Freight logistics distribution adjustment method and system

Abstract

The invention is suitable for the field of logistics supply chains, and provides a freight logistics distribution adjustment method and system, and the method comprises the steps: collecting the data of freight vehicles, freight sources and road conditions in real time based on Internet of Things equipment, a GPS positioning system and a traffic API interface, and building a three-dimensional freight chain database; based on the three-dimensional cargo chain database, gradient price incentive information and carpooling optimization information are generated and sent; based on the three-dimensional cargo chain database, trunk and branch line cooperative scheduling information and drop and pull management information are generated and sent; road condition information is monitored in real time, delay duration is predicted, and a delay notice is synchronously sent; according to the method and the system, the limitation of single link optimization of traditional logistics is broken through, a closed loop of'data acquisition-intelligent decision-execution feedback 'is utilized, the return no-load rate of the truck is reduced and the load rate of the truck is improved through accurate matching of intelligent spelling and return goods sources, a'data-driven' intelligent scheduling normal form is provided for the logistics industry, and the logistics efficiency is improved. And traditional freight transport is promoted to be transformed to precision and intelligence.

Description

Technical Field

[0001] The present invention belongs to the field of logistics supply chain, and in particular relates to a freight logistics distribution and adjustment method and system. Background Art

[0002] In today's freight logistics industry, low truck load rates and high rates of empty return trips are persistent and pressing issues. With rapid economic development and growing logistics demand, these issues not only waste resources but also increase logistics costs, negatively impacting the efficiency and sustainability of the entire industry.

[0003] On the one hand, low truck load factors mean that trucks aren't fully utilizing their cargo capacity during transportation, resulting in idle transportation resources. On the other hand, a high rate of empty return trips is another prominent issue in the freight logistics industry. After completing a transport mission, trucks often return empty to their point of departure. This process not only generates no economic benefits but also consumes resources such as fuel and manpower.

[0004] The existence of these problems not only affects the profitability of freight logistics companies, but also puts a certain amount of pressure on the environment. In order to increase the load factor of truck vehicles, reduce the empty return rate, reduce resource waste, and improve logistics efficiency, it is urgent to develop an innovative freight logistics allocation and adjustment method and system to achieve the optimal allocation and efficient utilization of logistics resources. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for adjusting freight logistics distribution, aiming to solve the problems raised in the above background technology.

[0006] The present invention is implemented as follows: on the one hand, a method for adjusting freight logistics distribution, the method comprising:

[0007] Based on IoT devices, GPS positioning systems, and traffic API interfaces, real-time data on freight vehicles, cargo sources, and road conditions is collected to establish a three-dimensional cargo chain database;

[0008] Generate and send gradient price incentive information and load-sharing optimization information based on the three-dimensional cargo chain database;

[0009] Generate and send trunk and branch line coordinated dispatch information and drop-and-hook management information based on the 3D cargo chain database;

[0010] Monitor traffic conditions in real time, predict delay duration, and send delay notifications simultaneously.

[0011] As a further solution of the present invention, the three-dimensional cargo chain database specifically includes:

[0012] Vehicle status information: vehicle mission, vehicle location, vehicle load, mileage, and return empty rate;

[0013] Cargo source information: cargo weight, cargo volume, cargo delivery location, cargo delivery time, cargo type, timeliness requirements, cargo owner ID, hub node location, terminal delivery requirements, cargo hard time window, cargo soft time window;

[0014] Trailer information: preset hub geofence location, trailer ID, trailer status, trailer pool status;

[0015] Traffic condition information: real-time traffic flow, congested roads, and weather warnings.

[0016] As a further solution of the present invention, the generation and transmission of gradient price incentive information and load-sharing optimization information based on the three-dimensional cargo chain database specifically includes:

[0017] Extract the real-time return empty rate information R from the three-dimensional cargo chain database;

[0018] Based on the preset time sensitivity T and cargo priority P, calculate and send the real-time freight discount rate F including the real-time return empty load rate information;

[0019] The time sensitivity T specifically includes:

[0020] Night transport period: T=1.2;

[0021] Legal holidays and the day before or after: T=1.1;

[0022] Normal period: T=1;

[0023] The cargo priority P specifically includes:

[0024] Fresh food and dangerous goods: P = 1.5;

[0025] Goods with a delivery time of less than three days: P = 1;

[0026] Lightweight cargo with a volume-to-weight ratio greater than the threshold: P = 0.8;

[0027] The calculation process of the real-time freight discount rate F including the real-time return empty load rate information is as follows:

[0028] F=0.6R+0.3T+0.1P.

