Vehicle scheduling management method and system based on artificial intelligence

Abstract

The invention relates to the technical field of vehicle scheduling management, and particularly discloses a vehicle scheduling management method and system based on artificial intelligence, which are particularly suitable for decomposition and distribution of complex tasks. The complex tasks are analyzed through an artificial intelligence technology, whether the complex tasks can be independently completed by a single vehicle is judged, and if not, the vehicle scheduling management is completed. The method comprises the following steps of: if the vehicle scheduling task is a task, decomposing the task into a plurality of sub-tasks according to the constraint condition of the task, and identifying an efficient cooperative driver combination by screening the state of the vehicle and the social network of the driver so as to optimally distribute each sub-task, thereby improving the efficiency and accuracy of the vehicle scheduling system. The method is especially suitable for scheduling scenes with high task complexity and uneven resource distribution.

Description

Technical Field

[0001] The invention belongs to the technical field of vehicle dispatching management, and relates to a vehicle dispatching management method and system based on artificial intelligence. Background Art

[0002] In the modern logistics and transportation industry, faced with increasingly complex and diverse mission requirements, a single vehicle often finds it difficult to independently complete certain large or complex tasks. Therefore, rationally splitting complex tasks and efficiently allocating them to multiple vehicles for execution has become the key to improving overall operational efficiency and flexibility. In this way, companies can not only optimize resource allocation and ensure that tasks are completed on time and efficiently, but also improve response speed and service quality, thereby maintaining a leading position in a highly competitive market environment.

[0003] In current technology, vehicle dispatch for complex orders usually relies on preset rules and simple algorithms, mainly based on the idle status and geographical location of the vehicle. However, this method is incapable of handling multiple constraints and dynamically changing tasks, often resulting in waste of resources, slow response speed and inability to effectively deal with emergencies. In addition, the lack of comprehensive evaluation of vehicles and drivers leads to inaccurate task allocation, affecting overall efficiency and customer satisfaction. Summary of the invention

[0004] In view of the above problems existing in the prior art, the present invention provides a vehicle dispatching management method and system based on artificial intelligence to solve the above technical problems.

[0005] In order to achieve the above purpose and other purposes, the technical solution adopted by the present invention is as follows: The first aspect of the present invention provides a vehicle dispatch management method based on artificial intelligence, the method comprising the following steps: The vehicle dispatching system obtains the task parameters of the task to be assigned, analyzes the task to be assigned, and determines whether the task to be assigned can be completed independently by a single idle vehicle; if it is determined that the task to be assigned can be completed independently by a single idle vehicle, the best execution vehicle for the task to be assigned is selected; If it is determined that the task to be assigned cannot be completed independently by a single idle vehicle, the vehicle dispatching system decomposes the task to be assigned into multiple subtasks according to the constraints of the task to be assigned; Based on the established screening conditions, the best execution vehicles corresponding to each subtask are comprehensively screened out; The license plate number of the best performing vehicle corresponding to each subtask is obtained, and the task information of each best performing vehicle that needs to complete the corresponding subtask is fed back to the display interface of each best performing vehicle through the background terminal.

[0006] The task parameters of the task to be assigned include the task starting point location, the task end point location, the task completion deadline and the task load requirement capacity.

[0007] Determine whether the task to be assigned can be completed independently by a single idle vehicle. The specific determination process is as follows: Obtain the maximum load capacity of each idle vehicle at the current time point, compare it with the task load requirement capacity of the task to be assigned, eliminate the idle vehicles whose maximum load capacity at the current time point is less than the task load requirement capacity of the task to be assigned, and then record the remaining idle vehicles as the initial screening idle vehicles; Get the position of each pre-screened idle vehicle at the current time point, compare it with the starting point of the task to be assigned, and get the distance between each pre-screened idle vehicle and the starting point of the task to be assigned; get the remaining power of each pre-screened idle vehicle at the current time point , and then judge that the power of each initially screened idle vehicle to complete the task to be assigned meets the evaluation value , 1 means compliance, 0 means non-compliance, c is the number of each initially screened idle vehicle, is the total estimated power consumption of the cth pre-screened idle vehicle from the current time point to the task end point of the task to be assigned. The specific calculation formula is: , is the power consumption of the cth initially screened idle vehicle during the unloaded segment of the task to be assigned, The power consumption of the cth initially screened idle vehicle corresponding to the load segment of the task to be assigned; in ; ; is the distance between the cth initially screened idle vehicle and the starting point corresponding to the task to be assigned, is the total distance that the cth initially screened idle vehicle needs to travel to complete the task to be assigned, The power consumption per unit distance of the cth initially screened idle vehicle when it is unloaded; The power consumption per unit distance when the cth vehicle is initially screened for idle load; is the influence coefficient of speed reduction when the cth primary screening idle vehicle is loaded, is the influence coefficient of the cth idle vehicle whose power consumption rate slows down as the power consumption rate slows down, and and All are less than 1; Then calculate the total time consumed by each initially screened idle vehicle to complete the task to be assigned , are the maximum driving speeds of the cth initially screened idle vehicle for the unloaded and fully loaded sections of the task to be assigned, respectively. is the minimum safe driving speed of the cth idle vehicle in the initial screening, is the estimated service time of the cth initially screened idle vehicle at the task starting point, The estimated charging time of the cth initially screened idle vehicle on the way; are the road condition influence coefficients of the cth initially screened idle vehicle for the unloaded section and the fully loaded section of the task to be assigned; The total time consumed by each initially screened idle vehicle to complete the task to be assigned is added to the current time point to obtain the completion time of each initially screened idle vehicle corresponding to the task to be assigned, and the completion time is compared with the task completion deadline of the task to be assigned. If there is an initially screened idle vehicle whose completion time corresponding to the task to be assigned is before the task completion deadline of the task to be assigned, it is determined that the task to be assigned can be completed independently by a single idle vehicle; if the completion time of each initially screened idle vehicle corresponding to the task to be assigned is after the task completion deadline of the task to be assigned, it is determined that the task to be assigned cannot be completed independently by a single idle vehicle.

