Vehicle and goods matching method, electronic equipment and readable storage medium

Abstract

The invention provides a vehicle and goods matching method, electronic equipment and a readable storage medium. The vehicle and goods matching method comprises the following steps: in response to a goods source matching request of a target driver on a logistics transportation platform, determining a target departure place, a target destination, a target vehicle length and a target vehicle type; generating a target screening condition based on the target departure place, the target destination, the target vehicle length and the target vehicle type; according to the target screening condition and the candidate goods sources, determining a first goods source set; based on the feature information of the target driver and the attribute information of the first goods sources, carrying out screening and first sorting processing on the M first goods sources to obtain a second goods source set; based on the target weight, performing second sorting processing on N second goods sources in a second goods source set to obtain a target goods source set; wherein the target weight is at least one of a behavior distance weight, a behavior time weight, a vehicle length consistency weight and a vehicle type consistency weight; and responding to the goods source matching request based on the target goods source set.

Description

Technical Field

[0001] The present application relates to the field of logistics and transportation technology, and in particular to a vehicle-cargo matching method, an electronic device, and a readable storage medium. Background Art

[0002] With the rapid development of Internet technology, the matching of vehicles and cargoes in the logistics industry has gradually transitioned from the traditional offline operation mode to a unified online platform operation. More and more drivers and cargo owners rely on comprehensive logistics and transportation platforms to achieve the connection between cargo sources and transportation needs. By integrating a large amount of cargo source information, the logistics and transportation platform provides cargo owners with a wide range of transportation capacity options, and also provides drivers with abundant freight opportunities.

[0003] However, existing vehicle-cargo matching mainly relies on a fixed rule-based matching system. This traditional rule-based matching system usually performs preliminary screening based on some simple conditions, such as region and cargo type, and fails to deeply consider the driver's personalized needs and transportation preferences, resulting in poor matching accuracy and difficulty for drivers to quickly find suitable cargo sources; at the same time, the needs of cargo owners are often not met in time due to the slow response of drivers, affecting the user experience. Summary of the invention

[0004] The present application provides a vehicle-cargo matching method, an electronic device and a readable storage medium to solve the problems of poor vehicle-cargo matching accuracy, slow response speed and poor user experience in the existing vehicle-cargo matching process.

[0005] In a first aspect, the present application provides a vehicle-cargo matching method, comprising:

[0006] In response to a cargo matching request from a target driver on a logistics transportation platform, a target departure point and a target destination of the target driver, as well as a target length and a target vehicle type of the target transportation vehicle are determined; wherein the target transportation vehicle is the vehicle of the target driver;

[0007] Generate a target screening condition based on the target departure place, the target destination, the target vehicle length and the target vehicle type;

[0008] Determine a first supply source set according to the target screening condition and the candidate supply sources; wherein the first supply source set includes M first supply sources, M is a positive integer, and M≥1; the first supply sources are candidate supply sources that meet the target screening condition;

[0009] Based on the characteristic information of the target driver and the attribute information of the first source of goods, the M first sources of goods are screened and sorted to obtain a second source of goods set; wherein the second source of goods set includes N second sources of goods, N is a positive integer, and N≤M;

[0010] Based on the target weight, performing a second sorting process on the N second cargo sources in the second cargo source set to obtain a target cargo source set; wherein the target weight is at least one of a behavior distance weight, a behavior time weight, a vehicle length consistency weight, and a vehicle type consistency weight;

[0011] Based on the target supply set, respond to the supply matching request.

[0012] In a possible design, generating a target screening condition based on the target departure place, the target destination, the target vehicle length and the target vehicle type includes:

[0013] Determining a first screening condition according to the target departure place and the target vehicle length;

[0014] Determining a second screening condition according to the target destination and the first historical behavior data of the target driver;

[0015] Determine a third screening condition according to a plurality of first historical orders corresponding to the target vehicle length; wherein the first historical order is an order with an actual vehicle length of the target vehicle length completed on the logistics transportation platform within the preset time period;

[0016] Determining a fourth screening condition according to the second historical behavior data corresponding to the target driver;

[0017] Determining a fifth screening condition according to a second historical order corresponding to the target vehicle model; wherein the second historical order is an order for the target vehicle model completed on the logistics transportation platform within the preset time period;

[0018] Determining a sixth screening condition according to the third historical behavior data corresponding to the target driver;

[0019] The target filtering condition is determined according to the first filtering condition, the second filtering condition, the third filtering condition, the fourth filtering condition, the fifth filtering condition and the sixth filtering condition.

[0020] In a possible design, determining the first screening condition according to the target departure place and the target vehicle length includes:

[0021] According to the target vehicle length, determining a basic empty driving distance, an empty driving coefficient and a first empty driving distance corresponding to the target vehicle length;

[0022] Determine the product of the empty driving coefficient and the transportation distance of the candidate cargo source;

[0023] Determine a second empty driving distance as a sum of the product and the basic empty driving distance;

[0024] The first screening condition is determined based on the target departure point, the first empty driving distance and the second empty driving distance; wherein the first screening condition is that the distance between the shipping place of the first source of goods and the target departure point is less than a third empty driving distance; and the third empty driving distance is the minimum value of the first empty driving distance and the second empty driving distance.

[0025] In a possible design, determining the second screening condition according to the target destination and the first historical behavior data of the target driver includes:

[0026] Obtain the number of searches, clicks, phone calls, and transactions of the target driver on the logistics transportation platform for each of P cities within the preset time period; wherein P is a positive integer, and P≥2;

[0027] Calculate the behavior score for each city based on the number of search behaviors, the number of click behaviors, the number of phone calls, the number of transaction behaviors, the weight of the search behaviors, the weight of the click behaviors, the weight of the phone calls, and the weight of the transaction behaviors;

[0028] Determine Q target cities according to the behavior score of each city in the P cities; wherein the target cities are cities with the behavior scores in the first Q cities among the P cities, Q is a positive integer, and Q≤P;

[0029] Based on the target destination and the Q target cities, the second screening condition is determined; wherein the second screening condition is that the delivery place of the first source of goods is the target destination or the target city.

[0030] In a possible design, determining the third screening condition according to the first historical order corresponding to the target vehicle length includes:

[0031] For each first historical order among the plurality of first historical orders, determining a minimum vehicle length of the first historical order;

[0032] Determine the number of the first historical orders corresponding to each minimum vehicle length among multiple minimum vehicle lengths;

[0033] Arrange the quantities of the first historical orders corresponding to the multiple minimum vehicle lengths in descending order and perform normalization processing to obtain normalized first values ​​corresponding to the multiple minimum vehicle lengths respectively;

[0034] Based on the normalized first values ​​corresponding to the multiple minimum vehicle lengths, a third filtering condition is determined; wherein, the third filtering condition is that the minimum vehicle length of the first source of cargo is the first target minimum vehicle length; the first target minimum vehicle length is the minimum vehicle length among the multiple minimum vehicle lengths whose normalized first value is greater than a first threshold.

[0035] In a possible design, based on the characteristic information of the target driver and the attribute information of the first source of goods, the M first sources of goods are screened and sorted to obtain a second source of goods set, including:

[0036] Based on the characteristic information of the target driver and the attribute information of the first source of goods, the M first sources of goods are screened to obtain a third source of goods set; wherein the third source of goods set includes the N second sources of goods arranged in the first order;

[0037] Based on the characteristic information of the target driver and the attribute information of the second source of goods, performing a first sorting process on the N second sources of goods in the third source of goods set to obtain the second source of goods set; wherein the second source of goods set includes the N second sources of goods sorted in a second order, the second order is different from the first order, and each second source of goods corresponds to a first score;

[0038] Among them, the characteristic information of the target driver includes the probability of the target driver going to each city, the probability of the target driver residing in each city, the straight-line distance between the target departure point of the target driver and the shipping place of the first source of cargo or the second source of cargo, and the time interval between the release time of the first source of cargo or the second source of cargo and the release time of the source of cargo matching request.

[0039] In a possible design, performing a second sorting process on the N second supply sources in the second supply source set based on the target weight to obtain a target supply source set includes:

[0040] For each second source of supply in the second source of supply set, when the second source of supply meets the target condition, obtaining a second score corresponding to the second source of supply based on the target weight and the first score corresponding to the second source of supply;

[0041] Sorting the N second sources of supply based on the second scores corresponding to the N second sources of supply in descending order to obtain the target source set;

[0042] Wherein, the target condition is at least one of the following:

[0043] The distance between the shipping place of the second source of goods and the target departure place is less than a preset distance;

[0044] The time interval between the release time of the second supply source and the release time of the supply source matching request is less than a preset time interval;

[0045] The required vehicle length of the second cargo source is the same as the target vehicle length;

[0046] The vehicle model required by the second source of goods is the same as the target vehicle model.

