Transport capacity resource recommendation method and device, equipment and storage medium

Abstract

The invention relates to the technical field of intelligent logistics, and provides a transport capacity resource recommendation method and device, equipment and a storage medium. The method comprises the following steps: acquiring transport capacity resource data and historical transport data of each transport capacity vehicle in the transport capacity resource data; generating a transport capacity portrait of each transport capacity vehicle according to the transport capacity resource data and the historical transport data; and receiving a transport capacity resource request of a user, matching a cargo source demand in the transport capacity resource request with the transport capacity portrait, and recommending the matched transport capacity resource to the user. By constructing the transport capacity portrait of the vehicle, the matching degree of the recommended transport capacity resource and the user demand can be improved, so that the demand of the user for the transport capacity resource is better met, and meanwhile, the recommendation efficiency and recommendation accuracy of the transport capacity resource are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent logistics technology, and in particular to a method, device, equipment and storage medium for recommending transportation resources. Background Art

[0002] With the rapid development of e-commerce and online freight services, how to effectively match transportation resources has become a key issue in improving logistics efficiency. Existing methods for selecting and matching transportation resources rely too much on manual experience and market information. For example, based on freight maps, they monitor and manage information such as goods, vehicles, and routes in the logistics transportation process in real time. When obtaining the user's transportation needs, they often use simple rule matching or directly rely on manual intervention to recommend transportation resources that meet their needs to users. The transportation resources recommended by this method have limited matching degree with user needs, and the recommended transportation resources often cannot reach the ideal state, which affects the efficiency and accuracy of the recommendation of transportation resources. Summary of the invention

[0003] The present invention provides a method, device, equipment and storage medium for recommending transportation resources, which are used to solve the defects of existing transportation resource recommendation methods that use simple rule matching or directly rely on manual intervention, have low efficiency in recommending transportation resources and insufficient accuracy in recommendation.

[0004] The present invention provides a method for recommending transport resources, comprising: Acquire transportation resource data and historical transportation data of each transportation vehicle in the transportation resource data; Generate a transport capacity profile of each transport vehicle according to the transport capacity resource data and the historical transport data; Receive a user's transport resource request, match the cargo demand in the transport resource request with the transport profile, and recommend the matched transport resources to the user.

[0005] According to the method for recommending transport resources provided by the present invention, generating a transport capacity profile of each transport vehicle according to the transport capacity resource data and the historical transport data includes: Extracting basic attribute information of each of the transport vehicles according to the transport resource data; the basic attribute information includes vehicle length, vehicle type, tonnage, load and location; Extracting the transport behavior preferences of each of the transport vehicles based on the historical transport data; the transport behavior preferences include frequently traveled routes, frequently traveled areas, transport mileage, transport time, cargo type and number of visits; The basic attribute information and the transportation behavior preference are used to construct the capacity label of the transportation vehicle to obtain the capacity portrait of the transportation vehicle.

[0006] According to the method for recommending transportation resources provided by the present invention, matching the cargo demand in the transportation resource request with the transportation profile, and recommending the matched transportation resources to the user, comprises: Acquire the transport demand information corresponding to the cargo demand in the transport resource request; the transport demand information at least includes the vehicle type and vehicle length; Based on the transport demand information, the transport profile is matched in multiple dimensions to obtain a transport capacity set under each matching dimension; the matching dimension includes at least one of a location dimension, a route dimension, a frequently traveled area dimension, and a cargo type dimension; The transport capacity sets under each of the matching dimensions are merged to obtain transport capacity resources that match the cargo demand, and the transport capacity resources are recommended to the user.

[0007] According to the method for recommending transport resources provided by the present invention, before merging the transport capacity sets under the matching dimensions to obtain transport resources matching the cargo demand, the method further includes: Obtaining the matching degree of each transport vehicle in the target set; the target set is a transport vehicle set under a target dimension, and the target dimension is any one of the matching dimensions; The transport vehicles in the target set are sorted according to the matching degree, and the transport vehicles in the target set are screened according to the sorting order.

[0008] According to the method for recommending transport resources provided by the present invention, the transport demand information further includes a loading location, and after merging the transport sets under the matching dimensions to obtain transport resources matching the cargo demand, the method further includes: Obtaining the availability factor of each target transport vehicle corresponding to the transport resource; the availability factor includes location information and load information; Extracting the driving characteristics of each of the target transport vehicles within a historical time period of a preset length according to the location information, and predicting a first probability that the target transport vehicle will drive to the loading place according to the driving characteristics; Extract the load characteristics of each target transport vehicle in the historical time period according to the load information, and predict the availability probability of the target transport vehicle according to the load characteristics; the load characteristics include load status, load duration, empty duration, heavy load continuous mileage and empty continuous mileage; The target transport vehicles are sorted based on the first probability and the available probability.

