Network freight platform intelligent scheduling system and scheduling method based on big data analysis

Abstract

The invention discloses a network freight platform intelligent scheduling system and scheduling method based on big data analysis, belongs to the field of network freight platform scheduling, and is used for solving the problems of how to accurately predict the dynamic change of transportation demands and vehicle resources and how to improve the operation efficiency and service quality of a network freight platform. The system comprises five basic modules, namely a data acquisition module, a data preprocessing module, a big data analysis module, an intelligent scheduling module and a user interaction module, and the intelligent scheduling module adopts an intelligent algorithm to generate a recommended scheduling scheme, monitors a vehicle transportation state in real time and dynamically adjusts the recommended scheduling scheme. According to the invention, through analysis of a large amount of transportation related data, intelligent and precise vehicle scheduling is realized, the transportation efficiency is improved, and the cost is reduced.

Description

Technical Field

[0001] The present invention belongs to the field of network freight platform scheduling, and specifically relates to an intelligent scheduling system and a scheduling method for a network freight platform based on big data analysis. Background Art

[0002] With the rapid development of Internet technology, online freight platforms have been widely used in the logistics industry. However, there are many problems in the current scheduling process of online freight platforms. On the one hand, traditional scheduling methods often rely on manual experience and cannot fully consider many complex transportation factors, such as the real-time location of vehicles, traffic conditions on transportation routes, cargo weight and volume, customer delivery time requirements, etc., resulting in low scheduling efficiency, high vehicle idle rate, and increased transportation costs. On the other hand, although some existing scheduling systems have introduced certain information technology means, they lack in-depth analysis and utilization of big data, cannot accurately predict the dynamic changes of transportation demand and vehicle resources, and are difficult to achieve intelligent and precise scheduling.

[0003] Patent document Urban smart traffic safety travel monitoring and analysis method, emergency evacuation decision-making platform, application number CN202210143402.1. Main classification number G06Q. This technical solution provides a decision-making platform. The urban smart traffic safety travel monitoring and analysis method and emergency evacuation decision-making platform include: traffic travel monitoring and analysis method, emergency decision-making platform, traffic data resource center, integrated traffic operation status monitoring platform, collaborative dispatching platform, traffic safety management operation platform, operation analysis and decision support platform and public service platform.

[0004] This technical solution can improve the scientific management level of road traffic, the modern management of police officers and the emergency alarm and rapid response capabilities of traffic accidents, thereby enhancing the advantage of decision-making capabilities.

[0005] Patent document: Capacity dispatching system and method for online-hailing freight vehicle platform based on order distribution density, application number CN201910006238.8. Main classification number G06Q. The technical solution discloses a customer handheld terminal, which is interconnected with a cloud control platform through an external network, and the cloud control platform is interconnected with the owner's handheld terminal through an external network. The cloud control platform is composed of an order data collection unit, an order data integration unit, an order data statistics unit, an order density analysis unit, a concentration point data collection unit, an online viewing unit for online-hailing vehicles, and a vehicle dispatch control unit. The vehicle dispatch control unit is composed of a vehicle information analysis module, an online active statistics module, a dispatch time calculation module, and a dispatch completion feedback module.

[0006] This technical solution strictly controls the scheduling time through the setting of a scheduling time calculation module using an algorithm, thereby improving the overall user experience of the placing customers as a whole.

[0007] However, none of the above technical solutions involve the network freight platform. The network freight platform needs to comprehensively collect multi-source data such as vehicles, goods, orders, and traffic in order to make accurate scheduling and effectively reduce the vehicle empty driving rate. On the one hand, the network freight platform needs to significantly shorten the scheduling time in terms of how to improve the scheduling efficiency. On the other hand, the network freight platform needs to control costs by reducing the vehicle empty driving rate and transportation mileage, and effectively cutting transportation costs.

[0008] Therefore, a system that can achieve intelligent scheduling using big data analysis is needed to improve the operation efficiency and service quality of the network freight platform. Summary of the Invention

[0009] The purpose of the present invention is to provide an intelligent scheduling system and scheduling method for a network freight platform based on big data analysis. Through the analysis of a large amount of transportation-related data, intelligent and accurate vehicle scheduling is achieved, improving transportation efficiency and reducing costs.

