Logistics city distribution automatic scheduling method and scheduling system

Abstract

The invention discloses a logistics city distribution automatic scheduling method and scheduling system, and relates to the field of logistics distribution, and the method comprises the steps: receiving a cargo form imported by a user, and extracting the cargo information of the cargo form, and the distribution time information and the distribution address information corresponding to the cargo information; screening the cargo information according to the delivery time information to obtain a plurality of to-be-delivered target cargoes, and determining target delivery time and target delivery addresses of the target cargoes; obtaining vehicle information in the scheduling system, and determining vehicle cargo carrying information and vehicle route information according to the vehicle information; determining a plurality of candidate delivery vehicles of each target cargo according to the target delivery address and the vehicle route information; screening the candidate delivery vehicles based on the target delivery time and the vehicle cargo carrying information to obtain a target delivery vehicle; and generating a delivery task list corresponding to the target delivery vehicle and the target cargo, and exporting the delivery task list. According to the invention, the processing efficiency of the distribution form of the logistics station can be improved.

Description

Technical Field

[0001] The present application relates to the field of logistics dispatching, and in particular to an automated dispatching method and dispatching system for urban logistics distribution. Background Art

[0002] In modern logistics and distribution, logistics sites, as key nodes for cargo transfer and distribution, need to efficiently process information on multiple links such as arrival, inventory, and distribution. Accurate and timely processing of cargo information and reasonable allocation of transportation resources are important links in ensuring the efficiency of logistics and distribution.

[0003] In related technologies, logistics sites mainly rely on manual methods to manage the distribution process. Workers need to manually organize multiple Excel forms, including the arrival list that records the information of incoming goods, the arrival inventory table that records the status of inventory goods, and the delivery inventory table of goods to be delivered. Workers need to perform data statistics, analysis and sorting based on the information in these forms, and then formulate distribution plans and allocate transportation resources.

[0004] However, the data between multiple forms need to be checked and transcribed repeatedly, which takes a long time for staff to process the data and is inefficient. Summary of the invention

[0005] The present application provides an automated scheduling method and scheduling system for logistics city distribution, which are used to improve the processing efficiency of delivery forms at logistics sites.

[0006] In the first aspect, the present application provides an automated scheduling method for logistics city distribution, which is applied to a scheduling system, and the method includes: receiving a cargo form imported by a user, extracting cargo information from the cargo form and delivery time information and delivery address information corresponding to the cargo information; screening the cargo information according to the delivery time information, obtaining multiple target cargoes to be delivered, and respectively determining the target delivery time and target delivery address of the target cargoes; obtaining vehicle information registered in the scheduling system, determining the vehicle cargo information and vehicle route information according to the vehicle information; determining multiple candidate delivery vehicles for each target cargo according to the target delivery address and vehicle route information; screening the candidate delivery vehicles based on the target delivery time and vehicle cargo information, and obtaining the target delivery vehicle; generating a delivery task list corresponding to the target delivery vehicle and the target cargo, and exporting the delivery task list.

[0007] In the above embodiment, the scheduling system realizes the automated scheduling of the logistics distribution process by automatically extracting cargo form information, screening according to time information, intelligently matching delivery vehicles according to vehicle information and routes, and automatically generating delivery task orders, thereby avoiding the tedious work of manual processing of forms and improving the processing efficiency and accuracy of logistics site distribution forms.

[0008] In combination with some embodiments of the first aspect, in some embodiments, the step of determining multiple candidate delivery vehicles for each target cargo based on the target delivery address and vehicle route information specifically includes: obtaining electronic map information of multiple preset geographical areas in the scheduling system; marking the target delivery address on the electronic map information to obtain the target delivery point; obtaining the historical delivery route in the vehicle route information, and determining the key delivery area for each delivery vehicle based on the historical delivery route; calculating the delivery distance value from the target delivery point to each key delivery area; sorting the delivery vehicles according to the delivery distance value, and selecting a preset number of vehicles with the smallest delivery distance value as candidate delivery vehicles.

[0009] In the above embodiment, the dispatching system realizes the visual display of logistics distribution and intelligent vehicle screening based on geographic location by marking distribution points on electronic maps, analyzing historical distribution routes to determine key distribution areas, and calculating distribution distance values, thereby improving the rationality of distribution routes and distribution efficiency.

[0010] In combination with some embodiments of the first aspect, in some embodiments, before generating a delivery task list corresponding to the target delivery vehicle and the target cargo, and exporting the delivery task list, the method also includes: calculating the estimated mileage of the target delivery vehicle from the current location to the target delivery point; extracting the vehicle information of the target delivery vehicle, and determining the mileage cost of the estimated mileage; obtaining the volume rate and weight rate of the target cargo; and calculating the delivery cost based on the volume rate, weight rate and mileage cost.

[0011] In the above embodiment, the dispatching system achieves accurate calculation of distribution costs by calculating the expected mileage, determining mileage costs, and considering factors such as freight rates, which helps to optimize resource allocation and cost control and improve the economic benefits of logistics distribution.

[0012] In combination with some embodiments of the first aspect, in some embodiments, before the step of screening the cargo information according to the delivery time information to obtain multiple target cargoes to be delivered, and respectively determining the target delivery time and target delivery address of the target cargoes, the method also includes: extracting the cargo information of the cargo form and the pickup information corresponding to the cargo information; when the pickup information is self-pickup, removing the corresponding cargo to be self-pickup from the cargo form; and counting the cargo information of the cargo to be self-pickup to generate a self-pickup cargo form.

[0013] In the above embodiment, the dispatching system generates a special self-pickup goods form through intelligent identification and classification of self-pickup goods, thereby avoiding the waste of distribution resources and improving the operational efficiency of logistics sites.

[0014] In combination with some embodiments of the first aspect, in some embodiments, based on the target delivery time and vehicle cargo information, the candidate delivery vehicles are screened to obtain the step of the target delivery vehicle, specifically including: obtaining the target volume and target weight of the target cargo; determining the remaining available volume and the remaining carryable weight of each candidate delivery vehicle according to the vehicle cargo information; calculating a first ratio of the target volume of the target cargo to the remaining available volume, and a second ratio of the target weight of the target cargo to the remaining carryable weight; if both the first ratio and the second ratio are less than a preset threshold, determining the corresponding candidate delivery vehicle as the target delivery vehicle; sorting multiple candidate delivery vehicles that meet the conditions according to the first ratio and the second ratio, and selecting the vehicle with the smallest ratio as the target delivery vehicle.

[0015] In the above embodiment, the dispatching system achieves accurate matching of the vehicle's cargo capacity by calculating the ratio of the cargo volume weight to the vehicle's remaining capacity, thereby avoiding overloading or underloading and improving vehicle utilization.

[0016] In combination with some embodiments of the first aspect, in some embodiments, after receiving a cargo form imported by a user and extracting the cargo information of the cargo form and the delivery time information and delivery address information corresponding to the cargo information, the method also includes: extracting the cargo information of the cargo form and the cargo type identification corresponding to the cargo information; determining special cargo in the cargo form according to the cargo type identification; special cargo includes customs declaration cargo, cold chain cargo, fragile cargo and dangerous goods; when the special cargo is customs declaration cargo, determining the time required for customs declaration, and adding the time required for customs declaration to the target delivery time of the customs declaration cargo; when the special cargo is cold chain cargo, screening vehicles with temperature control equipment as candidate delivery vehicles, and adjusting the delivery cost of the cold chain cargo according to the operating cost of the temperature control equipment; when the special cargo is fragile cargo or dangerous goods, obtaining the transportation qualification requirements of the delivery vehicle, and eliminating vehicles whose transportation qualification requirements do not meet the preset requirements from the candidate delivery vehicles.