[0009] According to the method for recommending transport resources provided by the present invention, the transport demand information further includes a place of unloading, and the multi-dimensional matching of the transport profile based on the transport demand information to obtain a transport capacity set under each matching dimension includes: According to the loading place in the transport demand information, a first candidate vehicle within a preset distance range of the loading place is screened out; wherein the model of the first candidate vehicle matches the model in the transport demand information, and the length of the first candidate vehicle matches the length of the vehicle in the transport demand information; If the number of the first candidate vehicles exceeds a preset threshold, the first candidate vehicles are sorted and screened according to the distances between the first candidate vehicles and the loading place to obtain a preset number of second candidate vehicles; Determine the transportation route information according to the loading place and unloading place in the transportation capacity demand information; Based on the transportation route information, multi-dimensional matching is performed on the capacity profile of the second candidate vehicle to obtain a capacity set under each matching dimension.

[0010] According to the method for recommending transportation resources provided by the present invention, after obtaining the transportation resource data and the historical transportation data of each transportation vehicle in the transportation resource data, the method further includes: Identify abnormal data in the transport resource data and the historical transport data; the abnormal data includes missing value data, invalid field data, and data with a field coverage rate less than a preset value; The abnormal data is preprocessed; the preprocessing operation includes at least one of data cleaning, data filtering and data filling.

[0011] The present invention also provides a transport resource recommendation device, comprising the following modules: A data collection module, used to obtain transport resource data and historical transport data of each transport vehicle in the transport resource data; A portrait construction module, used to generate a transport capacity portrait of each transport vehicle according to the transport capacity resource data and the historical transport data; The resource recommendation module is used to receive a user's transport resource request, match the cargo demand in the transport resource request with the transport profile, and recommend the matched transport resources to the user.

[0012] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the capacity resource recommendation method as described in any one of the above are implemented.

[0013] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above-mentioned transport resource recommendation methods are implemented.

[0014] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the steps of any of the above-mentioned transport resource recommendation methods are implemented.

[0015] The capacity resource recommendation method, device, equipment and storage medium provided by the present invention generate a capacity portrait of a capacity vehicle through capacity resource data and historical transportation data, match it with the user's demand for capacity resources, and recommend the matched capacity resources to the user. The capacity portrait improves the matching degree between the recommended capacity resources and the user's demand, which can better meet the user's demand for capacity resources while improving the recommendation efficiency and accuracy of capacity resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 It is a flow chart of the transportation resource recommendation method provided by the present invention.

[0018] Figure 2 It is a structural schematic diagram of the transportation resource recommendation device provided by the present invention.

[0019] Figure 3 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] The embodiment of the present invention provides a method for recommending transportation resources. By deeply integrating the freight map platform and the transportation capacity portrait capability, the transportation resources are deeply analyzed, the characteristics and laws of the transportation resources are excavated, and a portrait of the transportation resources is formed. Then, intelligent matching is performed according to the supply demand, and the optimal transportation resources are selected for recommendation, providing a decision-making basis for logistics and transportation, and improving the matching degree between the recommended transportation resources and user needs, thereby improving the selection accuracy and recommendation efficiency of transportation resources.

[0022] Specifically, Figure 1is a flow chart of the method for recommending transportation resources provided by the present invention, such as Figure 1 As shown, the method comprises the following steps: Step 100, obtaining transport resource data and historical transport data of each transport vehicle in the transport resource data; Step 200, generating a transport capacity profile of each transport vehicle according to the transport capacity resource data and the historical transport data; Step 300, receiving a user's transport resource request, matching the cargo demand in the transport resource request with the transport profile, and recommending the matched transport resources to the user.

[0023] The transport resource data and the historical transport data of each transport vehicle in the transport resource data are obtained, wherein the transport resource data is the relevant data of each transport vehicle, such as the size, transport capacity, driver information and place of affiliation of the vehicle.

[0024] Optionally, the capacity resource data and historical transportation data may be derived from a freight map platform, which collects vehicle-related data to obtain the capacity resource data and the vehicle's historical transportation data.

[0025] A transport capacity profile of each transport vehicle is generated based on the transport capacity resource data and historical transport data of each transport vehicle. Optionally, a transport capacity profile of any target vehicle among the transport vehicles is generated based on the transport capacity resource data and historical transport data of the target vehicle.

[0026] Furthermore, when generating the capacity profile of the capacity vehicle, the capacity resource characteristics of the capacity vehicle are extracted based on the capacity resource data of the capacity vehicle, and the transportation behavior characteristics of the capacity vehicle are extracted based on the historical transportation data of the capacity vehicle. Based on the capacity resource characteristics and transportation behavior characteristics of the capacity vehicle, the capacity profile of the capacity vehicle is constructed.

[0027] When receiving a user's request for transportation resources, the cargo demand in the user's transportation resource request is matched with the generated transportation profile to select transportation resources that meet the user's needs, and the matched transportation resources are recommended to the user.

[0028] Optionally, the user's request for transportation resources can be initiated based on the source of goods to be transported, that is, the user configures the goods to be transported, so that transportation resources that meet the goods transportation needs can be automatically matched.

[0029] Optionally, the user's request for transportation resources can also be initiated based on transportation vehicles, that is, the user configures the required transportation vehicles, so that the transportation vehicles that meet the user's needs can be automatically matched.

[0030] Optionally, the user's transport resource request can also be initiated based on the user's preference for transport resources. The user configures the transport resource preference, so that the transport vehicle that meets the user's resource preference can be automatically matched. The user's preference for transport resources may include, for example, the nearest transport vehicle, the transport vehicle with the same place of origin as the loading place, and the route that is often run between the loading place and the unloading place.