[0008] If it is determined that the task to be assigned can be completed independently by a single idle vehicle, the best vehicle to perform the task to be assigned is selected. The specific selection process is as follows: The completion time of each preliminarily screened idle vehicle corresponding to the task to be assigned is obtained, and they are arranged in chronological order from small to large, and the preliminarily screened idle vehicle with the first completion time of the task to be assigned is used as the best execution vehicle for the task to be assigned.

[0009] The screening conditions include vehicle screening conditions and driver screening conditions.

[0010] According to the established vehicle screening conditions, available vehicles corresponding to each subtask under the vehicle screening conditions are screened out. The specific screening process is as follows: Calculate the fitness score of each candidate vehicle for each subtask , i is the number of each candidate vehicle, j is the number of each subtask, are weight coefficients set to balance the importance of different factors. is the distance adaptation score of the i-th candidate vehicle corresponding to the j-th subtask, is the load adaptation score of the i-th candidate vehicle corresponding to the j-th subtask, is the time adaptation score of the i-th candidate vehicle corresponding to the j-th subtask, is the matching score between the vehicle status of the i-th candidate vehicle and the j-th subtask requirement; in , is the actual driving distance between the position of the i-th candidate vehicle and the starting point of the j-th subtask, is the normalization factor of the maximum distance in the subtask corresponding to the candidate vehicle position; , is the remaining load capacity of the i-th candidate vehicle, is the task load requirement capacity of the jth subtask; , is the task completion deadline of the j-th subtask, is the estimated travel time for the i-th candidate vehicle to reach the starting point of the j-th subtask, is the time required for the i-th candidate vehicle to execute the j-th subtask, is the current time; , are the remaining battery power and health score of the i-th candidate vehicle, is the minimum battery power required for the i-th candidate vehicle to complete the j-th subtask, The highest health score among all candidate vehicles; The fitness score of each candidate vehicle corresponding to each subtask is compared with the set score lower limit threshold, and each candidate vehicle whose fitness score of each subtask is greater than the score lower limit threshold is used as each available vehicle corresponding to each subtask under the vehicle screening condition.

[0011] According to the established driver screening conditions, available vehicles corresponding to each subtask under the driver screening conditions are screened out. The specific screening process is as follows: Obtain the driver's historical information corresponding to each candidate vehicle, and then obtain the driver's historical task completion rate corresponding to each candidate vehicle , Drivers’ historical task execution efficiency and the number of each task type that the driver has historically performed; Based on the task type of each subtask and the number of task types that the driver of each candidate vehicle has performed in the past, the matching degree of the driver's historical task type corresponding to each candidate vehicle is calculated. ; According to the total number of subtasks J, obtain the total number of candidate vehicles M, where M>J; Randomly select J candidate vehicles from the total number of candidate vehicles for permutation and combination, thereby obtaining the total number of permutations and combinations of candidate vehicles H, and obtain the historical social interaction information, historical collaboration information, and historical task dependency information of the driver corresponding to each candidate vehicle in each candidate vehicle permutation and combination; Based on the historical social interaction information of the drivers of each candidate vehicle in each candidate vehicle combination, the interaction frequency of the drivers of each candidate vehicle in each candidate vehicle combination is extracted. , interaction quality , positive degree of interaction emotion , reciprocity and speed of information dissemination ; Thus, the social interaction score of the driver corresponding to each candidate vehicle in each candidate vehicle arrangement combination is obtained , To weigh the frequency, quality, and positivity of interactions, and to adjust the impact of these factors on the overall score, To balance the weights of reciprocity and information diffusion speed, we used them to adjust the impact of these factors on the overall score; is the average network connectivity of all drivers; r is the number of each candidate vehicle combination, and ; Based on the historical cooperation information of the drivers of each candidate vehicle in each candidate vehicle combination, the task alternation frequency of the drivers of each candidate vehicle in each candidate vehicle combination is extracted. , Dependence and resource sharing ; Thus, the task dependency score of each candidate vehicle corresponding to the driver in each candidate vehicle combination is obtained , is the maximum frequency of historical task alternation of the candidate vehicle driver, The historical maximum dependency of the candidate vehicle driver, are the calculation coefficient factors set respectively, ; Based on the historical task dependency information of the drivers of each candidate vehicle in each candidate vehicle combination, the number of tasks completed by the drivers of each candidate vehicle in each candidate vehicle combination is extracted. , Joint completion success rate , Collaboration Time and task quality rating ; Thus, the cooperation efficiency score of the corresponding driver of each candidate vehicle in each candidate vehicle arrangement combination is obtained , are the calculation weight factors set respectively, ; are the average number of common tasks and average task quality score of all drivers, is the minimum collaboration time of all drivers; According to the driver's historical task completion rate, driver's historical task execution efficiency and driver's historical task type matching degree corresponding to each candidate vehicle, the historical task completion rate, driver's historical task execution efficiency and driver's historical task type matching degree corresponding to each candidate vehicle in each candidate vehicle arrangement combination are obtained; using the weighted average method, the historical task completion rate, driver's historical task execution efficiency and driver's historical task type matching degree corresponding to each candidate vehicle in each candidate vehicle arrangement combination are multiplied to obtain the historical score of the driver corresponding to each candidate vehicle in each candidate vehicle arrangement combination; Using the weighted average method, the social interaction score, task dependency score and collaboration efficiency score of the driver corresponding to each candidate vehicle in each candidate vehicle arrangement combination are multiplied to obtain the social score of the driver corresponding to each candidate vehicle in each candidate vehicle arrangement combination; The weighted average method is used again to multiply the historical scores of the drivers of each candidate vehicle in each candidate vehicle arrangement by the social scores, thereby obtaining the comprehensive scores of the drivers of each candidate vehicle in each candidate vehicle arrangement, and then summing and averaging them to obtain the comprehensive scores of each candidate vehicle arrangement; The comprehensive score of each candidate vehicle combination is compared with the set score lower limit threshold, and each candidate vehicle in each candidate vehicle combination with a comprehensive score greater than the score lower limit threshold is used as each available vehicle corresponding to each subtask under the driver screening condition.

