Method and device for determining scheduling strategy

Abstract

The invention discloses a scheduling strategy determination method and device, and relates to the technical field of computers. A specific embodiment of the method comprises the following steps: constructing a scheduling strategy set according to acquired sample data; performing the following operations in a loop iteration manner until a preset iteration ending condition is met, and taking a scheduling strategy in the scheduling strategy set as a target scheduling strategy: performing vehicle scheduling on the transportation order in the sample data by using each scheduling strategy in the scheduling strategy set, and further calculating to obtain a strategy evaluation index of each scheduling strategy; and screening the scheduling strategies in the scheduling strategy set according to the strategy evaluation indexes, evolving the screened scheduling strategies by adopting a genetic algorithm, and updating the scheduling strategy set by using the evolved scheduling strategies. According to the embodiment, the target scheduling strategy is automatically evolved and generated based on the thought of genetic evolution, then vehicle scheduling is dynamically carried out in real time according to the target scheduling strategy, and the method has high environmental adaptability.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and device for determining a scheduling strategy. Background Art

[0002] In transportation scenarios, vehicle scheduling mainly adopts active methods (such as operations research algorithms, heuristic algorithms, etc.), that is, based on expert experience, scheduling strategies that can perform well in various common environments are planned in advance, and the best scheduling plan is found through mathematical solutions and other methods.

[0003] In the process of implementing the present invention, the inventors found that the prior art has the following problems:

[0004] The existing methods of active vehicle scheduling rely on fixed scheduling strategies planned in advance. The calculation and solution process is complex, and the scheduling strategy has poor adaptability in an uncertain environment. It cannot be adjusted according to changes in the environment and demand, and cannot meet actual needs well. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a method and device for determining a scheduling strategy, which can automatically determine a scheduling strategy that meets the needs of the current environment without the need for expert experience. Based on the idea of ​​genetic evolution, it automatically evolves and generates a target scheduling strategy, and then dynamically, real-time and efficiently performs vehicle scheduling according to the target scheduling strategy. It has strong environmental adaptability and can better meet scheduling needs.

[0006] To achieve the above object, according to one aspect of an embodiment of the present invention, a method for determining a scheduling strategy is provided, comprising:

[0007] Acquire sample data, each sample data including order information of a transport order to be processed and status information of a vehicle to be dispatched corresponding to the transport order;

[0008] Constructing a scheduling strategy set according to the order information and the status information;

[0009] The following operations are performed in a loop iteratively until a preset iteration end condition is met, and the scheduling strategy in the scheduling strategy set is used as the target scheduling strategy: each scheduling strategy in the scheduling strategy set is used to perform vehicle scheduling on the transport order in the sample data, and the strategy evaluation index of each scheduling strategy in the scheduling strategy set is calculated according to the vehicle scheduling result; each scheduling strategy in the scheduling strategy set is screened according to the strategy evaluation index, a genetic algorithm is used to evolve the screened scheduling strategies, and the evolved scheduling strategies are used to update the scheduling strategy set.

[0010] Optionally, a scheduling strategy set is constructed based on the order information and the status information, including: extracting order elements from the order information and assigning corresponding order character representations to the order elements; extracting status elements from the status information and assigning corresponding status character representations to the status elements; generating a specified number of scheduling strategies based on the order character representation, the status character representation and a preset operator set to form a scheduling strategy set.

[0011] Optionally, a specified number of scheduling strategies are generated according to the order character representation, the status character representation and a preset operator set, including: forming a policy element set with the order character representation, the status character representation and the operator set; for each scheduling strategy, generating it according to the following method: extracting a policy element from the policy element set; in response to the extraction result being an operator, taking the operator as the root node, extracting other elements except the operator from the policy element set as leaf nodes of the root node; loopingly performing the following operations until the order character representation or the status character representation is extracted from the policy element set, or the level of the leaf node reaches a preset level threshold: in response to the leaf node being an operator, extracting other elements except the elements that have been extracted from the policy element set as leaf nodes of the next level of the leaf node.

[0012] Optionally, the sample data is sampled from the full historical data according to a preset extraction strategy; each scheduling strategy in the scheduling strategy set is used to perform vehicle scheduling on the transport order in the sample data, including: for each scheduling strategy in the scheduling strategy set: calculating the scheduling priority of each of the vehicles to be scheduled when the transport order executes the scheduling strategy; and according to the scheduling priority, selecting a scheduling vehicle corresponding to the highest scheduling priority from each of the vehicles to be scheduled.

[0013] Optionally, each scheduling strategy in the scheduling strategy set is screened according to the strategy evaluation index, and a genetic algorithm is used to evolve the screened scheduling strategies, including: selecting the scheduling strategy to be evolved from the scheduling strategy set according to the strategy evaluation index; performing strategy evolution on the scheduling strategy to be evolved using a preset genetic operator to obtain an evolved scheduling strategy.

[0014] Optionally, each scheduling strategy in the scheduling strategy set is stored in a tree structure; the scheduling strategy to be evolved is subjected to strategy evolution using a preset genetic operator, including: performing node evolution on some tree nodes in the scheduling strategy to be evolved using at least one of a crossover operator, a mutation operator, and a reconstruction operator.

[0015] Optionally, the strategy evaluation indicators include the transportation cost of the dispatched vehicle, the transportation efficiency of the dispatched vehicle and the dispatch waiting time; the iteration end conditions include that the strategy evaluation indicators of the evolved dispatching strategy do not reach a preset improvement threshold, and the number of loop iterations of the dispatching strategy set reaches a preset number threshold.

[0016] According to a second aspect of an embodiment of the present invention, a vehicle scheduling method is provided, comprising:

[0017] Acquire real-time order information of a real-time transport order to be processed, and real-time status information of a vehicle to be dispatched corresponding to the real-time transport order;

[0018] According to the target scheduling strategy, combined with the real-time order information and the real-time status information of the vehicle to be scheduled, a scheduling vehicle that meets the real-time transportation order requirements is determined from the vehicles to be scheduled, wherein the target scheduling strategy is determined according to any one of the methods provided in the first aspect of the embodiment of the present invention.

