore intended to limit the scope of protection of the present invention.

[0057] It should be noted that, unless otherwise stated, the technical or scientific terms used in this application should have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0058] like Figure 1As shown, this application provides an operational method for a digital logistics freight platform architecture. The digital logistics freight platform architecture includes a shipper's end, a driver's end, and an environment end. The operational method includes:

[0059] Step S10 involves collecting comprehensive data from the shipper, driver, and environmental perspectives. This comprehensive data is preprocessed to form a basic dispatch data pool. The comprehensive data includes various relevant data from the shipper, driver, and environmental perspectives. Shipper data includes cargo attributes, delivery and pickup addresses, and timeliness requirements. Driver data includes vehicle parameters, real-time operating status, and order preferences. Environmental data includes real-time road conditions and regional transportation capacity supply and demand. Data preprocessing involves systematically processing the collected data, including removing ambiguous, false, and invalid data; verifying the authenticity and completeness of valid data; intelligently completing missing key data; and establishing a data timeliness management mechanism to ensure the real-time nature and completeness of the data.

[0060] Step S20: Based on the basic data pool, a multi-dimensional dynamic intelligent scheduling engine is constructed according to a preset multi-objective optimization algorithm. A scheduling plan is generated based on the scheduling engine and the current transportation scenario. The basic data pool provides the data foundation for the construction of the scheduling engine. The multi-objective optimization algorithm focuses on the accurate matching of cargo sources, transportation capacity, and transportation scenarios, and refines the algorithm logic by combining objectives such as improving scheduling efficiency and rational utilization of transportation resources. The multi-dimensional dynamic intelligent scheduling engine is built on the multi-objective optimization algorithm. The scheduling engine can comprehensively integrate various types of data in the basic data pool and generate suitable scheduling plans based on the specific needs of the current actual transportation scenario. The current transportation scenario can be divided into different types such as intra-city short-distance transportation and medium- and long-distance transportation. The corresponding generated scheduling plans can better adapt to the scenario requirements, improving the targeting and accuracy of scheduling.

[0061] Multi-objective optimization algorithms include non-dominated sorting genetic algorithms, fast non-dominated sorting genetic algorithms with elitist strategies, multi-objective particle swarm optimization algorithms, and multi-objective simulated annealing algorithms.

[0062] Step S30 involves combining the scheduling plan with the publicly available order dispatch rules for scheduling execution control and providing feedback on the control implementation data. The publicly available order dispatch rules are standardized rules that are publicly searchable across the entire platform. These rules clearly define the priority logic and adaptation standards for order dispatch. Combining the scheduling plan with these rules for execution control standardizes the scheduling execution process and ensures that the scheduling plan proceeds in an orderly manner according to established standards. During the scheduling execution control process, the entire process, including order push, driver order confirmation, and transportation progress, is monitored. The execution progress and effectiveness of the scheduling plan are tracked in real time, and various relevant data are collected synchronously during the control process to form control implementation data and provide feedback.

[0063] Step S40 involves loading the control and management data into a pre-set dynamic pricing system. This dynamic pricing system interacts with the big data platform. The control and management data includes various actual data from the scheduling and execution process. This data reflects the scheduling and execution effectiveness and the current situation of transportation capacity and cargo availability. Loading it into the dynamic pricing system allows the pricing system to obtain real operational data support. The dynamic pricing system is a pre-set standardized pricing management system with the ability to interact with the big data platform. This data interaction enables the push of control and management data to the big data platform and the feedback of relevant analytical data from the big data platform to the pricing system.

[0064] Step S50: The big data platform integrates and analyzes the data from the implementation of management and control measures with multi-dimensional data, outputting standardized freight rate calculation results and dynamic supply and demand control strategies. The standardized freight rate calculation results are used to control pricing within the dynamic pricing system, while the control strategies are used to optimize the scheduling engine. The big data platform possesses data fusion and analysis capabilities, integrating the data from the implementation of management and control measures with previously collected multi-dimensional data, conducting systematic correlation analysis and in-depth mining of both types of data. By analyzing and outputting standardized freight rate calculation results, the calculation logic and detailed composition of freight rates are clarified, providing standardized and verifiable criteria for pricing. Simultaneously, the output of dynamic supply and demand control strategies accurately reflects the current supply and demand balance of transportation capacity and cargo. The standardized freight rate calculation results can be used to adjust the pricing standards of the dynamic pricing system, making pricing more standardized and reasonable; the dynamic supply and demand control strategies can be used to optimize the algorithm parameters and dispatch logic of the scheduling engine, further improving scheduling accuracy and achieving synergistic optimization of scheduling and pricing.