[0012] Comprehensively select the best performing vehicle for each subtask. The specific selection process includes: Record each candidate vehicle permutation combination whose comprehensive score is greater than the score lower limit threshold as each standard combination, arrange the comprehensive scores of each standard combination in descending order, obtain the license plate number of each candidate vehicle in the first standard combination, and obtain the license plate number of each available vehicle corresponding to each subtask under the vehicle screening condition. If the license plate number of each candidate vehicle in the first standard combination is within the set of license plate numbers of each available vehicle corresponding to each subtask under the vehicle screening condition, then each candidate vehicle in the first standard combination is used as the best execution vehicle corresponding to each subtask; otherwise, the second standard combination, the third, ... until the last standard combination are screened according to the screening method of each candidate vehicle in the first standard combination; If all standard combinations are not in the license plate number set of each available vehicle corresponding to each subtask under the vehicle screening condition after screening, then each available vehicle corresponding to each subtask under the vehicle screening condition will be used as the best execution vehicle corresponding to each subtask.

[0013] The second aspect of the present invention provides a vehicle dispatching and management system based on artificial intelligence, comprising: In the preliminary determination module, the vehicle dispatching system obtains the task parameters of the task to be assigned, analyzes the task to be assigned, and determines whether the task to be assigned can be completed independently by a single idle vehicle; if it is determined that the task to be assigned can be completed independently by a single idle vehicle, the best execution vehicle for the task to be assigned is selected; Task decomposition module: if it is determined that the task to be assigned cannot be completed independently by a single idle vehicle, the vehicle dispatching system will decompose the task to be assigned into multiple subtasks according to the constraints of the task to be assigned; The vehicle determination module comprehensively selects the best execution vehicle for each subtask based on the established screening conditions; The scheduling feedback module obtains the license plate number of the best performing vehicle corresponding to each subtask, and feeds back the task information of each best performing vehicle that needs to complete the corresponding subtask to the display interface of each best performing vehicle through the background terminal.

[0014] As described above, the vehicle dispatching management method and system based on artificial intelligence provided by the present invention have at least the following beneficial effects: The present invention provides an artificial intelligence-based vehicle dispatch management method and system, which helps to improve the efficiency and accuracy of task assignment by determining whether the task to be assigned can be completed independently by a single idle vehicle, and performing vehicle dispatch on this basis. If the task can be completed by a single vehicle, directly screening the best performing vehicle can simplify the dispatch process and save time and resources. When the task is too complex and needs to be decomposed into multiple subtasks, the system can reasonably allocate resources according to the specific constraints of the task to ensure that each subtask is executed by the most suitable vehicle. This method not only optimizes resource utilization, but also enhances the flexibility and reliability of task execution, ensuring that tasks can be completed efficiently under various circumstances, and reflects the intelligence and adaptability of the dispatch system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0016] Figure 1 It is a schematic diagram of the connection of each step of the method of the present invention.

[0017] Figure 2 It is a schematic diagram of the connection of various modules of the system of the present invention. DETAILED DESCRIPTION

[0018] The above contents in combination with the implementation of the present invention are merely examples and explanations of the concept of the present invention. The technical personnel in the relevant technical field may make various modifications or supplements to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they shall all fall within the protection scope of the present invention.

[0019] Example 1 See also Figure 1 As shown, a vehicle dispatching management method based on artificial intelligence comprises the following steps: The vehicle dispatching system obtains the task parameters of the task to be assigned, analyzes the task to be assigned, and determines whether the task to be assigned can be completed independently by a single idle vehicle; if it is determined that the task to be assigned can be completed independently by a single idle vehicle, the best execution vehicle for the task to be assigned is selected; The task parameters of the task to be assigned include the task starting point location, the task end point location, the task completion deadline and the task load requirement capacity.