[0010] To achieve the above technical objectives, the technical solution adopted by the present invention is as follows: This technical solution includes a data acquisition module for collecting vehicle information, cargo information, order information, historical transportation data, and real-time traffic data on the network freight platform; a data preprocessing module for cleaning, converting, and integrating the raw data collected by the data acquisition module; a big data analysis module for analyzing the data processed by the data preprocessing module using data mining, establishing a transportation demand prediction model, and analyzing road congestion conditions, vehicle driving states, and the matching degree between vehicles and goods; an intelligent scheduling module for generating a recommended scheduling plan using an intelligent algorithm based on the analysis results of the big data analysis module, combined with order information and vehicle resource conditions; and a user interaction module for providing an interaction interface for shippers, drivers, and platform managers, used for shippers to release transportation demands, query order status, drivers to receive scheduling tasks and feedback transportation situations, and platform managers to manage and monitor the system and view data reports.

[0011] Under the traditional scheduling method, manual scheduling cannot comprehensively consider complex factors such as the real-time position of vehicles, the characteristics of goods, order time requirements, and traffic conditions, resulting in low scheduling efficiency, increased costs, and difficult-to-guarantee service quality. For this reason, this technical solution aims to solve the problem of dynamic vehicle scheduling in the process of network freight platform scheduling by designing an intelligent scheduling module, considering multiple factors comprehensively.

[0012] The intelligent scheduling module uses intelligent algorithms to generate recommended scheduling plans, and monitors the vehicle transportation status in real time, dynamically adjusting the recommended scheduling plans. Among them, the intelligent scheduling module includes an initialization sub-module, a multi-objective evaluation sub-module, an intelligent algorithm optimization sub-module, a real-time monitoring sub-module, a dynamic adjustment sub-module, and a plan output sub-module.

[0013] The initialization sub-module is used to initially construct a vehicle-order allocation combination according to the basic information of vehicles and orders provided by the big data analysis module. The multi-objective evaluation sub-module is used to evaluate each vehicle-order allocation combination generated by the initialization sub-module according to the three optimization objectives of scheduling: transportation cost, delivery time, and vehicle utilization rate. The intelligent algorithm optimization sub-module uses the genetic algorithm as the intelligent algorithm to optimize the results output by the multi-objective evaluation sub-module. This sub-module encodes each vehicle-order allocation combination into a chromosome, and through selection, crossover, and mutation genetic operations, continuously iterates to generate new combinations and generate an initial scheduling plan. The real-time monitoring sub-module interacts with the data source of the on-vehicle device in real time to obtain the real-time location, driving speed, cargo status, transportation status information of the vehicle, and the real-time road traffic conditions, such as congestion conditions and temporary control information. The dynamic adjustment sub-module dynamically adjusts the generated initial scheduling plan according to the abnormal conditions feedback by the real-time monitoring sub-module. When a vehicle breaks down, this sub-module will re-evaluate the remaining vehicle resources and order requirements, select a new vehicle from the optimized set of vehicle-order allocation combinations to undertake the order task of the faulty vehicle, and re-plan the transportation route; if there is road congestion, it will adjust the driving route or delivery order of the affected vehicles according to the congestion degree and the expected clearance time, minimizing the impact on the overall transportation efficiency and delivery time. The plan output sub-module outputs the finally determined recommended scheduling plan and the dynamically adjusted plan to relevant users, such as drivers, shippers, and platform managers, in a visual form. The scheduling tasks of the vehicle are displayed through the user interaction interface.

[0014] The intelligent scheduling module effectively realizes the dynamic scheduling of vehicles through the coordinated operation of each sub-module, thereby improving the overall transportation efficiency. Among them, the initialization sub-module, the multi-objective evaluation sub-module, and the intelligent algorithm optimization sub-module cooperate and work together.

[0015] First, based on the basic information of vehicles and orders, a feasible allocation combination is quickly constructed to narrow the search scope. Secondly, key indicators such as transportation cost, delivery time, and vehicle utilization rate are quantified to provide a clear basis for subsequent optimization, solving the problem of unclear target measurement. Then, intelligent algorithms such as the genetic algorithm are used to iteratively optimize numerous combinations, seeking a balance between multiple objectives to form a recommended scheduling plan.