[0019] According to a third aspect of an embodiment of the present invention, a device for determining a scheduling strategy is provided, including:

[0020] A sample data acquisition module, used to acquire sample data, each sample data includes order information of a transport order to be processed and status information of a vehicle to be dispatched corresponding to the transport order;

[0021] A strategy set building module, used to build a scheduling strategy set according to the order information and the status information;

[0022] The target strategy determination module is used to iteratively perform the following operations until a preset iteration end condition is met, and use the scheduling strategy in the scheduling strategy set as the target scheduling strategy: use each scheduling strategy in the scheduling strategy set to perform vehicle scheduling on the transport order in the sample data, and calculate the strategy evaluation index of each scheduling strategy in the scheduling strategy set according to the vehicle scheduling result; screen each scheduling strategy in the scheduling strategy set according to the strategy evaluation index, use a genetic algorithm to evolve the screened scheduling strategies, and use the evolved scheduling strategies to update the scheduling strategy set.

[0023] According to a fourth aspect of an embodiment of the present invention, there is provided an electronic determination device for a scheduling strategy, comprising:

[0024] one or more processors;

[0025] a storage device for storing one or more programs,

[0026] When the one or more programs are executed by the one or more processors, the one or more processors implement the method provided in the first aspect of the embodiment of the present invention.

[0027] According to a fifth aspect of an embodiment of the present invention, there is provided a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method provided in the first aspect of the embodiment of the present invention is implemented.

[0028] According to a sixth aspect of an embodiment of the present invention, there is provided a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method provided in the first aspect of the embodiment of the present invention is implemented.

[0029] An embodiment of the invention has the following advantages or beneficial effects: by acquiring sample data, each sample data includes order information of a transport order to be processed and status information of a vehicle to be dispatched corresponding to the transport order; a dispatch strategy set is constructed according to the order information and status information; the following operations are performed in a loop until a preset iteration end condition is met, and a dispatch strategy in the dispatch strategy set is used as a target dispatch strategy: each dispatch strategy in the dispatch strategy set is used to dispatch vehicles for the transport orders in the sample data, and a strategy evaluation index of each dispatch strategy in the dispatch strategy set is calculated according to the vehicle dispatch result; each dispatch strategy in the dispatch strategy set is screened according to the strategy evaluation index, a genetic algorithm is used to evolve the screened dispatch strategies, and the evolved dispatch strategies are used to update the dispatch strategy set. A technical solution realizes a method for automatically determining a dispatch strategy that meets the needs of the current environment, which does not require expert experience and is based on the idea of ​​genetic evolution to automatically evolve and generate a target dispatch strategy, and then dynamically, real-time and efficiently dispatches vehicles according to the target dispatch strategy. It has strong environmental adaptability and can better meet dispatch needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0031] Figure 1 is a schematic diagram of the main process of the method for determining a scheduling strategy according to an embodiment of the present invention;

[0032] Figure 2 is a schematic diagram of a method for constructing a scheduling strategy set according to an embodiment of the present invention;

[0033] Figure 3 It is a schematic diagram of the composition of the scheduling strategy tree of an embodiment of the present invention;

[0034] Figure 4 A method for updating a scheduling strategy set based on a genetic algorithm according to an embodiment of the present invention;

[0035] Figure 5 is a schematic diagram of the evolution of a crossover operator according to an embodiment of the present invention;

[0036] Figure 6 is a flow chart of a method for determining a scheduling strategy according to an embodiment of the present invention;

[0037] Figure 7 is a schematic diagram of main modules of a device for determining a scheduling strategy according to an embodiment of the present invention;

[0038] Figure 8 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0039] Fig. 9 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0040] It should be noted that in the technical solution of the present invention, the collection / collection, update, analysis, use, transmission, storage and other aspects of user personal information involved are in compliance with the provisions of relevant laws and regulations, are used for legal and reasonable purposes, are not shared, leaked or sold outside of these legal uses, and are subject to supervision and management by national regulatory authorities. Necessary measures should be taken for user personal information to selectively block the use or access to personal information data to prevent illegal access to such personal information data, ensure that persons who have access to personal information data comply with the provisions of relevant laws and regulations, and ensure the security of user personal information. In addition, once these user personal information data are no longer needed, the risks should be minimized by limiting or even prohibiting data collection and / or deleting data.

[0041] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0042] The existing methods of active vehicle scheduling rely on fixed scheduling strategies planned in advance. The calculation and solution process is complex, and the scheduling strategy has poor adaptability in an uncertain environment. It cannot be adjusted according to changes in the environment and demand, and cannot meet actual needs well.

[0043] In order to solve the above problems existing in the prior art, the present invention proposes a method for determining a scheduling strategy, which utilizes a genetic algorithm, combined with the calculated strategy evaluation index and genetic operator of each scheduling strategy in the scheduling strategy set, to iteratively screen and evolve each scheduling strategy in the constructed scheduling strategy set, and obtain a target scheduling strategy that meets the needs to perform vehicle scheduling. This method realizes a method for automatically determining a scheduling strategy that meets environmental requirements, which does not require expert experience. Based on the idea of ​​genetic evolution, the target scheduling strategy is automatically evolved and generated, and then vehicle scheduling is performed dynamically, in real time and efficiently according to the target scheduling strategy. This method has strong environmental adaptability and can better meet scheduling needs.

[0044] Figure 1 is a schematic diagram of the main process of the method for determining a scheduling strategy according to an embodiment of the present invention, such as Figure 1 As shown, the method for determining a scheduling strategy in an embodiment of the present invention includes the following steps S101 to S103.

[0045] Step S101: Acquire sample data, each sample data including order information of a transport order to be processed and status information of a vehicle to be dispatched corresponding to the transport order.

[0046] Specifically, in the vehicle dispatching business, dispatching requirements mainly include planned requirements and unplanned requirements. Planned requirements usually report the pending transport orders in advance, and the dispatching system allocates vehicles for these transport orders. Unplanned requirements are temporarily submitted transport orders, and each transport order is processed once. In order to ensure the timeliness of vehicle dispatching, the embodiment of the present invention determines the target dispatching strategy in advance. For example, the target dispatching strategy can be determined at 23:00 on the previous night, and the transport orders after midnight on the same day and before midnight on the next day are all dispatched according to the target dispatching strategy.