[0017] In the above embodiment, the dispatching system ensures the compliant and safe transportation of different types of goods through intelligent identification and specialized processing procedures for special goods, thereby improving the professional level of logistics distribution.

[0018] In combination with some embodiments of the first aspect, in some embodiments, after the steps of generating a delivery task list corresponding to the target delivery vehicle and the target goods, and exporting the delivery task list, the method also includes: obtaining the current location information and the initial delivery route of the target delivery vehicle; calculating the connection distance and the estimated driving time between each delivery point in the initial delivery route based on the current location information, real-time road conditions information and historical traffic data; reordering each delivery point based on the connection distance and the estimated driving time, combined with the target delivery time, to generate an optimized delivery route; and sending the optimized delivery route to the on-board terminal of the target delivery vehicle.

[0019] In the above embodiment, the dispatching system optimizes the delivery routes through real-time traffic information and historical traffic data, thereby improving delivery efficiency and reducing delivery delays.

[0020] In a second aspect, an embodiment of the present application provides a scheduling system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the scheduling system to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0021] In a third aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when the computer program product is run on a scheduling system, enables the scheduling system to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, comprising instructions, which, when executed on a scheduling system, enable the scheduling system to execute the method described in the first aspect and any possible implementation of the first aspect.

[0023] It is understandable that the scheduling system provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the method provided in the embodiment of the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, which will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. Due to the use of automated cargo information extraction, intelligent matching of delivery vehicles and task order generation mechanisms, the distribution system can automatically process cargo forms, analyze delivery time and address information, and select the optimal delivery vehicle based on vehicle cargo capacity and route information. This greatly reduces the workload and error rate of manual form processing, effectively solving the problem of time-consuming and inefficient manual processing of multiple Excel forms in the prior art, realizing automated scheduling of the logistics distribution process and improving the operational efficiency of logistics sites.

[0025] 2. Due to the adoption of an intelligent self-pickup goods identification and classification mechanism, the pickup information of the goods can be automatically extracted, the self-pickup goods can be separated from the delivered goods, and a special self-pickup goods form can be generated, thus avoiding the situation where the self-pickup goods are incorrectly arranged for delivery. It effectively solves the problems of chaotic goods distribution methods and unreasonable resource allocation in the existing technology, and improves the operational efficiency and customer service quality of logistics sites.

[0026] 3. Due to the adoption of intelligent identification and professional processing procedures for special goods, the system can automatically identify special goods such as customs declaration, cold chain, fragile goods and dangerous goods according to the cargo type identification, and adjust the delivery time, select professional vehicles and calculate special costs in a targeted manner, thereby ensuring the compliant and safe transportation of special goods, effectively solving the problems of irregular distribution management and high safety risks of special goods in the existing technology, and improving the safety and service quality of logistics distribution. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a flow chart of the automated dispatching method for logistics city distribution in the embodiment of the present application; Figure 2 This is another flow chart of the automated dispatching method for urban logistics distribution in an embodiment of the present application; Figure 3 It is a schematic diagram of the structure of a physical device of the scheduling system in an embodiment of the present application. DETAILED DESCRIPTION

[0028] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification of the present application, the singular expressions "one", "a kind of", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more of the listed items.

[0029] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0030] The following is a description of the process of the method provided by this implementation. Figure 1 , which is a flow chart of the automated dispatching method for urban logistics distribution in an embodiment of the present application.

[0031] S101, receiving a goods form imported by a user, extracting goods information in the goods form and delivery time information and delivery address information corresponding to the goods information.

[0032] Among them, the cargo form refers to an electronic document that records information related to the cargo to be delivered, and is used to store and manage various types of data required for cargo delivery; cargo information represents the basic properties of the cargo to be delivered, including cargo number, name, specifications, quantity, etc.; delivery time information refers to the specific date and time period when the cargo needs to be delivered; delivery address information represents the target delivery location of the cargo, including detailed address, region, contact person and other information.

[0033] After receiving the cargo form uploaded by the user through the system interface, the dispatch system needs to parse and process the form content. Specifically, the dispatch system first verifies the correctness of the form format to ensure that the data structure meets the system requirements; then extracts the basic information of the cargo in the form through the data parsing module, and associates this information with the corresponding delivery time, delivery address and other related data; finally, converts the extracted information into the system standard format and stores it.

[0034] In some embodiments, the receipt of cargo forms and information extraction can be achieved in a variety of ways: Optionally, the system provides a Web interface for users to upload cargo forms in Excel format, identifies the table header and data columns through the table parsing engine, extracts the required information according to the preset mapping rules, and finally performs data format verification and conversion; Optionally, the system provides a standardized API interface, and the user system can directly call the interface to transmit cargo data in JSON format. After receiving the data, the system parses, verifies and stores it, and returns the processing results. It is understandable that other methods can also be used to receive cargo forms and extract information, such as scanning paper forms for OCR recognition, etc., which are not limited here.

[0035] S102, screening the cargo information according to the delivery time information to obtain a plurality of target cargoes to be delivered, and determining the target delivery time and target delivery address of the target cargoes respectively.

[0036] Among them, the target goods refer to the goods to be delivered that meet the delivery time requirements; the target delivery time refers to the actual delivery time node after optimization and adjustment; the target delivery address refers to the final delivery location information after standardized processing.

[0037] After obtaining complete cargo information, the dispatching system needs to screen and plan cargo delivery based on the time dimension. Specifically, the dispatching system first groups all cargo according to delivery time and removes cargo that is not within the current delivery cycle; then prioritizes cargo within each time period, taking into account factors such as urgent items and special requirements; finally, the screened cargo is identified as the target cargo, and a specific delivery time window is assigned to each target cargo.

[0038] In some embodiments, cargo screening and time planning can be achieved in a variety of ways: Optionally, the system performs multi-level screening according to time priority, first screening out the cargo that must be delivered on the same day, then sorting the remaining cargo according to the latest delivery time, and finally making adjustments based on the customer's special time requirements; Optionally, the system uses intelligent algorithms to optimize the delivery time, and reasonably allocates the delivery time window by considering factors such as cargo volume, delivery distance, and vehicle capacity. It is understandable that other methods can also be used to achieve cargo screening and delivery time planning, such as machine learning models based on historical delivery data, which are not limited here.

[0039] S103: Acquire vehicle information registered in the dispatching system, and determine vehicle cargo information and vehicle route information based on the vehicle information.

[0040] Among them, vehicle information refers to the basic attribute data of the delivery vehicles registered in the system; vehicle cargo information indicates the vehicle's carrying capacity and current cargo status; vehicle route information refers to the vehicle's historical delivery routes and regular delivery area data.

[0041] The dispatching system needs to have real-time information on the vehicle resources available for delivery. Specifically, the system first queries the basic information of registered vehicles, including vehicle type, load limit, temperature control function, etc.; then calculates the current available cargo capacity of each vehicle, taking into account the occupancy of the assigned tasks; and finally analyzes the historical delivery data of the vehicle to determine its advantageous service areas and commonly used delivery routes.

[0042] In some embodiments, vehicle information acquisition and analysis can be achieved in a variety of ways: Optionally, the system collects vehicle status data in real time through the vehicle terminal, including location information, cargo load, remaining power, etc., and updates it dynamically; Optionally, the system creates a vehicle portrait based on historical delivery records, and analyzes the vehicle's delivery efficiency, punctuality, service area preference, and other characteristics. It is understandable that other methods can also be used to achieve vehicle information management and analysis, such as through data access from a third-party vehicle monitoring platform, which is not limited here.