[0047] In a possible design, responding to the supply source matching request based on the target supply source set includes:

[0048] Based on the target departure place and the target destination, determine the precise source of goods and the generalized source of goods in the target source of goods set; wherein the precise source of goods is a second source of goods whose place of shipment is consistent with the target departure place and whose place of receipt is consistent with the target destination; the generalized source of goods is other second sources of goods in the target source of goods set except the precise source of goods;

[0049] Sorting the precise sources of goods in descending order of the second scores corresponding to the precise sources of goods to obtain a precise source sequence;

[0050] Sorting the generalized supply sources in descending order of the second scores corresponding to the generalized supply sources to obtain a generalized supply source sequence;

[0051] Based on the second score corresponding to the precise source of supply and the second score corresponding to the generalized source of supply, and a preset rule, the generalized source of supply sequence is inserted and displayed when the precise source of supply sequence is displayed.

[0052] In a second aspect, the present application provides a vehicle-cargo matching device, comprising: a memory and at least one processor;

[0053] The memory stores computer-executable instructions;

[0054] The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor performs the method described in the first aspect or various possible designs of the first aspect.

[0055] In a third aspect, the present application provides an electronic device, characterized in that it includes: a memory and at least one processor;

[0056] The memory stores computer-executable instructions;

[0057] The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor performs the method described in the first aspect or various possible designs of the first aspect.

[0058] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer execution instructions are stored. When the computer execution instructions are executed, the method described in the first aspect or various possible designs of the first aspect is implemented.

[0059] The vehicle-cargo matching method provided in the embodiment of the present application firstly responds to the cargo matching request of the target driver on the logistics transportation platform, determines the target departure place and target destination of the target driver, and the target vehicle length and target vehicle model of the target transportation vehicle, and generates target screening conditions based on the target departure place, target destination, target vehicle length and target vehicle model, so as to determine a first cargo source set according to the target screening conditions and the candidate cargo sources; secondly, based on the characteristic information of the target driver and the attribute information of the first cargo source, performs a first sorting process on the M first cargo sources to obtain a second cargo source set; finally, based on the target weight, performs a second sorting process on the N second cargo sources in the second cargo source set to obtain a target cargo source set, and responds to the cargo source matching request based on the target cargo source set. This application accurately obtains the target driver's departure point, destination, length and model information of the target vehicle, and through a multi-step cargo screening and sorting algorithm, ensures that the matched cargo is highly consistent with the target driver and the target vehicle, greatly improving the efficiency and accuracy of vehicle-cargo matching, facilitating rapid response to cargo sources released by cargo owners, and providing a good user experience. It effectively solves the problem of poor accuracy and slow response speed in vehicle-cargo matching in traditional rule-based matching systems, which leads to poor user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A schematic diagram of a vehicle-cargo matching method provided in an embodiment of the present application;

[0061] Figure 2 A flowchart of a method for generating target screening conditions provided in an embodiment of the present application;

[0062] Figure 3 A flowchart of a method for determining a first screening condition provided in an embodiment of the present application;

[0063] Figure 4 A flowchart of a method for determining a second screening condition provided in an embodiment of the present application;

[0064] Figure 5 A flowchart of a method for determining a third screening condition provided in an embodiment of the present application;

[0065] Figure 6 A schematic diagram of the structure of a vehicle-cargo matching device provided in an embodiment of the present application;

[0066] Figure 7A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0067] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0068] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used in the specification of the application are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification, claims and drawings of this application are intended to cover non-exclusive inclusions.

[0069] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase "embodiments" in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0070] The term "and / or" in this article is only a description of the association relationship of the associated objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists, A and B can exist at the same time, and B exists. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0071] In addition, the terms "first", "second", etc. in the specification and claims of this application or the above-mentioned drawings are used to distinguish different objects rather than to describe a specific order, and may explicitly or implicitly include one or more of the features.

[0072] In the description of this application, unless otherwise specified, "multiple" and "at least two" mean more than two (including two). Similarly, "multiple groups" and "at least two groups" mean more than two groups (including two).

[0073] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense. For example, "connected" or "connection" can refer to not only physical connection, but also electrical connection or signal connection. For example, it can be directly connected, that is, physically connected, or indirectly connected through at least one intermediate element, as long as the circuit is connected, or it can be the internal connection of two elements; signal connection can refer to signal connection through a circuit or through a media medium, such as radio waves. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0074] In order to make those skilled in the art better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the accompanying drawings. It should be noted that different technical features in the present application can be combined with each other without conflict.

[0075] Figure 1 A flow chart of a vehicle-cargo matching method provided in an embodiment of the present application. Figure 1 As shown, the vehicle-cargo matching method provided in the embodiment of the present application specifically includes S101 to S106, and S101 to S106 are described in detail below.

[0076] It should be noted that the executor of the vehicle-cargo matching method provided in the embodiment of the present application may be a processor.

[0077] S101. The processor determines the target departure point and target destination of the target driver, and the target length and target model of the target transport vehicle in response to the cargo matching request of the target driver on the logistics transportation platform.

[0078] Among them, the target driver is the driver who sends a cargo matching request on the logistics transportation platform, and the target transportation vehicle is the target driver's vehicle.

[0079] It should be noted that the target transport vehicle refers to the vehicle registered by the target driver on the logistics transportation platform and will be used to perform transportation tasks.

[0080] Among them, the logistics and transportation platform refers to an online platform that integrates cargo transportation demand and vehicle capacity resources through information technology and the Internet, and provides docking services between cargo owners and drivers. The core function of the logistics and transportation platform is to help cargo owners find suitable transportation vehicles and drivers find suitable cargo sources, thereby efficiently matching vehicles and cargo.

[0081] It should be noted that the logistics and transportation platform can exist in the form of a mobile phone APP, a web platform, or other forms, and this embodiment does not specifically limit this.

[0082] The target departure point refers to the location where the vehicle of the target driver is ready to depart, that is, the starting point of the vehicle transportation task.

[0083] It should be noted that the target departure point can be manually input by the target driver, or obtained through automatic GPS positioning, and this embodiment does not specifically limit this.

[0084] Among them, the target destination refers to the end point or unloading location where the target driver hopes to deliver the goods, that is, the transportation end point of the goods.

[0085] It should be noted that the target destination may be manually input by the target driver, or may be recommended by the processor based on the first historical behavior data of the target driver.

[0086] The target vehicle length refers to the length of the target transport vehicle, usually expressed in meters.

[0087] It should be noted that the vehicle length is one of the important factors that affect whether the cargo can be loaded. Different cargo sources have different requirements for vehicle length. For example, some large cargo sources require a longer vehicle compartment for transportation. Therefore, the platform needs to match the appropriate cargo source according to the target vehicle length to ensure that the driver's vehicle has enough space to load the cargo source.

[0088] Among them, the target vehicle type refers to the type of target transport vehicle (such as flatbed trucks, vans, refrigerated trucks, hazardous goods trucks, etc.).

[0089] It should be noted that the vehicle type determines the type of cargo that the target transport vehicle is suitable for transporting. The target vehicle type can be automatically determined by the vehicle information provided by the target driver when registering on the logistics transportation platform, or can be manually updated by the target driver, which is not specifically limited in this embodiment.

[0090] Among them, there is a "Cargo Matching" control on the main interface of the logistics and transportation platform. The target driver can enter the cargo matching interface by clicking the "Cargo Matching" control and publish a cargo matching request on the cargo matching interface.

[0091] In one implementation, the target driver may generate a cargo matching request by actively inputting the target departure point and the target destination in the cargo matching interface.

[0092] It should be noted that the target length and target model of the target transport vehicle corresponding to the target driver have been stored in the processor when the target driver registered. The processor can automatically use the previously stored target length and target model, in conjunction with the target departure point and target destination entered by the target driver, to form a cargo matching request.

[0093] In this embodiment, the processor can more accurately screen out the cargo sources that meet the needs of the target driver through the target departure point and target destination manually input by the target driver, avoid matching irrelevant cargo sources, and improve the efficiency and accuracy of vehicle-cargo matching.

[0094] In another embodiment, after the target driver enters the cargo matching interface, the processor may use the target driver's current location as the target departure point of the target driver and the city frequently clicked by the target driver as the target destination to generate a cargo matching request.

[0095] It should be noted that the processor can obtain the target driver's frequently visited destination based on the target driver's first historical behavior data on the logistics transportation platform. The method by which the processor obtains the target driver's frequently visited destination based on the target driver's first historical behavior data is an existing method, which will not be described in detail in this embodiment.

[0096] In this embodiment, the processor automatically obtains the current location of the target driver and the city frequently clicked by the target driver, and uses the current location of the target driver as the target departure point and the city frequently clicked by the target driver as the target destination, which significantly simplifies the operation steps in the matching process and improves the user experience.

[0097] S102: The processor generates a target screening condition based on the target departure place, the target destination, the target vehicle length and the target vehicle type.

[0098] It should be noted that generating target screening conditions is a key step to ensure that the matching cargo sources meet the needs of the target drivers. By setting clear target screening conditions, the processor can quickly screen out the first cargo source that meets the target screening conditions from a large number of candidate cargo sources, avoid interference from irrelevant cargo sources, and improve the efficiency of vehicle-cargo matching.