[0020] Determine whether the task to be assigned can be completed independently by a single idle vehicle. The specific determination process is as follows: Obtain the maximum load capacity of each idle vehicle at the current time point, compare it with the task load requirement capacity of the task to be assigned, eliminate the idle vehicles whose maximum load capacity at the current time point is less than the task load requirement capacity of the task to be assigned, and then record the remaining idle vehicles after elimination as the initial screening idle vehicles; Get the position of each pre-screened idle vehicle at the current time point, compare it with the starting point of the task to be assigned, and get the distance between each pre-screened idle vehicle and the starting point of the task to be assigned; get the remaining power of each pre-screened idle vehicle at the current time point , and then judge that the power of each initially screened idle vehicle to complete the task to be assigned meets the evaluation value , 1 means compliance, 0 means non-compliance, c is the number of each initially screened idle vehicle, is the total estimated power consumption of the cth pre-screened idle vehicle from the current time point to the task end point of the task to be assigned. The specific calculation formula is: , is the power consumption of the cth initially screened idle vehicle during the unloaded segment of the task to be assigned, The power consumption of the cth initially screened idle vehicle corresponding to the load segment of the task to be assigned; The definition of the unloaded segment is the distance between the current position of each pre-screened idle vehicle and the starting position of the task to be assigned, and the definition of the fully loaded segment is the distance between the starting position of the task to be assigned and the ending position of the task to be assigned; in ; ; is the distance between the cth initially screened idle vehicle and the starting point corresponding to the task to be assigned, is the total distance that the cth initially screened idle vehicle needs to travel to complete the task to be assigned, The power consumption per unit distance of the cth initially screened idle vehicle when it is unloaded; The power consumption per unit distance when the cth vehicle is initially screened for idle load; is the influence coefficient of speed reduction when the cth primary screening idle vehicle is loaded, is the influence coefficient of the cth idle vehicle whose power consumption rate slows down as the power consumption rate slows down, and and All are less than 1; Then calculate the total time consumed by each initially screened idle vehicle to complete the task to be assigned , are the maximum driving speeds of the cth initially screened idle vehicle for the unloaded and fully loaded sections of the task to be assigned, respectively. is the minimum safe driving speed of the cth idle vehicle in the initial screening, is the estimated service time of the cth initially screened idle vehicle at the task starting point, The estimated charging time of the cth initially screened idle vehicle on the way; are the road condition influence coefficients of the cth preliminarily screened idle vehicle for the unloaded and fully loaded sections of the task to be assigned, indicating the influence of the road condition on the speed, with a specific value range of [0,1]; The denominator of the first term in the above formula is the actual driving speed from each initially screened idle vehicle position to the task starting position, which is affected by the maximum unloaded speed of the vehicle and the road conditions; the denominator of the second term in the above formula is the actual driving speed from each initially screened idle vehicle task starting position to the task end position, which is also affected by the maximum unloaded speed of the vehicle and the road conditions; , is the charging power of the cth initially screened idle vehicle; In current technology applications, vehicle scheduling usually takes battery power into account, resulting in vehicles with insufficient battery power being directly excluded. However, this approach may ignore the reality that some vehicles with insufficient battery power only need a short time to charge and can complete the task faster than vehicles that meet the battery power requirements but are farther away. Therefore, in vehicle scheduling, the charging time and geographical location of the vehicle should be comprehensively considered to optimize the efficiency of task completion.

[0021] The total time consumed by each initially screened idle vehicle to complete the task to be assigned is added to the current time point to obtain the completion time of each initially screened idle vehicle corresponding to the task to be assigned, and the completion time is compared with the task completion deadline of the task to be assigned. If there is an initially screened idle vehicle whose completion time corresponding to the task to be assigned is before the task completion deadline of the task to be assigned, it is determined that the task to be assigned can be completed independently by a single idle vehicle; if the completion time of each initially screened idle vehicle corresponding to the task to be assigned is after the task completion deadline of the task to be assigned, it is determined that the task to be assigned cannot be completed independently by a single idle vehicle.

[0022] If it is determined that the task to be assigned can be completed independently by a single idle vehicle, the best vehicle to perform the task to be assigned is selected. The specific selection process is as follows: The completion time of each preliminarily screened idle vehicle corresponding to the task to be assigned is obtained, and they are arranged in chronological order from small to large, and the preliminarily screened idle vehicle with the first completion time of the task to be assigned is used as the best execution vehicle for the task to be assigned.

[0023] If it is determined that the task to be assigned cannot be completed independently by a single idle vehicle, the vehicle dispatching system decomposes the task to be assigned into multiple subtasks according to the constraints of the task to be assigned; For complex tasks that cannot be completed by a single vehicle, the vehicle dispatching system first conducts a comprehensive analysis of the overall attributes of the tasks to be assigned, clarifies the time range of the tasks to be assigned, including the earliest start time, the latest completion time and possible intermediate time nodes of the tasks to be assigned, and divides the time span of the entire tasks to be assigned into multiple continuous time intervals, so as to provide an accurate time benchmark for the subsequent assignment of tasks to be assigned; then, the vehicle dispatching system makes a detailed division of the geographical locations involved in the tasks to be assigned, and groups all the task points to be assigned according to spatial distance and regional characteristics based on the distribution of the execution locations of the tasks to be assigned, ensuring that the task points to be assigned in each group have a high degree of spatial aggregation, thereby reducing the path intersection and invalid driving of vehicles when performing tasks; then, the vehicle dispatching system allocates the tasks to be assigned according to the specific execution requirements of the tasks to be assigned, such as the quantity and volume of goods to be transported or the services to be completed. The system will accurately decompose the demand for the overall task based on the service content, ensure that the demand for each subtask meets the carrying capacity of a single vehicle, and consider the service nature and operational complexity of the tasks to be assigned, and give priority to classifying the tasks to be assigned with similar demand characteristics into the same subtask; after completing the task demand decomposition, the system will further refine the priority and execution order of the tasks to be assigned, and reasonably plan the execution time window of the subtasks in combination with the time requirements, geographical distribution and driving path of each subtask, to ensure that the dependencies and logical order between the tasks to be assigned are strictly followed; finally, the vehicle dispatching system will make an overall adjustment to the allocation scheme of all subtasks through a dynamic optimization algorithm to ensure the shortest execution path, the most reasonable time arrangement and the highest resource utilization rate of the tasks, while controlling the overall completion time and execution cost of the tasks within the optimal range, thereby achieving efficient decomposition and execution of complex tasks.

[0024] Based on the established screening conditions, the best execution vehicles corresponding to each subtask are comprehensively screened out; The screening conditions include vehicle screening conditions and driver screening conditions.