[0031] In one embodiment, by generating a capacity portrait of a transport vehicle, when matching it with the user's capacity needs, the transport behavior preferences of the transport vehicle can be met while meeting the user's demand for transport resources, thereby recommending experienced or interested transport vehicles to the user, which is conducive to improving user satisfaction, ensuring the match between the recommended transport resources and user needs, and improving the accuracy of the recommended transport resources.

[0032] Furthermore, by matching the capacity portraits, compared with the rule-based matching of each capacity vehicle corresponding to the capacity resources, the amount of data that needs to be matched is reduced, which is conducive to improving the efficiency of recommending capacity resources.

[0033] In this embodiment, a capacity portrait of a transport vehicle is generated through capacity resource data and historical transportation data, and matched with the user's demand for capacity resources, and the matched capacity resources are recommended to the user. The capacity portrait is used to improve the matching degree between the recommended capacity resources and the user's demand, which can better meet the user's demand for capacity resources while improving the recommendation efficiency and accuracy of capacity resources.

[0034] In one embodiment, after obtaining the transportation resource data and the historical transportation data of the transportation vehicles, it is necessary to pre-process the obtained data. Therefore, after step 100, the following may also be included: Step 101, identifying abnormal data in the transport resource data and the historical transport data; the abnormal data includes missing value data, invalid field data, and data with a field coverage rate less than a preset value; Step 102, performing a preprocessing operation on the abnormal data; the preprocessing operation includes at least one of data cleaning, data filtering and data filling.

[0035] For the acquired transport resource data and historical transport data, firstly, abnormal data therein is acquired, and the abnormal data includes but is not limited to missing value data, invalid field data and data with field coverage less than a preset value. A preprocessing operation is performed on the identified abnormal data, and the preprocessing operation includes at least one of data cleaning, data filtering and data filling.

[0036] Optionally, for the transport resource data and historical transport data collected from the freight map platform, the coverage of key fields, the data quality of the fields, etc. are used as screening factors to screen out valid data and perform preprocessing operations on abnormal data. Exemplarily, the preprocessing operation includes but is not limited to filtering and cleaning data with missing or invalid key fields such as the driver's mobile phone number and license plate number, or filtering invalid data that does not meet the requirements such as vehicle length and vehicle model, such as eliminating transport resources with a vehicle length of less than 2 meters and a car model, and filtering data with a field coverage rate of less than 60%.

[0037] When constructing a transport resource profile, the multi-dimensional features of transport vehicles are extracted based on the acquired transport resource data and historical transport data. Step 200 may specifically include: Step 210, extracting basic attribute information of each of the transport vehicles according to the transport resource data; the basic attribute information includes vehicle length, vehicle type, tonnage, load and location; Step 220, extracting the transport behavior preference of each of the transport vehicles based on the historical transport data; the transport behavior preference includes frequently traveled routes, frequently traveled areas, transport mileage, transport time, cargo type and number of visits; Step 230, using the basic attribute information and the transportation behavior preference to construct the capacity label of the transportation vehicle, and obtain the capacity portrait of the transportation vehicle.

[0038] When constructing a transport capacity portrait, basic attribute information of each transport vehicle is extracted based on the acquired transport resource data. The basic attribute information includes the length, vehicle type, tonnage, load and location of the transport vehicle. The location of the transport vehicle can be determined based on the license plate number of the transport vehicle.

[0039] Furthermore, based on the historical transportation data of each transport vehicle obtained, the transport behavior preference of each transport vehicle is extracted, and the transport behavior preference includes the frequently-running routes, frequently-running areas, transportation mileage, transportation time, cargo type and triggering times. By analyzing the historical transportation data of the transport vehicle, the transport characteristics of the transport vehicle are mined, including the cargo type of frequently-hauled cargo, the number of waybills and the transportation behavior in historical time periods of different lengths. Among them, the transportation behavior in time periods of different lengths includes the running mileage in the past 6 months, the running time in the past 6 months, the distribution of frequently-running routes in the past 6 months, the distribution of frequently-running areas in the past 6 months, the distribution of frequently-running routes in the past 2 months, the distribution of frequently-running areas in the past 2 months, the distribution of frequently-running routes in the past 1 month, the distribution of frequently-running areas in the past 1 month, the list of areas passed in the past 1 week, and the distribution of cargo types of frequently-hauled cargo in the past 6 months. The number of contacts is determined based on the number of times the driver of the transport vehicle is contacted and the number of successful contacts.

[0040] Optionally, the transport behavior preference of the transport vehicle may also include the transport intention reported by the driver, for example, the transport route whitelist, transport route blacklist, transport area whitelist, transport area blacklist, transport cargo type whitelist and transport cargo type blacklist reported by the driver. Among them, the transport route whitelist contains the routes that the driver wants to run, the transport route blacklist contains the routes that the driver does not want to run, the transport area whitelist contains the areas that the driver wants to run, the transport area blacklist contains the areas that the driver does not want to run, the transport acquisition type whitelist contains the types of cargo that the driver expects to transport, and the transport cargo type blacklist contains the types of cargo that the driver does not want to transport.

[0041] When constructing a capacity profile, the capacity label of the transport vehicle is generated based on the extracted basic attribute information and transportation behavior preferences to obtain the capacity profile of the transport vehicle.