[0099] S103. The processor determines a first supply source set according to the target screening condition and the candidate supply sources.

[0100] Among them, the logistics and transportation platform maintains a candidate cargo source database, which contains the cargo source information of all current drivers to be matched.

[0101] It should be noted that the processor can extract information of all candidate sources of supply by accessing the candidate source database, so as to facilitate subsequent screening of the candidate sources of supply according to target screening conditions.

[0102] Among them, the first source set includes M first sources, M is a positive integer, and M≥1; the first source is a candidate source that meets the target screening conditions.

[0103] It should be noted that the processor screens the candidate sources in the candidate source database according to the target screening conditions, filters out the candidate sources that do not meet the target screening conditions, and uses the candidate sources that meet the target screening conditions as the first source.

[0104] In this embodiment, candidate sources of cargo are screened by target screening conditions, candidate sources of cargo that do not meet the target screening conditions are deleted, and a broader first source of cargo set is generated based on candidate sources of cargo that meet the target screening conditions, which significantly reduces the complexity of subsequent vehicle-cargo matching and can ensure that the first source of cargo has a certain degree of match with the needs of the target driver.

[0105] S104. The processor screens and performs a first sorting process on the M first sources of goods based on the characteristic information of the target driver and the attribute information of the first sources of goods to obtain a second source of goods set.

[0106] Among them, the second source set includes N second sources of supply, N is a positive integer, and N≤M.

[0107] It should be noted that by introducing the characteristic information of the target driver, the M first sources of goods in the first source set are further screened and sorted to obtain the second source set. Compared with the first source set, the range of the second source set may be smaller, and the second source set sorts the N second sources of goods, which effectively improves the accuracy of personalized matching and can better meet the needs of the target driver.

[0108] S105. The processor performs a second sorting process on the N second sources of supply in the second source set based on the target weight to obtain a target source set.

[0109] Among them, the target weight is at least one of the behavior distance weight, the behavior time weight, the vehicle length consistency weight, and the vehicle type consistency weight.

[0110] It should be noted that the target weight can be any one of the behavior distance weight, behavior time weight, vehicle length consistency weight, and vehicle type consistency weight, or it can be the behavior distance weight and the behavior time weight, or it can be the behavior distance weight and the vehicle length consistency weight, or it can be the behavior distance weight and the vehicle type consistency weight, or it can be the behavior distance weight, the behavior time weight, and the vehicle length consistency weight, or it can be the behavior distance weight, the behavior time weight, and the vehicle type consistency weight, or it can be the behavior distance weight, the vehicle length consistency weight, and the vehicle type consistency weight, or it can be the behavior distance weight, the behavior time weight, the vehicle length consistency weight, and the vehicle type consistency weight. The specific weight can be determined according to the actual usage situation, and this embodiment does not make any specific limitation on this.

[0111] In this embodiment, by considering factors such as distance, time, and vehicle matching degree, the sorting of N second cargo sources is further optimized, so that the target driver can obtain the cargo matching result that best suits his actual situation, thereby improving the scientific nature of vehicle-cargo matching and the feasibility in actual operation.

[0112] S106. The processor responds to the supply matching request based on the target supply set.

[0113] It should be noted that after obtaining the target cargo source set, the processor displays the target cargo source set to the target driver to respond to the cargo source matching request issued by the target driver.

[0114] In one implementation, the processor may display the N second sources of cargo to the target driver in sequence according to the arrangement order of the N second sources of cargo in the target source set.

[0115] In an embodiment of the present application, first, in response to a cargo matching request from a target driver on a logistics transportation platform, the target departure point and target destination of the target driver, as well as the target vehicle length and target vehicle model of the target transportation vehicle are determined, and based on the target departure point, target destination, target vehicle length and target vehicle model, a target screening condition is generated to determine a first cargo source set according to the target screening condition and the candidate cargo sources; secondly, based on the characteristic information of the target driver and the attribute information of the first cargo source, the M first cargo sources are screened and sorted by a first process to obtain a second cargo source set; finally, based on the target weight, the N second cargo sources in the second cargo source set are sorted by a second process to obtain a target cargo source set, and based on the target cargo source set, a response is given to the cargo matching request. This application accurately obtains the target driver's departure point, destination, length and model information of the target vehicle, and through a multi-step cargo screening and sorting algorithm, ensures that the matched cargo is highly consistent with the target driver and the target vehicle, greatly improving the efficiency and accuracy of vehicle-cargo matching, facilitating rapid response to cargo sources released by cargo owners, and providing a good user experience. It effectively solves the problem of poor accuracy and slow response speed in vehicle-cargo matching in traditional rule-based matching systems, which leads to poor user experience.

[0116] In the above embodiment, the processor needs to generate the target screening condition based on the target departure place, the target destination, the target vehicle length and the target vehicle type. Next, the specific process of the processor generating the target screening condition based on the target departure place, the target destination, the target vehicle length and the target vehicle type is described in detail.

[0117] Figure 2 A flow chart of a method for generating target screening conditions provided in an embodiment of the present application. Figure 2As shown, in a possible embodiment, S102, the processor generates a target screening condition based on the target departure place, target destination, target vehicle length and target vehicle type, which can be implemented through S1021 to S1027. S1021 to S1027 are described in detail below.

[0118] S1021. The processor determines a first screening condition according to the target departure point and the target vehicle length.

[0119] It should be noted that the first screening condition is used to screen candidate cargo sources based on the distance between their shipping location and the target departure location, thereby reducing the number of candidate cargo sources, avoiding invalid long-distance dispatching, and improving vehicle-cargo matching efficiency.

[0120] S1022. The processor determines a second screening condition according to the target destination and the first historical behavior data of the target driver.

[0121] Among them, the first historical behavior data can reflect the preferences, habits, frequently visited places or driving routes of the target driver, so as to more personalized screen the sources of goods that meet the transportation habits of the target driver.

[0122] It should be noted that the second screening condition is used to screen candidate cargo sources based on their target destination and the first historical behavior data of the target driver, thereby improving the personalization of vehicle-cargo matching and ensuring that the cargo received by the driver not only meets the current target destination, but also meets his frequent routes or preferences, thereby improving the driver's satisfaction and the success rate of vehicle-cargo matching.

[0123] S1023. The processor determines a third screening condition according to multiple first historical orders corresponding to the target vehicle length.

[0124] Among them, the first historical order is an order with an actual vehicle length as the target vehicle length completed on the logistics transportation platform within a preset time period.

[0125] It should be noted that the first historical order refers to the historical transportation order completed on the logistics transportation platform within a preset time period using a vehicle with the same length as the target vehicle. By analyzing multiple first historical orders, the processor can more accurately select the source of goods that meet the requirements of the target vehicle length and improve the accuracy of vehicle-cargo matching.

[0126] It should be noted that the preset time period can be set by the management personnel of the logistics transportation platform, and this embodiment does not make any specific limitation to this.

[0127] For example, the preset time period is the past 90 days.

[0128] S1024. The processor determines a fourth filtering condition according to the second historical behavior data of the target driver.

[0129] Among them, the second historical behavior data refers to the number of search behaviors, click behaviors, phone calls and transaction behaviors of the target driver on the logistics transportation platform for each source of cargo with the minimum vehicle length within a preset time period.

[0130] It should be noted that the processor performs weighted summation based on the target driver's number of search behaviors, number of click behaviors, number of phone behaviors, number of transaction behaviors, as well as the weight of the search behavior, the weight of the click behavior, the weight of the phone behavior and the weight of the transaction behavior, to obtain the behavior score of each minimum vehicle length, and normalizes the behavior scores of different minimum vehicle lengths to obtain the normalized second values ​​corresponding to different minimum vehicle lengths.

[0131] The fourth screening condition is that the required vehicle length of the target driver is at least one of the second target minimum vehicle lengths. The second target minimum vehicle length is the minimum vehicle length whose second value after normalization corresponding to different minimum vehicle lengths is greater than the second threshold.

[0132] It should be noted that the second threshold value can be set by the management personnel of the logistics transportation platform, and this embodiment does not make any specific limitation to this.

[0133] For example, the second threshold is 0.1.

[0134] S1025. The processor determines a fifth screening condition according to the second historical order corresponding to the target vehicle model.

[0135] Among them, the second historical order is an order for the target vehicle model completed on the logistics transportation platform within a preset time period.

[0136] It should be noted that the second historical order refers to the historical transportation order completed on the logistics transportation platform within a preset time period using the same vehicle model as the target vehicle model. By analyzing multiple second historical orders, the processor can more accurately select the source of goods that meet the needs of the target vehicle model and improve the accuracy of vehicle-cargo matching.

[0137] It should be noted that the specific implementation method of S1025 is the same as the specific implementation method of S1023 in the following embodiment, and will not be repeated in this embodiment.

[0138] The fifth screening condition is that the demand model of the first source of goods is at least one of the first target models. The first target model is a model whose normalized value corresponding to different models is greater than the third threshold.

[0139] It should be noted that the third threshold value can be set by the management personnel of the logistics and transportation platform, and this embodiment does not make any specific limitation on this.