[0025] According to the established vehicle screening conditions, available vehicles corresponding to each subtask under the vehicle screening conditions are screened out. The specific screening process is as follows: Calculate the fitness score of each candidate vehicle for each subtask , i is the number of each candidate vehicle, j is the number of each subtask, are weight coefficients set to balance the importance of different factors. is the distance adaptation score of the i-th candidate vehicle corresponding to the j-th subtask, is the load adaptation score of the i-th candidate vehicle corresponding to the j-th subtask, is the time adaptation score of the i-th candidate vehicle corresponding to the j-th subtask, is the matching score between the vehicle status of the i-th candidate vehicle and the j-th subtask requirement; in , is the actual driving distance between the position of the i-th candidate vehicle and the starting point of the j-th subtask, is the normalization factor of the maximum distance in the subtask corresponding to the candidate vehicle position; , is the remaining load capacity of the i-th candidate vehicle, is the task load requirement capacity of the jth subtask; , is the task completion deadline of the j-th subtask, is the estimated travel time for the i-th candidate vehicle to reach the starting point of the j-th subtask, is the time required for the i-th candidate vehicle to execute the j-th subtask, is the current time; , are the remaining battery power and health score of the i-th candidate vehicle, is the minimum battery power required for the i-th candidate vehicle to complete the j-th subtask, The highest health score among all candidate vehicles; The fitness score of each candidate vehicle corresponding to each subtask is compared with the set score lower limit threshold, and each candidate vehicle whose fitness score of each subtask is greater than the score lower limit threshold is used as each available vehicle corresponding to each subtask under the vehicle screening condition.

[0026] According to the established driver screening conditions, available vehicles corresponding to each subtask under the driver screening conditions are screened out. The specific screening process is as follows: Obtain the driver's historical information corresponding to each candidate vehicle, and then obtain the driver's historical task completion rate corresponding to each candidate vehicle , Drivers’ historical task execution efficiency and the number of each task type that the driver has historically performed; Based on the task type of each subtask and the number of task types that the driver of each candidate vehicle has performed in the past, the matching degree of the driver's historical task type corresponding to each candidate vehicle is calculated. ; According to the total number of subtasks J, obtain the total number of candidate vehicles M, where M>J; Randomly select J candidate vehicles from the total number of candidate vehicles for permutation and combination, thereby obtaining the total number of permutations and combinations of candidate vehicles H, and obtain the historical social interaction information, historical collaboration information, and historical task dependency information of the driver corresponding to each candidate vehicle in each candidate vehicle permutation and combination; Based on the historical social interaction information of the drivers of each candidate vehicle in each candidate vehicle combination, the interaction frequency of the drivers of each candidate vehicle in each candidate vehicle combination is extracted. , interaction quality , positive degree of interaction emotion , reciprocity and speed of information dissemination ; Thus, the social interaction score of the driver corresponding to each candidate vehicle in each candidate vehicle arrangement combination is obtained , To weigh the frequency, quality, and positivity of interactions, and to adjust the impact of these factors on the overall score, To balance the weights of reciprocity and information diffusion speed, we used them to adjust the impact of these factors on the overall score; is the average network connectivity of all drivers; r is the number of each candidate vehicle combination, and ; The first term in the above formula combines interaction frequency, quality, and sentiment analysis, normalized to account for average network connectivity. It emphasizes high-frequency, high-quality, and sentimentally positive interactions; The numerator of the second term in the above formula is the square of reciprocity, which highlights the value of two-way interaction. Combined with the Sigmoid function of the denominator, the speed of information dissemination emphasizes reciprocity in efficient dissemination. The above social interaction scoring formula is a theoretical framework that aims to evaluate the efficiency of interaction between drivers through multi-dimensional social network analysis. Its practical basis is that frequent and high-quality interactions usually promote more efficient communication and collaboration, while reciprocity enhances the sustainability of interaction. Network connectivity and information dissemination speed reflect the influence and responsiveness of drivers in social networks, which are factors that contribute to the formation of an efficient work team in practical applications. By combining these indicators, the formula strives to identify a combination of drivers with efficient communication in a complex social environment; Based on the historical cooperation information of the drivers of each candidate vehicle in each candidate vehicle combination, the task alternation frequency of the drivers of each candidate vehicle in each candidate vehicle combination is extracted. , Dependence and resource sharing ; Thus, the task dependency score of each candidate vehicle corresponding to the driver in each candidate vehicle combination is obtained , is the maximum frequency of historical task alternation of the candidate vehicle driver, The historical maximum dependency of the candidate vehicle driver, are the calculation coefficient factors set respectively, ; The first term of the above formula uses normalization to process the frequency of task alternation, reflecting the flexibility of drivers in tasks; the second term of the formula is the squared historical dependence, emphasizing the depth of historical cooperation; the third term of the formula is the logarithmic resource sharing degree, smoothing the impact of extreme values; Based on the historical task dependency information of the drivers of each candidate vehicle in each candidate vehicle combination, the number of tasks completed by the drivers of each candidate vehicle in each candidate vehicle combination is extracted. , Joint completion success rate , Collaboration Time and task quality rating ; Thus, the cooperation efficiency score of the corresponding driver of each candidate vehicle in each candidate vehicle arrangement combination is obtained , are the calculation weight factors set respectively, ; are the average number of common tasks and average task quality score of all drivers, is the minimum collaboration time of all drivers; The calculation formula for the Collaboration Efficiency Score (CE) combines multiple real-world factors. By considering the number of common tasks and the success rate, it can measure the effectiveness of drivers in past collaboration, which reflects their collaboration history and successful experience. The task quality score provides a direct evaluation of the effectiveness of the collaboration. High-quality task completion indicates good collaboration and communication between drivers. The average collaboration time is a key indicator of efficiency. A shorter time usually means better coordination and higher efficiency. Overall, these factors are often used in reality to evaluate the effectiveness of teamwork. The formula evaluates the collaborative efficiency of driver combinations through nonlinear processing and smoothing of extreme values. According to the driver's historical task completion rate, driver's historical task execution efficiency and driver's historical task type matching degree corresponding to each candidate vehicle, the historical task completion rate, driver's historical task execution efficiency and driver's historical task type matching degree corresponding to each candidate vehicle in each candidate vehicle arrangement combination are obtained; using the weighted average method, the historical task completion rate, driver's historical task execution efficiency and driver's historical task type matching degree corresponding to each candidate vehicle in each candidate vehicle arrangement combination are multiplied to obtain the historical score of the driver corresponding to each candidate vehicle in each candidate vehicle arrangement combination; Using the weighted average method, the social interaction score, task dependency score and collaboration efficiency score of the driver corresponding to each candidate vehicle in each candidate vehicle arrangement combination are multiplied to obtain the social score of the driver corresponding to each candidate vehicle in each candidate vehicle arrangement combination; The weighted average method is used again to multiply the historical scores of the drivers of each candidate vehicle in each candidate vehicle arrangement by the social scores, thereby obtaining the comprehensive scores of the drivers of each candidate vehicle in each candidate vehicle arrangement, and then summing and averaging them to obtain the comprehensive scores of each candidate vehicle arrangement; The comprehensive score of each candidate vehicle combination is compared with the set score lower limit threshold, and each candidate vehicle in each candidate vehicle combination with a comprehensive score greater than the score lower limit threshold is used as each available vehicle corresponding to each subtask under the driver screening condition.