[0047] In order to ensure that the determined scheduling strategy meets the needs of the current environment as much as possible, the embodiment of the present invention uses the system data of the day before the scheduling day as sample data. Of course, the system data of the most recent three days can also be selected as sample data as needed. The embodiment of the present invention does not limit the specific sample extraction time. Of course, it can also be fine-grained to the business scope of the sample data. For example, if scheduling services are provided to medium-sized trucks, the order information of the historical transportation orders of medium-sized trucks and the corresponding status information of the medium-sized trucks to be scheduled are obtained from the system data. That is, according to business needs, the order information of the transportation orders within the specified historical time period and / or within the specified business scope and the status information of the corresponding vehicles to be scheduled at that time are obtained from the system data. The order information of the embodiment of the present invention mainly includes: attribute information such as the starting point, end point, and demand time of the transportation; the status information mainly includes: the current geographical location of the vehicle to be scheduled, the vehicle configuration of the vehicle to be scheduled, and other status attribute information.

[0048] According to an embodiment of the present invention, the sample data is sampled from the entire historical data according to a preset extraction strategy.

[0049] Specifically, in order to reduce computing consumption and improve the efficiency of determining the target scheduling strategy, the embodiment of the present invention uses small sample data to determine the target scheduling strategy, and extracts from the historical full data according to a preset extraction strategy. The specific extraction strategy can use simple random sampling, extracting through a random number table or computer-generated random numbers; it can also extract a sample every sample interval at a fixed interval; it can also use stratified sampling, stratify the full data according to certain features, and randomly extract from each layer; it can also use clustering extraction, use a clustering algorithm to divide the full data into several clusters, select the target cluster from it, and use the data in the target cluster as sample data; it can also select the above-mentioned extraction strategies for combination; it can also select the most easily accessible individuals from the full data as sample data; it can also artificially set specific standards for extraction; it can also purposefully set quotas for various features, and extract sample data according to the proportion of the quota. The embodiment of the present invention does not specifically limit the specific extraction strategy, and it can be selected and used according to the needs of the scene, or used in combination.

[0050] Step S102: construct a scheduling strategy set according to the order information and the status information.

[0051] Specifically, each attribute information contained in the above-extracted order information and each state attribute information contained in the state information are used as the construction elements of the scheduling strategy, and several attribute information are randomly selected, and then these attribute information and operators are arranged and combined with randomly extracted operators (such as min, +, -, ×, ÷, etc.) to obtain multiple scheduling strategies, and multiple scheduling strategies are combined together to obtain an initial scheduling strategy set. It can be understood that each scheduling strategy in the scheduling strategy set constructed at this time is only the initial scheduling strategy of genetic evolution, so its specific rationality does not need to be paid too much attention, and it will be screened and eliminated in the subsequent iterative evolution process.

[0052] Step S103, perform the following operations in a loop iteratively until the preset iteration end condition is met, and use the scheduling strategy in the scheduling strategy set as the target scheduling strategy: use each scheduling strategy in the scheduling strategy set to perform vehicle scheduling on the transport order in the sample data, and calculate the strategy evaluation index of each scheduling strategy in the scheduling strategy set according to the vehicle scheduling result; screen each scheduling strategy in the scheduling strategy set according to the strategy evaluation index, use a genetic algorithm to evolve the screened scheduling strategies, and use the evolved scheduling strategies to update the scheduling strategy set.

[0053] Specifically, the embodiment of the present invention adopts a genetic algorithm, which requires multiple training iterative evolutions based on each scheduling strategy in the scheduling strategy set constructed above. Each evolution needs to eliminate scheduling strategies with poor performance and select excellent scheduling strategies as the objects of genetic evolution for strategy evolution to obtain scheduling strategies with better performance. This cycle is repeated in this way, and the scheduling strategy obtained at the end of the cycle iteration is the target scheduling strategy that meets the environmental requirements.

[0054] From the above, it can be seen that selecting a scheduling strategy suitable for evolution from the scheduling strategy set plays an important role in determining the subsequent target scheduling strategy of the embodiment of the present invention. In order to take into account the comprehensiveness of the calculation strategy evaluation index and the rationality of the computing resource occupancy, the embodiment of the present invention samples a small amount of sample data that can represent the full amount of data from the historical full amount of data according to the preset extraction strategy. The embodiment of the present invention also refers to small sample data to save unnecessary computing resources. Using each scheduling strategy in the scheduling strategy set, vehicle scheduling is performed for each transport order in the small sample data, and the dispatched vehicle that completes all the transport orders in the small sample under the scheduling strategy is obtained, which is the vehicle scheduling result under one scheduling strategy, and then the vehicle scheduling results corresponding to all scheduling strategies in the scheduling strategy set are obtained.

[0055] According to an embodiment of the present invention, the strategy evaluation index includes the transportation cost of the dispatched vehicle, the transportation efficiency of the dispatched vehicle and the dispatch waiting time.

[0056] Specifically, based on the vehicle scheduling results of the transport orders in the small sample data using various scheduling strategies, the embodiment of the present invention mainly comprehensively calculates the strategy evaluation index of each scheduling strategy from three aspects: the transportation cost of the scheduling vehicle, the transportation efficiency of the scheduling vehicle, and the scheduling waiting time of the transport order. According to the transportation starting point and transportation end point of the transport order in the vehicle scheduling result, and the distance of the scheduling vehicle from the transportation starting point, combined with information such as the model of the scheduling vehicle, the transportation cost of the scheduling vehicle is calculated; according to the transportation volume of the transport order in the vehicle scheduling result, and the transportation time of the scheduling vehicle, the transportation efficiency of the scheduling vehicle is calculated; according to the time when the transport order initiates the vehicle scheduling request and the time when the scheduling vehicle starts to perform the transportation task of the transport order, the scheduling waiting time of the transport order is calculated; finally, the transportation cost, transportation efficiency and scheduling waiting time are weighted and fused to obtain the strategy evaluation index corresponding to the scheduling strategy.

[0057] Furthermore, based on the strategy evaluation index of each scheduling strategy, the scheduling strategies that can be used for evolution are screened from the scheduling strategy set, the screened scheduling strategies are evolved using a genetic algorithm, and the evolved scheduling strategies are used to update the scheduling strategy set to obtain the scheduling strategy set after the first round of evolution.

[0058] Then, according to the above method, each scheduling strategy in the scheduling strategy set after the first round of evolution is used to perform vehicle scheduling on the transport orders in the small sample data, and the vehicle scheduling results corresponding to all the transport orders in the small sample data are obtained, and then the vehicle scheduling results corresponding to all scheduling strategies in the scheduling strategy set after the first round of evolution are obtained. Then, according to the vehicle scheduling results of these scheduling strategies, the corresponding strategy evaluation index is calculated. Finally, based on the calculated strategy evaluation index, the scheduling strategies that can be used for the second round of evolution are selected from the scheduling strategy set after the first round of evolution, and the selected scheduling strategies are evolved using a genetic algorithm, and the scheduling strategy set is updated using the evolved scheduling strategies to obtain the scheduling strategy set after the second round of evolution. The subsequent loop iterations are similar to the above, until the preset iteration end condition is met, the scheduling strategy in the scheduling strategy set is the target scheduling strategy.