[0065] In this real-time example, by collecting and preprocessing comprehensive data from cargo owners, drivers, and the environment, a precise and effective basic data pool for scheduling is formed, reducing scheduling deviations caused by incomplete or inaccurate data. Based on this basic data pool, a multi-dimensional dynamic intelligent scheduling engine is constructed using a multi-objective optimization algorithm. This engine generates scheduling plans in conjunction with the current transportation scenario, achieving precise matching of cargo sources, transportation capacity, and transportation scenarios, thus reducing scheduling inaccuracies. Simultaneously, the feedback data from scheduling execution control provides a basis for optimizing the scheduling engine, forming a closed loop for scheduling optimization and continuously improving scheduling accuracy. Regarding pricing rationality, the data from dispatch execution and control is loaded into the dynamic pricing system and interacts with the big data platform. This allows the big data platform to integrate multi-dimensional data with control data for comprehensive analysis, outputting standardized freight rate calculation results. This ensures that freight rate calculation is systematic and based on evidence, reducing the problem of chaotic pricing standards. At the same time, the dynamic supply and demand control strategies output by the big data platform enable pricing to adapt to real-time supply and demand of transportation capacity, balancing the logistics cost control of shippers with reasonable operating income for drivers. This achieves fair and reasonable pricing, further reducing the problem of unreasonable pricing, realizing the synergistic optimization of platform dispatch and pricing, improving the platform's operational quality, and meeting the core needs of the platform, shippers, and drivers.

[0066] In one embodiment of this application, the step of preprocessing full-dimensional data includes:

[0067] Step S110 involves eliminating vague, false, and invalid data to verify the authenticity and completeness of the data. Vague data refers to data that is unclear or ambiguous, such as cargo owners not specifying exact pick-up and drop-off addresses, or drivers not specifying vehicle parameters. False data refers to data that has been artificially altered or fabricated, such as falsely reporting cargo weight or falsifying vehicle operation records. Invalid data refers to data irrelevant to scheduling needs and unable to support scheduling decisions. Data authenticity verification checks whether the collected data matches the actual situation, such as verifying whether driver vehicle parameters match vehicle registration information. Data completeness verification checks for missing data, such as confirming whether cargo owners have provided complete information on cargo delivery timeliness requirements and loading / unloading details.

[0068] Step S120 involves retrieving historical source data from the platform to complete the missing key data. Historical source data refers to data collected during the platform's past operations that is consistent with the currently missing data type and related in origin. Examples include past transportation data for similar goods from the same shipper, historical vehicle operating parameters for the same driver, and historical supply and demand data for the same region. Key data refers to data that significantly impacts scheduling decisions, such as the precise weight and volume of goods, the driver's real-time location, and real-time road condition details. By retrieving this type of historical data to complete the currently missing key data, the impact of data loss on subsequent scheduling processes can be reduced, making the scheduling foundation data pool more comprehensive.

[0069] Step S130: Establish data timeliness tags and update data that exceeds the preset time limit. Data timeliness tags are time identifiers added to each piece of collected data, clearly defining the data collection time and validity period. The validity period for different types of data can be set according to actual operational needs. For example, the validity period for real-time traffic data from the environmental perspective can be set to a shorter time, while the validity period for cargo attribute data from the cargo owner perspective can be set to a longer time. Data exceeding the preset time limit refers to data that exceeds the set validity period and is inconsistent with the current actual situation, such as traffic data from several hours ago or transportation capacity supply and demand data from several days ago. Timely updates to this type of data ensure that the data in the scheduling basic data pool always aligns with the current operational scenario, reducing the interference of outdated data on scheduling decisions.