[0042] When selecting the transport resources recommended to the customer, the user demand is matched with the transport profile based on the constructed transport profile. In step 300, the cargo demand in the transport resource request is matched with the transport profile, and the matched transport resources are recommended to the user, which specifically includes: Step 310, obtaining the transport demand information corresponding to the cargo demand in the transport resource request; the transport demand information at least includes the vehicle type and vehicle length; Step 320: performing multi-dimensional matching on the transport capacity profile based on the transport capacity demand information to obtain a transport capacity set under each matching dimension; the matching dimension includes at least one of a location dimension, a route dimension, a frequently traveled area dimension, and a cargo type dimension; Step 330, merging the transport capacity sets under each of the matching dimensions to obtain transport capacity resources that match the cargo demand, and recommending the transport capacity resources to the user.

[0043] When matching the cargo demand with the transport capacity profile, first obtain the transport capacity demand information corresponding to the cargo demand in the transport capacity resource request, and the transport capacity demand information at least includes the vehicle type and length of the required transport capacity resource. Optionally, the transport capacity demand information may be included in the cargo demand, or may be calculated based on the cargo demand. For example, if the cargo demand includes the type and quantity of the goods to be transported, the required vehicle type and length and other transport capacity demand information may be determined based on the goods to be transported.

[0044] Based on the acquired capacity demand information, multi-dimensional matching is performed on the capacity portrait to obtain a capacity set under each matching dimension, wherein the matching dimension includes at least one of the location dimension, route dimension, frequently traveled area dimension and cargo type dimension, and the capacity set includes one or more capacity vehicles that match the capacity demand information under the corresponding matching dimension.

[0045] Optionally, based on the constructed capacity portrait, which includes basic attribute information of the capacity vehicles and capacity labels corresponding to the transportation behavior preferences, each capacity label can be used as a matching dimension to perform multi-dimensional matching between the capacity demand information and the capacity portrait, so as to obtain the capacity set under each matching dimension.

[0046] Furthermore, for one or more transport capacity sets that are matched, the transport vehicles therein are merged to obtain transport resources that match the cargo demand, and the transport resources are recommended to the user. Optionally, the transport resources recommended to the user correspond to one or more transport vehicles, and through multi-dimensional matching of transport portraits, different needs of users for transport resources can be met.

[0047] In some embodiments, when multi-dimensional matching is performed on the capacity portrait, the next matching dimension is matched based on the matching result of the previous matching dimension, so that multi-capacity vehicles are screened and matched layer by layer in multiple dimensions. In other embodiments, when multi-dimensional matching is performed on the capacity portrait, each matching dimension is matched separately, and then the capacity sets under each matching dimension are merged. The merging method includes but is not limited to taking the union of the capacity sets under each dimension, or taking the intersection of the capacity sets under each matching dimension.

[0048] Furthermore, before merging the transport capacity sets under each matching dimension, it is necessary to sort and filter the transport vehicles in each transport capacity set. Based on this, in step 330, before merging the transport capacity sets under each matching dimension to obtain the transport capacity resources matching the cargo demand, the following steps may also be included: Step 301, obtaining the matching degree of each transport vehicle in a target set; the target set is a transport vehicle set under a target dimension, and the target dimension is any one of the matching dimensions; Step 302: sort the transport vehicles in the target set according to the matching degree, and screen the transport vehicles in the target set according to the sorting order.

[0049] For the transport capacity set under each matching dimension, the matching degree of each transport vehicle in the target set is obtained, where the target set is the transport capacity set under the target dimension, and the target dimension is any one of the matching dimensions. That is, for the transport capacity set under any matching dimension, the matching degree of each transport vehicle in the target set under the matching dimension is obtained.

[0050] Furthermore, the transport vehicles are sorted according to the matching degree of each transport vehicle, and the sorting method can be descending or ascending, and there is no specific limitation on this. According to the sorting order of each transport vehicle in the target set, the transport vehicles in the target set are screened, and the transport vehicles with matching degrees greater than a preset matching degree threshold are retained; or, in a descending sorting method, the transport vehicles with sorting orders before a preset percentile are retained; or, in a descending sorting method, a preset number of transport vehicles are retained according to the sorting order.

[0051] After sorting and filtering the capacity sets under each matching dimension, the capacity sets are merged. If the same capacity vehicle is included in different capacity sets, the capacity vehicle is deduplicated and the matching degree of the capacity vehicle under different matching dimensions is merged and marked.

[0052] Optionally, the transport demand information also includes loading and unloading locations. In one embodiment, the sorting and screening of the matched transport vehicles includes two stages. The first stage is to perform preliminary sorting and screening of the transport vehicles in each transport set before merging the transport sets; the second stage is to perform further fine sorting and screening of the transport vehicles after merging the transport sets. Therefore, in step 330, after merging the transport sets under each matching dimension to obtain the transport resources that match the cargo demand, it can also include: Step 303, obtaining the availability factor of each target transport vehicle corresponding to the transport resource; the availability factor includes location information and load information; Step 304, extracting the driving characteristics of each of the target transport vehicles within a historical time period of a preset length according to the location information, and predicting a first probability of the target transport vehicle driving to the loading place according to the driving characteristics; Step 305, extracting the load characteristics of each of the target transport vehicles in the historical time period according to the load information, and predicting the availability probability of the target transport vehicles according to the load characteristics; the load characteristics include load status, load duration, empty duration, heavy load continuous mileage and empty continuous mileage; Step 306: sort the target transport vehicles based on the first probability and the available probability.