[0059] According to another embodiment of the present invention, the iteration end condition includes that the policy evaluation index of the evolved scheduling policy does not reach a preset improvement threshold, and the number of iterations of the scheduling policy set reaches a preset number threshold.

[0060] Specifically, the iteration termination condition of the embodiment of the present invention can be a threshold of the number of iterations set according to an empirical value. Each time the scheduling strategy set is updated, the number of iterations is increased by 1 until the preset number threshold is reached. Or determine whether to stop iteration by comparing the degree of improvement of the new policy evaluation index of the new scheduling strategy generated after evolution when executing small sample data relative to the old policy evaluation index of the scheduling strategy before evolution when executing the same small sample data. If the new policy evaluation index is not improved relative to the old policy evaluation index or the improvement is less than the preset improvement threshold, then the iteration is stopped, otherwise, a new round of screening and iterative evolution is started. Of course, the two iteration termination conditions of the improvement threshold and the number threshold can also be combined. As long as one of the two is met, the iteration is stopped; it can also be stopped when the number threshold is reached and the improvement threshold is not reached.

[0061] It can be understood that if there are multiple scheduling strategies in the scheduling strategy set when the iteration stops, then any one of them can be selected as the target scheduling strategy, or the sample data can be used to calculate the strategy evaluation indicators of each scheduling strategy, and the scheduling strategy with the best strategy evaluation indicator can be selected as the target scheduling strategy.

[0062] The embodiment of the present invention also provides a method for constructing a detailed scheduling strategy set. Figure 2 Schematic diagram of a method for constructing a scheduling strategy set according to an embodiment of the present invention. Figure 2 The method for constructing a scheduling strategy set may include:

[0063] Step S201: extract order elements from the order information, and assign corresponding order character representations to the order elements.

[0064] Specifically, in order to facilitate the construction of scheduling strategies, the embodiment of the present invention extracts key order elements, that is, key order attributes, from the order information, and assigns corresponding order character representations to these order elements, using abstract characters to represent order elements. For example: OSLNG is used to represent the longitude of the order starting point, OSLAT is used to represent the latitude of the order starting point, ODLNG is used to represent the longitude of the order end point, ODLAT is used to represent the latitude of the order end point, the order vehicle time is OT, and the order vehicle model is OV.

[0065] Step S202: extracting status elements from the status information, and assigning corresponding status character representations to the status elements.

[0066] Similarly, the embodiment of the present invention extracts key status elements from the status information of the vehicle to be dispatched, that is, the key status attributes of the vehicle to be dispatched, and assigns corresponding status character representations to these status elements, using abstract characters to represent the status elements. For example: LNG represents the current position longitude of the vehicle to be dispatched, LAT represents the current position latitude of the vehicle to be dispatched, RON represents the number of tasks to be completed by the vehicle to be dispatched, VT represents the vehicle type, VL represents the vehicle length, VW represents the vehicle width, VH represents the vehicle height, VO represents the oil vehicle, and VE represents the electric vehicle.

[0067] Step S203: Generate a specified number of scheduling strategies according to the order character representation, the status character representation and a preset operator set to form a scheduling strategy set.

[0068] Specifically, based on the above-mentioned abstracted order character representation and status character representation, in order to construct a scheduling strategy, a set of operators required for constructing a scheduling strategy is preset in advance. The embodiment of the present invention constructs a scheduling strategy with a tree structure, establishes a logical operation relationship between each character representation through an operator, and sets the number of scheduling strategies corresponding to tree structures of different levels according to the total number of preset scheduling strategies. For example, if a scheduling strategy of a tree structure of 1 to 10 levels is constructed, the total number is 1000, then the scheduling strategies of each level of the tree structure can be averaged to 100, or a large proportion of the scheduling strategies corresponding to the tree structure of the middle levels 5, 6, and 7 can be allocated, and a small proportion of the scheduling strategies corresponding to the tree structure of other levels can be allocated.

[0069] Then, according to the number of scheduling strategies corresponding to the tree structure of each level, character representations and operators are randomly extracted from the set of order character representations, the set of status character representations, and the set of operators to generate the corresponding scheduling strategies. It can be understood that if one level requires an operator, the operator can be extracted first, and then the character representation can be extracted from the order character representation set and the status character representation set.

[0070] For example, for a three-level scheduling strategy, an operator is first extracted from the operator set, such as min, and then a character representation is extracted from the order character representation set and the status character representation set as an operand of the operator (X), and then a secondary operator is extracted from the operator set as another operand of the operator ( ), and then extract two character representations (Y, Z) from the order character representation set and the status character representation set as the operands of the secondary operator, then a logical operation relationship of min(X, (Y Z)) is the scheduling strategy tree for calculating the minimum value.

[0071] According to one embodiment of the present invention, the order character representation, the status character representation and the operator set form a policy element set; for each scheduling strategy, it is generated according to the following method: extracting policy elements from the policy element set; in response to the extraction result being an operator, taking the operator as the root node, extracting other elements except the operator from the policy element set as leaf nodes of the root node; looping through the following operations until the order character representation or the status character representation is extracted from the policy element set, or the level of the leaf node reaches a preset level threshold: in response to the leaf node being an operator, extracting other elements except the elements that have been extracted from the policy element set as leaf nodes of the next level of the leaf node.

[0072] Specifically, an embodiment of the present invention also provides another method for completely randomly constructing a scheduling strategy, which summarizes the above-mentioned order character representation, status character representation and preset operator set into a strategy element set, and the specific order character representation, status character representation and operators in the operator set can be stored arbitrarily in the strategy element set without restricting their storage location.