[0016] After breaking through the dilemma that traditional methods are difficult to balance multiple objectives, the intelligent scheduling module ensures the stability and timeliness of the transportation process through the real-time monitoring sub-module, dynamic adjustment sub-module, and solution output sub-module, solving the problem of scheduling failure caused by unexpected situations during the transportation process.

[0017] On the one hand, it real-time tracks the vehicle transportation status and traffic information, and according to the monitoring feedback, timely adjusts the scheduling plan for emergencies. On the other hand, it clearly presents the optimized scheduling plan to all parties of users through the solution output sub-module to achieve effective information transmission.

[0018] The intelligent scheduling module of the present invention has brought remarkable effects. In terms of improving the scheduling efficiency, it significantly shortens the scheduling time and realizes fast and accurate matching. In terms of cost control, it reduces the vehicle empty running rate and transportation mileage, effectively reducing the transportation cost. At the level of service quality, it ensures the on-time delivery of goods and improves customer satisfaction. In terms of management decision-making, it provides scientific data support for platform managers, helps optimize operation strategies, enhances the overall competitiveness of the platform, and promotes the intelligent and efficient development of the network freight platform.

[0019] Compared with the prior art, the present invention has the following four major advantages.

[0020] First, it has efficient and accurate scheduling, improving the overall efficiency. Through the data collection module, it comprehensively collects multi-source data such as vehicles, goods, orders, and traffic. After preprocessing and in-depth analysis, the intelligent scheduling module can quickly and accurately match the optimal vehicle for the order based on the big data analysis results and intelligent algorithms. For example, in the traditional scheduling method, it may take several hours to match a suitable vehicle for an order in a remote area, while this system can instantly generate a scheduling plan with the help of intelligent algorithms, greatly shortening the scheduling time and reducing the average vehicle response time, which greatly improves the overall transportation efficiency.

[0021] Second, it reduces costs and improves economic benefits. Precise scheduling effectively reduces the vehicle empty running rate. The system reasonably arranges vehicles according to the transportation demand prediction to avoid the empty running situation of vehicles without goods to transport, which can reduce the empty running rate. At the same time, it optimizes the transportation route, reduces unnecessary driving mileage and fuel consumption, combines vehicle load capacity for reasonable cargo distribution, makes full use of vehicle transport capacity, and reduces the unit transportation cost. Through comprehensive calculation, the overall transportation cost is reduced, saving a large amount of operating funds for the enterprise and improving economic benefits.

[0022] Third, it ensures on-time delivery and improves service quality. It real-time monitors the vehicle transportation status and road conditions, and dynamically adjusts the scheduling plan according to the delivery time requirements. In case of sudden road congestion, the system timely plans a new route for the vehicle to ensure the goods are delivered on time. After actual verification, after adopting this system, the on-time delivery rate of orders has increased from 70%-80% in the traditional method to over 90%, greatly improving customer satisfaction, enhancing the competitiveness of the network freight platform in the market, and establishing a good brand image.

[0023] Fourth, data-driven decision-making enhances management science. The big data analysis module provides rich decision-making basis for platform management. Through detailed data reports, managers can deeply understand the situation of each link of the transportation business, such as the changes in transportation demand in different regions, vehicle utilization efficiency, etc. Based on these data, operational strategies can be scientifically adjusted, such as reasonably increasing or decreasing vehicles, optimizing the layout of service areas, making the platform operation more in line with market demand, and achieving sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The present invention can be further illustrated by the non-limiting embodiments given in the drawings; Figure 1 It is a schematic diagram of the scheduling system module of the present invention; Figure 2 It is a schematic diagram of the intelligent scheduling module of the present invention; Figure 3 It is a partial process schematic diagram of the intelligent algorithm optimization sub-module of the present invention; Figure 4 It is a flowchart of the scheduling method of the intelligent scheduling system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0025] In order to enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be further described below with reference to the drawings and embodiments.