[0070] In one embodiment of this application, the step of generating a scheduling scheme based on the scheduling engine and the current transportation scenario includes:

[0071] Step S210: Combining real-time capacity data and historical capacity change patterns from the environmental perspective, a pre-defined capacity supply and demand forecasting model is used to predict the capacity situation in each region and transportation scenario in advance. Real-time capacity data from the environmental perspective refers to the current available capacity in each region and transportation scenario, including information such as the number of available vehicles and driver qualifications. Historical capacity change patterns are the patterns of capacity changes in each region and transportation scenario over time and seasonal factors during the platform's past operations, such as the surge in capacity demand in a certain region during holidays. The capacity supply and demand forecasting model is a pre-defined dedicated forecasting model that can systematically analyze real-time data and historical patterns. The prediction results include capacity shortages or surpluses in each region and transportation scenario, such as predicting insufficient short-distance intra-city transportation capacity in a certain region and excessive long-distance transportation capacity in another region.

[0072] In step S220, the scheduling engine generates a scheduling plan based on the available capacity and the current transportation scenario. The available capacity includes pre-predicted capacity shortages or surpluses in various regions and transportation scenarios. The current transportation scenarios include different types such as intra-city short-distance transportation, medium- and long-distance transportation, and special-category transportation. Different transportation scenarios have different requirements; for example, intra-city short-distance transportation emphasizes timeliness, while medium- and long-distance transportation emphasizes capacity stability. By combining these two factors, the scheduling engine generates a scheduling plan that adapts to the current capacity distribution while meeting the specific needs of the transportation scenario, thus improving the targeting of the scheduling plan.

[0073] Step S230 involves matching return trip cargo sources with the corresponding regions while generating the scheduling plan. Step S230 primarily aims to match return trip cargo sources, unlocking additional value from transportation resources and improving capacity utilization. Return trip cargo sources are those that match the vehicle return routes in the scheduling plan. For example, if a vehicle transports goods from region A to region B according to the scheduling plan, the return trip cargo source is the cargo transported from region B to region A or surrounding regions. Simultaneously with generating the scheduling plan, the platform is searched for return trip cargo sources that meet the criteria, achieving an initial match between return trip cargo sources and vehicle return demand. This lays the foundation for subsequent synchronous integration of the two and reduces empty runs.

[0074] Step S240 involves synchronizing and matching return trip cargo with the dispatch plan. The previously matched return trip cargo is integrated with the currently generated dispatch plan, and the details of the dispatch plan are adjusted to ensure that the transportation needs of the return trip cargo are compatible with the transportation tasks of the original dispatch plan, avoiding conflicts. For example, the transportation routes and times of vehicles are rationally planned so that vehicles can smoothly take on return trip cargo after completing their original dispatch tasks, improving the utilization rate of transportation resources, reducing vehicle empty running rates, increasing drivers' operating income, and simultaneously improving the platform's capacity allocation efficiency.

[0075] In one embodiment of this application, the step of combining the scheduling scheme with the published order dispatch rules for scheduling execution control includes:

[0076] Step S310 involves optimizing the order priority weights in real time based on the supply and demand dynamic control strategy output by the big data platform, resulting in at least two types of orders: high-level and low-level. Step S310 primarily focuses on dynamically optimizing the order priority weights to ensure that the order dispatch logic aligns with the supply and demand situation. The supply and demand dynamic control strategy output by the big data platform is derived from the fusion analysis of full-dimensional data and control implementation data, reflecting the current supply and demand situation for various types of orders in different regions. The order priority weights are adjusted in real time according to the control strategy, resulting in two or more types of orders: high-level and low-level. High-level orders typically represent goods from regions with tight supply and demand or goods with high timeliness requirements, while low-level orders typically represent goods from regions with ample supply and demand or goods with relatively lower timeliness requirements.

[0077] Step S320 embeds a real-time driver credit score verification process during order dispatch, setting a preset credit score threshold. Step S320 primarily implements real-time verification of driver credit scores, standardizing driver eligibility for accepting orders and ensuring order fulfillment quality. Driver credit scores are calculated based on data such as the driver's historical performance records and service attitude, and are used to evaluate the driver's operational reputation and service quality. The preset credit score threshold is a credit qualification standard set according to the platform's operational needs; the preset threshold may vary for different types of orders. During order dispatch, the driver's current credit score is retrieved in real-time for immediate verification of the driver's credit status.

[0078] Step S330: Drivers with credit scores below a preset threshold are automatically blocked from accepting high-priority orders. After real-time verification, drivers with credit scores below the preset threshold are automatically restricted from accepting high-priority orders, and are only allowed to accept low-priority orders. This prevents drivers with poor credit from accepting high-level orders with high performance requirements, reducing the occurrence of order delays and cargo damage, ensuring a better service experience for cargo owners, and maintaining the platform's operational order.