[0043] S104: Determine multiple candidate delivery vehicles for each target cargo based on the target delivery address and vehicle route information.

[0044] Among them, candidate delivery vehicles represent a set of vehicles that may be suitable for performing a specific delivery task.

[0045] The dispatching system needs to match the appropriate delivery vehicle for each item to be delivered. Specifically, the system first analyzes the matching degree between the target delivery address and the historical delivery area of ​​each vehicle; then calculates the distance and estimated travel time from the current location of each vehicle to the target address; finally, based on factors such as matching degree and distance, multiple most suitable vehicles are selected as candidate delivery vehicles.

[0046] In some embodiments, the candidate delivery vehicles can be determined in a variety of ways: Optionally, the system uses a geographic information system to perform regional matching, analyzes the vehicle's historical delivery heat map, and selects vehicles with high regional overlap as candidates; Optionally, the system uses real-time road condition analysis and multi-objective optimization algorithms to comprehensively consider factors such as distance, time, and cost to dynamically determine the candidate vehicle list. It is understandable that other methods can also be used to select candidate vehicles, such as intelligent recommendations based on vehicle scoring models, which are not limited here.

[0047] S105. Based on the target delivery time and vehicle cargo information, candidate delivery vehicles are screened to obtain a target delivery vehicle.

[0048] Among them, the target delivery vehicle refers to the vehicle that is finally determined to perform the delivery task; the vehicle loading information refers to the vehicle's load capacity and current load status.

[0049] The dispatching system needs to select the best delivery vehicle from the candidate vehicles. Specifically, the system first verifies the availability of the candidate vehicles during the target delivery time period; then checks whether the vehicle's remaining cargo capacity meets the current cargo demand; and finally determines the final target delivery vehicle by comprehensively considering factors such as the vehicle's delivery efficiency and cost.

[0050] In some embodiments, the target delivery vehicle can be selected in a variety of ways: Optionally, the system uses a multi-level screening mechanism to evaluate time matching, cargo capacity, historical service quality and other indicators in turn, and gradually narrow the candidate range; Optionally, the system uses an intelligent scheduling algorithm to optimize the objective function by establishing a mathematical model to select the vehicle with the highest comprehensive score. It is understandable that other methods can also be used to determine the target vehicle, such as intelligent decision-making based on reinforcement learning, which is not limited here.

[0051] S106: Generate a delivery task list corresponding to the target delivery vehicle and the target goods, and export the delivery task list.

[0052] Among them, the delivery task order refers to the electronic document containing detailed information of the delivery task; the target delivery vehicle and target goods are the delivery resources that are finally matched successfully.

[0053] The dispatching system needs to convert the delivery plan into an executable task order. Specifically, the system first integrates the detailed information of the target delivery vehicle and goods, including delivery time, address, contact information, etc.; then generates a delivery task order in a standard format, adds the task number and delivery requirements; finally, exports the task order and distributes it to the relevant executors.

[0054] The following is a more detailed description of the process of the method provided by this implementation. Figure 2 , which is another flow chart of the automated scheduling method for urban logistics distribution in an embodiment of the present application.

[0055] S201, receiving a goods form imported by a user, extracting goods information in the goods form and delivery time information and delivery address information corresponding to the goods information.

[0056] Referring to step S101, the dispatching system extracts the delivery time information and delivery address information of the cargo form.

[0057] In some embodiments, the scheduling system extracts the cargo information of the cargo form and the cargo type identification corresponding to the cargo information; determines the special cargo in the cargo form according to the cargo type identification; special cargo includes customs declaration cargo, cold chain cargo, fragile cargo and dangerous goods; when the special cargo is customs declaration cargo, determines the time required for customs declaration, and adds the time required for customs declaration to the target delivery time of the customs declaration cargo; when the special cargo is cold chain cargo, screens vehicles with temperature control equipment as candidate delivery vehicles, and adjusts the delivery cost of the cold chain cargo according to the operating cost of the temperature control equipment; when the special cargo is fragile cargo or dangerous goods, obtains the transportation qualification requirements of the delivery vehicle, and eliminates vehicles whose transportation qualification requirements do not meet the preset requirements from the candidate delivery vehicles.

[0058] Among them, cargo form refers to a data set that records the basic information of the goods to be delivered; cargo information represents the basic attributes of the goods such as volume, weight, and quantity; cargo type identification is used to indicate the classification attributes of the goods; special cargo refers to cargo that requires special handling or special transportation conditions; customs declaration cargo refers to imported and exported cargo that needs to go through customs procedures or other cargo that requires certificates to enter and exit special places; cold chain cargo refers to perishable goods that require temperature control; fragile cargo refers to fragile cargo that requires special protection; dangerous goods are used to indicate special cargo with dangerous characteristics; the time required for customs declaration refers to the estimated time to complete the procedures; the target delivery time refers to the estimated delivery time; temperature control equipment refers to special equipment used to maintain the temperature of cargo; transportation qualification requirements refer to the qualifications required for vehicles to transport special cargo.

[0059] After receiving a cargo delivery request, the dispatching system needs to identify and process special cargo. Specifically, the system first extracts detailed information and corresponding type identification of each cargo from the cargo form; then screens out cargo that needs special treatment based on the type identification; then takes corresponding treatment measures for different types of special cargo: calculates customs clearance time for declared cargo and adjusts the delivery plan, generally choosing to arrange designated vehicles for delivery; matches professional transportation capacity for cold chain cargo and calculates additional costs, and strictly reviews vehicle qualifications for fragile and dangerous goods; finally, generates a delivery plan that meets various special requirements.

[0060] In some embodiments, the identification and processing of special goods can be achieved in a variety of ways: the system adopts a rule engine method to classify goods and match processing strategies by configuring a multi-level rule tree structure. The rule engine includes a condition matching module and a behavior execution module. The condition matching builds a decision tree based on the attributes of the goods, and the behavior execution module is responsible for calling the corresponding processing function, such as adjusting the time, screening vehicles, etc. The system can also use machine learning to train a classification model to identify special goods. First, a feature vector is constructed based on historical order data, including dimensions such as goods description, packaging method, and transportation requirements; then multiple classifiers are trained using algorithms such as support vector machines (SVM); finally, new orders are classified and predicted and the corresponding processing flow is triggered. For example, the system may automatically identify goods containing keywords such as "fragile" and "moisture-proof" as fragile goods and limit their stacking height. It is understandable that other methods can also be used to implement the identification and processing flow of special goods, which are not limited here.

[0061] S202: Filter the cargo information according to the delivery time information to obtain a plurality of target cargoes to be delivered, and respectively determine the target delivery time and target delivery address of the target cargoes.

[0062] Referring to step S102, the dispatching system determines the target delivery time and target delivery address of the target goods.

[0063] In some embodiments, the scheduling system extracts the cargo information of the cargo form and the pickup information corresponding to the cargo information; when the pickup information is self-pickup, the corresponding cargo to be self-pickup is removed from the cargo form; the cargo information of the cargo to be self-pickup is counted to generate a self-pickup cargo form.

[0064] Among them, pickup information refers to the instructions on how to pick up the goods; self-pickup means that the customer picks up the goods at the designated location by themselves; goods to be picked up by themselves are used to indicate goods that need to be picked up by the customer by themselves; and the self-pickup goods form refers to a special data set that records all self-pickup goods information.