[0026] As Figure 1 , Figure 2 , Figure 3 , Figure 4 As shown, the intelligent scheduling system of the network freight platform based on big data analysis includes a data collection module for collecting vehicle information, cargo information, order information, historical transportation data, and real-time traffic data on the network freight platform.

[0027] Among them, the data collection module is used to collect various data on the network freight platform. This module conducts data interaction with in-vehicle terminals, platform databases, and traffic department data interfaces to ensure the timeliness and accuracy of the data.

[0028] A data preprocessing module for cleaning, transforming, and integrating the raw data collected by the data collection module.

[0029] Among them, the data preprocessing module is used to remove duplicate, incorrect, or incomplete data, uniformly convert data in different formats into a standard format recognizable by the system, and integrate data from multiple data sources for subsequent analysis.

[0030] The big data analysis module uses data mining to analyze the data processed by the data preprocessing module, establish a transportation demand prediction model, and analyze road congestion conditions, vehicle driving status, and the matching degree between vehicles and goods.

[0031] Among them, the big data analysis module provides a reference for vehicle scheduling by establishing a transportation demand prediction model; at the same time, it analyzes the matching degree between vehicles and goods to determine the best vehicle and goods combination.

[0032] The intelligent scheduling module generates a recommended scheduling plan using intelligent algorithms according to the analysis results of the big data analysis module, combined with order information and vehicle resource conditions.

[0033] Among them, the intelligent scheduling module comprehensively considers multiple factors such as transportation costs, delivery times, and vehicle utilization rates to achieve precise matching and reasonable scheduling of vehicles and goods. In particular, in the present invention, the recommended scheduling plan is generated by the intelligent scheduling module encoding vehicles and orders and optimizing through selection, crossover, and mutation genetic operations. Moreover, the present invention uses the intelligent scheduling module to monitor the vehicle transportation status in real time and dynamically adjust the recommended scheduling plan.

[0034] The user interaction module provides an interaction interface for shippers, drivers, and platform administrators, which is used for shippers to release transportation demands, query order status, drivers to receive scheduling tasks and feedback transportation situations, and platform administrators to manage and monitor the system and view data reports.

[0035] Among them, through this user interaction module, shippers, drivers, and platform administrators can timely track the implementation of the plan to ensure that the goods are delivered to the destination on time and safely.

[0036] In the intelligent scheduling module, the recommended scheduling plan uses the linear weighted method to integrate the following three objectives into a comprehensive objective function to generate the recommended scheduling plan. The three objectives are the transportation cost objective, the delivery punctuality objective, and the vehicle utilization rate objective; among them, the objective of minimizing transportation costs, the objective of maximizing delivery punctuality, and the objective of maximizing vehicle utilization rate.

[0037] Let be a binary decision variable. If vehicle is assigned to order , then , otherwise , where , is the total number of vehicles, , is the total number of orders; Transportation cost objective is: ; Among them, For a vehicle to execute an order the transportation cost; it should be noted that the transportation cost is related to the vehicle driving distance, fuel consumption, and driver fees.

[0038] The on-time delivery target is: ; where, let be the time when the vehicle executes the order is expected to arrive at the delivery location, be the delivery deadline of the order . Define as the on-time penalty factor for the vehicle to execute the order . If , ; if , ; The vehicle utilization target is: ; where, is the rated load capacity of the vehicle , is the weight of the goods of the order ; the utilization rate of the vehicle is ; The comprehensive objective function is: ; where, for the three objectives, a linear weighted method is used to construct the comprehensive objective function , , , are the weight coefficients, and ， ; the values of the weight coefficients are adjusted according to the actual business needs and priorities. For example, if more emphasis is placed on transportation cost control, has a relatively large value.

[0039] The intelligent scheduling module uses intelligent algorithms to generate recommended scheduling plans and monitors the vehicle transportation status in real time, dynamically adjusting the recommended scheduling plans. Among them, the intelligent scheduling module includes an initialization sub-module, a multi-objective evaluation sub-module, an intelligent algorithm optimization sub-module, a real-time monitoring sub-module, a dynamic adjustment sub-module, and a plan output sub-module.