[0140] For example, the third threshold is 0.1.

[0141] S1026. The processor determines a sixth screening condition based on the third historical behavior data of the target driver.

[0142] Among them, the third historical behavior data refers to the number of search behaviors, click behaviors, phone behaviors and transaction behaviors of the target driver on the logistics transportation platform for each type of cargo within a preset time period.

[0143] It should be noted that the processor obtains the behavior score for each model based on the weighted summation of the target driver's search times, click times, phone calls, and transaction times for each model of cargo, as well as the weight of the search behavior, the weight of the click behavior, the weight of the phone call behavior, and the weight of the transaction behavior, and normalizes the behavior scores of different models to obtain the normalized third values ​​corresponding to different models.

[0144] The sixth screening condition is that the vehicle model required by the target driver is at least one of the second target vehicle models. The second target vehicle model is a vehicle model whose third value after normalization processing corresponding to different vehicle models is greater than the fourth threshold value.

[0145] It should be noted that the fourth threshold value can be set by the management personnel of the logistics and transportation platform, and this embodiment does not make any specific limitation to this.

[0146] For example, the fourth threshold is 0.1.

[0147] S1027. The processor determines a target filtering condition according to the first filtering condition, the second filtering condition, the third filtering condition, the fourth filtering condition, the fifth filtering condition, and the sixth filtering condition.

[0148] Among them, the target screening condition may include any one of the first screening condition, the second screening condition, the third screening condition, the fourth screening condition, the fifth screening condition and the sixth screening condition, or may include multiple of the first screening condition, the second screening condition, the third screening condition, the fourth screening condition, the fifth screening condition and the sixth screening condition, and this embodiment does not make any specific limitations on this.

[0149] It should be noted that the target filtering condition includes a first filtering condition, a second filtering condition, a third filtering condition, a fourth filtering condition, a fifth filtering condition and a sixth filtering condition. The processor integrates the first filtering condition, the second filtering condition, the third filtering condition, the fourth filtering condition, the fifth filtering condition and the sixth filtering condition to generate a comprehensive target filtering condition.

[0150] In an embodiment of the present application, by combining the first filtering condition, the second filtering condition, the third filtering condition, the fourth filtering condition, the fifth filtering condition and the sixth filtering condition in the four dimensions of departure place, destination, vehicle length and vehicle type, a more accurate target filtering condition can be generated, so as to facilitate screening out the first set of sources of cargo that meet the needs of the target driver from the candidate sources of cargo according to the target filtering condition, thereby greatly improving the accuracy of vehicle-cargo matching. At the same time, it also optimizes the recommendation capability of the logistics and transportation platform, ensuring that the sources of cargo provided by the logistics and transportation platform for the target driver not only meet the basic requirements of the target driver, but also are highly consistent with the personalized needs and historical behaviors of the target driver.

[0151] It should be noted that, through the third and fifth filtering conditions, more suitable choices can be provided for newly registered users on the logistics and transportation platform, thereby improving user retention rate; through the fourth and sixth filtering conditions, more personalized sources of goods of the length and type of vehicles most frequently pulled by target users can be introduced, thereby increasing user stickiness.

[0152] In the above embodiment, the processor needs to determine the first screening condition according to the target departure place and the target vehicle length. Next, the specific process of the processor determining the first screening condition according to the target departure place and the target vehicle length is described in detail.

[0153] Figure 3 A flowchart of a method for determining a first screening condition provided in an embodiment of the present application. Figure 3 As shown, in a possible embodiment, S1021, the processor determines the first screening condition according to the target departure place and the target vehicle length, which can be implemented through steps a1 to a4. Steps a1 to a4 are described in detail below.

[0154] Step a1: The processor determines the basic idle driving distance, the idle driving coefficient and the first idle driving distance corresponding to the target vehicle length according to the target vehicle length.

[0155] The basic empty driving distance refers to the reasonable empty driving distance between the target departure point and the source of goods when the vehicle is empty. The basic empty driving distance can be a standard value pre-set based on industry experience, market data or vehicle operating costs, representing the acceptable time and cost range for a vehicle to travel a certain distance when empty.

[0156] The idle driving coefficient is a measure of the relative cost or efficiency of a vehicle when it is traveling without a load. The value of the idle driving coefficient is used to calculate the actual cost of traveling without a load and to evaluate the impact of the empty driving distance on the driver's economy and efficiency during the vehicle-cargo matching process.

[0157] The first empty driving distance refers to the maximum distance allowed for a vehicle to travel when it is empty, set by the management personnel of the logistics transportation platform.

[0158] It should be noted that different vehicle lengths correspond to different basic empty distances, empty travel coefficients, and first empty travel distances. The basic empty distances, empty travel coefficients, and first empty travel distances corresponding to different vehicle lengths are pre-stored on the logistics transportation platform. When the processor knows the target vehicle length, it can directly obtain the basic empty distance, empty travel coefficient, and first empty travel distance corresponding to the target vehicle length.

[0159] The basic empty driving distance, empty driving coefficient and first empty driving distance corresponding to each vehicle length stored on the logistics and transportation platform can all be configured by the management personnel of the logistics and transportation platform according to actual conditions, and this embodiment does not make specific limitations on this.

[0160] Step a2: The processor determines the product of the idle driving coefficient and the transportation distance of the candidate cargo source.

[0161] It should be noted that the transportation distance of a candidate source of goods refers to the distance of the transportation route between the shipping place and the receiving place of the candidate source of goods.

[0162] Step a3: The processor determines that the sum of the product and the basic empty driving distance is the second empty driving distance.

[0163] It should be noted that the second empty driving distance represents the empty driving distance required by the candidate cargo source.

[0164] Step a4: The processor determines a first screening condition according to the target departure point, the first empty driving distance and the second empty driving distance.

[0165] Among them, the first screening condition is that the distance between the shipping place of the first source of goods and the target departure place is less than the third empty driving distance; the third empty driving distance is the minimum value of the first empty driving distance and the second empty driving distance.

[0166] It should be noted that the processor limits the shipping location range of the first source of cargo according to the smaller empty distance between the first empty distance and the second empty distance, so as to ensure that the empty distance corresponding to the first source of cargo is not too long, thereby ensuring that the first source of cargo recommended to the target driver is within a reasonable empty distance, which can significantly reduce the target driver's idle driving and improve the efficiency and rationality of vehicle-cargo matching.

[0167] For example, the basic empty driving distance corresponding to a vehicle length of 4.2 meters is 30 kilometers, the empty driving coefficient is 0.2, and the first empty driving distance is 100 kilometers. When the target vehicle length is 4.2 meters, the first empty driving distance is 100 kilometers, and the second empty driving distance is 30+0.2×the transportation distance of the candidate cargo source. When screening the candidate sources of cargo based on the first screening condition, if the transportation distance of candidate source A is 50 kilometers, and the distance between the departure point and the target departure point of candidate source A is 35 kilometers, the second empty driving distance of candidate source A is 40 kilometers. Since the distance between the departure point and the target departure point of candidate source A is less than the second empty driving distance of candidate source A, candidate source A is determined as the first source of cargo; if the transportation distance of candidate source B is 60 kilometers, and the distance between the departure point and the target departure point of candidate source B is 50 kilometers, the second empty driving distance of candidate source B is 42 kilometers. Since the distance between the departure point and the target departure point of candidate source B is greater than the second empty driving distance of candidate source B, it is determined that candidate source B is not the first source of cargo.

[0168] In an embodiment of the present application, the processor comprehensively considers the target vehicle length, basic empty driving distance, empty driving coefficient, first empty driving distance and the transportation distance of each candidate cargo source, and generates a reasonable first screening condition to screen the candidate cargo sources according to the first screening condition, thereby ensuring that the matched cargo sources meet the actual transportation capacity of the target driver, effectively improving the matching efficiency, solving the problem of excessively high empty driving costs for drivers, and improving the overall economy of logistics transportation.

[0169] In the above embodiment, the processor needs to determine the second screening condition according to the target destination and the first historical behavior data of the target driver. Next, the specific process of the processor determining the second screening condition according to the target destination and the first historical behavior data of the target driver is described in detail.

[0170] Figure 4 A flow chart of a method for determining a second screening condition provided in an embodiment of the present application. Figure 4 As shown, in a possible embodiment, S1022, the processor determines the second screening condition based on the first historical behavior data of the target destination and the target driver, which can be achieved through steps b1 to b4. Steps b1 to b4 are described in detail below.

[0171] Step b1, the processor obtains the number of search behaviors, click behaviors, phone behaviors and transaction behaviors of the target driver on the logistics transportation platform for each of the P cities within a preset time period.

[0172] Wherein, P is a positive integer and P≥2.

[0173] It should be noted that the processor collects the target driver's behavior data related to P cities on the logistics transportation platform within a preset time period. The behavior data includes search behavior frequency data, click behavior frequency data, phone behavior frequency data, and transaction frequency behavior data.