[0027] When a complex task needs to be broken down into multiple subtasks, it is very necessary to screen the drivers for each subtask based on the importance of driver collaboration, because good collaboration can significantly improve the overall execution efficiency and accuracy of the task. By selecting a combination of drivers with a good historical collaboration record, it is possible to ensure seamless connection between subtasks and reduce communication barriers and misunderstandings. At the same time, drivers with strong collaboration capabilities are more adaptable to dynamic changes and can coordinate and solve emergencies in a timely manner, thereby reducing the risk of task delays and errors. This collaboration-oriented screening strategy not only improves the quality of task completion, but also promotes a more harmonious and efficient working environment.

[0028] Comprehensively select the best performing vehicle for each subtask. The specific selection process includes: Record each candidate vehicle permutation combination whose comprehensive score is greater than the score lower limit threshold as each standard combination, arrange the comprehensive scores of each standard combination in descending order, obtain the license plate number of each candidate vehicle in the first standard combination, and obtain the license plate number of each available vehicle corresponding to each subtask under the vehicle screening condition. If the license plate number of each candidate vehicle in the first standard combination is within the set of license plate numbers of each available vehicle corresponding to each subtask under the vehicle screening condition, then each candidate vehicle in the first standard combination is used as the best execution vehicle corresponding to each subtask; otherwise, the second standard combination, the third, ... until the last standard combination are screened according to the screening method of each candidate vehicle in the first standard combination; If all standard combinations are not in the license plate number set of each available vehicle corresponding to each subtask under the vehicle screening condition after screening, then each available vehicle corresponding to each subtask under the vehicle screening condition will be used as the best execution vehicle corresponding to each subtask.

[0029] In the vehicle dispatching process, if all standard combinations do not meet the available vehicle set under the vehicle screening conditions, it is extremely rare to select these available vehicles as the best execution vehicles for each subtask. This is because in most cases, through a carefully designed scoring model and combination screening process, it is possible to effectively match task requirements with vehicle resources, ensuring that vehicle combinations with strong collaboration capabilities and high comprehensive scores are given priority. This approach not only improves the efficiency and accuracy of task completion, but also emphasizes the importance of collaboration between drivers, ensuring the smooth execution of tasks under complex conditions.

[0030] The license plate number of the best performing vehicle corresponding to each subtask is obtained, and the task information of each best performing vehicle that needs to complete the corresponding subtask is fed back to the display interface of each best performing vehicle through the background terminal.

[0031] Example 2 See also Figure 2 As shown, an artificial intelligence-based vehicle dispatch management system includes a preliminary determination module, a task decomposition module, a vehicle determination module and a dispatch feedback module, wherein the above modules are connected by wired and / or wireless connections to achieve data transmission between the modules; In the preliminary determination module, the vehicle dispatching system obtains the task parameters of the task to be assigned, analyzes the task to be assigned, and determines whether the task to be assigned can be completed independently by a single idle vehicle; if it is determined that the task to be assigned can be completed independently by a single idle vehicle, the best execution vehicle for the task to be assigned is selected; Task decomposition module: if it is determined that the task to be assigned cannot be completed independently by a single idle vehicle, the vehicle dispatching system will decompose the task to be assigned into multiple subtasks according to the constraints of the task to be assigned; The vehicle determination module comprehensively selects the best execution vehicle for each subtask based on the established screening conditions; The scheduling feedback module obtains the license plate number of the best performing vehicle corresponding to each subtask, and feeds back the task information of each best performing vehicle that needs to complete the corresponding subtask to the display interface of each best performing vehicle through the background terminal.

[0032] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0033] It should be understood that determining B based on A does not mean determining B only based on A. B can also be determined based on A and / or other information.

[0034] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0035] Finally: 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, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A vehicle dispatching management method based on artificial intelligence, characterized in that: include: The vehicle dispatching system obtains the task parameters of the task to be assigned, analyzes the task to be assigned, and determines whether the task to be assigned can be completed independently by a single idle vehicle; if it is determined that the task to be assigned can be completed independently by a single idle vehicle, the best execution vehicle for the task to be assigned is selected; If it is determined that the task to be assigned cannot be completed independently by a single idle vehicle, the vehicle dispatching system decomposes the task to be assigned into multiple subtasks according to the constraints of the task to be assigned; Based on the established screening conditions, the best execution vehicles corresponding to each subtask are comprehensively screened out; The license plate number of the best performing vehicle corresponding to each subtask is obtained, and the task information of each best performing vehicle that needs to complete the corresponding subtask is fed back to the display interface of each best performing vehicle through the background terminal.

2. The method for vehicle dispatching management based on artificial intelligence according to claim 1 is characterized in that: The task parameters of the task to be assigned include the task starting point location, the task end point location, the task completion deadline and the task load requirement capacity.