[0040] It should be noted that when the intelligent algorithm optimization sub-module uses the genetic algorithm as the intelligent algorithm and wants to meet the conditions to achieve the balance and optimization among the three goals, it is necessary to control the reasonable setting of the weights.

[0041] When the weight coefficients 、 、 are set to meet the actual needs and focuses of the current business, the genetic algorithm can be optimized in the direction of meeting the business expectations during the iteration process, so that the comprehensive objective function reaches the optimum, thus achieving the balance among the three goals. For example, in a freight scenario that is more sensitive to costs, if 、 、 are reasonably set, the genetic algorithm will find a better balance solution that takes into account the on-time delivery and vehicle utilization rate while controlling costs.

[0042] When the weight setting is reasonable, when the genetic algorithm converges after enough iteration times, that is, the individuals in the population no longer change significantly, the optimal solution obtained at this time represents the balance and optimization among the three goals under the current conditions.

[0043] Convergence means that the algorithm has found a relatively stable and optimal region in the search space, and the corresponding recommended scheduling scheme reaches the best state when considering the three goals comprehensively.

[0044] The initialization sub-module is used to initially construct a vehicle-order allocation combination according to the basic information of vehicles and orders provided by the big data analysis module. Among them, the basic information includes the current location of the vehicle, the loading capacity, the type and quantity of order goods, the place of dispatch and the place of receipt. It should be noted that this sub-module will filter out obviously mismatched combinations, such as the situation where the vehicle load cannot meet the weight requirement of the order goods. It provides a feasible starting point for the subsequent optimization process. By setting some basic rules and screening conditions, it quickly narrows the search space and improves the efficiency of generating the scheduling scheme.

[0045] The multi-objective evaluation sub-module is used to evaluate each vehicle-order allocation combination generated by the initialization sub-module according to the three optimization goals of scheduling: transportation cost, delivery time, and vehicle utilization rate.

[0046] Among them, for the transportation cost, the cost of each combination is calculated by combining data such as vehicle driving mileage, fuel consumption, and driver salary; for the delivery time, based on real-time traffic data, vehicle driving speed, and the delivery deadline required by the order, it is evaluated whether each combination can deliver on time and the possible early or late time; regarding vehicle utilization rate, it is measured by calculating the ratio of the total volume or weight of the goods assigned to each vehicle to the rated load capacity of the vehicle. This sub-module quantifies the three objectives into specific numerical indicators, providing a clear evaluation basis for the subsequent scheme optimization.

[0047] The intelligent algorithm optimization sub-module uses the genetic algorithm as the intelligent algorithm to optimize the results output by the multi-objective evaluation sub-module. This sub-module encodes each vehicle-order allocation combination as a chromosome, and through selection, crossover, and mutation genetic operations, continuously iterates to generate new combinations and generate an initial scheduling scheme.

[0048] Among them, in each iteration, according to the evaluation indicators given by the multi-objective evaluation sub-module, combinations with higher fitness, considering each objective comprehensively and closer to the optimal solution, are selected as parents for crossover and mutation to generate a new generation of combinations. Through repeated iterations, the scheduling scheme is continuously optimized to achieve a balance and optimality among multiple objectives.

[0049] The real-time monitoring sub-module interacts with the data source of in-vehicle devices in real time to obtain the real-time position, driving speed, cargo status, transportation status information of the vehicle, and the real-time traffic conditions of the road, such as congestion conditions and temporary control information.

[0050] Among them, the real-time monitoring sub-module can continuously track the actual progress of each vehicle in executing the scheduling scheme, compare the real-time data with the expected data in the scheduling scheme, and timely discover abnormal situations that may cause the scheduling scheme to need adjustment, such as sudden vehicle failures and temporary road closures. It provides real-time and accurate data support for the dynamic adjustment sub-module to ensure that the scheduling scheme can respond in a timely manner according to the actual transportation situation.

[0051] The dynamic adjustment sub-module dynamically adjusts the generated initial scheduling scheme according to the abnormal situations feedback by the real-time monitoring sub-module. When a vehicle failure occurs, this sub-module will re-evaluate the remaining vehicle resources and order requirements, select a new vehicle from the optimized set of vehicle-order allocation combinations to undertake the order tasks of the failed vehicle, and re-plan the transportation route; if there is road congestion, it will adjust the driving route or delivery sequence of the affected vehicles according to the congestion degree and the expected clearance time, minimizing the impact on the overall transportation efficiency and delivery time.