[0079] In one embodiment of this application, the steps of combining the scheduling scheme with the publicized order dispatch rules for scheduling execution control and feeding back the control landing data include:

[0080] Step S301 involves the complete documentation of the dispatch plan, order push records, driver order confirmation information, transportation trajectory data, and anomaly handling process. This documentation includes all key aspects of dispatch execution and control. The dispatch plan includes specific transportation tasks, vehicle allocation, route planning, and other relevant data. Order push records include the time and content of order pushes. Driver order confirmation information includes the time and status of driver order acceptance. Transportation trajectory data includes the real-time location and route of vehicles during transportation. Anomaly handling process data includes any anomalies that occur during transportation and the methods and results of their handling. This complete documentation of the dispatch execution process provides a basis for subsequent verification and dispute resolution.

[0081] Step S302: When feeding back the control and management implementation data, the blockchain evidence hash value is embedded into the control and management implementation data. The blockchain evidence hash value is a unique identifier generated after encrypting the evidence data of the entire scheduling and execution process using blockchain technology. Each piece of evidence data corresponds to a unique hash value, which has the characteristic of being tamper-proof. When feeding back the control and management implementation data, the corresponding hash value is embedded into the control and management implementation data, thus associating the control and management implementation data with the blockchain evidence data and improving the credibility and security of the control and management implementation data.

[0082] Step S303 involves reverse verification of the authenticity of the control data based on the hash value. This reverse hash value verification involves retrieving the original hash value from the blockchain storage and comparing it with the hash value embedded in the control data. If they match, the control data has not been tampered with and is authentic and valid; if they do not match, the control data may have been tampered with. This step verifies the authenticity of the control data, providing reliable support for subsequent pricing control and scheduling optimization based on the control data, and reducing the adverse effects of data tampering.

[0083] In one embodiment of this application, the step of data interaction between the dynamic pricing system and the big data platform includes:

[0084] Step S410: Based on the dynamic road consumption data and driver operating cost data output by the big data platform, the standardized fare calculation formula is adaptively adjusted in real time. The dynamic road consumption data is derived from factors such as real-time road conditions and regional differences, reflecting the actual losses on different transportation routes, such as data related to increased road consumption due to congestion on a certain route. The driver operating cost data includes data related to drivers' fuel costs, labor costs, and vehicle wear and tear, reflecting the actual operating burden on drivers. By adaptively adjusting the preset standardized fare calculation formula in real time based on these two types of data, the fare calculation logic is made to fit the actual operating conditions, improving the rationality and fairness of the fare calculation.

[0085] Step S420: Establish a data interaction anomaly early warning mechanism and set anomaly judgment criteria. This mechanism is a pre-set dedicated early warning system used to monitor the data transmission status between the dynamic pricing system and the big data platform. Anomaly judgment criteria are set based on normal data interaction parameters, including data transmission delay thresholds and data transmission interruption judgment conditions. For example, a data transmission delay exceeding a preset time is considered an anomaly, as is a signal interruption during data transmission. By establishing this anomaly early warning mechanism, potential anomalies during data interaction can be detected promptly, providing sufficient time for subsequent emergency handling.

[0086] Step S430: When data transmission is delayed or interrupted, a temporary call to locally cached data is automatically triggered. When abnormal situations such as data transmission delays or interruptions are detected, a temporary call to locally cached data is automatically triggered. This locally cached data consists of pre-stored recent relevant data, including recent freight rate data, subsidy data, and supply and demand data, which can temporarily replace real-time transmitted data. By temporarily calling the cached data, the normal operation of the dynamic pricing system and the big data platform can be ensured, avoiding the stagnation of scheduling and pricing coordination due to abnormal data interaction, thus improving the stability of platform operation.

[0087] In one embodiment of this application, the steps of the big data platform performing fusion analysis on control and management implementation data and multi-dimensional data to output standardized freight rate calculation results and dynamic supply and demand adjustment strategies include:

[0088] Step S510 involves building a big data fusion analysis framework using both real-time analysis and offline review modes. The real-time analysis mode is designed for rapid processing of various types of real-time data, emphasizing efficiency and enabling data processing and result output within a short timeframe, thus meeting the needs of real-time scheduling and pricing control. The offline review mode is used for in-depth analysis of historical data, focusing on depth and comprehensiveness, and capable of uncovering patterns and problems within historical data. Building a fusion analysis framework that includes both modes balances data processing efficiency and depth, improving the overall effectiveness of the fusion analysis.