[0053] For the transport resources obtained after merging, obtain the availability factor of each corresponding target transport vehicle, which includes location information and load information. Extract the driving characteristics of the target transport vehicle within a preset time period based on the location information of the target transport vehicle, and predict the first probability of the target transport vehicle driving to the loading place based on the driving characteristics. Among them, the driving characteristics include driving state, driving road type, driving speed and driving direction, etc. The driving state is used to characterize whether the target transport vehicle is driving.

[0054] Furthermore, the load characteristics of the target transport vehicle in the above historical time period are extracted according to the load information of the target transport vehicle, and the availability probability of the target transport vehicle is predicted according to the load characteristics. The load characteristics include load status, load duration, empty duration, heavy load continuous mileage and empty continuous mileage, etc. The load status is used to characterize whether the target transport vehicle is heavily loaded during driving.

[0055] According to the predicted first probability and available probability, each target transport vehicle is ranked, and availability factors such as the probability of the transport vehicle heading for the loading place and the available probability of the transport vehicle are comprehensively introduced to further refine the candidate transport vehicles screened out in the preliminary ranking stage, thereby improving the rationality and accuracy of the selected transport resources.

[0056] Optionally, after finely sorting the candidate transport vehicles, transport resources are recommended to the user in the sorting order. When recommending transport resources to the user, recommendations may be made in the sorting order, including but not limited to recommending the first transport vehicle in descending order, or recommending a given number of transport vehicles in descending order, or recommending a preset proportion of transport vehicles in descending order.

[0057] In one embodiment, the transport vehicles corresponding to the matched transport resources are recommended to the user in the form of a recommendation list. The probability of the transport vehicle heading to the loading place and the probability of borrowing the transport vehicle can be predicted based on a pre-trained machine learning model. Here, there are no excessive restrictions on the adopted machine learning model and its pre-training method.

[0058] In one embodiment, when performing multi-dimensional matching on the transport capacity profile, firstly, screening is performed based on the real-time location of each transport vehicle, and then multi-dimensional matching is performed on the transport capacity profile based on the transportation route information corresponding to the loading and unloading locations in the transport capacity demand information to obtain a transport capacity set under each matching dimension. Based on this, step 320 further includes: Step 321, based on the loading place in the transport demand information, screen out a first candidate vehicle within a preset distance range of the loading place; wherein the model of the first candidate vehicle matches the model in the transport demand information, and the length of the first candidate vehicle matches the length of the vehicle in the transport demand information; Step 322, if the number of the first candidate vehicles exceeds a preset threshold, sorting and screening the first candidate vehicles according to the distance between the first candidate vehicles and the loading place to obtain a preset number of second candidate vehicles; Step 323, determining the transportation route information according to the loading place and unloading place in the transportation capacity demand information; Step 324: Perform multi-dimensional matching on the capacity profile of the second candidate vehicle based on the transportation route information to obtain a capacity set under each matching dimension.

[0059] According to the loading place in the transportation demand information, a first candidate vehicle within a preset distance range of the loading place is screened out, and the first candidate vehicles include multiple vehicles, and each of the screened out first candidate vehicles is a transportation vehicle whose vehicle length matches the vehicle length corresponding to the transportation demand information, and each of the screened out first candidate vehicles is a transportation vehicle whose vehicle model matches the vehicle model corresponding to the transportation demand information. That is, according to the loading place in the transportation demand information, the transportation vehicles whose vehicle model and vehicle length match the vehicle model and vehicle length in the transportation demand information and are located within the preset distance range of the loading place are screened out.

[0060] The number of first candidate vehicles is counted. If the number of first candidate vehicles exceeds a preset threshold, the first candidate vehicles are sorted and screened according to the distance between the first candidate vehicles and the loading place to obtain a preset number of second candidate vehicles. If there are many transport vehicles with models and lengths that meet the requirements within the preset distance range of the loading place, a preset number of transport vehicles closest to the loading place are screened out according to the distance.

[0061] According to the loading and unloading places in the capacity demand information, the transportation route information is determined, and based on the transportation route information, multi-dimensional matching is performed on the capacity profile of the screened second candidate vehicle to obtain the capacity set under each matching dimension.

[0062] In other embodiments, the transport vehicles can be sorted according to the distance between each transport vehicle and the loading site, and then a preset number of transport vehicles closest to the loading site are selected in the sorted order. Then, based on the transportation route information between the loading site and the unloading site, the transport portraits of the screened transport vehicles are multi-dimensionally matched, and the matched transport vehicles are recommended to the user.

[0063] In one embodiment, a capacity resource pool is constructed based on the capacity portraits of all capacity vehicles. When recommending capacity resources, the capacity portraits in the capacity resource pool are matched based on multiple dimensions such as geographical location, transportation routes between loading and unloading locations, and types of transported goods, and capacity vehicles that meet the needs are selected and recommended to users.