[0052] Among them, during the adjustment process, the evaluation criteria set by the multi-objective evaluation sub-module and the optimization logic of the intelligent algorithm optimization sub-module will still be followed to ensure that the adjusted plan still maintains a good balance among multiple objectives and meets the actual transportation requirements.

[0053] The solution output sub-module outputs the finally determined recommended scheduling plan and the dynamically adjusted plan to relevant users in a visual form, such as drivers, shippers, and platform managers. It displays the vehicle scheduling tasks through the user interaction interface, including information such as loading location, unloading location, estimated driving route, and estimated arrival time; provides the overall transportation arrangement and estimated status of the order for the shipper; and provides a comprehensive summary of the scheduling plan to the platform managers for management and decision-making. At the same time, this sub-module is also responsible for recording the data related to the scheduling plan into the system database to provide data support for subsequent data analysis and performance evaluation.

[0054] The intelligent algorithm optimization sub-module will collect a large amount of historical transportation data through the historical transportation data obtained by the data collection module and analyze the distribution of data errors. Among them, for the vehicle driving speed data, the error range from the actual speed is statistically calculated, and a fault tolerance range is set; When the vehicle speed data error is within the set fault tolerance range of ±5%, it is considered that the vehicle-order allocation combination code conforms to the scheduling plan; When the vehicle speed data error is not within the set fault tolerance range of ±5%, it is considered that the vehicle-order allocation combination code does not conform to the scheduling plan, and a new combination is iteratively generated.

[0055] The data collection module collects data through data interaction with in-vehicle terminals, platform databases, and traffic department data interfaces.

[0056] The data preprocessing module removes duplicate, incorrect, or incomplete data from the original data and integrates the data in different formats after converting them into a standard format.

[0057] The big data analysis module uses historical transportation data to train a transportation demand prediction model, analyzes the road congestion level and the estimated arrival time of vehicles by combining real-time traffic data and vehicle location information, and calculates the matching degree between vehicles and goods according to the attribute information of vehicles and goods.

[0058] The present invention also discloses a scheduling method for an intelligent scheduling system of an online freight transportation platform based on big data analysis, which specifically includes the following steps: S101 Data collection step, using the data collection module to collect vehicle information, cargo information, order information, historical transportation data, and real-time traffic data on the online freight transportation platform; S102 Data preprocessing step: Transmit the collected raw data to the data preprocessing module, and perform cleaning, transformation, and integration processing on the data to make it into standard format data to be analyzed. S103 Big data analysis step: The big data analysis module deeply analyzes the preprocessed data, trains a transportation demand prediction model using historical transportation data to predict the transportation demands in different regions within a certain period in the future; combines real-time traffic data and vehicle location information to analyze the road congestion level and the expected arrival time of vehicles; calculates the matching degree between vehicles and goods according to the attribute information of vehicles and goods; where the specific period in the future for prediction is 1 to 2 hours.

[0059] S104 Initial scheduling plan generation step: The intelligent scheduling module generates an initial scheduling plan according to the analysis results of the big data analysis module using intelligent algorithms. Among them, the initial scheduling plan aims to minimize transportation costs, meet delivery time requirements, and maximize vehicle utilization rate, etc. During the process of generating the plan, factors such as the current location of the vehicle, loading capacity, weight and volume of the goods, and the delivery time of the order are fully considered to achieve the best matching between vehicles and goods. S105 Scheduling plan optimization and execution step: Real-time monitor the transportation status of vehicles and road condition changes, send the finally determined scheduling plan to drivers and shippers through the user interaction module, and track the execution of the plan to ensure that the goods are delivered to the destination on time.

[0060] Among them, if an unexpected situation occurs, the intelligent scheduling module timely optimizes and adjusts the scheduling plan. Among them, in the S101 data collection step, the data collection module collects data in real time according to the set time interval or event trigger mechanism.

[0061] Among them, in the S103 big data analysis step, the prediction model is trained using time series analysis algorithms, and the road congestion status is judged through the road congestion index.