[0029] As a further solution of the present invention, the generating and sending of gradient price incentive information and load-sharing optimization information based on the three-dimensional cargo chain database specifically further includes:

[0030] Extract cargo weight, cargo volume, cargo delivery location, and cargo delivery time information from the three-dimensional cargo chain database;

[0031] Based on the DBSCAN algorithm, the spatiotemporal distance measurement formula is defined as follows:

[0032]

[0033] Where, d ij is the comprehensive time-space distance between the i-th source and the j-th source, x i 、y i is the horizontal and vertical coordinates of the shipping place (or receiving place) of the i-th source of goods, x j 、y j is the horizontal and vertical coordinates of the place of shipment (or place of receipt) of the jth source of goods, t i 、j i is the delivery time between the i-th source and the j-th source, α, β, and λ are the corresponding weight coefficients, which are determined by historical training data;

[0034] Set the neighborhood radius κ and the minimum number of points ε, cluster the sources whose distance is less than the neighborhood radius κ and greater than the minimum number of points ε into one category, and generate several adjacent spatiotemporal source clusters D k ;

[0035] Based on several nearby spatiotemporal supply clusters D k , calculate and generate the comprehensive vehicle loading rate L;

[0036] The calculation process of the vehicle comprehensive loading rate L is:

[0037] And i∈D k w i ≤w c , i∈D k V i ≤V c ;

[0038] Where w i is the nearby space-time supply cluster D k The weight of the i-th source, v i is the nearby space-time supply cluster D k The volume of the i-th source of goods, μ is the weight loading coefficient, λ is the volume loading coefficient, w c is the vehicle load, V c is the vehicle volume;

[0039] Match the comprehensive vehicle loading rate L to the fitness f(L) = L;

[0040] Each adjacent spatiotemporal supply cluster D k The supply combination is coded as a chromosome;

[0041] Based on the two-point crossover method, the gene fragments at the chromosome crossover point were exchanged, the crossover probability P1 was set to 0.8, the chromosome values were randomly changed, and the mutation probability P2 was set to 0.05;

[0042] After multiple generations of iteration, the loading combination S with the maximum loading rate is screened and sent.

[0043] As a further solution of the present invention, the generation and transmission of trunk and branch line coordinated dispatching information and drop-and-hook management information based on the three-dimensional cargo chain database specifically includes:

[0044] Extract the real-time location of vehicles, hub node locations, terminal delivery requirements, cargo hard time windows, and cargo soft time windows from the three-dimensional cargo chain database;

[0045] Based on GIS and logistics hub distribution, a trunk and branch line network is generated and a visual topology model is constructed that includes node processing capabilities and section constraints.

[0046] Analyze historical data based on the LSTM algorithm to predict and generate cargo demand for each time period and route, as well as the terminal order concentration area;

[0047] Generate and send trunk and branch line vehicle matching and route optimization solutions based on genetic algorithms.

[0048] As a further solution of the present invention, the generating and sending of trunk and branch line coordinated dispatching information and drop and hook management information based on the three-dimensional cargo chain database specifically further includes:

[0049] Extract vehicle real-time tasks, hub geofence locations, trailer IDs, trailer status, and trailer pool status information from the 3D cargo chain database;

[0050] Based on GPS positioning, when the trailer and tractor enter the preset hub geo-fence, it generates a drop-and-hook preparation instruction and predicts vehicle demand;

[0051] Calculate the compatibility of trailers and tractors based on the Hungarian algorithm;

[0052] Generate and send a drop-and-hook plan based on the compatibility of the trailer and tractor.

[0053] As a further embodiment of the present invention, in another aspect, a freight logistics distribution and adjustment system is provided, the system comprising:

[0054] The collection module is used to collect real-time data on freight vehicles, cargo sources, and road conditions based on IoT devices, GPS positioning systems, and traffic API interfaces to establish a three-dimensional cargo chain database;

[0055] The first generation and sending module is used to generate and send gradient price incentive information and load-sharing optimization information based on the three-dimensional cargo chain database;

[0056] The second generation and sending module is used to generate and send trunk and branch line coordinated scheduling information and drop and hook management information based on the three-dimensional cargo chain database;

[0057] The synchronous sending module is used to monitor traffic conditions in real time, predict delay duration, and send delay notifications synchronously.

[0058] As a further solution of the present invention, the first generating and sending module specifically includes:

[0059] The first extraction unit is used to extract the real-time return empty rate information R from the three-dimensional cargo chain database;

[0060] The calculation and sending unit is used to calculate and send a real-time freight discount rate F including real-time return empty load rate information based on a preset time sensitivity T and cargo priority P.

[0061] As a further solution of the present invention, the first generating and sending module specifically further includes:

[0062] The second extraction unit is used to extract the cargo weight, cargo volume, cargo delivery location, and cargo delivery time information from the three-dimensional cargo chain database;

[0063] Definition unit, used to define the spatiotemporal distance measurement formula based on the DBSCAN algorithm;

[0064] Setting unit, used to set the neighborhood radius κ and the minimum number of points ε;

[0065] The generation unit is used to group the sources of goods whose distance is less than the neighborhood radius κ and greater than the minimum number of points ε into one category, and generate several adjacent spatiotemporal source clusters D k ;

[0066] A matching unit for matching the vehicle comprehensive loading rate L to a fitness f(L)=L;

[0067] The combined coding unit is used to combine each adjacent spatiotemporal supply cluster D k The supply combination is coded as a chromosome;

[0068] Crossover unit, used to exchange gene segments at chromosome crossover points based on the two-point crossover method;

[0069] Mutation unit, used to randomly change the chromosome value.

[0070] As a further solution of the present invention, the second generating and sending module specifically includes:

[0071] The third extraction unit is used to extract the real-time location of vehicles, hub node locations, terminal delivery requirements, cargo hard time windows, and cargo soft time windows from the three-dimensional cargo chain database;

[0072] The generation unit is used to generate the trunk and branch line network based on GIS and logistics hub distribution and build a visual topology model that includes node processing capabilities and road section constraints;

[0073] The prediction generation unit is used to analyze historical data based on the LSTM algorithm to predict and generate cargo demand for each time period and route, as well as the terminal order concentration area;

[0074] The generating and sending unit is used to generate and send trunk and branch line vehicle matching and route optimization solutions based on genetic algorithms.