[0065] The dispatching system needs to identify and separate self-pickup goods before starting distribution planning. Specifically, the system first extracts the basic information and pickup method of each record from the goods form; then identifies the goods records with the pickup method as self-pickup; then removes these self-pickup goods from the delivery goods form to avoid including them in the distribution plan; then organizes and classifies the information of all self-pickup goods; and finally generates a special self-pickup goods form for subsequent self-pickup business processing.

[0066] In some embodiments, the processing of self-pickup goods can be achieved in a variety of ways: the system adopts a diversion processing mechanism to build an independent self-pickup business processing flow. First, the state machine model is used to define the state transition rules of the goods processing, including the initial state S0, the self-pickup mark state S1, the separation completion state S2, etc. The state transition function T (Si, Event) determines the next processing state, where Si is the state and Event is the event. Then the state transition is triggered based on the event-driven mode, such as transferring the goods to a special processing queue when the self-pickup mark is received. The system can also adopt a batch processing mode to regularly scan the goods form for classification processing. By setting the processing time window [t0, t1], the accumulated goods records are collected in each processing cycle, and the classification and conversion processing are unified to generate a phased self-pickup goods report. For example, the system may perform a self-pickup goods separation task at 2 a.m. every day to sort and classify the self-pickup orders added the previous day. It can be understood that other methods can also be used to implement the processing flow of self-pickup goods, which is not limited here.

[0067] S203: Acquire vehicle information registered in the dispatching system, and determine vehicle cargo information and vehicle route information based on the vehicle information.

[0068] Referring to step S103, the dispatching system determines the vehicle cargo information and vehicle route information.

[0069] S204: Obtain electronic map information of multiple preset geographical areas in the dispatching system.

[0070] Among them, electronic map information refers to a digitized map data set containing geographic coordinates, road networks and regional divisions. The road network refers to the actual drivable paths connecting various locations; the preset geographic area represents the service area divided according to administrative divisions or distribution needs.

[0071] Before the dispatch system matches the delivery vehicles, it needs to obtain complete geographic information data support. Specifically, the system first connects to the geographic information database to obtain the electronic map base data of the current city; then loads the preset regional division scheme, including the boundary coordinates and attribute information of each area; then imports the real-time updated road network data, including road type, traffic status and other information; finally, integrates these data into a complete electronic map information.

[0072] In some embodiments, the construction and management of electronic map information can be achieved in a variety of ways. The system uses a hierarchical data structure to store geographic information, with the bottom layer being rasterized map data, the middle layer being a vectorized road network, and the top layer being a regional division represented by polygons. At the road network level, the system abstracts the road into a weighted directed graph G (V, E), where the vertex V represents an intersection or a key node, the edge E represents a road segment, and the weights include attributes such as distance and travel time. Regional division uses an improved K-means clustering algorithm to determine the regional center and boundaries through iterative optimization. For example, in actual applications, the system may divide a city into several distribution partitions, each of which contains multiple roads and multiple distribution points, and dynamically adjusts the regional division based on historical distribution data to improve distribution efficiency. It is understandable that other methods can also be used to implement the processing and application of electronic map information, which are not limited here.

[0073] S205: Mark the target delivery address on the electronic map information to obtain the target delivery point.

[0074] Among them, the target delivery address refers to the specific location information where goods need to be delivered; the target delivery point refers to the actual delivery location coordinate point after geocoding, and geocoding refers to the process of converting a text address into geographic coordinates.

[0075] After obtaining the electronic map information, the dispatch system needs to convert the delivery address into calculable geographic coordinates. Specifically, the system first cleans and standardizes the target delivery address to ensure that the address format is unified; then the standardized address is converted into longitude and latitude coordinates through the geocoding service; then the converted coordinate points are projected onto the electronic map and the accuracy is verified; finally, a visual delivery point mark is generated on the map.

[0076] In some embodiments, address labeling and coordinate conversion can be achieved in a variety of ways. The system uses a multi-level address resolution algorithm to decompose the address text into a hierarchical structure such as provinces, cities, districts, and streets, and establish an address tree model. The geocoding process uses an improved string matching algorithm to calculate the similarity score S=Σ(wi×si) between the address to be resolved and the standard address in the address library, where wi is the weight of each level and si is the matching score of the corresponding level. When the matching degree exceeds the threshold, the coordinates corresponding to the standard address are used; when multiple candidate results appear, they are further screened in combination with the surrounding POI information. For example, for an address such as "No. 20, a certain street in a certain district of a certain city", the system will parse each address element step by step and finally locate the specific coordinate position. It can be understood that other methods can also be used to achieve address labeling, which is not limited here.

[0077] S206: Obtain historical delivery routes in the vehicle route information, and determine the key delivery area for each delivery vehicle based on the historical delivery routes.

[0078] Among them, the historical delivery route refers to the trajectory data of the delivery tasks completed by the vehicle in the past; the key delivery area refers to the core service area where the vehicle frequently delivers.

[0079] The dispatching system needs to analyze historical data to determine the vehicle's advantageous service area. Specifically, the system first extracts the recent delivery trajectory data of each vehicle from the database; then cleans the trajectory points to remove abnormal points and noise data; then calculates the delivery frequency and stay time of each area based on the cleaned data; and finally determines the key delivery area of ​​each vehicle through cluster analysis.

[0080] In some embodiments, the identification of key delivery areas can be achieved in a variety of ways. The system uses a spatiotemporal clustering algorithm to analyze historical trajectories. First, the trajectory points are grouped by time window T, and the point density D(x, y) = Σ(K((x-xi) / h)) in each time window is calculated, where K is the kernel function and h is the bandwidth parameter. The DBSCAN density clustering algorithm is then used to identify high-density areas, and the minimum number of points MinPts and the neighborhood radius Eps threshold are set, and the clustering results are obtained through iterative expansion. For example, if a vehicle has completed a large number of delivery tasks in a specific area in the past month, the area will be identified as the key delivery area of ​​the vehicle. It can be understood that other methods can also be used to determine the delivery area, which is not limited here.

[0081] S207: Calculate the delivery distance from the target delivery point to each key delivery area.

[0082] Among them, the delivery distance value refers to the actual driving distance from the target delivery point to the key delivery area.

[0083] The dispatching system needs to calculate the accurate delivery distance to support vehicle matching decisions. Specifically, the system first obtains the coordinates of the target delivery point and the center point of each key delivery area; then builds the road network topology based on the road network data; then considers the real-time road conditions and road grades for path planning; and finally calculates the actual delivery distance value.

[0084] In some embodiments, the calculation of the delivery distance can be achieved in a variety of ways. The system uses an improved Dijkstra shortest path algorithm for distance calculation, defines the road network graph G=(V, E), and the edge weight w(e)=α×l(e)+β×t(e), where l(e) is the length of the road segment, t(e) is the travel time, and α and β are weight coefficients. The optimal path is solved by the dynamic programming method, and the A* heuristic search is introduced to improve the calculation efficiency. In practical applications, for example, from distribution point A to key distribution area B, the system will consider multiple possible paths and select the route with the lowest comprehensive cost as the final distance value. It can be understood that other methods can also be used to achieve distance calculation, which is not limited here.

[0085] S208. Sort the delivery vehicles according to the delivery distance values, and select a preset number of vehicles with the smallest delivery distance values ​​as candidate delivery vehicles.

[0086] Among them, the preset number represents the upper limit of the number of candidate vehicles set by the system; candidate delivery vehicles refer to vehicles that have been preliminarily screened out as being likely to perform delivery tasks.