[0089] Step S520 involves processing control-implemented data and multi-dimensional data through real-time analysis, outputting standardized freight rate calculation results and supply-demand control strategies. Real-time analysis rapidly integrates control-implemented data and multi-dimensional data, performing correlation analysis and shallow data mining on both types of data without requiring in-depth historical review, ensuring rapid output of analysis results. The output standardized freight rate calculation results clearly define the specific calculation details and standards for freight rates, providing a basis for pricing control in the dynamic pricing system; the output supply-demand control strategies reflect the current balance of transport capacity supply and demand, supporting strategy optimization for the scheduling engine.

[0090] Step S530: Using offline review mode, conduct review analysis based on historical management and control implementation data and comprehensive data. Optimize the parameters of the multi-objective optimization algorithm based on the review results. Offline review mode performs systematic and in-depth analysis of historical management and control implementation data and historical comprehensive data stored on the platform, uncovering patterns of change, identifying problems, and identifying areas for optimization. Based on the results of the review analysis, adjust and optimize the relevant parameters of the multi-objective optimization algorithm, refine the algorithm's optimization logic, and enable the algorithm to better adapt to the actual operational situation of the platform.

[0091] In one embodiment of this application, the step of optimizing the scheduling engine's strategy includes:

[0092] Step S501: Extract relevant data on empty-run rate and order fulfillment rate from the control and management landing data. The control and management landing data includes various data reflecting the platform's operational status. Empty-run rate data reflects vehicle empty-running conditions, including overall empty-run rate, empty-run rate for different regions, and empty-run rate for different vehicle types. Order fulfillment rate data reflects order completion status, including overall fulfillment rate, on-time fulfillment rate, and goods intact fulfillment rate. Extracting these two core data types allows for a clear understanding of the shortcomings of the current scheduling strategy.

[0093] Step S502: Based on relevant data, implement targeted subsidies for drivers of different levels and different types of cargo. Based on extracted empty-running rate and order fulfillment rate data, implement targeted subsidies for drivers of different levels and different types of cargo. For example, provide higher subsidies to high-quality drivers with lower empty-running rates and higher fulfillment rates; provide subsidies for cargo in areas with capacity shortages; and provide subsidies for special categories of cargo that are difficult to fulfill. Through differentiated subsidies, drivers can be precisely incentivized to improve operational quality, guiding capacity to flow to reasonable areas.

[0094] Step S503 involves binding the subsidy standards to the dispatch engine, guiding capacity flow to areas with capacity shortages based on subsidies; integrating the set differentiated subsidy standards into the dispatch engine's order dispatch logic, allowing the dispatch engine to fully consider subsidy factors when generating dispatch plans. For example, in areas with capacity shortages, priority is given to dispatching drivers who can enjoy subsidies, guiding drivers to actively go to areas with capacity shortages to accept orders, alleviating the problem of capacity supply and demand imbalance, and improving the capacity allocation efficiency of the dispatch engine.

[0095] Step S504 uses subsidies as an incentive to guide drivers to improve fulfillment quality and optimize the dispatch engine's order allocation strategy. Differentiated subsidies are used to incentivize drivers to proactively reduce empty-running rates, improve order fulfillment on-time and cargo integrity rates, and accumulate a high-quality fulfillment record. Based on changes in driver fulfillment quality and adjustments in capacity distribution, the dispatch engine continuously optimizes its order allocation strategy, improves the matching logic between cargo and capacity, and makes dispatch solutions more aligned with platform operational needs, further enhancing dispatch accuracy and the platform's overall operational quality.

[0096] In one embodiment of this application, the step of collecting comprehensive data from the cargo owner, driver, and environment includes:

[0097] Step S101 involves establishing data collection channels for the cargo owner, driver, and environment terminals, and defining the collection scope. These data collection channels are dedicated data transmission links between the platform and the cargo owner, driver, and environment terminals, used to acquire relevant data from each terminal. The collection channels for different terminals are independent yet collaborative, ensuring stable and efficient data transmission. The collection scope is defined based on the actual needs of scheduling decisions, clarifying the types and specific content to be collected from each terminal. This avoids wasting resources by collecting irrelevant data and ensures that no core data required for scheduling is missed.