[0075] The present invention provides a freight logistics distribution and adjustment method and system. This method and system break through the optimization limitations of a single link in traditional logistics, utilize the "data collection-intelligent decision-making-execution feedback" closed loop, and accurately match intelligent loading and return cargo sources to reduce the empty return rate of trucks and increase the truck load rate. This provides a "data-driven" intelligent scheduling paradigm for the logistics industry and promotes the transformation of traditional freight to precision and intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 It is a main flow chart of a freight logistics distribution and adjustment method.

[0077] Figure 2 The present invention is a flowchart of a first embodiment of a freight logistics distribution and adjustment method for generating and sending gradient price incentive information and load optimization information based on a three-dimensional freight chain database.

[0078] Figure 3 The present invention is a flowchart of the second embodiment of a freight logistics distribution and adjustment method for generating and sending gradient price incentive information and load optimization information based on a three-dimensional freight chain database.

[0079] Figure 4 The present invention is a flowchart of the first embodiment of a freight logistics distribution and adjustment method for generating and sending trunk and branch line coordinated scheduling information and drop-and-hook management information based on a three-dimensional freight chain database.

[0080] Figure 5 The present invention is a flow chart of the second embodiment of a freight logistics distribution and adjustment method for generating and sending trunk and branch line coordinated scheduling information and drop-and-hook management information based on a three-dimensional freight chain database.

[0081] Figure 6 It is a main structural diagram of a freight logistics distribution and regulation system.

[0082] Figure 7 The present invention is a structural block diagram of a first embodiment of a first generation and sending module in a freight logistics distribution and regulation system.

[0083] Figure 8The present invention is a structural block diagram of the second embodiment of the first generation and sending module in the freight logistics distribution and regulation system.

[0084] Figure 9 The present invention is a structural block diagram of the first embodiment of the second generation and sending module in the freight logistics distribution and regulation system. DETAILED DESCRIPTION

[0085] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0086] The specific implementation of the present invention is described in detail below with reference to specific embodiments.

[0087] The present invention provides a freight logistics distribution and adjustment method and system, which solve the technical problems in the background technology.

[0088] like Figure 1 FIG. 1 is a main flow chart of a freight logistics distribution and adjustment method provided by one embodiment of the present invention, wherein the freight logistics distribution and adjustment method includes:

[0089] Step S100: Based on IoT devices, GPS positioning systems, and traffic API interfaces, real-time data on freight vehicles, cargo sources, and road conditions is collected to establish a three-dimensional cargo chain database;

[0090] Step S200: Generate and send gradient price incentive information and load optimization information based on the three-dimensional cargo chain database;

[0091] Step S300: Generate and send trunk and branch line coordinated dispatch information and drop-and-hook management information based on the three-dimensional cargo chain database;

[0092] Step S400: Real-time monitoring of road conditions, prediction of delay duration, and simultaneous sending of delay notifications. During the cargo transportation process, the actual arrival time of the cargo is continuously monitored and road conditions are monitored in real time to predict the delay duration, and delay notifications are simultaneously sent to the cargo owner and the vehicle owner.

[0093] When this embodiment is applied, the present invention provides a data-driven freight logistics distribution and adjustment method, which collects vehicle status, cargo information, and road condition data in real time through Internet of Things devices, GPS positioning systems, and traffic API interfaces, and builds a distributed cargo chain database to achieve full-chain data fusion and real-time synchronization; designs a gradient price incentive model and an intelligent loading algorithm based on the database to enhance the attractiveness of return cargo sources and integrate the needs of scattered cargo owners; through the coordinated scheduling of trunk and branch lines combined with the drop-and-hook transportation mode, uses the Dijkstra algorithm and time window constraints to generate a time-space matching path, cooperates with the hub trailer pool to achieve rapid trailer switching and shorten the loading and unloading waiting time; at the same time, uses GPS and traffic APIs to collect vehicle location and road condition data in real time, dynamically predicts the delay threshold based on historical data and real-time congestion coefficient, continuously compares the actual driving time with the planned time, and when the delay duration exceeds the threshold, synchronously sends a delay notification to the vehicle owner and the cargo owner, so that the cargo owner can adjust the time window in time and reduce losses.

[0094] As a preferred embodiment of the present invention, the three-dimensional cargo chain database specifically includes:

[0095] Vehicle status information: vehicle mission, vehicle location, vehicle load, mileage, and return empty rate;

[0096] Cargo source information: cargo weight, cargo volume, cargo delivery location, cargo delivery time, cargo type, timeliness requirements, cargo owner ID, hub node location, terminal delivery requirements, cargo hard time window, cargo soft time window;

[0097] Trailer information: preset hub geofence location, trailer ID, trailer status, trailer pool status;

[0098] Traffic condition information: real-time traffic flow, congested roads, and weather warnings.