[0062] The scheduling method of the intelligent scheduling system of the online freight platform based on big data analysis provided by the present invention has the following advantages: First, it improves the scheduling efficiency. Through big data analysis and intelligent algorithms, the most suitable vehicle is matched for each order, reducing the scheduling time and improving the overall transportation efficiency. Second, it reduces transportation costs. The accurate scheduling plan can effectively reduce the vehicle empty driving rate, reasonably plan the transportation route, reduce fuel consumption and transportation mileage, thereby reducing transportation costs. Third, it improves service quality. Dynamically adjusting the scheduling plan according to the delivery time requirements and real-time road conditions can ensure that the goods are delivered on time. Fourth, it enhances the scientific nature of decision-making. Big data analysis provides rich decision-making basis, helping platform managers better understand the transportation business situation, make scientific and reasonable decisions, and optimize the platform operation strategy.

[0063] The above has introduced in detail the intelligent scheduling system and scheduling method of the network freight platform based on big data analysis provided by the present invention. The description of specific embodiments is only used to help understand the method and its core idea of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

Claims

1. The intelligent dispatching system of the network freight platform based on big data analysis is characterized by: include: Data collection module, used to collect vehicle information, cargo information, order information, historical transportation data and real-time traffic data on the online freight platform; A data preprocessing module cleans, converts and integrates the raw data collected by the data collection module; A big data analysis module uses data mining to analyze the data processed by the data preprocessing module, establish a transportation demand prediction model, and analyze road congestion, vehicle driving status, and the matching degree of vehicles and goods; An intelligent scheduling module, which generates a recommended scheduling plan using an intelligent algorithm based on the analysis results of the big data analysis module, combined with order information and vehicle resource conditions; The user interaction module provides an interactive interface for shippers, drivers and platform managers, allowing shippers to publish transportation requirements and query order status, drivers to receive scheduling tasks and provide feedback on transportation status, and platform managers to manage and monitor the system and view data reports.

2. The intelligent dispatching system for online freight platform based on big data analysis according to claim 1 is characterized in that: In the intelligent scheduling module, the recommended scheduling scheme uses a linear weighted method to integrate the following three objectives into a comprehensive objective function to generate a recommended scheduling scheme, wherein the three objectives are transportation cost objective, delivery punctuality objective, and vehicle utilization objective; set up is a binary decision variable. If the vehicle Assigned to order ,but ,otherwise ,in , is the total number of vehicles, , is the total number of orders; Shipping cost target for: ; in, For vehicles Execution of Orders transportation costs; On-time delivery target for: ; Among them, For vehicles Execution of Orders Estimated time of arrival at the delivery location, For orders Delivery deadline; definition For vehicles Execution of Orders The punctuality penalty factor is , ;like , ; Vehicle utilization target for: ; in, For vehicles Rated cargo capacity, For orders The weight of the cargo; vehicle Utilization rate ; Comprehensive objective function for: ; Among them, for the three objectives, the linear weighted method is used to construct the comprehensive objective function , , , is the weight coefficient, and ， 。 3. The intelligent dispatching system for network freight platform based on big data analysis according to claim 2 is characterized in that: The intelligent scheduling module uses an intelligent algorithm to generate a recommended scheduling plan, monitors the vehicle transportation status in real time, and dynamically adjusts the recommended scheduling plan, wherein the intelligent scheduling module includes an initialization submodule, a multi-objective evaluation submodule, an intelligent algorithm optimization submodule, a real-time monitoring submodule, a dynamic adjustment submodule, and a plan output submodule; The initialization submodule is used to preliminarily construct a vehicle-order allocation combination based on the basic information of vehicles and orders provided by the big data analysis module; The multi-objective evaluation submodule is used to evaluate each vehicle-order allocation combination generated by the initialization submodule according to the three optimization objectives of the scheduling, namely, transportation cost, delivery time, and vehicle utilization; The intelligent algorithm optimization submodule uses a genetic algorithm as an intelligent algorithm to optimize the results output by the multi-objective evaluation submodule, encodes each vehicle-order allocation combination as a chromosome, and continuously iterates to generate new combinations through selection, crossover and mutation genetic operations to generate an initial scheduling plan; The real-time monitoring submodule interacts with the data source of the vehicle-mounted device in real time to obtain the vehicle's real-time location, driving speed, cargo status and transportation status information, as well as the real-time traffic conditions of the road; The dynamic adjustment submodule dynamically adjusts the generated initial dispatch plan according to the abnormal situation fed back by the real-time monitoring submodule; when a vehicle failure occurs, it will re-evaluate the remaining vehicle resources and order requirements, select a new vehicle from the optimized vehicle-order allocation combination set to undertake the order task of the faulty vehicle, and re-plan the transportation route; if there is road congestion, it will adjust the driving route or delivery order of the affected vehicles according to the degree of congestion and the estimated clearing time; The solution output submodule outputs the final recommended scheduling solution and the dynamically adjusted solution to relevant users, such as drivers, cargo owners and platform managers, in a visual form, and displays the vehicle scheduling tasks through a user interaction interface.