[0073] First, a policy element is randomly extracted from the policy element set. It can be understood that if the extracted policy element is an order character representation or a status character representation, then it means that the extraction is terminated, because there is no other logical relationship. This order character representation or status character representation is a scheduling policy. If the extracted policy element is an operator, then it is necessary to perform the next level of extraction, extract the elements of the left and right leaf nodes for the root node operator, and extract two other elements except the operator from the policy element set as leaf nodes. Then determine whether the extracted leaf node is an operator. If both leaf nodes are not operators, then this extraction ends and generates a scheduling policy consisting of two leaf nodes and an operator; if one of the two leaf nodes is an operator or both are operators, then it is necessary to recursively perform the next level of extraction, and extract the next level of leaf nodes for the operator leaf node. The method is similar. Extract two other elements except the elements that have been extracted from the policy element set as the next level of leaf nodes. After that, it is determined whether the newly extracted leaf node is an operator. If it is an operator, the corresponding next-level leaf node is extracted for the operator, and the cycle is repeated recursively until all the extracted strategy elements are represented by order characters or status characters.

[0074] In addition, considering that the level of the generated scheduling strategy tree should not be too deep, the embodiment of the present invention can also preset the level threshold in advance. If the level of the recursive leaf node reaches the level threshold, the generation of the scheduling strategy tree is also terminated. It can be understood that if the level of the leaf node reaches the level threshold, and the leaf node of the maximum level is still an operator, the order character representation or state character representation other than the extracted elements is extracted from the strategy element set to replace the operator to build a complete scheduling strategy tree.

[0075] According to the above method, a specified number of scheduling policy trees are generated, and these scheduling policy trees constitute a scheduling policy set.

[0076] Figure 3 : is a schematic diagram of the composition of the scheduling strategy tree of the embodiment of the present invention. As can be seen from the figure, the scheduling strategy tree has undergone four extractions of strategy elements. First, the operator min is extracted from the strategy element set for the first time, and min is the root node. Due to the extracted operator, the second extraction is performed to extract the operators and the status character represents WIQ (vehicle load data in the supply queue), then the operator is the left leaf node of the root node, and WIQ is the right leaf node of the root node. Since the right leaf node is represented by the status character, the right leaf node ends, and the left leaf node is the operator, so a third extraction is required, and the extraction result is the left leaf node. The next level left and right leaf nodes. By analogy, the third extraction is performed from the strategy element set. First, the state character table NIQ (the number of required tasks for vehicles to be dispatched) is extracted as the left leaf node, and then the operator + is extracted as the right leaf node. Since the right leaf node is also an operator, a fourth extraction is required. The fourth extraction is performed from the strategy element set. The order character representation MWT (the waiting time for the transport order) is extracted as the left leaf node, and the state character representation WKR (the remaining time to execute the current transport task) is extracted as the right leaf node. Since neither the left nor the right leaf nodes are operators, the extraction ends, and the illustrated scheduling strategy tree is obtained.

[0077] In the above process of using the genetic algorithm to iteratively update the scheduling strategy set, for each iterative operation, the embodiment of the present invention also provides a detailed method for updating the scheduling strategy set. Figure 4 The method for updating the scheduling strategy set based on the genetic algorithm in an embodiment of the present invention may specifically include:

[0078] Step S401, for each scheduling strategy in the scheduling strategy set: calculating the scheduling priority of each of the vehicles to be scheduled when the transport order executes the scheduling strategy; and selecting a scheduling vehicle corresponding to the highest scheduling priority from each of the vehicles to be scheduled according to the scheduling priority.

[0079] Specifically, based on each scheduling strategy, the calculation result of the scheduling strategy is calculated when the transportation order of each sample data is executed according to the scheduling strategy, that is, the scheduling priority. For example, there is a scheduling strategy min(X, Y Z), X is obtained as 15 from the order information of the transport order, and the status information Y of the E vehicle to be dispatched corresponding to the transport order is 3, and Z is 6, then the transport order is calculated according to the dispatching strategy, and the dispatching priority of the E vehicle to be dispatched is 15. Considering that the application scenario of the embodiment of the present invention is the matching degree between the transport order and each vehicle to be dispatched, the dispatching priority characterizes the matching degree between the two. Then, the dispatching vehicle corresponding to the highest dispatching priority is selected from each vehicle to be dispatched to obtain the dispatching vehicle for the transport order of the sample data, so as to determine the vehicle dispatching results of the whole sample of the transport orders of all sample data under a fixed dispatching strategy in a loop according to the above method, and finally loop through all the dispatching strategies to obtain the vehicle dispatching results of the whole sample of each dispatching strategy in the dispatching strategy set, and one dispatching strategy corresponds to one vehicle dispatching result of the whole sample.

[0080] Step S402: Calculate the strategy evaluation index of each scheduling strategy in the scheduling strategy set according to the vehicle scheduling result.

[0081] Step S403: According to the strategy evaluation index, the scheduling strategy to be evolved is selected from the scheduling strategy set; and the scheduling strategy to be evolved is evolved using a preset genetic operator to obtain an evolved scheduling strategy.

[0082] Specifically, based on the preset screening ratio and the policy evaluation index corresponding to each scheduling policy, the scheduling policies that meet the screening ratio are selected from the scheduling policy set. For example, the top 10 scheduling policies with the policy evaluation index are selected, and the selected scheduling policies with relatively good performance are selected as the objects to be evolved. Then, these scheduling policies to be evolved are encoded according to the policy elements contained therein to obtain the policy codes corresponding to each scheduling policy to be evolved, and each policy code is evolved using a genetic operator to obtain the evolved policy code, and finally the evolved policy code is subjected to element parsing to obtain the evolved scheduling policy.

[0083] According to one embodiment of the present invention, each scheduling strategy in the scheduling strategy set is stored in a tree structure; the scheduling strategy to be evolved is subjected to strategy evolution using a preset genetic operator, including: performing node evolution on some tree nodes in the scheduling strategy to be evolved using at least one of a crossover operator, a mutation operator, and a reconstruction operator.

[0084] Specifically, based on the above-constructed scheduling strategy tree, the embodiment of the present invention uses a more efficient tree node strategy evolution to evolve some tree nodes in the scheduling strategy tree to be evolved. Specifically, at least one of the crossover operator, mutation operator, and reconstruction operator in the genetic algorithm can be selected to perform node evolution on randomly selected tree nodes in the scheduling strategy tree to be evolved to obtain the evolved scheduling strategy tree. Of course, other genetic operators can also be used. The embodiment of the present invention only uses the crossover, mutation, and reconstruction operators as examples.