[0174] For any city among the P cities, the number of search behaviors in this city refers to the number of times the target driver searches for the cargo sources in this city on the logistics and transportation platform; the number of click behaviors in this city refers to the number of times the target driver clicks on the cargo source details in this city; the number of phone calls in this city refers to the number of times the target driver contacts the cargo owners in this city through phone calls or messages on the platform; the number of transaction behaviors in this city refers to the number of orders related to this city that are completed by the target driver.

[0175] In this embodiment, by collecting the behavioral data of the target drivers in different cities, the processor can fully understand the target drivers' activity level and behavioral preferences on the logistics and transportation platform. Different behaviors reflect the driver's attention to a certain city and actual transaction situation, providing data support for subsequent city-based candidate source screening.

[0176] Step b2: The processor calculates the behavior score for each city based on the number of search behaviors, the number of click behaviors, the number of phone behaviors, the number of transaction behaviors, the weight of search behaviors, the weight of click behaviors, the weight of phone behaviors and the weight of transaction behaviors.

[0177] The processor performs a weighted summation of all behaviors in each city based on the number of search behaviors, click behaviors, phone behaviors, and transaction behaviors in each city, combined with the weight corresponding to each behavior, to obtain a behavior score for each city.

[0178] It should be noted that the weight of each behavior may be different or the same, and this embodiment does not specifically limit this.

[0179] For example, the weight of transaction behavior is greater than the weight of search behavior.

[0180] In this embodiment, by assigning weights, the processor can comprehensively evaluate the behavior data of the target driver in each city and quantify it into a score. A higher behavior score means that the target driver is more active in the city, indicating that the city has a higher transportation attractiveness to the target driver, which helps to screen out cities that are more valuable to the target driver.

[0181] Step b3: The processor determines Q target cities according to the behavior score of each city in the P cities.

[0182] Among them, the target city is the city with the top Q behavior scores among P cities; Q is a positive integer, and Q≤P.

[0183] It should be noted that the Q target cities are the optimal transportation cities obtained by the processor based on the historical behavior analysis of the target drivers, that is, the target drivers have frequent transportation activities in these cities and have strong preferences.

[0184] In this embodiment, by screening out the Q target cities with the highest behavior scores, the processor can more accurately identify the target driver's transportation preference cities, ensuring that these target cities that are highly correlated with the target driver's behavior are given priority in subsequent vehicle-cargo matching, further improving the accuracy and success rate of vehicle-cargo matching.

[0185] Step b4: The processor determines a second screening condition based on the target destination and the Q target cities.

[0186] Among them, the second screening condition is that the delivery place of the first source of goods is the target destination or target city.

[0187] It should be noted that the second screening condition is that the delivery address of the first source of cargo must be the target destination or one of the Q target cities, which means that in the subsequent vehicle-cargo matching, priority will be given to cargo sources whose delivery addresses match the target destination or target city of the target driver to ensure that the cargo matching results are highly consistent with the driver's preferred cities and transportation needs.

[0188] In the embodiment of the present application, by extracting the target driver's behavioral data of different dimensions such as search, click, phone and transaction in each city on the logistics and transportation platform, the behavior score of each city is quantified, and Q target cities are screened out based on the behavior score of each city, so as to generate a more personalized and accurate second screening condition based on the Q target cities and target destinations. This solves the problem of insufficient consideration of the historical preferences of the target driver in the existing vehicle-cargo matching system, allowing the logistics and transportation platform to more intelligently recommend the source of goods that meets the preferences and historical behaviors of the target driver, significantly improving the efficiency and accuracy of vehicle-cargo matching.

[0189] In the above embodiment, the processor needs to determine the third screening condition according to the multiple first historical orders corresponding to the target vehicle length. Next, the specific process of the processor determining the third screening condition according to the multiple first historical orders corresponding to the target vehicle length is described in detail.

[0190] Figure 5 A flow chart of a method for determining a third screening condition provided in an embodiment of the present application. Figure 5 As shown, in a possible embodiment, S1023, the processor determines the third screening condition according to the multiple first historical orders corresponding to the target vehicle length, which can be implemented through steps c1 to c4. Steps c1 to c4 are described in detail below.

[0191] Step c1: The processor determines the minimum vehicle length of each first historical order among the multiple first historical orders.

[0192] The first historical order refers to a historical transportation order completed on the logistics transportation platform within a preset time period using a vehicle with the same length as the target vehicle.

[0193] Among them, the minimum vehicle length refers to the minimum vehicle length required for the cargo in a specific order, that is, the minimum vehicle length requirement required to load the cargo.

[0194] For example, a historical order may indicate that a certain source of goods can be loaded using a 4.2-meter or 6-meter long vehicle, and the minimum vehicle length of the historical order is 4.2 meters.

[0195] Step c2: determine the number of first historical orders corresponding to each minimum vehicle length among multiple minimum vehicle lengths.

[0196] Step c3: Arrange the quantities of the first historical orders corresponding to the multiple minimum vehicle lengths in descending order and normalize them to obtain the normalized first values ​​corresponding to the multiple minimum vehicle lengths.

[0197] The processor arranges the first historical order quantities corresponding to the multiple minimum vehicle lengths in descending order, that is, the minimum vehicle length with the largest number of first historical orders is arranged in front, and the minimum vehicle length with the smallest number of first historical orders is arranged in the back.

[0198] The processor normalizes the number of first historical orders corresponding to multiple minimum vehicle lengths, and standardizes the number of first historical orders corresponding to different minimum vehicle lengths to a value between 0 and 1, so that different minimum vehicle lengths can be compared within the same range.

[0199] Step c4: determining a third screening condition based on the normalized first values ​​corresponding to the multiple minimum vehicle lengths.

[0200] Among them, the third screening condition is that the minimum vehicle length of the first source of cargo is the first target minimum vehicle length; the first target minimum vehicle length is the minimum vehicle length among multiple minimum vehicle lengths whose normalized first value is greater than the first threshold.

[0201] It should be noted that the first threshold value can be set by the management personnel of the logistics transportation platform, and this embodiment does not make any specific limitation to this.

[0202] For example, the first threshold is 0.1.

[0203] By analyzing the number of first historical orders corresponding to multiple normalized minimum vehicle lengths, we ensure that the third screening condition is determined based on the most reasonable and most commonly used minimum vehicle length, thereby ensuring that the matched cargo source is suitable for the vehicle length of the target driver, thereby improving the accuracy of the matching.

[0204] In the embodiment of the present application, by analyzing the first historical order corresponding to the target vehicle length, the minimum vehicle length when the source of goods is actually loaded is extracted, and a third screening condition is generated through normalization. The third screening condition can effectively ensure that the source of goods matched by the logistics transportation platform meets the specification requirements of the target vehicle, avoid recommending inappropriate sources of goods, improve the efficiency of vehicle-cargo matching, and reduce resource waste.

[0205] In the above embodiment, the processor needs to screen and sort the M first sources of goods based on the characteristic information of the target driver and the attribute information of the first source of goods to obtain a second source of goods set. Next, the specific process of the processor screening and sorting the M first sources of goods based on the characteristic information of the target driver and the attribute information of the first source of goods to obtain the second source of goods set is described in detail.

[0206] In a possible embodiment, S104, the processor screens and performs a first sorting process on M first sources of goods based on the characteristic information of the target driver and the attribute information of the first source of goods to obtain a second source of goods set, which can be achieved through S1041 and S1042. S1041 and S1042 are described in detail below.

[0207] S1041. The processor screens the M first sources of goods based on the characteristic information of the target driver and the attribute information of the first sources of goods to obtain a third source of goods set.

[0208] Among them, the third source set includes N second sources of supply arranged in the first order.

[0209] The processor screens the M first sources of goods through the first model based on the characteristic information of the target driver and the attribute information of the first sources of goods to obtain a third source of goods set.

[0210] Among them, the first model can be a gradient boosting framework (Light Gradient Boosting Machine, LightGBM) model, or it can be other models, which is not specifically limited in this embodiment.

[0211] The LightGBM model is designed to handle large-scale data sets and has great advantages in speed and memory consumption. The working principle of the LightGBM model is as follows: Gradient boosting: The LightGBM model uses a gradient boosting algorithm to optimize model performance by gradually training a series of decision trees. Each new tree is built based on the residual of the previous tree, thereby continuously reducing the prediction error. Histogram algorithm: The LightGBM model uses a histogram algorithm to speed up the training process. It discretizes continuous feature values ​​into discrete intervals (i.e., histogram buckets), which greatly reduces the computational complexity. Category feature processing: The LightGBM model supports efficient processing of category features. It can automatically encode category features and use these features to improve the model effect. Leaf-wise growth: Unlike the traditional depth-wise growth method, the LightGBM model uses a leaf-wise growth strategy, which will minimize the error on all leaf nodes of each growing tree, thereby improving the accuracy of the model.

[0212] It should be noted that the LightGBM model is used to roughly sort the M first sources of goods in the first source of goods set. The LightGBM model can screen and roughly sort the M first sources of goods according to the characteristic information of the target driver and the attribute information of the first source of goods to obtain the third source of goods set.