3. The method for vehicle dispatching management based on artificial intelligence according to claim 2 is characterized in that: Determine whether the task to be assigned can be completed independently by a single idle vehicle. The specific determination process is as follows: Obtain the maximum load capacity of each idle vehicle at the current time point, compare it with the task load requirement capacity of the task to be assigned, eliminate the idle vehicles whose maximum load capacity at the current time point is less than the task load requirement capacity of the task to be assigned, and then record the remaining idle vehicles as the initial screening idle vehicles; Get the position of each pre-screened idle vehicle at the current time point, compare it with the starting point of the task to be assigned, and get the distance between each pre-screened idle vehicle and the starting point of the task to be assigned; get the remaining power of each pre-screened idle vehicle at the current time point , and then judge that the power of each initially screened idle vehicle to complete the task to be assigned meets the evaluation value , 1 means compliance, 0 means non-compliance, c is the number of each initially screened idle vehicle, is the total estimated power consumption of the cth pre-screened idle vehicle from the current time point to the task end point of the task to be assigned. The specific calculation formula is: , is the power consumption of the cth initially screened idle vehicle during the unloaded segment of the task to be assigned, The power consumption of the cth initially screened idle vehicle corresponding to the load segment of the task to be assigned; in ; ; is the distance between the cth initially screened idle vehicle and the starting point corresponding to the task to be assigned, is the total distance that the cth initially screened idle vehicle needs to travel to complete the task to be assigned, The power consumption per unit distance of the cth initially screened idle vehicle when it is unloaded; The power consumption per unit distance when the cth vehicle is initially screened for idle load; is the influence coefficient of speed reduction when the cth primary screening idle vehicle is loaded, is the influence coefficient of the cth idle vehicle whose power consumption rate slows down as the power consumption rate slows down, and and All are less than 1; Then calculate the total time consumed by each initially screened idle vehicle to complete the task to be assigned , are the maximum driving speeds of the cth initially screened idle vehicle for the unloaded and fully loaded sections of the task to be assigned, respectively. is the minimum safe driving speed of the cth idle vehicle in the initial screening, is the estimated service time of the cth initially screened idle vehicle at the task starting point, The estimated charging time of the cth initially screened idle vehicle on the way; are the road condition influence coefficients of the cth initially screened idle vehicle for the unloaded section and the fully loaded section of the task to be assigned; The total time consumed by each initially screened idle vehicle to complete the task to be assigned is added to the current time point to obtain the completion time of each initially screened idle vehicle corresponding to the task to be assigned, and the completion time is compared with the task completion deadline of the task to be assigned. If there is an initially screened idle vehicle whose completion time corresponding to the task to be assigned is before the task completion deadline of the task to be assigned, it is determined that the task to be assigned can be completed independently by a single idle vehicle; if the completion time of each initially screened idle vehicle corresponding to the task to be assigned is after the task completion deadline of the task to be assigned, it is determined that the task to be assigned cannot be completed independently by a single idle vehicle.

4. The artificial intelligence-based vehicle dispatching management method according to claim 3 is characterized in that: If it is determined that the task to be assigned can be completed independently by a single idle vehicle, the best vehicle to perform the task to be assigned is selected. The specific selection process is as follows: The completion time of each preliminarily screened idle vehicle corresponding to the task to be assigned is obtained, and they are arranged in chronological order from small to large, and the preliminarily screened idle vehicle with the first completion time of the task to be assigned is used as the best execution vehicle for the task to be assigned.

5. The method for vehicle dispatching management based on artificial intelligence according to claim 1 is characterized in that: The screening conditions include vehicle screening conditions and driver screening conditions.

6. The method for vehicle dispatching management based on artificial intelligence according to claim 5 is characterized in that: According to the established vehicle screening conditions, available vehicles corresponding to each subtask under the vehicle screening conditions are screened out. The specific screening process is as follows: Calculate the fitness score of each candidate vehicle for each subtask , i is the number of each candidate vehicle, j is the number of each subtask, are weight coefficients set to balance the importance of different factors. is the distance adaptation score of the i-th candidate vehicle corresponding to the j-th subtask, is the load adaptation score of the i-th candidate vehicle corresponding to the j-th subtask, is the time adaptation score of the i-th candidate vehicle corresponding to the j-th subtask, is the matching score between the vehicle status of the i-th candidate vehicle and the j-th subtask requirement; in , is the actual driving distance between the position of the i-th candidate vehicle and the starting point of the j-th subtask, is the normalization factor of the maximum distance in the subtask corresponding to the candidate vehicle position; , is the remaining load capacity of the i-th candidate vehicle, is the task load requirement capacity of the jth subtask; , is the task completion deadline of the j-th subtask, is the estimated travel time for the i-th candidate vehicle to reach the starting point of the j-th subtask, is the time required for the i-th candidate vehicle to execute the j-th subtask, is the current time; , are the remaining battery power and health score of the i-th candidate vehicle, is the minimum battery power required for the i-th candidate vehicle to complete the j-th subtask, The highest health score among all candidate vehicles; The fitness score of each candidate vehicle corresponding to each subtask is compared with the set score lower limit threshold, and each candidate vehicle whose fitness score of each subtask is greater than the score lower limit threshold is used as each available vehicle corresponding to each subtask under the vehicle screening condition.