[0085] Figure 5 Schematic diagram of the evolution of the crossover operator of the embodiment of the present invention. Two scheduling strategy trees A and B are selected from the scheduling strategy tree to be evolved, and the dark nodes in B are cross-replaced with the dark nodes in A to obtain a new individual after the evolution of the crossover operator: scheduling strategy tree C.

[0086] Step S404: Use the evolved scheduling strategy to update the scheduling strategy set.

[0087] Finally, the evolved scheduling strategies form a new scheduling strategy set, which updates and replaces the scheduling strategies in the previous scheduling strategy set. A new round of calculating the strategy evaluation index of the evolved scheduling strategies begins, and a decision is made whether to start the next round of strategy evolution based on the strategy evaluation index and the number of iterations.

[0088] Figure 6It is a flow chart of the method for determining the scheduling strategy of an embodiment of the present invention. First, an initial scheduling strategy set is constructed based on the order information of the transport order in the acquired sample data and the status information of the corresponding vehicle to be scheduled. Then, each scheduling strategy in the scheduling strategy set is used to perform vehicle scheduling for the transport order in the sample data, and the strategy evaluation index of each scheduling strategy in the scheduling strategy set is calculated based on the vehicle scheduling result. Based on the configured strategy evaluation index as the iteration end condition, when the iteration end condition is first judged for the scheduling strategy set that has not been updated, it is assumed that the end condition is not met, and each scheduling strategy in the scheduling strategy set is directly screened according to the strategy evaluation index. Then, the genetic operators such as crossover, mutation, and reconstruction in the genetic algorithm are used to evolve the screened scheduling strategies, and finally the evolved scheduling strategies are used to update the scheduling strategy set.

[0089] Furthermore, according to a method similar to the above, each scheduling strategy in the updated scheduling strategy set is used to perform vehicle scheduling on the transport orders in the sample data, and the strategy evaluation index of each scheduling strategy in the current scheduling strategy set is calculated based on the vehicle scheduling result. According to the strategy evaluation index of each current scheduling strategy, the strategy evaluation index of the current scheduling strategy set is calculated by weighted average, and the initial strategy evaluation index obtained by weighted average calculation of the strategy evaluation index and the strategy evaluation index of each scheduling strategy in the initial scheduling strategy set is used for improvement comparison. If the improvement value reaches the improvement threshold, then continue to evolve, and screen each scheduling strategy in the current scheduling strategy set according to the strategy evaluation index of each scheduling strategy in the current scheduling strategy set. Then, genetic operators such as crossover, mutation, and reconstruction in the genetic algorithm are used to evolve the screened scheduling strategies, and finally the evolved scheduling strategies are used to update the scheduling strategy set.

[0090] The iteration is repeated again, similar to the above, and the strategy evaluation index of each scheduling strategy in the updated scheduling strategy set is calculated, and whether the iteration end condition is met is determined based on the strategy evaluation index. If the iteration end condition is not met, the scheduling strategy to be evolved is continued to be screened for the evolution of the genetic operator and the update of the scheduling strategy set, and the strategy evaluation index of each scheduling strategy in the updated scheduling strategy set is calculated until the strategy evaluation index meets the iteration end condition. At this time, the scheduling strategy in the scheduling strategy set is the target scheduling strategy after screening and iterative evolution.

[0091] Based on the above-mentioned method for determining the scheduling strategy, an embodiment of the present invention further provides a method for vehicle scheduling, comprising: obtaining real-time order information of the real-time transport order to be processed, and real-time status information of the vehicle to be scheduled corresponding to the real-time transport order; according to the target scheduling strategy, combining the real-time order information and the real-time status information of the vehicle to be scheduled, determining a scheduling vehicle that meets the requirements of the real-time transport order from the vehicles to be scheduled, wherein the target scheduling strategy is determined by any one of the methods for determining the scheduling strategy of the embodiments of the present invention.

[0092] Specifically, in the online vehicle dispatching business, various transport orders of planned and unplanned demands are received, and the target dispatching strategy newly determined by the above method is obtained, and the real-time order information of the real-time transport order and the real-time status information of the corresponding vehicle to be dispatched are input into the vehicle dispatching system built based on the target dispatching strategy. The vehicle dispatching system uses the target dispatching strategy to match the vehicle to be dispatched with the real-time transport order, and can quickly determine the corresponding dispatching vehicle for the real-time transport order. If there are multiple dispatching vehicles, multiple dispatching vehicles and corresponding vehicle information can also be pushed to the dispatching staff, and the dispatching staff will select a target dispatching vehicle to realize vehicle dispatching.

[0093] The vehicle dispatching method and the method for determining the dispatching strategy provided by the embodiment of the present invention determine the target dispatching strategy based on a genetic algorithm, and automatically evolve the target dispatching strategy that best suits the current environmental needs, avoiding the general strategies such as the dispatching strategy that minimizes the fulfillment cost and the dispatching strategy that minimizes the global running distance, making the dispatching more reasonable and achieving the global optimum in terms of effect, rather than the dispatching strategy that relies solely on expert experience, and based on the target dispatching strategy that is constructed and evolved from order information and status information, it is easier to quickly and efficiently determine the dispatching vehicle from each vehicle to be dispatched. In addition, the embodiment of the present invention uses small sample data to evaluate the strategy evaluation index of the dispatching strategy based on the sample extraction strategy, which can save computing resources and improve the efficiency of determining the target dispatching strategy.

[0094] Figure 7 Schematic diagram of the main modules of the device for determining the scheduling strategy according to an embodiment of the present invention. Figure 7 As shown, the scheduling strategy determination device 700 mainly includes a sample data acquisition module 701, a strategy set construction module 702 and a target strategy determination module 703.

[0095] The sample data acquisition module 701 is used to acquire sample data, each sample data includes order information of a transport order to be processed and status information of a vehicle to be dispatched corresponding to the transport order;

[0096] A strategy set building module 702, configured to build a scheduling strategy set according to the order information and the status information;

[0097] The target strategy determination module 703 is used to iteratively perform the following operations until a preset iteration end condition is met, and use the scheduling strategy in the scheduling strategy set as the target scheduling strategy: use each scheduling strategy in the scheduling strategy set to perform vehicle scheduling on the transport order in the sample data, and calculate the strategy evaluation index of each scheduling strategy in the scheduling strategy set according to the vehicle scheduling result; according to the strategy evaluation index, screen each scheduling strategy in the scheduling strategy set, use a genetic algorithm to evolve the screened scheduling strategies, and use the evolved scheduling strategies to update the scheduling strategy set.