[0099] When this embodiment is applied, through the IoT device cluster, the on-board OBD sensor collects vehicle load, fuel consumption, engine status in real time, and the RFID reader automatically identifies the trailer ID and load history; both the cargo owner and the vehicle owner are installed with an APP, and the APP enters the cargo attributes and timeliness requirements. The high-precision positioning system GPS / Beidou dual-mode positioning has an accuracy of ≤10 meters and a sampling frequency of not less than 1 time / minute, as well as a third-party API interface, which can synchronize traffic department congestion data, weather warning information, and third-party logistics platform cargo data in real time, with an update frequency of not less than 2 minutes / time, thereby building a three-dimensional cargo chain database that includes vehicle status, cargo attributes, and road conditions.

[0100] like Figure 2 As shown, as a preferred embodiment of the present invention, the generation and transmission of gradient price incentive information and load-sharing optimization information based on the three-dimensional cargo chain database specifically includes:

[0101] Step S201: extracting real-time return empty rate information R from the three-dimensional cargo chain database;

[0102] Step S202: Calculate and send a real-time freight discount rate F including real-time return empty load rate information based on the preset time sensitivity T and cargo priority P;

[0103] The time sensitivity T specifically includes:

[0104] Night transport period: T=1.2;

[0105] Legal holidays and the day before or after: T=1.1;

[0106] Normal period: T=1;

[0107] The cargo priority P specifically includes:

[0108] Fresh food and dangerous goods: P = 1.5;

[0109] Goods with a delivery time of less than three days: P = 1;

[0110] Lightweight cargo with a volume-to-weight ratio greater than the threshold: P = 0.8;

[0111] The calculation process of the real-time freight discount rate F including the real-time return empty load rate information is as follows:

[0112] F = 0.6R + 0.3T + 0.1P;

[0113] When this embodiment is applied, the real-time return empty rate information R is extracted from the three-dimensional cargo chain database, and a comprehensive calculation is performed in combination with the preset time sensitivity T and cargo priority P. The time sensitivity T specifically includes: night transportation period (22:00-6:00), T=1.2, which aims to encourage staggered transportation, alleviate daytime traffic pressure, and improve vehicle utilization at night; statutory holidays and the day before and after, T=1.1, which is used to cope with logistics demand fluctuations during holidays and balance transportation resources; ordinary period T=1; cargo priority P specifically includes: fresh and dangerous goods, P=1.5, because of their high transportation requirements and high risks, higher price incentives are given to ensure timely and safe transportation; goods with a time limit of less than three days, P=1; volume-to-weight ratio greater than the threshold (volume-to-weight ratio>6000cm 3For lightweight cargo ( / kg), P = 0.8, balancing vehicle space utilization and transportation costs by reducing priority. A comprehensive real-time freight discount rate F is calculated and transmitted. The weight coefficients corresponding to each factor in this real-time freight discount rate F are obtained through training based on historical transportation data. This weight distribution highlights the core impact of the return empty load rate R on resource utilization efficiency, while also taking into account the regulatory effects of transportation time and cargo type. Dynamically adjusting freight discounts through multi-dimensional parameters achieves precise price incentives, guides shippers to rationally arrange transportation, and optimizes the allocation of freight logistics resources.

[0114] like Figure 3 As shown, as a preferred embodiment of the present invention, the generating and sending of gradient price incentive information and load-sharing optimization information based on the three-dimensional cargo chain database specifically further includes:

[0115] Step S211: extracting cargo weight, cargo volume, cargo delivery location, and cargo delivery time information from the three-dimensional cargo chain database;

[0116] Step S212: Based on the DBSCAN algorithm, define the spatiotemporal distance measurement formula:

[0117]

[0118] Where, d ij is the comprehensive time-space distance between the i-th source and the j-th source, x i 、y i is the horizontal and vertical coordinates of the shipping place (or receiving place) of the i-th source of goods, x j 、y j is the horizontal and vertical coordinates of the place of shipment (or place of receipt) of the jth source of goods, t i 、j i is the delivery time between the i-th source and the j-th source, α, β, and λ are the corresponding weight coefficients, which are determined by historical training data;

[0119] Step S213: setting the neighborhood radius κ and the minimum number of points ε;

[0120] Step S214: Group the goods sources whose distance is less than the neighborhood radius κ and greater than the minimum number of points ε into one category, and generate several adjacent spatiotemporal goods source clusters D k ;

[0121] Step S215: Based on several adjacent time and space supply clusters D k , calculate and generate the comprehensive vehicle loading rate L;

[0122] The calculation process of the vehicle comprehensive loading rate L is:

[0123] And i∈D k w i≤w c , i∈D k V i ≤V c ;

[0124] Where w i is the nearby space-time supply cluster D k The weight of the i-th source, v i is the nearby space-time supply cluster D k The volume of the i-th source of goods, μ is the weight loading coefficient, λ is the volume loading coefficient, w c is the vehicle load, V c is the vehicle volume;

[0125] Step S216: Match the vehicle comprehensive loading rate L to the fitness f(L)=L;

[0126] Step S217: Each adjacent time and space supply cluster D k The supply combination is coded as a chromosome;

[0127] Step S218: Based on the two-point crossover method, the gene segments at the chromosome crossover points are exchanged; the crossover probability P1 is set to 0.8;

[0128] Step S219: Randomly change the chromosome value; set the mutation probability P2 = 0.05;