[0064] For example, based on the geographic location information of each transport vehicle, transport vehicles within 200 kilometers of the loading site are selected from the transport resource pool, and then these vehicles are sorted from near to far according to the distance between them and the loading site, and the first 2,000 vehicles closest to the loading site are selected. When calculating the distance between each transport vehicle and the loading site, the loading site is first parsed from the cargo demand, so that the address of the loading site is parsed into the corresponding longitude and latitude geographic coordinates, and then the geographic coordinates corresponding to the real-time position of each transport vehicle are obtained. Based on the geographic coordinates of the transport vehicle and the geographic coordinates of the loading site, the distance between the transport vehicle and the loading site is calculated using the Haversine algorithm.

[0065] When performing multi-dimensional matching on the transport profiles of the selected transport vehicles, the transport route information between the loading and unloading locations is specifically matched with the frequently-run routes of the selected transport vehicles, and the top 2,000 transport vehicles are selected in descending order of matching degree from high to low to obtain the transport set under the frequently-run route dimension. Optionally, the frequently-run route matching degree of the transport vehicle = the frequency of passing through the transport route in the past 6 months / the running mileage in the past 6 months.

[0066] In the same manner as the frequently-run route dimension, each capacity set under other matching dimensions such as the frequently-run area dimension and the preferred route dimension is obtained. In some embodiments, when sorting the capacity vehicles in the capacity set, the capacity vehicles can be scored according to the matching degree, and the capacity vehicles can be sorted according to the evaluation score. For example, the scoring logic of the frequently-run area dimension is that if the frequently-run area of ​​the capacity vehicle matches the unloading site, it will get 1 point, and if it does not match, it will get 0 points. In some embodiments, when sorting the capacity vehicles in the capacity set, the capacity vehicles in any capacity set can be comprehensively scored in multiple dimensions in this way, and sorted and filtered according to the comprehensive score.

[0067] Optionally, based on the indicators of the transport vehicle's transport profile, a logistic regression model is used to score the transport vehicle, as shown in the following formula 1-2: ; (1)

[0068] ; (2)

[0069] in, represents the score of the i-th indicator in the capacity profile of the transport vehicle, Is an indicator The corresponding weight coefficient, B is a configurable regression coefficient. Optionally, the weight coefficients of each indicator of the capacity profile can be configured and adjusted according to user preferences.

[0070] In some embodiments, the indicators of the transport portrait include route matching, location matching, frequently traveled area matching, license plate location matching, waybill matching, reach friendliness and information reliability, etc. Among them, route matching represents the similarity between the trajectory of the transport vehicle and the transport route; location matching represents the distance score between the real-time location of the transport vehicle and the loading place, the longer the distance, the lower the score; frequently traveled area matching represents the correlation coefficient between the frequently traveled area of ​​the transport vehicle and the loading place / unloading place, which can be determined based on the distance and regional affiliation between the frequently traveled area and the loading place / unloading place; license plate location matching represents the correlation coefficient between the unloading place of the transport vehicle and the vehicle location, which can be determined based on the distance or regional affiliation between the vehicle location and the unloading place; waybill matching represents the degree of matching between the historical waybill route of the transport vehicle and the required transport route; reach friendliness represents the frequency score of the recommended contact of the transport vehicle, the more frequent the contact, the lower the score; information reliability represents the completeness score of the relevant information of the transport vehicle, the more complete the information, the higher the score.

[0071] On the basis of sorting by comprehensive score, availability factor is introduced to further sort the selected transport vehicles, so as to recommend transport resources to users in the sorting order. Optionally, when initiating a transport resource request, the user can carry preference information for transport resources in the request, for example, giving priority to vehicles with a specific location, vehicles with a frequently used unloading area, or vehicles that are closest, etc. When recommending transport resources to users, personalized recommendations can be made based on the user's preference information for transport resources when the user's preference information for transport resources is obtained. Specifically, on the basis of fine sorting, the vehicles are re-sorted according to the user's preference information to improve the recommendation priority of the transport vehicles that meet the user's preference. For example, when the user's preference information is to give priority to recommending vehicles with the A region and its surrounding areas, the transport vehicles in the region and its surrounding areas are scored for their territorial intention. The vehicle with the A region has an intention score of 1.0, the vehicle with the A region's surrounding area has an intention score of 0.8, and the vehicle with the other regions has an intention score of 0.5. Finally, the final score of each transport vehicle to be recommended is calculated according to the territorial intention score. The calculation method is as follows: ; (3)

[0072] is the final recommended score of the transport vehicle. The score for the intention to belong to the place of origin is is the matching score of the capacity profile, a and b are configurable weights. When the user's preference information includes multiple There can be multiple preference scores.

[0073] Finally, based on different cargo demand sources, differentiated recommendation strategies are selected to obtain a recommended list of required transport vehicles, which includes the recommended serial number, license plate number, driver information, the matching degree between the transport portrait and user demand, the distance to the loading site, frequently used routes and frequently used areas, etc.

[0074] In this embodiment, by mining the basic attribute information and transportation behavior preferences of vehicles, a capacity profile of the transport vehicle is constructed, and it is matched with the user's demand for transport resources to screen out vehicles that meet the user's needs. On this basis, the screened transport vehicles are sorted and screened according to the matching degree, thereby improving the matching degree between the recommended transport vehicles and user needs.