[0213] The attribute information of the first source of cargo includes the required vehicle length of the first source of cargo, the required vehicle model of the first source of cargo, the tons of the first source of cargo, the transportation route of the first source of cargo, the longitude and latitude of the shipping place of the first source of cargo, the longitude and latitude of the receiving place of the first source of cargo, the loading time of the first source of cargo, the transportation distance of the first source of cargo, the credibility of the owner corresponding to the first source of cargo, the transportation price of the first source of cargo, etc.

[0214] In S1041, the characteristic information of the target driver includes the probability of the target driver going to each city, the probability of the target driver residing in each city, the straight-line distance between the target departure point of the target driver and the shipping point of the first source of cargo, and the time interval between the release time of the first source of cargo and the release time of the cargo matching request.

[0215] It should be noted that for each first source of supply, the LightGBM model outputs the fulfillment probability and non-fulfillment probability corresponding to the first source of supply, and sorts the fulfillment probabilities corresponding to the M first sources of supply from high to low to obtain the sorted M first sources of supply, and determines the first N first sources of supply among the sorted M first sources of supply as the second source of supply to obtain the third source set.

[0216] S1042. The processor performs a first sorting process on the N second sources of goods in the third source of goods set based on the characteristic information of the target driver and the attribute information of the second source of goods to obtain a second source of goods set.

[0217] The second source set includes N second sources sorted in a second order, the second order is different from the first order, and each second source corresponds to a first score.

[0218] Based on the characteristic information of the target driver and the attribute information of the first source of goods, the processor performs a first sorting process on the N second sources of goods in the third source of goods set through the second model to obtain a second source of goods set.

[0219] Among them, the second model can be an Entire Space Multi-task Model (ESMM). The ESMM model is an improved click-through rate prediction model used to solve the sample sparsity problem in click-through rate (CTR) prediction and conversion rate (CVR) prediction. The ESMM model establishes an association between the two tasks of click and conversion and trains them simultaneously in a multi-task learning manner, thereby improving the prediction effect of the model in the full-link scenario. The working principle of the ESMM model is as follows: Click-through rate prediction: The ESMM model predicts the probability of a user clicking on an advertisement or recommended content through a deep learning network. It combines the features of users and items and uses a neural network model to capture complex feature interactions. Multi-task learning: The ESMM model usually uses multi-task learning to handle related prediction tasks, such as click-through rate (CTR) and conversion rate (CVR) prediction. This method improves the overall performance of the model by learning shared features. Feature interaction: The ESMM model can capture the complex interactive relationships between features. By using a multi-layer neural network or a deep learning framework, it can model nonlinear relationships between different features and improve the accuracy of prediction.

[0220] It should be noted that the ESMM model is used to finely sort the N second sources of supply in the third source of supply set to obtain the second source of supply set.

[0221] In the refinement stage, the goal of the ESMM model is to consider the detailed preferences and contextual information of the target driver and further optimize the sorting of the N second sources of cargo in the third source set, so as to provide the target driver with a more accurate cargo matching result.

[0222] Among them, the ESMM model determines the first score of each source in the second source set through complex feature interaction and click-through rate prediction.

[0223] The attribute information of the second source of cargo includes the required vehicle length of the second source of cargo, the required vehicle model of the second source of cargo, the tons of the second source of cargo, the transportation route of the second source of cargo, the longitude and latitude of the shipping place of the second source of cargo, the longitude and latitude of the receiving place of the second source of cargo, the loading time of the second source of cargo, the transportation distance of the second source of cargo, the credibility of the owner corresponding to the second source of cargo, the transportation price of the second source of cargo, etc.

[0224] In S1042, the characteristic information of the target driver includes the probability of the target driver going to each city, the probability of the target driver residing in each city, the straight-line distance between the target departure point of the target driver and the shipping point of the second source of cargo, and the time interval between the release time of the second source of cargo and the release time of the cargo matching request.

[0225] The method for calculating the probability of the target driver going to each city is the same as the method for calculating the behavior score of each city in the above embodiment S1022, and this embodiment will not be repeated here.

[0226] When calculating the probability that the target driver resides in each city, the location data reported by the target driver every day within a preset time period can be sampled, the number of occurrences of the city corresponding to each location data in the sampling results can be counted, and normalized as the probability that the target driver resides in each city.

[0227] In this embodiment, the ESMM model will perform a more detailed evaluation of the N second sources of goods in the third source set after rough sorting and screening. By conducting in-depth analysis of the detailed characteristics of the target driver and the second source of goods, the ESMM model can further improve the accuracy of the second source set.

[0228] In the above embodiment, the processor needs to perform a second sorting process on the N second sources of goods in the second source set based on the target weight to obtain the target source set. Next, the specific process of the processor performing a second sorting process on the N second sources of goods in the second source set based on the target weight to obtain the target source set is described in detail.

[0229] In a possible embodiment, S105, the processor performs a second sorting process on the N second sources of supply in the second source set based on the target weight to obtain a target source set, which can be achieved through S1051 and S1052. S1051 and S1052 are described in detail below.

[0230] S1051. For each second source of supply in the second source set, if the second source of supply meets the target condition, the processor obtains a second score corresponding to the second source of supply based on the target weight and the first score corresponding to the second source of supply.

[0231] Among them, the target condition is at least one of the following: the distance between the shipping place of the second source of cargo and the target departure place is less than the preset distance; the time interval between the release time of the second source of cargo and the release time of the cargo matching request is less than the preset time interval; the required vehicle length of the second source of cargo is the same as the target vehicle length; the required vehicle model of the second source of cargo is the same as the target vehicle model.

[0232] It should be noted that, for each second source of supply in the second source set, the processor first determines whether the second source of supply meets the target conditions. If the second source of supply meets the target conditions, the processor updates the first score corresponding to the second source of supply according to the target weight to obtain the second score corresponding to the second source of supply.

[0233] When the distance between the shipping place of the second source of goods and the target departure place is less than the preset distance, the first score corresponding to the second source of goods is updated based on the behavior distance weight to obtain the second score corresponding to the second source of goods.

[0234] Among them, the preset distance and the behavior distance weight can be set by the management personnel of the logistics and transportation platform, and this embodiment does not make specific limitations on this.

[0235] For example, the preset distance may be 30 kilometers and the behavior distance weight may be 1.5.

[0236] For example, when the distance between the shipping place of the second source C and the target departure place is 20 kilometers, the first score corresponding to the second source C is 0.8, and the behavior distance weight is 1.5, the second score corresponding to the second source C is 0.8×1.5=1.2.

[0237] When the time interval between the release time of the second source of supply and the release time of the source of supply matching request is less than the preset time interval, the first score corresponding to the second source of supply is updated based on the behavior time weight to obtain the second score corresponding to the second source of supply.

[0238] Among them, the preset time interval and the behavior time weight can be set by the management personnel of the logistics and transportation platform, and this embodiment does not make specific limitations on this.

[0239] For example, the preset time interval may be 1 minute, and the behavior time weight may be 2.

[0240] Exemplarily, when the time interval between the release time of the second source D and the release time of the source matching request is 30 seconds, the first score corresponding to the second source D is 0.7, and the behavior time weight is 2, the second score corresponding to the second source D is 0.7×2=1.4.

[0241] When the required vehicle length of the second source of cargo is the same as the target vehicle length, the first score corresponding to the second source of cargo is updated based on the vehicle length consistency weight to obtain a second score corresponding to the second source of cargo.

[0242] Among them, the vehicle length consistency weight can be set by the management personnel of the logistics transportation platform, and this embodiment does not make any specific limitation to this.

[0243] For example, the vehicle length consistency weight can be 1.2.

[0244] Exemplarily, when the required vehicle length of the second cargo source E is the same as the target vehicle length, the first score corresponding to the second cargo source E is 0.5, and the vehicle length consistency weight is 1.2, the second score corresponding to the second cargo source E is 0.5×1.2=0.6.

[0245] When the demand vehicle model of the second source of supply is the same as the target vehicle model, the first score corresponding to the second source of supply is updated based on the vehicle model consistency weight to obtain a second score corresponding to the second source of supply.

[0246] Among them, the vehicle model consistency weight can be set by the management personnel of the logistics and transportation platform, and this embodiment does not make any specific limitation to this.

[0247] For example, the vehicle model consistency weight may be 1.2.

[0248] For example, when the demand vehicle model of the second source F is the same as the target vehicle model, the first score corresponding to the second source F is 0.4, and the vehicle model consistency weight is 1.2, the second score corresponding to the second source E is 0.4×1.2=0.48.

[0249] It should be noted that, for the same second source of cargo, the behavior distance weight, behavior time weight, vehicle length consistency weight and vehicle type consistency weight can be superimposed to update the first score of the second source of cargo and obtain the second score corresponding to the second source of cargo.

[0250] Exemplarily, when the time interval between the release time of the second source of cargo G and the release time of the cargo matching request is 30 seconds, the required vehicle length of the second source of cargo G is the same as the target vehicle length, the required vehicle model of the second source of cargo G is the same as the target vehicle model, and the first score corresponding to the second source of cargo G is 0.6, the second score corresponding to the second source of cargo G is 0.6×2×1.2×1.2=1.728.