[0129] Step S2110: After multiple iterations, the loading combination S with the maximum loading rate is selected and sent;

[0130] It should be understood that the weight, volume, receiving and sending location and time information of the goods in the three-dimensional cargo chain database are first extracted, and then the DBSCAN algorithm (Chinese interpretation is "density-based noise application spatial clustering algorithm") is applied. The core function of this algorithm is to divide the density-connected areas into clusters by measuring the local density of data points. It can effectively identify the core points, boundary points and noise points in the data set, define the spatiotemporal distance measurement formula, and generate several adjacent spatiotemporal cargo source clusters D. k This process can gather cargo sources with close temporal and spatial distribution, avoid circuitous loading routes, and build a cluster of cargo sources D based on the proximity of temporal and spatial distribution. k , calculate the comprehensive loading rate of the vehicle, and encode the cargo source combination in each adjacent time-space cargo source cluster into a chromosome in binary form, with each bit corresponding to a cargo source. Among them, "1" means that the cargo source is selected for loading, and "0" means not selected. For example, if the adjacent time-space cargo source cluster D kThe algorithm contains five cargo sources, coded [1,0,1,1,0], indicating that the first, third, and fourth cargo sources are selected for the load-sharing combination. Two crossover points are randomly selected on the chromosomes, and the gene segments between them are swapped to generate a new chromosome combination. The crossover probability is set to P1 = 0.8, indicating that 80% of chromosome pairs will undergo a crossover operation. This promotes gene recombination and diversity, expands the search space, and prevents the algorithm from prematurely falling into a local optimum. The value of a single bit on the chromosome is randomly changed (from 1 to 0 or 0 to 1) with a mutation probability of P2 = 0.05 to introduce new genetic features. Finally, after multiple generations of iterations, the fitness of the chromosomes (i.e., the vehicle's overall loading rate, L) is calculated in each generation. High-fitness chromosomes are retained, while low-fit individuals are eliminated. Through this "survival of the fittest" evolutionary mechanism, the load-sharing combination S with the highest loading rate is gradually selected and delivered.

[0131] like Figure 4 As shown, as a preferred embodiment of the present invention, the generation and transmission of trunk and branch line coordinated dispatching information and drop-and-hook management information based on the three-dimensional cargo chain database specifically includes:

[0132] Step S221: extracting the real-time location of vehicles, hub node locations, terminal delivery requirements, cargo hard time windows, and cargo soft time windows from the three-dimensional cargo chain database;

[0133] Step S222: Based on GIS and logistics hub distribution, a trunk and branch line network is generated and a visual topology model including node processing capabilities and road section constraints is constructed;

[0134] Step S223: Analyze historical data based on the LSTM algorithm to predict and generate cargo demand for each time period and route, as well as the terminal order concentration area;

[0135] Step S224: Generate and send a trunk and branch line vehicle matching and route optimization solution based on a genetic algorithm;

[0136] When this embodiment is applied, the real-time position of vehicles, hub node positions, terminal delivery demand, cargo hard time window, and cargo soft time window information are first extracted from the three-dimensional cargo chain database. Based on the distribution of logistics hubs, historical transportation traffic and geographic information system (GIS), the trunk line (main route for long-distance transportation between cities) and the branch line (short-distance delivery route from the hub to the terminal) are defined to form a visual trunk and branch line transportation topology structure. Based on the LSTM algorithm, combined with historical order data, seasonal factors, e-commerce promotion activities, etc., the cargo volume demand for each time period and each route is predicted. With the lowest total vehicle transportation cost and the fastest timeliness as the goal, combined with vehicle load, volume restrictions, cargo timeliness requirements, and time window constraints, a genetic algorithm is used to generate and send the matching and path planning of trunk vehicles and branch vehicles to the vehicle owner end, and simultaneously update them to the cargo owner end.

[0137] like Figure 5 As shown, as a preferred embodiment of the present invention, the generation and transmission of trunk and branch line coordinated dispatching information and drop-and-hook management information based on the three-dimensional cargo chain database specifically includes:

[0138] Step S231: extracting vehicle real-time tasks, hub geo-fence locations, trailer IDs, trailer status, and trailer pool status information from the three-dimensional cargo chain database;

[0139] Step S232: Based on GPS positioning, when the trailer and the tractor enter the preset hub geo-fence, a drop-and-hook preparation instruction is generated and vehicle demand is predicted;

[0140] Step S233: Calculating the compatibility of the trailer and tractor based on the Hungarian algorithm;

[0141] Step S234: Generate and send a drop-and-hook plan based on the compatibility of the trailer and tractor;

[0142] When this embodiment is applied, the real-time tasks of the vehicle, the location of the hub geographical fence, the trailer ID, the trailer status, and the trailer pool status information in the three-dimensional cargo chain database are first extracted, and the location of the tractor is tracked in real time through the vehicle-mounted GPS positioning system. When the vehicle enters the set range of the hub, the system automatically triggers the drop-and-hook preparation process. Combined with the estimated arrival time in the transport task list, the vehicle demand is predicted in advance to reduce waiting time. Based on the current tasks of the tractor and the trailer pool status in the cargo chain database, the Hungarian algorithm is used to calculate the adaptability of the trailer and the tractor, and a matching solution with the lowest total cost is calculated. Priority is given to trailers that are load-adaptable and on the way to the destination, and a drop-and-hook solution is generated and sent to the vehicle owner and the cargo owner. The drop-and-hook solution includes the recommended replacement trailer ID, parking location, and estimated replacement time.