[0075] Furthermore, by introducing the availability factor, the driving direction and availability probability of the transport vehicles are predicted based on the real-time location and load information of the transport vehicles, so that the selected transport vehicles can be sorted and screened more finely, reducing the lag of data collection in a predictive manner and improving the efficiency of recommending transport resources. At the same time, the availability of the recommended transport resources is improved through availability prediction, ensuring the effectiveness and accuracy of the transport resource recommendation.

[0076] The transportation resource recommendation device provided by the present invention is described below. The transportation resource recommendation device described below and the transportation resource recommendation method described above can be referenced to each other.

[0077] Reference Figure 2 , the transport resource recommendation device provided by the embodiment of the present invention includes: The data collection module 10 is used to obtain the transport resource data and the historical transport data of each transport vehicle in the transport resource data; A portrait construction module 20, for generating a transport capacity portrait of each transport vehicle according to the transport capacity resource data and the historical transport data; The resource recommendation module 30 is used to receive a user's transport resource request, match the cargo demand in the transport resource request with the transport profile, and recommend the matched transport resource to the user.

[0078] In one embodiment, the portrait building module 20 is further used to: Extracting basic attribute information of each of the transport vehicles according to the transport resource data; the basic attribute information includes vehicle length, vehicle type, tonnage, load and location; Extracting the transport behavior preferences of each of the transport vehicles based on the historical transport data; the transport behavior preferences include frequently traveled routes, frequently traveled areas, transport mileage, transport time, cargo type and number of visits; The basic attribute information and the transportation behavior preference are used to construct the capacity label of the transportation vehicle to obtain the capacity portrait of the transportation vehicle.

[0079] In one embodiment, the resource recommendation module 30 is further used to: Acquire the transport demand information corresponding to the cargo demand in the transport resource request; the transport demand information at least includes the vehicle type and vehicle length; Based on the transport demand information, the transport profile is matched in multiple dimensions to obtain a transport capacity set under each matching dimension; the matching dimension includes at least one of a location dimension, a route dimension, a frequently traveled area dimension, and a cargo type dimension; The transport capacity sets under each of the matching dimensions are merged to obtain transport capacity resources that match the cargo demand, and the transport capacity resources are recommended to the user.

[0080] In one embodiment, the resource recommendation module 30 is further used to: Obtaining the matching degree of each transport vehicle in the target set; the target set is a transport vehicle set under a target dimension, and the target dimension is any one of the matching dimensions; The transport vehicles in the target set are sorted according to the matching degree, and the transport vehicles in the target set are screened according to the sorting order.

[0081] In one embodiment, the transport capacity demand information also includes a loading location, and the resource recommendation module 30 is further used to: Obtaining the availability factor of each target transport vehicle corresponding to the transport resource; the availability factor includes location information and load information; Extracting the driving characteristics of each of the target transport vehicles within a historical time period of a preset length according to the location information, and predicting a first probability that the target transport vehicle will drive to the loading place according to the driving characteristics; Extract the load characteristics of each target transport vehicle in the historical time period according to the load information, and predict the availability probability of the target transport vehicle according to the load characteristics; the load characteristics include load status, load duration, empty duration, heavy load continuous mileage and empty continuous mileage; The target transport vehicles are sorted based on the first probability and the available probability.

[0082] In one embodiment, the transport capacity demand information also includes a discharge location, and the resource recommendation module 30 is further used to: According to the loading place in the transport demand information, a first candidate vehicle within a preset distance range of the loading place is screened out; wherein the model of the first candidate vehicle matches the model in the transport demand information, and the length of the first candidate vehicle matches the length of the vehicle in the transport demand information; If the number of the first candidate vehicles exceeds a preset threshold, the first candidate vehicles are sorted and screened according to the distances between the first candidate vehicles and the loading place to obtain a preset number of second candidate vehicles; Determine the transportation route information according to the loading place and unloading place in the transportation capacity demand information; Based on the transportation route information, multi-dimensional matching is performed on the capacity profile of the second candidate vehicle to obtain a capacity set under each matching dimension.

[0083] In one embodiment, the data acquisition module 10 is further used for: Identify abnormal data in the transport resource data and the historical transport data; the abnormal data includes missing value data, invalid field data, and data with a field coverage rate less than a preset value; The abnormal data is preprocessed; the preprocessing operation includes at least one of data cleaning, data filtering and data filling.

[0084] Figure 3 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 3 As shown, the electronic device may include: a processor 310, a communications interface 320, a memory 330 and a communication bus 340, wherein the processor 310, the communications interface 320 and the memory 330 communicate with each other through the communication bus 340. The processor 310 may call the logic instructions in the memory 330 to execute the steps of the capacity resource recommendation method, for example including:

[0085] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0086] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored in a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer can perform the steps of the capacity resource recommendation method provided in the above embodiments, for example, including: Acquire transportation resource data and historical transportation data of each transportation vehicle in the transportation resource data; Generate a transport capacity profile of each transport vehicle according to the transport capacity resource data and the historical transport data; Receive a user's transport resource request, match the cargo demand in the transport resource request with the transport profile, and recommend the matched transport resources to the user.