[0087] The dispatch system needs to select the best candidate vehicle based on the distance value. Specifically, the system first sorts all delivery vehicles in ascending order according to the delivery distance value; then removes vehicles that exceed the maximum delivery distance threshold; then intercepts according to the candidate number N preset by the system; and finally marks the selected vehicle as a candidate delivery vehicle.

[0088] In some embodiments, the selection of candidate vehicles can be achieved in a variety of ways. The system adopts a multi-objective sorting algorithm, taking into account the distance value d and other factors, and constructs a scoring function S=w1×(1 / d)+w2×f(h)+w3×g(r), where h is the historical service quality indicator, r is the real-time response capability indicator, and w1, w2, and w3 are weight coefficients. Sort the vehicles according to the score, and select the N vehicles with the highest scores as candidates. For example, when the system presets the number of candidates to 3, the 3 vehicles with the highest comprehensive scores will be selected from all vehicles that meet the distance requirements as candidate delivery vehicles. It can be understood that other methods can also be used to screen candidate vehicles, which are not limited here.

[0089] S209: Based on the target delivery time and vehicle cargo information, candidate delivery vehicles are screened to obtain a target delivery vehicle.

[0090] Referring to step S105, the dispatching system determines the target delivery vehicle.

[0091] In some embodiments, the scheduling system obtains the target volume and target weight of the target cargo; determines the remaining available volume and remaining carry-weight of each candidate delivery vehicle based on the vehicle cargo information; calculates a first ratio of the target volume of the target cargo to the remaining available volume, and a second ratio of the target weight of the target cargo to the remaining carry-weight; if the first ratio and the second ratio are both less than a preset threshold, the corresponding candidate delivery vehicle is determined as the target delivery vehicle; and multiple candidate delivery vehicles that meet the conditions are sorted according to the first ratio and the second ratio, and the vehicle with the smallest ratio is selected as the target delivery vehicle.

[0092] Among them, the target volume refers to the space occupied by the goods to be delivered; the target weight represents the actual mass value of the goods; the cargo information refers to the vehicle's loading capacity parameters; the remaining available volume represents the remaining space in the compartment; the remaining carryable weight is used to represent the vehicle's remaining load-bearing capacity; the first ratio refers to the volume occupancy ratio; the second ratio represents the weight occupancy ratio.

[0093] The dispatching system needs to verify the feasibility of loading when matching vehicles. Specifically, the system first obtains the volume and weight data of the goods to be delivered; then queries the current loading status of the candidate vehicles, calculates the remaining available space and load-bearing capacity; then calculates the occupancy ratio of the cargo volume and weight relative to the remaining capacity of the vehicle; then compares these ratios with the safety thresholds preset by the system; finally, based on the comparison results, the system selects suitable vehicles and prioritizes them according to the matching degree of the remaining capacity.

[0094] In some embodiments, vehicle matching can be achieved in a variety of ways: the system uses a multi-dimensional constraint optimization model for vehicle-cargo matching. A decision variable xij is established to indicate whether cargo i is assigned to vehicle j. The constraints include volume constraint Σ(vi×xij)≤Vj and weight constraint Σ(wi×xij)≤Wj, where vi and wi are the volume and weight of the cargo, respectively, and Vj and Wj are the remaining capacity of vehicle j. The objective function is designed to minimize the overall loading rate difference min(Σ|vi / Vj-wi / Wj|).

[0095] The system can also use heuristic algorithms for fast matching. Using the best fitness priority strategy, calculate the fitness score S = α × (1-v / V) + β × (1-w / W) of each candidate vehicle, where α and β are weight coefficients. For example, for a batch of 1.5 cubic meters and 0.8 tons of goods, the system will give priority to vehicles with a remaining volume of 2 cubic meters and a remaining load of 1 ton, rather than large vehicles with a volume much larger than the demand. It is understandable that other methods can also be used to achieve vehicle-cargo matching, which is not limited here.

[0096] S210: Calculate the estimated mileage of the target delivery vehicle from the current location to the target delivery point.

[0097] Among them, the estimated mileage refers to the total distance the delivery vehicle is expected to travel to complete the delivery task; the current location refers to the real-time geographic coordinate point of the vehicle.

[0098] The dispatching system needs to calculate the precise mileage for the determined delivery vehicles. Specifically, the system first obtains the real-time GPS positioning information of the vehicle; then builds a road network model based on the electronic map and real-time traffic data; then performs multi-scheme path planning and optimization comparison; and finally outputs the optimal path plan and the corresponding estimated mileage.

[0099] In some embodiments, the calculation of mileage can be achieved in a variety of ways. The system uses a dynamic path planning algorithm to establish a time-varying road network model G(t). The road segment travel time function T(e, t) changes with time, and the traffic status at different time periods is described by a piecewise function. Combined with real-time road conditions, the comprehensive driving cost C=Σ(di×vi(t)) is calculated, where di is the road segment distance and vi(t) is the average driving speed of the road segment at time t. For example, the system may plan different driving routes for the same delivery task based on the difference in road conditions during peak hours in the morning and evening, so as to obtain a more accurate mileage estimate. It is understandable that other methods can also be used to achieve mileage calculation, which is not limited here.

[0100] S211. Extract vehicle information of the target delivery vehicle and determine the mileage cost of the estimated mileage.

[0101] Among them, vehicle information refers to the basic attributes and operating parameters of the delivery vehicle; mileage cost refers to the operating expenditure per unit distance.

[0102] The dispatching system needs to calculate accurate mileage costs based on vehicle characteristics. Specifically, the system first reads the basic information of the vehicle, including the model, energy consumption standard, etc., then analyzes the actual energy consumption level in combination with historical operating data, then calculates the comprehensive cost including fuel, maintenance, and depreciation, and finally derives the standard mileage cost per kilometer.

[0103] In some embodiments, the calculation of mileage costs will be different based on the ownership of the vehicle. For vehicles provided by logistics stations, the system uses a full life cycle cost calculation model, with total cost TC=FC+MC+DC+OC, where FC is the fuel cost function (related to mileage and load), MC is the maintenance cost function (related to the length of use), DC is the depreciation cost function (related to the value of the vehicle), and OC is other operating costs; the total cost of renting a car is the total rental.

[0104] For example, for a 4.2-meter box truck that has been in operation for two years at a logistics site, the system will take into account factors such as its actual fuel consumption level, regular maintenance expenses, vehicle depreciation, etc., and calculate a reasonable mileage cost standard; for a short-term leased truck, the rental fee only needs to be calculated based on the rental time.

[0105] S212. Obtain the volume rate and weight rate of the target cargo.

[0106] Among them, the volume rate refers to the base price for delivery of goods per unit volume; the weight rate refers to the base price for delivery of goods per unit weight.

[0107] The dispatching system needs to determine reasonable charging standards based on the characteristics of the goods. Specifically, the system first reads the basic attributes of the goods, such as volume and weight; then queries the benchmark rate standard for the corresponding type of goods; then adjusts the rate based on factors such as season and region; and finally determines the specific rate applicable to the current delivery task.

[0108] In some embodiments, the rate can be determined in a variety of ways. The system adopts a multi-dimensional pricing model, and the base rate BR = (max (v × rv, w × rw)) × K, where v is the volume of the goods, w is the weight of the goods, rv is the volume unit price, rw is the weight unit price, and K is the comprehensive adjustment coefficient. The adjustment coefficient K is further subdivided into K = k1 × k2 × k3, corresponding to seasonal, regional and special requirements and other influencing factors.