[0143] like Figure 6 As shown, as another preferred embodiment of the present invention, on the other hand, a freight logistics distribution and regulation system includes:

[0144] The collection module 100 is used to collect freight vehicles, cargo sources, and road condition data in real time based on IoT devices, GPS positioning systems, and traffic API interfaces to establish a three-dimensional cargo chain database;

[0145] The first generating and sending module 200 is used to generate and send gradient price incentive information and load-sharing optimization information based on the three-dimensional cargo chain database;

[0146] The second generation and transmission module 300 is used to generate and transmit trunk and branch line coordinated scheduling information and drop and hook management information based on the three-dimensional cargo chain database;

[0147] The synchronous sending module 400 is used to monitor traffic information in real time, predict delay duration, and synchronously send delay notifications.

[0148] When this embodiment is applied, based on the Internet of Things devices, GPS positioning system, and traffic API interface, the collection module 100 collects freight vehicles, cargo sources, and road condition data in real time to establish a three-dimensional cargo chain database. Based on the three-dimensional cargo chain database, the first generation and sending module 200 generates and sends gradient price incentive information and loading optimization information. Based on the three-dimensional cargo chain database, the second generation and sending module 300 generates and sends trunk and branch line coordinated scheduling information and drop-and-hook management information. The synchronous sending module 400 monitors road condition information in real time, predicts delay duration, and synchronously sends delay notifications.

[0149] like Figure 7 As shown, as another preferred embodiment of the present invention, the first generating and sending module 200 specifically includes:

[0150] The first extraction unit 201 is used to extract the real-time return empty rate information R from the three-dimensional cargo chain database;

[0151] The calculation and sending unit 202 is used to calculate and send a real-time freight discount rate F including real-time return empty load rate information based on a preset time sensitivity T and a cargo priority P;

[0152] When this embodiment is applied, the first extraction unit 201 extracts the real-time return empty rate information R from the three-dimensional cargo chain database. Based on the preset time sensitivity T and cargo priority P, the calculation and sending unit 202 calculates and sends the real-time freight discount rate F including the real-time return empty rate information.

[0153] like Figure 8 As shown, as another preferred embodiment of the present invention, the first generating and sending module 200 specifically further includes:

[0154] The second extraction unit 211 is used to extract the cargo weight, cargo volume, cargo delivery location, and cargo delivery time information from the three-dimensional cargo chain database;

[0155] A definition unit 212 is used to define a spatiotemporal distance measurement formula based on the DBSCAN algorithm;

[0156] A setting unit 213 is used to set a neighborhood radius κ and a minimum number of points ε;

[0157] The generating unit 214 is used to group the goods sources whose distance is less than the neighborhood radius κ and greater than the minimum number of points ε into one category, and generate several adjacent spatiotemporal goods source clusters D k ;

[0158] A matching unit 215 is used to match the vehicle comprehensive loading rate L to a fitness f(L)=L;

[0159] The combined coding unit 216 is used to combine each adjacent spatiotemporal supply cluster D kThe supply combination is coded as a chromosome;

[0160] A crossover unit 217 is used to exchange gene segments at chromosome crossover points based on a two-point crossover method;

[0161] The mutation unit 218 is used to randomly change the chromosome value.

[0162] When this embodiment is applied, the second extraction unit 211 extracts the cargo weight, cargo volume, cargo delivery location, and cargo delivery time information from the three-dimensional cargo chain database. Based on the DBSCAN algorithm, the definition unit 212 defines the spatiotemporal distance measurement formula. The setting unit 213 sets the neighborhood radius κ and the minimum number of points ε. The cargo sources with a distance less than the neighborhood radius κ and greater than the minimum number of points ε are clustered into one category. The generation unit 214 generates a number of adjacent spatiotemporal cargo source clusters D. k The matching unit 215 matches the vehicle comprehensive loading rate L to the fitness f(L)=L, and the combined encoding unit 216 combines each adjacent spatiotemporal cargo source cluster D k The supply combination is encoded as a chromosome. Based on the two-point crossover method, the crossover unit 217 exchanges the gene fragments of the chromosome crossover point, and the mutation unit 218 randomly changes the chromosome value.

[0163] like Figure 9 As shown, as another preferred embodiment of the present invention, the second generating and sending module 300 specifically includes:

[0164] The third extraction unit 301 is used to extract the real-time location of vehicles, hub node locations, terminal delivery requirements, cargo hard time windows, and cargo soft time windows from the three-dimensional cargo chain database;

[0165] A generation unit 302 is used to generate a trunk and branch line network based on GIS and logistics hub distribution and to construct a visual topology model including node processing capabilities and road section constraints;

[0166] The prediction generation unit 303 is used to analyze historical data based on the LSTM algorithm to predict and generate cargo demand for each time period and each route and the terminal order concentration area;

[0167] The generating and sending unit 304 is used to generate and send a trunk line and branch line vehicle matching and route optimization solution based on a genetic algorithm.