[0098] According to one embodiment of the present invention, the strategy set construction module 702 is also used to: extract order elements from the order information and assign corresponding order character representations to the order elements; extract status elements from the status information and assign corresponding status character representations to the status elements; generate a specified number of scheduling strategies based on the order character representation, the status character representation and a preset operator set to form a scheduling strategy set.

[0099] According to another embodiment of the present invention, the policy set construction module 702 is further used to: combine the order character representation, the status character representation and the operator set into a policy element set; for each scheduling policy, generate it according to the following method: extract policy elements from the policy element set; in response to the extraction result being an operator, take the operator as the root node, and extract other elements except the operator from the policy element set as leaf nodes of the root node; loop through the following operations until the order character representation or the status character representation is extracted from the policy element set, or the level of the leaf node reaches a preset level threshold: in response to the leaf node being an operator, extract other elements except the elements that have been extracted from the policy element set as leaf nodes of the next level of the leaf node.

[0100] According to another embodiment of the present invention, the sample data is sampled from the historical full data according to a preset extraction strategy; the target strategy determination module 703 is also used for: for each scheduling strategy in the scheduling strategy set: calculating the scheduling priority of each of the vehicles to be scheduled when the transport order executes the scheduling strategy; according to the scheduling priority, selecting the scheduling vehicle corresponding to the highest scheduling priority from each of the vehicles to be scheduled.

[0101] According to another embodiment of the present invention, the target strategy determination module 703 is also used to: select the scheduling strategy to be evolved from the scheduling strategy set according to the strategy evaluation index; use a preset genetic operator to perform strategy evolution on the scheduling strategy to be evolved to obtain an evolved scheduling strategy.

[0102] According to another embodiment of the present invention, each scheduling strategy in the scheduling strategy set is stored in a tree structure; the target strategy determination module 703 is also used to: use at least one of a crossover operator, a mutation operator and a reconstruction operator to perform node evolution on some tree nodes in the scheduling strategy to be evolved.

[0103] According to another embodiment of the present invention, the strategy evaluation indicators include the transportation cost of the dispatched vehicle, the transportation efficiency of the dispatched vehicle and the dispatch waiting time; the iteration end conditions include that the strategy evaluation indicators of the evolved dispatching strategy do not reach a preset improvement threshold, and the number of loop iterations of the dispatching strategy set reaches a preset number threshold.

[0104] Figure 8 is an exemplary system architecture diagram to which embodiments of the present invention can be applied.

[0105] like Figure 8 As shown, system architecture 800 may include terminal devices 801, 802, 803, a network 804, and a server 805. Network 804 is used to provide a medium for communication links between terminal devices 801, 802, 803 and server 805. Network 804 may include various connection types, such as wired, wireless communication links, or optical fiber cables, etc.

[0106] Users can use terminal devices 801, 802, 803 to interact with server 805 through network 804 to receive or send messages, etc. Terminal devices 801, 802, 803 can be installed with various communication client applications, such as policy determination applications, etc. (only as an example).

[0107] The terminal devices 801 , 802 , and 803 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0108] The server 805 may be a server that provides various services, such as a background management server (only as an example) that supports the determination of the dispatching strategy by the user using the terminal devices 801, 802, and 803. The background management server may obtain sample data, each of which includes order information of the transport order to be processed and status information of the vehicle to be dispatched corresponding to the transport order; a dispatching strategy set is constructed according to the order information and the status information; the following operations are performed iteratively in a loop until the preset iteration end condition is met, and the dispatching strategy in the dispatching strategy set is used as the target dispatching strategy: each dispatching strategy in the dispatching strategy set is used to dispatch the vehicle for the transport order in the sample data, and the strategy evaluation index of each dispatching strategy in the dispatching strategy set is calculated according to the vehicle dispatching result; each dispatching strategy in the dispatching strategy set is screened according to the strategy evaluation index, the screened dispatching strategy is evolved using a genetic algorithm, the dispatching strategy set is updated using the evolved dispatching strategy, and the processing result (such as the target dispatching strategy, etc. - only as an example) is fed back to the terminal device.

[0109] It should be noted that the method for determining the scheduling strategy provided in the embodiment of the present invention is generally executed by the server 805 , and accordingly, the device for determining the scheduling strategy is generally arranged in the server 805 .

[0110] It should be understood that Figure 8 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to the implementation requirements.

[0111] Reference below Fig. 9 , Fig. 9 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. Fig. 9 The terminal device or server 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.

[0112] like Fig. 9 As shown, the computer system 900 includes a central processing unit (CPU) 901, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 902 or a program loaded from a storage part 908 into a random access memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the system 900 are also stored. The CPU 901, the ROM 902, and the RAM 903 are connected to each other through a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0113] The following components are connected to the I / O interface 905: an input section 906 including a keyboard, a mouse, etc.; an output section 907 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, a modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the I / O interface 905 as needed. A removable medium 911, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 910 as needed, so that a computer program read therefrom is installed into the storage section 908 as needed.

[0114] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 909, and / or installed from the removable medium 911. When the computer program is executed by the central processing unit (CPU) 901, the above-mentioned functions defined in the system of the present invention are executed.

[0115] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More 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 or 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 of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination thereof.

[0116] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0117] The units involved in the embodiments of the present invention may be implemented by software or hardware. The units described may also be set in a processor, for example, it may be described as: a processor includes: a sample data acquisition module, a strategy set construction module and a target strategy determination module.

[0118] The names of these modules do not, in some cases, limit the modules themselves. For example, the strategy set building module can also be described as "a module for building a scheduling strategy set according to the order information and the status information."

[0119] On the other hand, the present invention also provides a computer-readable medium, which may be included in the device described in the embodiment; or it may exist independently and not be assembled into the device. The computer-readable medium carries one or more programs, and when the one or more programs are executed by a device, the device includes: obtaining sample data, each sample data includes order information of a transport order to be processed and status information of a vehicle to be dispatched corresponding to the transport order; constructing a scheduling strategy set according to the order information and the status information; iteratively performing the following operations until a preset iteration end condition is met, and taking the scheduling strategy in the scheduling strategy set as the target scheduling strategy: using each scheduling strategy in the scheduling strategy set to dispatch vehicles for the transport order in the sample data, and calculating the strategy evaluation index of each scheduling strategy in the scheduling strategy set according to the vehicle dispatch result; screening each scheduling strategy in the scheduling strategy set according to the strategy evaluation index, evolving the screened scheduling strategy using a genetic algorithm, and updating the scheduling strategy set using the evolved scheduling strategy.