[0098] Step S102: Collect data from the shipper regarding cargo attributes, delivery and receipt addresses, timeliness requirements, and value-added service needs. Cargo attribute data includes the cargo's weight, volume, type, and fragility. Different types of cargo have different transportation needs; for example, fragile items require special protective transportation, and bulk cargo has specific requirements for vehicle load capacity. Delivery and receipt address data includes detailed information on the cargo's originating transportation address and destination delivery address. Timeliness requirement data reflects the shipper's specific expectations for the cargo's transportation completion time. Value-added service needs data represents additional service requests from the shipper based on their own needs, such as door-to-door pickup and delivery.

[0099] Step S103 involves collecting data from the driver's end, including vehicle parameters, real-time operating status, order preferences, and historical performance records. Vehicle parameter data includes specific information such as vehicle load capacity and model; for example, different truck load capacities and cargo box types are suitable for different types of goods. Real-time operating status data includes the vehicle's current location and driving status. Order preference data reflects the driver's order-taking tendencies based on their operational needs; for example, some drivers prefer short-distance intra-city transport orders, while others prefer medium- to long-distance transport. Historical performance record data details the driver's past completed transport orders, including on-time delivery rate and cargo integrity rate.

[0100] Step S104 involves collecting real-time traffic conditions, regional transportation capacity supply and demand, and road consumption data from the environmental perspective. Real-time traffic data includes information on congestion along transportation routes, road construction, and weather conditions. For example, heavy rain may cause traffic disruptions on some road sections, and congested sections will extend transportation time. Regional transportation capacity supply and demand data compares the total current freight volume with the total available transportation capacity in each region. For instance, some regions may have sufficient freight but few available vehicles, while other regions may have surplus capacity. Road consumption data includes cost-related information such as fuel and electricity consumption for different transportation routes.

[0101] Step S105 involves aggregating the collected data to form an initial full-dimensional dataset. This involves systematically aggregating various data collected from the shipper, driver, and environmental terminals, organizing them according to a unified format standard, and eliminating differences in data format to form a complete initial full-dimensional dataset. This dataset includes all the basic data required for scheduling decisions, providing comprehensive and robust data support for subsequent data preprocessing stages.

[0102] See Figure 2 As shown, this application also provides a digital logistics freight platform architecture management system. The digital logistics freight platform architecture includes a cargo owner end, a driver end, and an environment end. The management system includes: a data collection module 10, a generation module 20, a combination module 30, an interaction module 40, and a fusion analysis module 50.

[0103] The data acquisition module 10 is used to collect comprehensive data from the shipper, driver, and environment terminals, and preprocesses this data to form a basic scheduling data pool. The acquisition module 10 primarily undertakes the collection and preprocessing of comprehensive data to form the basic scheduling data pool. It connects to the shipper, driver, and environment terminals respectively, establishing dedicated data acquisition links to obtain relevant data from each terminal. Shipper data includes cargo attributes such as shipping and receiving addresses; driver data includes vehicle parameters and historical performance records; and environment data includes real-time road conditions and road consumption. Data preprocessing involves systematically processing the collected raw data, including data cleaning, verification, and dynamic completion. Invalid data is removed, data validity is verified, and missing data is completed to ensure the accuracy and comprehensiveness of the resulting basic scheduling data pool.

[0104] The generation module 20 is used to construct a multi-dimensional dynamic intelligent scheduling engine based on a basic data pool and a preset multi-objective optimization algorithm. It then generates scheduling schemes based on the scheduling engine and the current transportation scenario. The generation module 20 primarily relies on the scheduling basic data pool to complete the construction of the intelligent scheduling engine and the generation of scheduling schemes. The scheduling basic data pool provides comprehensive data support, and the multi-objective optimization algorithm considers multiple objectives, including improving capacity utilization and scheduling efficiency. The optimization logic is further refined by introducing a capacity supply and demand prediction model. The multi-dimensional dynamic intelligent scheduling engine, built upon this algorithm, can integrate and analyze various types of data in the basic data pool. Combined with the specific needs of the current actual transportation scenario, it generates highly adaptable scheduling schemes. Current transportation scenarios can be categorized into different types, such as intra-city short-distance, medium-distance, and long-distance travel. The generated scheduling schemes can fit the scenario requirements, improving the accuracy and feasibility of scheduling.