[0168] When this embodiment is applied, the third extraction unit 301 extracts the real-time position of vehicles, hub node positions, terminal distribution requirements, cargo hard time windows, and cargo soft time windows from the three-dimensional cargo chain database. Based on GIS and logistics hub distribution, the generation unit 302 generates a trunk and branch line network and constructs a visual topology model including node processing capabilities and road section constraints. Based on the LSTM algorithm, the historical data is analyzed. The prediction generation unit 303 predicts and generates cargo volume requirements for each time period and route, as well as the terminal order concentration area. Based on the genetic algorithm, the generation and sending unit 304 generates and sends trunk and branch line vehicle matching and path optimization solutions.

[0169] The above embodiment of the present invention provides a method for distributing and regulating freight logistics, and provides a system for distributing and regulating freight logistics. Through the Internet of Things devices, GPS positioning systems and traffic API interfaces, vehicle status, cargo information and road condition data are collected in real time, and a distributed cargo chain database is constructed to realize data fusion and real-time synchronization of the entire chain. A gradient price incentive model and an intelligent loading algorithm are designed based on the database to enhance the attractiveness of return cargo sources and integrate the needs of scattered cargo owners. Through the coordinated dispatch of trunk and branch lines combined with the drop-and-hook transport mode, the Dijkstra algorithm and time window constraints are used to generate a time-space matching path, and the hub trailer pool is used to realize rapid trailer switching, shortening the loading and unloading waiting time. , through GPS and traffic API, vehicle location and road condition data are collected in real time, and the delay threshold is dynamically predicted based on historical data and real-time congestion coefficient. The actual driving time is continuously compared with the planned time. When the delay time exceeds the threshold, a delay notification is sent to the vehicle owner and the cargo owner at the same time, so that the cargo owner can adjust the time window in time to reduce losses. This method and system breaks through the optimization limitations of a single link in traditional logistics, and uses the "data collection-intelligent decision-execution feedback" closed loop. Through intelligent loading and precise matching of return cargo sources, the empty return rate of trucks is reduced and the load rate of trucks is increased, providing the logistics industry with a "data-driven" intelligent scheduling paradigm and promoting the transformation of traditional freight to precision and intelligence.

[0170] In order to enable the above-mentioned method and system to be loaded and run smoothly, in addition to the various modules mentioned above, the system may also include more or fewer components than described above, or a combination of certain components, or different components, for example, it may include input and output devices, network access devices, buses, processors and memories, etc.

[0171] The processor may be a central processing unit, other general-purpose processors, digital signal processors, application-specific integrated circuits, off-the-shelf programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor. The processor is the control center of the system, connecting various components using various interfaces and lines.

[0172] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0173] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

[0174] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for adjusting freight logistics distribution, characterized in that: The method comprises: Based on IoT devices, GPS positioning systems, and traffic API interfaces, real-time data on freight vehicles, cargo sources, and road conditions is collected to establish a three-dimensional cargo chain database; Vehicle status information: vehicle mission, vehicle location, vehicle load, mileage, and return empty rate; Cargo source information: cargo weight, cargo volume, cargo delivery location, cargo delivery time, cargo type, timeliness requirements, cargo owner ID, hub node location, terminal delivery requirements, cargo hard time window, cargo soft time window; Trailer information: preset hub geofence location, trailer ID, trailer status, trailer pool status; Traffic condition information: real-time traffic flow, congested roads, and weather warnings; Generate and send gradient price incentive information and load-sharing optimization information based on the three-dimensional cargo chain database; Generate and send trunk and branch line coordinated dispatch information and drop-and-hook management information based on the 3D cargo chain database; Monitor traffic conditions in real time, predict delay duration, and send delay notifications simultaneously.

2. The freight logistics distribution and adjustment method according to claim 1, characterized in that: The generation and transmission of gradient price incentive information and load-sharing optimization information based on the three-dimensional cargo chain database specifically includes: Extract the real-time return empty rate information R from the three-dimensional cargo chain database; Based on the preset time sensitivity T and cargo priority P, calculate and send the real-time freight discount rate F including the real-time return empty load rate information; The time sensitivity T specifically includes: Night transport period: T=1.2; Legal holidays and the day before or after: T=1.1; Normal period: T=1; The cargo priority P specifically includes: Fresh food and dangerous goods: P = 1.5; Goods with a delivery time of less than three days: P = 1; Lightweight cargo with a volume-to-weight ratio greater than the threshold: P = 0.8; The calculation process of the real-time freight discount rate F including the real-time return empty load rate information is as follows: F=0.6R+0.3T+0.1P.