[0120] The technical solution according to the embodiment of the present invention has the following advantages or beneficial effects: by obtaining sample data, each sample data includes order information of the transport order to be processed and status information of the vehicle to be dispatched corresponding to the transport order; constructing a scheduling strategy set according to the order information and the status information; iteratively performing the following operations until the preset iteration end condition is met, and taking the scheduling strategy in the scheduling strategy set as the target scheduling strategy: using each scheduling strategy in the scheduling strategy set to perform vehicle scheduling for the transport order in the sample data, and calculating the strategy evaluation index of each scheduling strategy in the scheduling strategy set according to the vehicle scheduling result; screening each scheduling strategy in the scheduling strategy set according to the strategy evaluation index, evolving the screened scheduling strategy using a genetic algorithm, and updating the scheduling strategy set using the evolved scheduling strategy. The technical solution realizes a method for automatically determining the scheduling strategy that meets the needs of the current environment, does not require expert experience, and automatically evolves and generates the target scheduling strategy based on the idea of ​​genetic evolution, and then dynamically, real-time and efficiently performs vehicle scheduling according to the target scheduling strategy. It has strong environmental adaptability and can better meet scheduling needs.

[0121] The specific implementation methods described herein do not constitute limitations on the scope of protection of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention shall be included in the scope of protection of the present invention.

Claims

1. A method for determining a scheduling strategy, characterized in that: include: Acquire sample data, each sample data including order information of a transport order to be processed and status information of a vehicle to be dispatched corresponding to the transport order; Constructing a scheduling strategy set according to the order information and the status information; The following operations are performed in a loop iteratively until a preset iteration end condition is met, and the scheduling strategy in the scheduling strategy set is used as the target scheduling strategy: each scheduling strategy in the scheduling strategy set is used to perform vehicle scheduling on the transport order in the sample data, and the strategy evaluation index of each scheduling strategy in the scheduling strategy set is calculated according to the vehicle scheduling result; each scheduling strategy in the scheduling strategy set is screened according to the strategy evaluation index, a genetic algorithm is used to evolve the screened scheduling strategies, and the evolved scheduling strategies are used to update the scheduling strategy set.

2. The method according to claim 1, characterized in that: Constructing a scheduling strategy set according to the order information and the status information, including: Extracting order elements from the order information and assigning corresponding order character representations to the order elements; Extracting status elements from the status information and assigning corresponding status character representations to the status elements; According to the order character representation, the status character representation and a preset operator set, a specified number of scheduling strategies are generated to form a scheduling strategy set.

3. The method according to claim 2, characterized in that Generate a specified number of scheduling strategies according to the order character representation, the status character representation and a preset operator set, including: Combining the order character representation, the status character representation and the operator set into a strategy element set; For each scheduling strategy, generate it as follows: Extracting policy elements from the policy element set; In response to the extraction result being an operator, taking the operator as a root node, extracting other elements except the operator from the policy element set as leaf nodes of the root node; The following operations are performed cyclically until the order character representation or the status character representation is extracted from the policy element set, or the level of the leaf node reaches a preset level threshold: In response to the leaf node being an operator, other elements except the elements that have been extracted are extracted from the policy element set as leaf nodes of the next level of the leaf node.

4. The method according to claim 1, characterized in that The sample data is sampled from the full amount of historical data according to a preset extraction strategy; Using each scheduling strategy in the scheduling strategy set to perform vehicle scheduling on the transport order in the sample data includes: For each scheduling strategy in the scheduling strategy set: calculating the scheduling priority of each of the to-be-scheduled vehicles when the transport order executes the scheduling strategy; According to the dispatching priority, a dispatching vehicle corresponding to the highest dispatching priority is selected from each of the vehicles to be dispatched.

5. The method according to claim 1, characterized in that: The scheduling strategies in the scheduling strategy set are screened according to the strategy evaluation index, and the screened scheduling strategies are evolved using a genetic algorithm, including: According to the strategy evaluation index, the scheduling strategy to be evolved is selected from the scheduling strategy set; and the scheduling strategy to be evolved is evolved by using a preset genetic operator to obtain an evolved scheduling strategy.

6. The method according to claim 5, characterized in that Each scheduling strategy in the scheduling strategy set is stored in a tree structure; The scheduling strategy to be evolved is evolved by using a preset genetic operator, including: At least one of a crossover operator, a mutation operator and a reconstruction operator is used to perform node evolution on some tree nodes in the scheduling strategy to be evolved.

7. The method according to claim 1, characterized in that The strategy evaluation indicators include the transportation cost of the dispatched vehicle, the transportation efficiency of the dispatched vehicle and the dispatch waiting time; the iteration end conditions include that the strategy evaluation indicators of the evolved dispatching strategy do not reach the preset improvement threshold, and the number of loop iterations of the dispatching strategy set reaches the preset number threshold.

8. A vehicle dispatching method, characterized in that: include: Acquire real-time order information of a real-time transport order to be processed, and real-time status information of a vehicle to be dispatched corresponding to the real-time transport order; According to the target scheduling strategy, combined with the real-time order information and the real-time status information of the vehicle to be scheduled, a scheduling vehicle that meets the real-time transportation order requirements is determined from the vehicles to be scheduled, wherein the target scheduling strategy is determined according to any one of the methods in claims 1 to 7.

9. A device for determining a scheduling strategy, characterized in that: include: A sample data acquisition module, used to acquire sample data, each sample data includes order information of a transport order to be processed and status information of a vehicle to be dispatched corresponding to the transport order; A strategy set building module, used to build a scheduling strategy set according to the order information and the status information; The target strategy determination module is used to iteratively perform the following operations until a preset iteration end condition is met, and use the scheduling strategy in the scheduling strategy set as the target scheduling strategy: use each scheduling strategy in the scheduling strategy set to perform vehicle scheduling on the transport order in the sample data, and calculate the strategy evaluation index of each scheduling strategy in the scheduling strategy set according to the vehicle scheduling result; screen each scheduling strategy in the scheduling strategy set according to the strategy evaluation index, use a genetic algorithm to evolve the screened scheduling strategies, and use the evolved scheduling strategies to update the scheduling strategy set.

10. A mobile electronic device terminal, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

11. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images