[0109] For example, for a batch of fresh products that need to be transported at a constant temperature, the system will add the temperature control service rate to the base rate to arrive at the final billing standard. It is understandable that other methods can also be used to determine the rate, which is not limited here.

[0110] S213. Calculate the delivery cost based on the volume rate, weight rate and mileage cost.

[0111] Among them, distribution cost refers to the overall economic expenditure to complete the distribution task.

[0112] The dispatching system needs to combine multiple factors to calculate the accurate delivery cost. Specifically, the system first calculates the basic freight based on the cargo rate; then calculates the path cost based on the estimated mileage and mileage cost; then adds other fees such as fuel surcharges and special service fees; and finally summarizes the complete delivery cost.

[0113] In some embodiments, the calculation of distribution costs can be achieved in a variety of ways. The system adopts a hierarchical cost calculation model, with total cost TC=BF+DC+AC, where the basic freight BF=max(V×Rv, W×Rw), mileage cost DC=L×CL, and additional cost AC=Σ(Si×Pi). Where V is volume, W is weight, Rv is volume rate, Rw is weight rate, L is mileage, CL is mileage cost, Si is various special service indicators, and Pi is the corresponding rate.

[0114] For example, for a shipment that requires cold chain delivery, the system will calculate the basic freight and mileage costs, and then add additional fees such as temperature control service fees and insurance fees to ultimately arrive at the complete delivery cost. It is understandable that other methods can also be used to achieve cost calculation, which is not limited here.

[0115] S214: Generate a delivery task list corresponding to the target delivery vehicle and the target goods, and export the delivery task list.

[0116] Referring to step S106, the dispatching system generates a delivery task list.

[0117] In some embodiments, the generation of the distribution plan can be achieved in a variety of ways. The system uses a multi-objective optimization model to evaluate the quality of the plan and constructs a scoring function S=Σ(wi×fi(x)), where wi is the weight of each indicator and fi(x) is the normalized indicator score function.

[0118] Evaluation indicators include cost-effectiveness ratio: R=Revenue / Cost; Timeliness: T=Delivery_Time / Standard_Time; Vehicle Compatibility: V=Match_Score / Max_Score; …and other dimensions.

[0119] For example, when the system matches delivery resources for a batch of fresh goods, it will give priority to vehicles with temperature control functions, reasonable delivery costs, and that can meet time requirements, forming a delivery plan that balances the needs of all parties. It is understandable that other methods can also be used to achieve plan generation, which is not limited here.

[0120] The system uses the analytic hierarchy process (AHP) to determine the weights of each indicator in the scoring function. First, construct the judgment matrix A, where the matrix element aij represents the importance ratio of indicator i to indicator j. Through the eigenvalue equation: (A-λI)W=0 solves the eigenvector W corresponding to the maximum eigenvalue λmax, and obtains the weights of each indicator after normalization. To ensure the rationality of the weights, it is necessary to calculate the consistency ratio CR=CI / RI, where: CI = (λmax-n) / (n-1) is the consistency index, and RI is the random consistency index.

[0121] When CR<0.1, the judgment matrix is ​​considered to have satisfactory consistency. For example, for a scoring system that includes three indicators: cost-effectiveness, timeliness, and vehicle adaptability, the judgment matrix obtained through expert evaluation may be [[1, 2, 3], [1 / 2, 1, 2], [1 / 3, 1 / 2, 1]], and the weight vector [0.5, 0.3, 0.2] is calculated, indicating that cost-effectiveness is the most important. The system will also regularly adjust the weight value based on actual operating data to ensure the scientific nature of the scoring results.

[0122] In some embodiments, the system also formats the data of the delivery plan, and then pushes it to the target vehicle terminal through the wireless communication network, then waits for the terminal device to confirm receipt, and finally updates the plan status to have been issued.

[0123] Specifically, the system implements reliable message push based on the publish-subscribe model. The message producer encapsulates the delivery plan into a message body and sends it to the message queue cluster. The message body contains fields such as the unique identifier messageId, business data payload, and priority. The consistent hashing algorithm is used to shard and route the messageId to ensure that messages from the same vehicle are delivered to a fixed queue partition. The message proxy cluster uses the Raft protocol to ensure data consistency, elects a Leader node to be responsible for message writing, and configures at least 3 copies. The message consumer (on-board terminal) uses long polling to obtain new messages, and sets a batch confirmation mechanism to reduce network overhead. Message persistence and the ACK mechanism ensure that each message is successfully processed at least once. For important delivery plan messages, the system sets the message priority to "high" to ensure priority processing.

[0124] In addition, the system uses a multi-level fault-tolerant mechanism to handle abnormal reception of the solution. When it is detected that the terminal does not send an ACK confirmation on time, the message is first retransmitted. The retransmission interval uses an exponential backoff algorithm with an initial interval of 1 second and a maximum interval of 5 minutes, with a maximum of 5 retries. At the same time, the backup communication channel is started, including SMS notification and manual confirmation by the dispatch center. If the retransmission still fails, the system will trigger the task fuse, transfer the delivery plan to the pending queue and notify the emergency handling personnel. The system will record a detailed error log, including the cause of the failure, network status, terminal status and other information for subsequent analysis and prevention. When the network is restored, the system will give priority to the previously interrupted delivery plan to ensure that the task can be delivered in time. For example, a vehicle may experience a signal interruption when passing through a tunnel. The system will immediately re-push the delivery plan after the vehicle exits the tunnel and ask the driver to confirm.

[0125] In some embodiments, the dispatch system obtains the current location information and initial delivery route of the target delivery vehicle; calculates the connection distance and estimated driving time between each delivery point in the initial delivery route based on the current location information, real-time road conditions and historical traffic data; reorders each delivery point based on the connection distance and estimated driving time, combined with the target delivery time, to generate an optimized delivery route; and sends the optimized delivery route to the on-board terminal of the target delivery vehicle.

[0126] Among them, the current location information refers to the real-time geographic coordinates of the vehicle; the initial delivery route refers to the initially planned delivery route; the connection distance refers to the actual mileage between delivery points; the estimated driving time is used to indicate the estimated time required to pass a road section; the real-time traffic information represents the current traffic status data; the historical traffic data refers to the accumulated traffic pattern information; the on-board terminal refers to the information display device on the vehicle, which can be a device such as the driver’s mobile phone, with a specific App installed on the device for information interaction with the dispatch system.

[0127] After determining the delivery vehicle, the dispatch system needs to optimize the delivery route. Specifically, the system first obtains the vehicle's GPS location information and the initially planned delivery route; then obtains road condition information from multiple data sources, including real-time traffic status, historical traffic data, and weather impacts; then calculates the actual travel distance and time consumption between each delivery point; then, combined with the delivery time window required by the customer, uses an algorithm to optimize and adjust the delivery sequence; finally, generates a new delivery route plan and pushes it to the vehicle-mounted device.

[0128] In some embodiments, the scheduling system also receives a return list, extracts the return information and the original delivery order information from the return list; matches the corresponding delivery vehicle and delivery route according to the original delivery order information; calculates the distance and estimated travel time from the return location to the current location of the vehicle; if the route optimization condition is met, inserts the return task into the current delivery route and re-plans the route; simultaneously updates the load status and available volume of the vehicle; generates a return handling plan and pushes it to the on-vehicle terminal.