3. The freight logistics distribution and adjustment method according to claim 1, characterized in that: The generating and sending of gradient price incentive information and load-sharing optimization information based on the three-dimensional cargo chain database specifically includes: Extract cargo weight, cargo volume, cargo delivery location, and cargo delivery time information from the three-dimensional cargo chain database; Based on the DBSCAN algorithm, the spatiotemporal distance measurement formula is defined as follows: Where, d ij is the comprehensive time-space distance between the i-th source and the j-th source, x i 、y i is the horizontal and vertical coordinates of the shipping place (or receiving place) of the i-th source of goods, x j 、y j is the horizontal and vertical coordinates of the place of shipment (or place of receipt) of the jth source of goods, t i 、j i is the delivery time between the i-th source and the j-th source, α, β, and λ are the corresponding weight coefficients, which are determined by historical training data; Set the neighborhood radius κ and the minimum number of points ε, cluster the sources whose distance is less than the neighborhood radius κ and greater than the minimum number of points ε into one category, and generate several adjacent spatiotemporal source clusters D k ; Based on several nearby spatiotemporal supply clusters D k , calculate and generate the comprehensive vehicle loading rate L; The calculation process of the vehicle comprehensive loading rate L is: And i∈D k w i ≤w c , i∈D k V i ≤V c ; Where w i is the nearby space-time supply cluster D k The weight of the i-th source, v i is the nearby space-time supply cluster D k The volume of the i-th source of goods, μ is the weight loading coefficient, λ is the volume loading coefficient, w c is the vehicle load, V c is the vehicle volume; Match the comprehensive vehicle loading rate L to the fitness f(L) = L; Each adjacent spatiotemporal supply cluster D k The supply combination is coded as a chromosome; Based on the two-point crossover method, the gene fragments at the chromosome crossover point were exchanged, the crossover probability P1 was set to 0.8, the chromosome values were randomly changed, and the mutation probability P2 was set to 0.05; After multiple generations of iteration, the loading combination S with the maximum loading rate is screened and sent.

4. The freight logistics distribution and adjustment method according to claim 1, characterized in that: The generation and transmission of trunk and branch line coordinated dispatch information and drop-and-hook management information based on the three-dimensional cargo chain database specifically includes: Extract the real-time location of vehicles, hub node locations, terminal delivery requirements, cargo hard time windows, and cargo soft time windows from the three-dimensional cargo chain database; Based on GIS and logistics hub distribution, a trunk and branch line network is generated and a visual topology model is constructed that includes node processing capabilities and section constraints. Analyze historical data based on the LSTM algorithm to predict and generate cargo demand for each time period and route, as well as the terminal order concentration area; Generate and send trunk and branch line vehicle matching and route optimization solutions based on genetic algorithms.

5. The freight logistics distribution and adjustment method according to claim 1, characterized in that: The generating and sending of trunk and branch line coordinated dispatching information and drop-and-hook management information based on the three-dimensional cargo chain database specifically includes: Extract vehicle real-time tasks, hub geofence locations, trailer IDs, trailer status, and trailer pool status information from the 3D cargo chain database; Based on GPS positioning, when the trailer and tractor enter the preset hub geo-fence, it generates a drop-and-hook preparation instruction and predicts vehicle demand; Calculate the compatibility of trailers and tractors based on the Hungarian algorithm; Generate and send a drop-and-hook plan based on the compatibility of the trailer and tractor.

6. A freight logistics distribution and adjustment system, characterized in that: The method for allocating and regulating freight logistics according to any one of claims 1 to 6 is applied, wherein the system comprises: The collection module is used to collect real-time data on freight vehicles, cargo sources, and road conditions based on IoT devices, GPS positioning systems, and traffic API interfaces to establish a three-dimensional cargo chain database; The first generation and sending module is used to generate and send gradient price incentive information and load-sharing optimization information based on the three-dimensional cargo chain database; The second generation and sending module is used to generate and send trunk and branch line coordinated scheduling information and drop and hook management information based on the three-dimensional cargo chain database; The synchronous sending module is used to monitor traffic conditions in real time, predict delay duration, and send delay notifications synchronously.

7. The freight logistics distribution and adjustment system according to claim 6, characterized in that: The first generating and sending module specifically includes: The first extraction unit is used to extract the real-time return empty rate information R from the three-dimensional cargo chain database; The calculation and sending unit is used to calculate and send a real-time freight discount rate F including real-time return empty load rate information based on a preset time sensitivity T and cargo priority P.

8. The freight logistics distribution and adjustment system according to claim 6, characterized in that: The first generating and sending module specifically further includes: The second extraction unit is used to extract the cargo weight, cargo volume, cargo delivery location, and cargo delivery time information from the three-dimensional cargo chain database; Definition unit, used to define the spatiotemporal distance measurement formula based on the DBSCAN algorithm; Setting unit, used to set the neighborhood radius κ and the minimum number of points ε; The generation unit is used to group the sources of goods whose distance is less than the neighborhood radius κ and greater than the minimum number of points ε into one category, and generate several adjacent spatiotemporal source clusters D k ; A matching unit for matching the vehicle comprehensive loading rate L to a fitness f(L)=L; The combined coding unit is used to combine each adjacent spatiotemporal supply cluster D k The supply combination is coded as a chromosome; Crossover unit, used to exchange gene segments at chromosome crossover points based on the two-point crossover method; Mutation unit, used to randomly change the chromosome value.

9. The freight logistics distribution and adjustment system according to claim 6, characterized in that: The second generation and sending module specifically includes: The third extraction unit is used to extract the real-time location of vehicles, hub node locations, terminal delivery requirements, cargo hard time windows, and cargo soft time windows from the three-dimensional cargo chain database; The generation unit is used to generate the trunk and branch line network based on GIS and logistics hub distribution and build a visual topology model that includes node processing capabilities and road section constraints; The prediction generation unit is used to analyze historical data based on the LSTM algorithm to predict and generate cargo demand for each time period and route, as well as the terminal order concentration area; The generating and sending unit is used to generate and send trunk and branch line vehicle matching and route optimization solutions based on genetic algorithms.

Images