[0087] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the steps of the capacity resource recommendation method provided in the above embodiments, for example, including: Acquire transportation resource data and historical transportation data of each transportation vehicle in the transportation resource data; Generate a transport capacity profile of each transport vehicle according to the transport capacity resource data and the historical transport data; Receive a user's transport resource request, match the cargo demand in the transport resource request with the transport profile, and recommend the matched transport resources to the user.

[0088] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0089] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for recommending transport resources, characterized in that: include: Acquire transportation resource data and historical transportation data of each transportation vehicle in the transportation resource data; Generate a transport capacity profile of each transport vehicle according to the transport capacity resource data and the historical transport data; Receive a user's transport resource request, match the cargo demand in the transport resource request with the transport profile, and recommend the matched transport resources to the user.

2. The method for recommending transportation resources according to claim 1, characterized in that: Generating a transport capacity profile of each transport vehicle according to the transport capacity resource data and the historical transport data includes: Extracting basic attribute information of each of the transport vehicles according to the transport resource data; the basic attribute information includes vehicle length, vehicle type, tonnage, load and location; Extracting the transport behavior preferences of each of the transport vehicles based on the historical transport data; the transport behavior preferences include frequently traveled routes, frequently traveled areas, transport mileage, transport time, cargo type and number of visits; The basic attribute information and the transportation behavior preference are used to construct the capacity label of the transportation vehicle to obtain the capacity portrait of the transportation vehicle.

3. The method for recommending transportation resources according to claim 1, characterized in that: The matching of the cargo demand in the transport resource request with the transport profile and recommending the matched transport resources to the user includes: Acquire the transport demand information corresponding to the cargo demand in the transport resource request; the transport demand information at least includes the vehicle type and vehicle length; Based on the transport demand information, the transport profile is matched in multiple dimensions to obtain a transport capacity set under each matching dimension; the matching dimension includes at least one of a location dimension, a route dimension, a frequently traveled area dimension, and a cargo type dimension; The transport capacity sets under each of the matching dimensions are merged to obtain transport capacity resources that match the cargo demand, and the transport capacity resources are recommended to the user.

4. The method for recommending transportation resources according to claim 3, characterized in that: Before merging the transport capacity sets under the matching dimensions to obtain transport capacity resources matching the cargo demand, the method further includes: Obtaining the matching degree of each transport vehicle in the target set; the target set is a transport vehicle set under a target dimension, and the target dimension is any one of the matching dimensions; The transport vehicles in the target set are sorted according to the matching degree, and the transport vehicles in the target set are screened according to the sorting order.

5. The method for recommending transportation resources according to claim 3, characterized in that: The transport capacity demand information also includes a loading location. After merging the transport capacity sets under the matching dimensions to obtain transport capacity resources matching the cargo demand, the following is further included: Obtaining the availability factor of each target transport vehicle corresponding to the transport resource; the availability factor includes location information and load information; Extracting the driving characteristics of each of the target transport vehicles within a historical time period of a preset length according to the location information, and predicting a first probability that the target transport vehicle will drive to the loading place according to the driving characteristics; Extract the load characteristics of each target transport vehicle in the historical time period according to the load information, and predict the availability probability of the target transport vehicle according to the load characteristics; the load characteristics include load status, load duration, empty duration, heavy load continuous mileage and empty continuous mileage; The target transport vehicles are sorted based on the first probability and the available probability.

6. The method for recommending transportation resources according to claim 5, characterized in that: The transport demand information also includes a place of unloading, and the multi-dimensional matching of the transport profile based on the transport demand information to obtain a transport capacity set under each matching dimension includes: According to the loading place in the transport demand information, a first candidate vehicle within a preset distance range of the loading place is screened out; wherein the model of the first candidate vehicle matches the model in the transport demand information, and the length of the first candidate vehicle matches the length of the vehicle in the transport demand information; If the number of the first candidate vehicles exceeds a preset threshold, the first candidate vehicles are sorted and screened according to the distances between the first candidate vehicles and the loading place to obtain a preset number of second candidate vehicles; Determine the transportation route information according to the loading place and unloading place in the transportation capacity demand information; Based on the transportation route information, multi-dimensional matching is performed on the capacity profile of the second candidate vehicle to obtain a capacity set under each matching dimension.

7. The method for recommending transportation resources according to claim 1, characterized in that: After obtaining the transport resource data and the historical transport data of each transport vehicle in the transport resource data, the method further includes: Identify abnormal data in the transport resource data and the historical transport data; the abnormal data includes missing value data, invalid field data, and data with a field coverage rate less than a preset value; The abnormal data is preprocessed; the preprocessing operation includes at least one of data cleaning, data filtering and data filling.

8. A transport resource recommendation device, characterized in that: include: A data collection module, used to obtain transport resource data and historical transport data of each transport vehicle in the transport resource data; A portrait construction module, used to generate a transport capacity portrait of each transport vehicle according to the transport capacity resource data and the historical transport data; The resource recommendation module is used to receive a user's transport resource request, match the cargo demand in the transport resource request with the transport profile, and recommend the matched transport resources to the user.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the transportation resource recommendation method according to any one of claims 1 to 7 are implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the transportation resource recommendation method according to any one of claims 1 to 7 are implemented.

Images