[0251] By introducing multiple target conditions and target weights corresponding to the multiple target conditions, the first score of each second source of cargo is updated to obtain the second score, and the second score is used to sort the N second sources of cargo to obtain a target source set, which helps target drivers quickly find the source of cargo that best meets their needs.

[0252] S1052. The processor sorts the N second sources of supply based on the second scores corresponding to the N second sources of supply in descending order to obtain a target source set.

[0253] It should be noted that by sorting the N second sources of cargo according to their corresponding second scores, the second sources of cargo that best meet the needs of the target driver can be displayed to the user in priority, which helps to reduce the time for the target driver to screen the sources of cargo and improve the efficiency of vehicle-cargo matching.

[0254] In this embodiment, the target source set is obtained by rearranging the N second sources of goods in the second source set in S1051 and S1052. In the vehicle-cargo matching method of the present application, rearrangement plays a vital role. The accuracy of the recommendation result can be improved by rearranging the N second sources of goods in the second source set.

[0255] By further adjusting the second source set after the initial sorting, it can better adapt to specific business goals and user needs. For example, considering the user's real-time behavior and context information, the recommendation results can be fine-tuned to make them more in line with the user's current actual needs.

[0256] In addition, business rules and constraints can be considered when re-ranking. Adjustments can be made based on business rules, such as giving priority to recommending products from specific brands or specific promotions, or ensuring that certain important content is displayed in specific locations. At the same time, it can also avoid recommending similar items continuously, improving the user experience.

[0257] When re-ranking, you can also introduce different types of recommendation items to increase the diversity of recommendation results and meet the different interests of users. For example, in e-commerce recommendations, both popular products and niche specialty products are displayed to increase the chances of users discovering new items.

[0258] In this embodiment, the purpose of re-arrangement is mainly two, one is business support, and the other is driver experience. For example, the platform can increase the exposure of goods that have just been shipped by setting a higher behavior time weight to promote fulfillment efficiency and improve the owner's experience. On the other hand, since the driver's demand conditions are generalized during the recall stage, not all recalled goods can perfectly match the driver's needs, so it is necessary to increase the weight of some particularly precise goods.

[0259] In the above embodiment, the processor needs to respond to the supply matching request based on the target supply set. Next, the specific process of the processor responding to the supply matching request based on the target supply set is described in detail.

[0260] In a possible embodiment, S106, the processor responds to the supply matching request based on the target supply set, which can be implemented through S1061 to S1064, and S1061 to S1064 are described in detail below.

[0261] S1061. The processor determines the precise supply source and the generalized supply source in the target supply source set based on the target departure place and the target destination.

[0262] Among them, the precise source of goods is the second source of goods whose place of shipment is consistent with the target departure place, and whose place of receipt is consistent with the target destination; the generalized source of goods is the other second source of goods in the target source set except the precise source of goods.

[0263] It should be noted that the processor divides the second cargo source in the target cargo source set into precise cargo source and generalized cargo source by matching the target departure place and target destination of the target driver.

[0264] Accurate source of goods: The second source of goods whose place of shipment is the same as the target departure place and whose place of receipt is the same as the target destination. The accurate source of goods fully meets the transportation route needs of the target driver.

[0265] Generalized cargo sources: Secondary cargo sources that do not completely match the target departure point or target destination, but are still within a reasonable range. Generalized cargo sources may have some conditions that are slightly deviated, such as the shipping place or the receiving place not being completely consistent with the transportation route requirements of the target driver, but they are still considered as cargo sources.

[0266] S1062. The processor sorts the precise sources of goods in descending order according to the second scores corresponding to the precise sources of goods to obtain a precise source sequence.

[0267] S1063. The processor sorts the generalized sources of goods in descending order of the second scores corresponding to the generalized sources of goods to obtain a generalized source of goods sequence.

[0268] S1064. The processor inserts and displays the generalized source sequence when displaying the precise source sequence based on the second score corresponding to the precise source of supply and the second score corresponding to the generalized source of supply, as well as preset rules.

[0269] It should be noted that in order to improve fulfillment efficiency while ensuring the target driver experience, the logistics and transportation platform includes the display logic of precise and generalized sources of goods.

[0270] When the target departure place is the target departure place actively input by the user, the target source set is displayed through the display logic of multiple precise sources of goods and generalized sources of goods; otherwise, the target source set is displayed in the order of the second scores corresponding to the N second sources of goods in the target source set determined by S105 above from high to low.

[0271] Specifically, the display logic of inserting multiple precise sources of supply and generalized sources of supply means that when displaying a precise source sequence, based on the second score corresponding to the precise source of supply, the second score corresponding to the generalized source of supply and preset rules, a certain proportion of generalized sources of supply are inserted when displaying the precise source of supply.

[0272] The preset rules can include a variety of strategies, such as inserting a generalized source of goods after displaying several precise sources of goods, or dynamically adjusting the display order according to the difference in the source of goods scores. The purpose is to allow drivers to see suitable generalized sources of goods without interfering with the precise sources of goods.

[0273] The condition for inserting a generalized source of supply when displaying a precise source of supply is that the second score corresponding to the inserted generalized source of supply is higher than the second score corresponding to the precise source of supply following the inserted generalized source of supply.

[0274] It should be noted that, when any of the precise sources of goods and the generalized sources of goods are in surplus or insufficient, the second scores corresponding to the remaining sources of goods are naturally added to the display list in descending order.

[0275] In this embodiment, by reasonably inserting generalized sources of goods when displaying precise sources of goods, the diversity of the source matching results displayed to the target driver is ensured, avoiding reliance on precise sources of goods alone, and increasing the choice space for the target driver.

[0276] Figure 6 This is a schematic diagram of the structure of a vehicle-cargo matching device provided in an embodiment of the present application. Figure 6 As shown, the vehicle-cargo matching device 600 provided in this embodiment can exist independently and be used to implement the operations corresponding to the processor in the above method embodiment.

[0277] The vehicle-cargo matching device 600 may include: a transceiver module 601 and a processing module 602. The processing module 602 is used for data processing, and the transceiver module 601 may implement corresponding communication functions. The transceiver module 601 may also be called a communication interface or a communication unit.

[0278] Optionally, the vehicle-cargo matching device 600 may further include a storage unit, which may be used to store instructions and / or data, and the processing module 602 may read the instructions and / or data in the storage unit so that the vehicle-cargo matching device 600 implements the steps implemented by the processor in the aforementioned method embodiment.

[0279] The transceiver module 601 is used to execute the reception-related operations of the processor in the above method embodiment, and the processing module 602 is used to execute the processing-related operations of the processor in the above method embodiment.

[0280] Optionally, the transceiver module 601 may include a sending module and a receiving module. The sending module is used to perform the sending operation in the above method embodiment. The receiving module is used to perform the receiving operation in the above method embodiment.

[0281] It should be noted that the vehicle-cargo matching device 600 may include a sending module but not a receiving module. Alternatively, the vehicle-cargo matching device 600 may include a receiving module but not a sending module. Specifically, it depends on whether the above solution executed by the vehicle-cargo matching device 600 includes a sending action and a receiving action.

[0282] As an example, the vehicle-cargo matching device 600 is used to execute the above Figure 1 The actions performed by the processor in the illustrated embodiment.

[0283] The vehicle-cargo matching device 600 may include: a transceiver module 601 and a processing module 602 .

[0284] Processing module 602 is used to respond to the cargo matching request of the target driver on the logistics transportation platform, determine the target departure point and target destination of the target driver, and the target length and target model of the target transportation vehicle; wherein the target transportation vehicle is the vehicle of the target driver.

[0285] The processing module 602 is further used to generate a target screening condition based on the target departure place, the target destination, the target vehicle length and the target vehicle type.

[0286] Processing module 602 is also used to determine a first source set according to the target screening conditions and candidate sources; wherein the first source set includes M first sources, M is a positive integer, and M≥1; the first sources are candidate sources that meet the target screening conditions.

[0287] The processing module 602 is also used to screen and perform a first sorting process on the M first sources of goods based on the characteristic information of the target driver and the attribute information of the first source of goods to obtain a second source of goods set; wherein the second source of goods set includes N second sources of goods, N is a positive integer, and N≤M.

[0288] The processing module 602 is also used to perform a second sorting process on the N second sources of goods in the second source of goods set based on the target weight to obtain a target source of goods set; wherein the target weight is at least one of the behavior distance weight, the behavior time weight, the vehicle length consistency weight, and the vehicle model consistency weight.

[0289] The processing module 602 is further configured to respond to the supply matching request based on the target supply set.

[0290] It should be understood that the execution of the above corresponding processes by each module has been described in detail in the above method embodiment, and for the sake of brevity, it will not be repeated here.

[0291] The processing module 602 in the above embodiment can be implemented by at least one processor or processor-related circuit. The transceiver module 601 can be implemented by a transceiver or a transceiver-related circuit. The transceiver module 601 can also be called a communication unit or a communication interface. The storage unit can be implemented by at least one memory.