[0129] Among them, the return list refers to a list of information recording the goods to be returned; the return information represents the basic attributes of the returned goods; the original delivery order information refers to the relevant records of the first delivery of the goods; the route optimization condition is used to represent the judgment criteria for adjusting the delivery route; the load status refers to the current loading situation of the vehicle; the return handling plan represents the specific implementation plan for the return task.

[0130] After receiving the return request, the scheduling system needs to reasonably arrange the return handling. Specifically, the system first receives the return application and extracts the relevant information; then matches the original delivery record to find the corresponding delivery vehicle; then evaluates whether the return location is within the vehicle delivery range and meets the route optimization condition; then calculates the impact on the existing delivery plan after adding the return task; finally generates a new delivery plan including the return handling and notifies the relevant personnel.

[0131] In some embodiments, the return handling can be implemented in various ways: the system uses an insertion planning method to handle the return task. Construct an insertion cost function: IC(i, j, k) = δd + γt + λw, where δd represents the additional travel distance increment, γt represents the time delay impact, λw represents the waiting time cost, and find the optimal insertion position in the existing delivery sequence that satisfies: IC(i, j, k) < threshold.

[0132] threshold is a dynamically adjusted acceptance threshold used to judge whether the return point insertion is feasible. Let me explain in detail the determination and usage mechanism of this threshold: the calculation method of threshold is: threshold = base_cost × (1 + α × urgency_factor) × (1 + β × load_factor) × (1 + γ × time_factor); Among them, base_cost is the benchmark cost, which is obtained through statistical analysis of historical delivery data, and indicates the upper limit of acceptable insertion cost under normal circumstances, such as 10 kilometers or 30 minutes; urgency_factor is the urgency factor, which is related to the processing time limit of the return task. The more urgent it is, the larger the threshold is, and a higher insertion cost is allowed; load_factor is the load factor, which is related to the current load factor of the vehicle. The lower the load factor, the larger the threshold is, indicating that there is more room for adjustment; time_factor is the time factor, which is related to the busyness of delivery in the current period. The threshold is larger during non-peak hours and tighter during peak hours; α, β, and γ are weight coefficients used to adjust the influence of each factor.

[0133] The system can also handle returns by reorganizing tasks. All the delivery points and return points to be processed are reconstructed into a task network G (V, E), where the vertex set V contains all service points and the edge set E represents a feasible connection path. A new delivery plan is generated by solving the vehicle routing problem with a time window constraint (VRPTW).

[0134] For example, when receiving a return request from a certain community, the system will give priority to arranging vehicles that need to deliver in the community on the same day to handle the return, so as to improve the efficiency of transportation capacity utilization. It is understandable that other methods can also be used to implement the return processing process, which is not limited here.

[0135] The insertion planning method achieves the optimal insertion of the return task through iterative evaluation and optimization. First, define the insertion position evaluation function: E(p,q)=w1×ΔD+w2×ΔT+w3×ΔW, where p represents the return point and q represents the adjacent point pair (i, j) in the existing delivery sequence; ΔD represents the total driving distance increment caused by the insertion, which is D(i, p)+D(p, j)-D(i, j); D(i, p) represents the actual driving distance from the delivery point i to the return point p; D(p, j) represents the actual driving distance from the return point p to the next delivery point j; D(i, j) represents the distance from point i directly to point j in the original delivery route; it can be obtained that ΔD represents the additional driving distance after the return point is inserted, that is, the detour distance.

[0136] ΔT represents the time delay impact on subsequent delivery points, which is Σmax(0, AT(k)-LT(k)); ΔW represents the additional waiting time cost, which is max(0, ET(p)-AT(p)); AT(k) represents the actual arrival time of delivery point k; LT(k) represents the latest delivery time required by delivery point k; ET(p) represents the earliest time when return point p can start service; AT(p) represents the estimated time for the vehicle to arrive at return point p.

[0137] For example, assuming that the original planned delivery route is A→B→C, and now it is necessary to insert a return point P between A and B, if D(A, B) = 10km, D(A, P) = 7km, D(P, B) = 5km, then: ΔD=7+5-10=2km (additional 2km driving distance); If the original planned arrival time of point B is AT(B) = 10:00, and the latest required time is LT(B) = 10:30, the actual arrival time becomes AT(B) = 10:15 after inserting point P; then the delay effect on point B is max(0, 10:15-10:30) = 0 (within the latest deadline); If the earliest service time of point P is ET(P) = 9:30 and the actual arrival time is AT(P) = 9:15, then the waiting time cost ΔW = max(0, 9:30-9:15) = 15 minutes.

[0138] w1, w2, and w3 are weight coefficients of various indicators; w1 is the weight coefficient of distance increment, usually set between 0.3-0.5, which is related to fuel consumption cost; w2 is the weight coefficient of time delay, usually set between 0.3-0.4, reflecting service quality requirements; w3 is the weight coefficient of waiting time, usually set between 0.2-0.3, reflecting time utilization efficiency. The specific values ​​of these weight coefficients can be adjusted according to actual business conditions. For example, when oil prices rise, the value of w1 can be appropriately increased; for high-priority customers, the value of w2 can be increased; in areas with high labor costs, the value of w3 can be increased.

[0139] The system calculates the evaluation values ​​of all feasible insertion positions in turn, and selects the position with the smallest evaluation value and that satisfies the vehicle load constraint C(v)≤Cmax and the time window constraint AT(k)≤LT(k) as the insertion point.

[0140] For example, when a return task needs to be processed within 5 kilometers of the current location of the delivery vehicle, the system will calculate the incremental cost of inserting the return point between adjacent delivery points. If an insertion plan only increases the driving distance by 3 kilometers and the impact on the existing delivery time is within 10 minutes, then this plan is likely to be selected.

[0141] The system uses a multi-layer network structure to represent the relationship between delivery and return tasks. The vertex set V includes the starting point depot, the delivery point delivery_set and the return point return_set. Each vertex v has attributes such as service time window [ETv, LTv], service time STv and cargo volume Qv (return is a negative value). The edge set E represents the feasible connection between vertices, and the edge weight W (vi, vj) is represented by a composite function: W(vi, vj) = α×D(vi, vj) / Dmax+β×|ATj-PTj| / Tmax+γ×Load(vi, vj) / Qmax, where D(vi, vj) represents the actual driving distance, ATj represents the arrival time, PTj represents the expected time, Load(vi, vj) represents the interval load rate, and α, β, and γ are weight coefficients. Based on this network structure, the system obtains the optimal distribution solution by solving the shortest path problem with resource constraints.

[0142] For example, for a task network consisting of 10 delivery points and 2 return points, the system will find a traversal path with the lowest total cost and meeting all service requirements based on multi-dimensional constraints such as distance, time and load.

[0143] Among them, the genetic algorithm (GA) optimizes the delivery route by simulating the natural evolution process. Each individual in the population represents a feasible delivery sequence, and the fitness function fitness (x) comprehensively considers the total distance, time window default degree and load balance. The crossover operator uses sequence order crossover (OX) to maintain the relative order of delivery points, and the mutation operator improves the quality of the solution through 2-opt local search.

[0144] The algorithm parameter settings include population size N=100, crossover probability pc=0.8, mutation probability pm=0.1, and the termination condition is reaching the maximum number of iterations or there is no improvement in the optimal solution for consecutive K generations. Taboo search (TS) avoids local optimality through a memory mechanism. During the search process, a taboo table tabu_list is maintained to record recently visited solutions. Neighborhood actions include point exchange move (i, j) and segment rearrangement move (i, j, k). The amnesty rule allows tabooed excellence to be released to avoid missing the global optimal solution.