4. The intelligent dispatching system for network freight platform based on big data analysis according to claim 3 is characterized in that: The intelligent algorithm optimization submodule collects a large amount of historical transportation data through the historical transportation data obtained by the data acquisition module, analyzes the distribution of data errors, and, for vehicle speed data, calculates the error range between the vehicle speed data and the actual speed, and sets the fault tolerance range. When the vehicle speed data error is within the set tolerance range of ±5%, the vehicle-order allocation combination code is considered to be in compliance with the scheduling plan; When the vehicle speed data error is not within the set tolerance range of ±5%, it is considered that the vehicle-order allocation combination coding does not conform to the scheduling plan, and a new combination is generated iteratively.

5. The intelligent dispatching system for network freight platform based on big data analysis according to claim 4 is characterized in that: The data acquisition module collects data by interacting with the vehicle terminal, the platform database and the traffic department data interface.

6. The intelligent dispatching system for network freight platform based on big data analysis according to claim 5 is characterized in that: The data preprocessing module removes duplicate, erroneous or incomplete data from the original data, and converts data in different formats into a standard format for integration.

7. The intelligent dispatching system for network freight platform based on big data analysis according to claim 6 is characterized in that: The big data analysis module uses historical transportation data to train a transportation demand prediction model, combines real-time traffic data and vehicle location information to analyze road congestion and vehicle estimated arrival time, and calculates the matching degree of vehicles and goods based on the attribute information of vehicles and goods.

8. The dispatching method of the intelligent dispatching system of the network freight platform based on big data analysis according to claim 7 is characterized in that: The specific steps include: S101 data collection step, using a data collection module to collect vehicle information, cargo information, order information, historical transportation data and real-time traffic data on the online freight platform; S102 data preprocessing step, transmitting the collected raw data to the data preprocessing module, cleaning, converting and integrating the data to make it into standard format data to be analyzed; S103 big data analysis step: the big data analysis module conducts in-depth analysis on the preprocessed data, uses historical transportation data to train the transportation demand prediction model, and predicts the transportation demand of different regions in the next 1 to 2 hours; combines real-time traffic data and vehicle location information to analyze the road congestion level and the estimated arrival time of the vehicle; and calculates the matching degree of the vehicle and the cargo based on the attribute information of the vehicle and the cargo; S104, a scheduling plan generation step, in which the intelligent scheduling module generates an initial scheduling plan using an intelligent algorithm according to the analysis results of the big data analysis module; S105 dispatch plan optimization and execution step, real-time monitoring of the vehicle's transportation status and road condition changes, sending the finalized dispatch plan to the driver and cargo owner through the user interaction module, and tracking the execution of the plan to ensure that the goods are delivered to the destination on time.

9. The dispatching method of the intelligent dispatching system of the network freight platform based on big data analysis according to claim 8 is characterized in that: in, In the data collection step S101, the data collection module collects data in real time according to a set time interval or event triggering mechanism.

10. The dispatching method of the intelligent dispatching system of the network freight platform based on big data analysis according to claim 9 is characterized in that: in, In the big data analysis step S103, the prediction model is trained using a time series analysis algorithm, and the road congestion status is determined by the road congestion index.

Images