[0292] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 7 As shown, the electronic device 700 provided in this embodiment includes: a memory 701 and a processor 702 .

[0293] The memory 701 may be an independent physical unit, and may be connected to the processor 702 via a bus 703. The memory 701 and the processor 702 may also be integrated together and implemented by hardware. The memory 701 is used to store program instructions, and the processor 702 calls the program instructions to execute the operations executed by the processor in any of the above method embodiments.

[0294] Optionally, when part or all of the methods of the above embodiments are implemented by software, the electronic device 700 may also include only a processor 702. The memory 701 for storing programs is located outside the electronic device 700, and the processor 702 is connected to the memory through circuits / wires to read and execute the programs stored in the memory. The processor 702 may be a central processing unit (CPU), a network processor (NP) or a combination of a CPU and a NP. The processor 702 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD) or a combination thereof. The PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL) or any combination thereof.

[0295] The memory 701 may include a volatile memory, such as a random-access memory (RAM); the memory may also include a non-volatile memory, such as a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD); the memory may also include a combination of the above types of memory.

[0296] Illustratively, the present application provides a chip, comprising: an interface circuit and a logic circuit, wherein the interface circuit is used to receive signals from other chips outside the chip and transmit them to the logic circuit, or to send signals from the logic circuit to other chips outside the chip, and the logic circuit is used to execute the operations performed by the processor in the above method embodiment.

[0297] Illustratively, the present application provides a computer-readable storage medium on which computer program instructions are stored. The computer program instructions are executed by a processor of an electronic device so that the electronic device performs the operations performed by the processor in the above method embodiments.

[0298] Illustratively, the present application provides a computer program product. When the computer program product is executed on an electronic device, the electronic device executes the operations executed by the processor in the above method embodiment.

[0299] The above description is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A vehicle-cargo matching method, characterized in that: The method comprises: In response to a cargo matching request from a target driver on a logistics transportation platform, a target departure point and a target destination of the target driver, as well as a target length and a target vehicle type of the target transportation vehicle are determined; wherein the target transportation vehicle is the vehicle of the target driver; Generate a target screening condition based on the target departure place, the target destination, the target vehicle length and the target vehicle type; Determine a first supply source set according to the target screening condition and the candidate supply sources; wherein the first supply source set includes M first supply sources, M is a positive integer, and M≥1; the first supply sources are candidate supply sources that meet the target screening condition; Based on the characteristic information of the target driver and the attribute information of the first source of goods, the M first sources of goods are screened and sorted to obtain a second source of goods set; wherein the second source of goods set includes N second sources of goods, N is a positive integer, and N≤M; Based on the target weight, performing a second sorting process on the N second cargo sources in the second cargo source set to obtain a target cargo source set; wherein the target weight is at least one of a behavior distance weight, a behavior time weight, a vehicle length consistency weight, and a vehicle type consistency weight; Based on the target supply set, respond to the supply matching request.

2. The method according to claim 1, characterized in that The generating of target screening conditions based on the target departure place, the target destination, the target vehicle length and the target vehicle type includes: Determining a first screening condition according to the target departure place and the target vehicle length; Determining a second screening condition according to the target destination and the first historical behavior data of the target driver; Determine a third screening condition according to a plurality of first historical orders corresponding to the target vehicle length; wherein the first historical order is an order with an actual vehicle length of the target vehicle length completed on the logistics transportation platform within the preset time period; Determining a fourth screening condition according to the second historical behavior data corresponding to the target driver; Determining a fifth screening condition according to a second historical order corresponding to the target vehicle model; wherein the second historical order is an order for the target vehicle model completed on the logistics transportation platform within the preset time period; Determining a sixth screening condition according to the third historical behavior data corresponding to the target driver; The target filtering condition is determined according to the first filtering condition, the second filtering condition, the third filtering condition, the fourth filtering condition, the fifth filtering condition and the sixth filtering condition.

3. The method according to claim 2, characterized in that The determining of the first screening condition according to the target departure place and the target vehicle length includes: According to the target vehicle length, determining a basic empty driving distance, an empty driving coefficient and a first empty driving distance corresponding to the target vehicle length; Determine the product of the empty driving coefficient and the transportation distance of the candidate cargo source; Determine a second empty driving distance as a sum of the product and the basic empty driving distance; The first screening condition is determined based on the target departure point, the first empty driving distance and the second empty driving distance; wherein the first screening condition is that the distance between the shipping place of the first source of goods and the target departure point is less than a third empty driving distance; and the third empty driving distance is the minimum value of the first empty driving distance and the second empty driving distance.

4. The method according to claim 2, characterized in that: The determining of the second screening condition according to the target destination and the first historical behavior data of the target driver includes: Obtain the number of searches, clicks, phone calls, and transactions of the target driver on the logistics transportation platform for each of P cities within the preset time period; wherein P is a positive integer, and P≥2; Calculate the behavior score for each city based on the number of search behaviors, the number of click behaviors, the number of phone calls, the number of transaction behaviors, the weight of the search behaviors, the weight of the click behaviors, the weight of the phone calls, and the weight of the transaction behaviors; Determine Q target cities according to the behavior score of each city in the P cities; wherein the target cities are cities with the behavior scores in the first Q cities among the P cities, Q is a positive integer, and Q≤P; Based on the target destination and the Q target cities, the second screening condition is determined; wherein the second screening condition is that the delivery place of the first source of goods is the target destination or the target city.

5. The method according to claim 2, characterized in that: The determining of the third screening condition according to the first historical order corresponding to the target vehicle length includes: For each first historical order among the plurality of first historical orders, determining a minimum vehicle length of the first historical order; Determine the number of the first historical orders corresponding to each minimum vehicle length among multiple minimum vehicle lengths; Arrange the quantities of the first historical orders corresponding to the multiple minimum vehicle lengths in descending order and perform normalization processing to obtain normalized first values ​​corresponding to the multiple minimum vehicle lengths respectively; Based on the normalized first values ​​corresponding to the multiple minimum vehicle lengths, a third filtering condition is determined; wherein, the third filtering condition is that the minimum vehicle length of the first source of cargo is the first target minimum vehicle length; the first target minimum vehicle length is the minimum vehicle length among the multiple minimum vehicle lengths whose normalized first value is greater than a first threshold.

6. The method according to claim 1, characterized in that The method of screening and sorting the M first sources of goods based on the characteristic information of the target driver and the attribute information of the first source of goods to obtain a second source of goods set includes: Based on the characteristic information of the target driver and the attribute information of the first source of goods, the M first sources of goods are screened to obtain a third source of goods set; wherein the third source of goods set includes the N second sources of goods arranged in the first order; Based on the characteristic information of the target driver and the attribute information of the second source of goods, performing a first sorting process on the N second sources of goods in the third source of goods set to obtain the second source of goods set; wherein the second source of goods set includes the N second sources of goods sorted in a second order, the second order is different from the first order, and each second source of goods corresponds to a first score; Among them, the characteristic information of the target driver includes the probability of the target driver going to each city, the probability of the target driver residing in each city, the straight-line distance between the target departure point of the target driver and the shipping place of the first source of cargo or the second source of cargo, and the time interval between the release time of the first source of cargo or the second source of cargo and the release time of the source of cargo matching request.

7. The method according to claim 6, characterized in that The step of performing a second sorting process on the N second supply sources in the second supply source set based on the target weight to obtain a target supply source set includes: For each second source of supply in the second source of supply set, when the second source of supply meets the target condition, obtaining a second score corresponding to the second source of supply based on the target weight and the first score corresponding to the second source of supply; Sorting the N second sources of supply based on the second scores corresponding to the N second sources of supply in descending order to obtain the target source set; Wherein, the target condition is at least one of the following: The distance between the shipping place of the second source of goods and the target departure place is less than a preset distance; The time interval between the release time of the second supply source and the release time of the supply source matching request is less than a preset time interval; The required vehicle length of the second cargo source is the same as the target vehicle length; The vehicle model required by the second source of goods is the same as the target vehicle model.

8. The method according to claim 1, characterized in that The responding to the supply source matching request based on the target supply source set includes: Based on the target departure place and the target destination, determine the precise source of goods and the generalized source of goods in the target source of goods set; wherein the precise source of goods is a second source of goods whose place of shipment is consistent with the target departure place and whose place of receipt is consistent with the target destination; the generalized source of goods is other second sources of goods in the target source of goods set except the precise source of goods; Sorting the precise sources of goods in descending order of the second scores corresponding to the precise sources of goods to obtain a precise source sequence; Sorting the generalized supply sources in descending order of the second scores corresponding to the generalized supply sources to obtain a generalized supply source sequence; Based on the second score corresponding to the precise source of supply and the second score corresponding to the generalized source of supply, and a preset rule, the generalized source of supply sequence is inserted and displayed when the precise source of supply sequence is displayed.

9. An electronic device, characterized in that: include: memory and at least one processor; The memory stores computer-executable instructions; The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor performs the method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed, the method according to any one of claims 1 to 8 is implemented.

Images