[0105] Module 30 is used to integrate the scheduling plan with the publicly announced order dispatch rules for scheduling execution control and to provide feedback on the control implementation data. Module 30 is primarily responsible for combining the generated scheduling plan with the publicly announced order dispatch rules to achieve standardized control over scheduling execution and to provide feedback on the control implementation data. The publicly announced order dispatch rules are standardized rules published by the platform, clearly defining core content such as order priority logic and driver qualification matching standards, ensuring fairness and transparency in the order dispatch process. Module 30 deeply integrates the scheduling plan with these rules, controlling the entire execution process of the scheduling plan, including order push, driver order acceptance, and transportation trajectory tracking, to monitor the execution progress and effectiveness of the scheduling plan in real time. Simultaneously, Module 30 collects various relevant data during the control process, forming control implementation data and providing timely feedback.

[0106] The interaction module 40 is used to load the control and management data into a preset dynamic pricing system, which then interacts with the big data platform. The interaction module 40, combined with the control and management data fed back from module 30, loads the data into the preset dynamic pricing system, allowing the pricing system to obtain actual operational data during the scheduling process and ensuring that pricing adjustments align with actual operational conditions. The dynamic pricing system is a standardized pricing management system preset by the platform, capable of interacting with the big data platform. The interaction process employs standardized data transmission methods, enabling bidirectional transmission of control and management data and big data platform analysis data, providing data interaction support for fare calculation and pricing adjustments.

[0107] The Fusion Analysis Module 50 is used by the big data platform to perform fusion analysis on control and management data and multi-dimensional data, outputting standardized freight rate calculation results and dynamic supply and demand control strategies. The standardized freight rate calculation results are used for pricing control within the dynamic pricing system, while the control strategies are used for strategy optimization of the scheduling engine. The Fusion Analysis Module 50 is primarily responsible for the fusion analysis work of the big data platform, outputting standardized freight rate calculation results and dynamic supply and demand control strategies. The Fusion Analysis Module 50 integrates the control and management data fed back by Module 30 with the multi-dimensional data collected by Module 10, using fusion analysis methods to perform correlation analysis and in-depth mining of the two types of data, balancing data processing efficiency and depth. The standardized freight rate calculation results clarify the calculation logic and detailed standards of freight rates, which can be directly used for pricing control within the dynamic pricing system, making pricing more standardized and fair. The dynamic supply and demand control strategies can accurately reflect the current supply and demand balance of transportation capacity in various regions and transportation scenarios, used for strategy optimization of the scheduling engine, further improving scheduling accuracy.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A digital logistics freight platform architecture operation method, characterized in that, The digital logistics freight platform architecture includes a shipper's end, a driver's end, and an environment end, and the operation method includes: Collect comprehensive data from cargo owners, drivers, and the environment; preprocess the comprehensive data to form a basic data pool for scheduling. Based on the aforementioned basic data pool, a multi-dimensional dynamic intelligent scheduling engine is constructed according to a preset multi-objective optimization algorithm, and a scheduling scheme is generated based on the scheduling engine and the current transportation scenario. The aforementioned scheduling scheme is combined with the published order dispatch rules for scheduling execution control, and the control implementation data is fed back. The control and management data is loaded into a preset dynamic pricing system, which interacts with the big data platform. The big data platform integrates and analyzes the control and management data with the full-dimensional data to output standardized freight rate calculation results and dynamic supply and demand control strategies. The standardized freight rate calculation results are used to control the pricing of the dynamic pricing system, and the control strategies are used to optimize the scheduling engine.

2. The operating method according to claim 1, characterized in that, The steps for preprocessing the full-dimensional data include: Eliminate vague, false, and invalid data, and verify the authenticity and completeness of the data; Use historical data from the same source on the platform to fill in any missing key data; Establish data timeliness tags to update data that exceeds the preset time limit.

3. The operating method according to claim 1, characterized in that, The steps for generating a scheduling scheme based on the scheduling engine and the current transportation scenario include: By combining real-time capacity data and historical capacity change patterns from the environmental perspective, the capacity situation in each region and transportation scenario can be predicted in advance based on a pre-set capacity supply and demand prediction model. The scheduling engine generates a scheduling plan based on the available capacity and the current transportation scenario. While generating the scheduling plan, the system matches return freight sources in the corresponding regions. The return trip cargo is synchronized and matched with the scheduling scheme.