7. The artificial intelligence-based vehicle dispatching management method according to claim 6 is characterized in that: According to the established driver screening conditions, available vehicles corresponding to each subtask under the driver screening conditions are screened out. The specific screening process is as follows: Obtain the driver's historical information corresponding to each candidate vehicle, and then obtain the driver's historical task completion rate corresponding to each candidate vehicle , Drivers’ historical task execution efficiency and the number of each task type that the driver has historically performed; Based on the task type of each subtask and the number of task types that the driver of each candidate vehicle has performed in the past, the matching degree of the driver's historical task type corresponding to each candidate vehicle is calculated. ; According to the total number of subtasks J, obtain the total number of candidate vehicles M, where M>J; Randomly select J candidate vehicles from the total number of candidate vehicles for permutation and combination, thereby obtaining the total number of permutations and combinations of candidate vehicles H, and obtain the historical social interaction information, historical collaboration information, and historical task dependency information of the driver corresponding to each candidate vehicle in each candidate vehicle permutation and combination; Based on the historical social interaction information of the drivers of each candidate vehicle in each candidate vehicle combination, the interaction frequency of the drivers of each candidate vehicle in each candidate vehicle combination is extracted. , interaction quality , positive degree of interaction emotion , reciprocity and speed of information dissemination ; Thus, the social interaction score of the driver corresponding to each candidate vehicle in each candidate vehicle arrangement combination is obtained , To weigh the frequency, quality, and positivity of interactions, and to adjust the impact of these factors on the overall score, To balance the weights of reciprocity and information diffusion speed, we used them to adjust the impact of these factors on the overall score; is the average network connectivity of all drivers; r is the number of each candidate vehicle combination, and ; Based on the historical cooperation information of the drivers of each candidate vehicle in each candidate vehicle combination, the task alternation frequency of the drivers of each candidate vehicle in each candidate vehicle combination is extracted. , Dependence and resource sharing ; Thus, the task dependency score of each candidate vehicle corresponding to the driver in each candidate vehicle combination is obtained , is the maximum frequency of historical task alternation of the candidate vehicle driver, The historical maximum dependency of the candidate vehicle driver, are the calculation coefficient factors set respectively, ; Based on the historical task dependency information of the drivers of each candidate vehicle in each candidate vehicle combination, the number of tasks completed by the drivers of each candidate vehicle in each candidate vehicle combination is extracted. , Joint completion success rate , Collaboration Time and task quality rating ; Thus, the cooperation efficiency score of the corresponding driver of each candidate vehicle in each candidate vehicle arrangement combination is obtained , are the calculation weight factors set respectively, ; are the average number of common tasks and average task quality score of all drivers, is the minimum collaboration time of all drivers; According to the driver's historical task completion rate, driver's historical task execution efficiency and driver's historical task type matching degree corresponding to each candidate vehicle, the historical task completion rate, driver's historical task execution efficiency and driver's historical task type matching degree corresponding to each candidate vehicle in each candidate vehicle arrangement combination are obtained; using the weighted average method, the historical task completion rate, driver's historical task execution efficiency and driver's historical task type matching degree corresponding to each candidate vehicle in each candidate vehicle arrangement combination are multiplied to obtain the historical score of the driver corresponding to each candidate vehicle in each candidate vehicle arrangement combination; Using the weighted average method, the social interaction score, task dependency score and collaboration efficiency score of the driver corresponding to each candidate vehicle in each candidate vehicle arrangement combination are multiplied to obtain the social score of the driver corresponding to each candidate vehicle in each candidate vehicle arrangement combination; The weighted average method is used again to multiply the historical scores of the drivers of each candidate vehicle in each candidate vehicle arrangement by the social scores, thereby obtaining the comprehensive scores of the drivers of each candidate vehicle in each candidate vehicle arrangement, and then summing and averaging them to obtain the comprehensive scores of each candidate vehicle arrangement; The comprehensive score of each candidate vehicle combination is compared with the set score lower limit threshold, and each candidate vehicle in each candidate vehicle combination with a comprehensive score greater than the score lower limit threshold is used as each available vehicle corresponding to each subtask under the driver screening condition.

8. The artificial intelligence-based vehicle dispatching management method according to claim 7 is characterized in that: Comprehensively select the best performing vehicle for each subtask. The specific selection process includes: Record each candidate vehicle permutation combination whose comprehensive score is greater than the score lower limit threshold as each standard combination, arrange the comprehensive scores of each standard combination in descending order, obtain the license plate number of each candidate vehicle in the first standard combination, and obtain the license plate number of each available vehicle corresponding to each subtask under the vehicle screening condition. If the license plate number of each candidate vehicle in the first standard combination is within the set of license plate numbers of each available vehicle corresponding to each subtask under the vehicle screening condition, then each candidate vehicle in the first standard combination is used as the best execution vehicle corresponding to each subtask; otherwise, the second standard combination, the third, ... until the last standard combination are screened according to the screening method of each candidate vehicle in the first standard combination; If all standard combinations are not in the license plate number set of each available vehicle corresponding to each subtask under the vehicle screening condition after screening, then each available vehicle corresponding to each subtask under the vehicle screening condition will be used as the best execution vehicle corresponding to each subtask.

9. A vehicle dispatching and management system based on artificial intelligence, characterized in that: It is implemented based on an artificial intelligence-based vehicle dispatching management method according to any one of claims 1 to 8, including: In the preliminary determination module, the vehicle dispatching system obtains the task parameters of the task to be assigned, analyzes the task to be assigned, and determines whether the task to be assigned can be completed independently by a single idle vehicle; if it is determined that the task to be assigned can be completed independently by a single idle vehicle, the best execution vehicle for the task to be assigned is selected; Task decomposition module: if it is determined that the task to be assigned cannot be completed independently by a single idle vehicle, the vehicle dispatching system will decompose the task to be assigned into multiple subtasks according to the constraints of the task to be assigned; The vehicle determination module comprehensively selects the best execution vehicle for each subtask based on the established screening conditions; The scheduling feedback module obtains the license plate number of the best performing vehicle corresponding to each subtask, and feeds back the task information of each best performing vehicle that needs to complete the corresponding subtask to the display interface of each best performing vehicle through the background terminal.

Images