[0145] Simulated annealing (SA) achieves global optimization through a temperature control mechanism. The temperature decay function is T(k) = T0×αk, and the acceptance probability is P(ΔE) = exp(-ΔE / T).

[0146] Preferably, an appropriate optimization strategy can be selected according to the specific problem scale and time requirements. For example, for dynamic distribution scenarios that require real-time response, a taboo search algorithm with higher computational efficiency can be selected; while for static large-scale path planning problems, a genetic algorithm with better solution quality can be selected.

[0147] In the embodiment of the present application, due to the use of an intelligent dispatching system based on multi-dimensional constraints, including special cargo identification, vehicle intelligent matching, dynamic path optimization and return task processing, etc., the intelligent level of logistics distribution can be comprehensively improved, and the problems existing in the traditional dispatching system in special cargo processing, vehicle matching, path planning and dynamic task adjustment can be effectively solved, thereby achieving the improvement of distribution efficiency, effective control of operating costs, comprehensive improvement of customer satisfaction and optimal configuration of logistics resources. Not only the quality of distribution services is ensured, but also the reasonable control of logistics costs is achieved.

[0148] The scheduling system in the embodiment of the present invention is described below from the perspective of hardware processing. Figure 3 , which is a schematic diagram of a physical device structure of a scheduling system in an embodiment of the present application.

[0149] It should be noted that Figure 3 The structure of the scheduling system shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0150] like Figure 3 As shown, the scheduling system includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage part 308 to the random access memory (RAM) 303, such as executing the method described in the above embodiment. In RAM 303, various programs and data required for system operation are also stored. CPU 301, ROM 302 and RAM 303 are connected to each other through bus 304. Input / output (I / O) interface 305 is also connected to bus 304.

[0151] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a button switch, etc.; an output section 307 including a liquid crystal display (LCD) and an audio output device, an indicator light, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read therefrom is installed into the storage section 308 as needed.

[0152] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 309, and / or installed from a removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the present invention are performed.

[0153] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device.

[0154] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from that marked in the accompanying drawings.

[0155] Specifically, the scheduling system of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, the logistics city distribution automated scheduling method provided in the above embodiment is implemented.

[0156] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the scheduling system described in the above embodiment; or may exist independently without being assembled into the scheduling system. The above storage medium carries one or more computer programs, and when the above one or more computer programs are executed by a processor of the scheduling system, the scheduling system implements the logistics city distribution automatic scheduling method provided in the above embodiment.

[0157] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0158] As used in the above embodiments, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted to mean "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0159] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

Claims

1. A logistics city distribution automated dispatching method, characterized in that: Applied to a dispatching system, the method comprises: Receive a goods form imported by a user, extract goods information in the goods form and delivery time information and delivery address information corresponding to the goods information; Filtering the cargo information according to the delivery time information to obtain a plurality of target cargoes to be delivered, and respectively determining the target delivery time and target delivery address of the target cargoes; Acquire vehicle information registered in the dispatching system, and determine vehicle cargo information and vehicle route information based on the vehicle information; Determining a plurality of candidate delivery vehicles for each target cargo according to the target delivery address and the vehicle route information; Based on the target delivery time and the vehicle cargo information, the candidate delivery vehicles are screened to obtain a target delivery vehicle; Generate a delivery task list corresponding to the target delivery vehicle and the target goods, and export the delivery task list.

2. The method according to claim 1, characterized in that The step of determining a plurality of candidate delivery vehicles for each target cargo according to the target delivery address and the vehicle route information specifically includes: Obtain electronic map information of multiple preset geographical areas in the dispatching system; Marking the target delivery address on the electronic map information to obtain a target delivery point; Acquire the historical delivery routes in the vehicle route information, and determine the key delivery area of ​​each delivery vehicle according to the historical delivery routes; Calculate the delivery distance from the target delivery point to each of the key delivery areas; The delivery vehicles are sorted according to the delivery distance values, and a preset number of vehicles with the smallest delivery distance values ​​are selected as candidate delivery vehicles.

3. The method according to claim 2, characterized in that Before the step of generating a delivery task list corresponding to the target delivery vehicle and the target goods and exporting the delivery task list, the method further includes: Calculate the estimated mileage of the target delivery vehicle from the current location to the target delivery point; Extracting vehicle information of the target delivery vehicle and determining the mileage cost of the estimated mileage; Obtaining the volume rate and weight rate of the target cargo; The delivery cost is calculated based on the volume rate, the weight rate and the mileage cost.

4. The method according to claim 1, characterized in that: Before the step of screening the cargo information according to the delivery time information to obtain a plurality of target cargoes to be delivered, and respectively determining the target delivery time and target delivery address of the target cargoes, the method further includes: Extracting cargo information from the cargo form and pickup information corresponding to the cargo information; When the pickup information is self-pickup, the corresponding goods to be picked up by self-pickup are removed from the goods list; The cargo information of the cargo to be picked up is counted, and a self-pickup cargo form is generated.

5. The method according to claim 1, characterized in that The step of screening the candidate delivery vehicles based on the target delivery time and the vehicle cargo information to obtain the target delivery vehicle specifically includes: Obtaining a target volume and a target weight of the target cargo; Determining the remaining available volume and remaining loadable weight of each candidate delivery vehicle according to the vehicle cargo information; Calculating a first ratio of the target volume of the target cargo to the remaining available volume, and a second ratio of the target weight of the target cargo to the remaining loadable weight; If both the first ratio and the second ratio are less than a preset threshold, the corresponding candidate delivery vehicle is determined as a target delivery vehicle; A plurality of candidate delivery vehicles that meet the conditions are sorted according to the first ratio and the second ratio, and a vehicle with the smallest ratio is selected as a target delivery vehicle.

6. The method according to claim 1, characterized in that After the step of receiving the goods form imported by the user and extracting the goods information of the goods form and the delivery time information and delivery address information corresponding to the goods information, the method further includes: Extracting cargo information from the cargo form and cargo type identification corresponding to the cargo information; Determine the special cargo in the cargo form according to the cargo type identifier; the special cargo includes customs declaration cargo, cold chain cargo, fragile cargo and dangerous cargo; When the special cargo is customs declaration cargo, determining the time required for customs declaration, and adding the time required for customs declaration to the target delivery time of the customs declaration cargo; When the special goods are cold chain goods, vehicles with temperature control equipment are selected as candidate delivery vehicles, and the delivery cost of the cold chain goods is adjusted according to the operating cost of the temperature control equipment; When the special cargo is fragile cargo or dangerous cargo, the transportation qualification requirements of the delivery vehicle are obtained, and the vehicles whose transportation qualification requirements do not meet the preset requirements are eliminated from the candidate delivery vehicles.

7. The method according to claim 1, characterized in that After the step of generating a delivery task list corresponding to the target delivery vehicle and the target goods, and exporting the delivery task list, the method further includes: Obtaining the current location information and initial delivery route of the target delivery vehicle; Calculate the connection distance and estimated travel time between each delivery point in the initial delivery route based on the current location information, real-time traffic information and historical traffic data; Reorder the delivery points according to the connection distance and the estimated travel time, combined with the target delivery time, to generate an optimized delivery route; The optimized delivery route is sent to the on-board terminal of the target delivery vehicle.

8. A scheduling system, characterized in that: The scheduling system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the scheduling system to execute the method described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that: When the instruction is executed on a scheduling system, the scheduling system is caused to execute the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product is executed on a scheduling system, the scheduling system is caused to execute the method according to any one of claims 1 to 7.

Images