4. The operating method according to claim 1, characterized in that, The steps for scheduling execution control by combining the aforementioned scheduling scheme with the published order dispatch rules include: Based on the supply and demand dynamic adjustment strategy output by the big data platform, the priority weight of order dispatch is optimized in real time to form at least two types of orders: high-level and low-level. A real-time driver credit score verification process is embedded in the dispatching process, and a preset threshold for the credit score is set. Drivers with credit scores below the preset threshold are automatically blocked from receiving high-priority orders.

5. The operating method according to claim 1, characterized in that, The steps of combining the aforementioned scheduling scheme with the published order dispatch rules for scheduling execution control and feeding back the control data include: The entire process of recording and documenting dispatch plans, order push records, driver order confirmation information, transportation trajectory data, and anomaly handling is documented. When providing feedback on the implementation data of management and control, the blockchain-based evidence hash value is embedded into the implementation data. The authenticity of the control data is verified by reverse verification based on the hash value.

6. The operating method according to claim 1, characterized in that, The steps for the dynamic pricing system to interact with the big data platform include: Based on the dynamic road consumption data and driver operating cost data output by the big data platform, the standardized fare calculation formula is adaptively adjusted in real time. Establish a data interaction anomaly early warning mechanism and set anomaly judgment criteria; When data transmission is delayed or interrupted, a temporary call to the local cached data is automatically triggered.

7. The operating method according to claim 1, characterized in that, The big data platform integrates and analyzes the control and management data with the multi-dimensional data to output standardized freight rate calculation results and dynamic supply and demand control strategies, including the following steps: A big data fusion analysis framework is built by adopting both real-time analysis mode and offline review mode; Through the real-time analysis mode, the control and management implementation data and the full-dimensional data are processed to output standardized freight rate calculation results and supply and demand regulation strategies; Through the offline review mode, a review analysis is conducted based on historical management and control implementation data and the full-dimensional data, and the parameters of the multi-objective optimization algorithm are optimized based on the review results.

8. The operating method according to claim 1, characterized in that, The steps for optimizing the scheduling engine's strategy include: Extract relevant data on empty run rate and order fulfillment rate from the aforementioned control and management implementation data; Based on the relevant data, targeted subsidies will be implemented for drivers of different levels and for different types of cargo. The subsidy standards are linked to the scheduling engine, and the capacity is guided to flow to the gap areas based on the subsidies; Subsidies are used as incentives to guide drivers to improve their performance and optimize the dispatching strategy of the scheduling engine.

9. The operating method according to claim 1, characterized in that, The steps for collecting comprehensive data from the shipper, driver, and environmental perspectives include: Establish separate data collection channels for cargo owners, drivers, and the environment, and determine the scope of data collection. Collect data from cargo owners regarding cargo attributes, shipping and receiving addresses, timeliness requirements, and value-added service needs; Collect data from the driver's end, including vehicle parameters, real-time operating status, order preferences, and historical fulfillment records; Collect real-time road conditions, regional transportation capacity supply and demand, and road consumption data from the environmental perspective; The collected data are aggregated to form an initial full-dimensional data set.

10. A digital logistics freight platform architecture management system, characterized in that, The digital logistics freight platform architecture includes a shipper's end, a driver's end, and an environment end, and the management system includes: The data acquisition module is used to collect full-dimensional data from the cargo owner, driver, and environment, and preprocess the full-dimensional data to form a basic data pool for scheduling. The generation module is used to construct a multi-dimensional dynamic intelligent scheduling engine based on the basic data pool and according to a preset multi-objective optimization algorithm, and to generate a scheduling scheme based on the scheduling engine and the current transportation scenario. The module is used to combine the scheduling scheme with the published order dispatch rules for scheduling execution control and to feed back the control implementation data; The interaction module is used to load the control and management data into a preset dynamic pricing system, which interacts with the big data platform. The fusion analysis module is used by the big data platform to perform fusion analysis on the control and management landing data and the full-dimensional data, and output standardized freight rate calculation results and supply and demand dynamic adjustment strategies. The standardized freight rate calculation results are used to adjust the pricing of the dynamic pricing system, and the adjustment strategies are used to optimize the scheduling engine.

Images