Logistics platform freight vehicle selection method, device, equipment and storage medium

Abstract

The present invention provides a method, device, equipment and storage medium for optimizing freight vehicles on a logistics platform, including decomposing the type and weight of a target freight task, obtaining the body structure and load capacity of a required truck and publishing corresponding bidding information, collecting bidding trucks according to the bidding information, establishing a recursive hierarchical structure model to obtain all evaluation indicators, selecting subjective evaluation indicators from the evaluation indicators and obtaining the weights of all subjective evaluation indicators, selecting objective evaluation indicators from the evaluation indicators and obtaining the weights of all objective evaluation indicators, normalizing the weights of all subjective evaluation indicators and objective evaluation indicators, scoring the comprehensive performance of all bidding trucks, sorting the bidding trucks from high to low according to the scores, and selecting the bidding trucks based on the sorting. The present invention can comprehensively consider from multiple aspects and select a solution with a high matching degree between the target freight task and the freight car.

Description

Technical Field

[0001] The present invention relates to the technical field of freight vehicle allocation on a network freight platform, and in particular to a method, device, equipment and storage medium for optimizing freight vehicles on a logistics platform. Background Art

[0002] With the advent of the information age, more and more truck drivers choose to accept orders on online freight logistics platforms. Usually, after the online freight logistics platform releases a transport order, the driver who has registered and logged in to the online freight logistics platform will receive a transport order reminder. The driver will receive the transport order according to his own needs, and the online freight logistics platform will then screen the drivers and assign the transport order to relatively suitable drivers.

[0003] At present, there are usually two ways for online freight logistics platforms to select drivers. One is to select freight trucks based on price, but this selection method cannot guarantee service quality, often resulting in poor timeliness, weak reliability, low safety, and high customer complaint rates; the other is to select based on the location of the freight truck and assign the transport order to the freight truck closest to the place of shipment, but this method cannot guarantee the rationality of the price and often increases the cost of the shipper of the transport order. In summary, the existing selection method has the phenomenon of low matching between transport orders and freight trucks, which reduces work efficiency and wastes the waiting time of the consignee. Summary of the invention

[0004] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a method, device, equipment and storage medium for optimizing freight vehicles for a logistics platform, which can comprehensively consider multiple aspects, select solutions with a high degree of match between the target freight task and the freight truck, and construct a freight vehicle optimization evaluation system for a network freight logistics platform that is universal and objective.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A method for selecting a freight vehicle for a logistics platform, comprising:

[0007] Decompose the type and weight of the target freight task, obtain the body structure and load capacity of the required truck, publish bidding information based on the body structure and load capacity of the required truck, and collect bidding trucks based on the bidding information;

[0008] Establish a hierarchical structure model to obtain all evaluation indicators;

[0009] Select the first part of all evaluation indicators as subjective evaluation indicators, and obtain the weights of all subjective evaluation indicators;

[0010] Select the second part of all evaluation indicators as objective evaluation indicators, and obtain the weights of all objective evaluation indicators;

[0011] After the weights of all subjective and objective evaluation indicators are normalized, the comprehensive performance of all bidding trucks is scored and the bidding trucks are ranked from high to low according to the scores;

[0012] Based on the ranking, the bidding truck is selected. When the weight of the target freight task is less than or equal to the first weight threshold, the first truck is selected according to the ranking. When bidding for trucks, the weight of the target freight task is greater than the first weight threshold, the improved genetic algorithm is used to select the first truck in the sorting order. of the bidding trucks Bidding truck.

[0013] Furthermore, the hierarchical structure model includes primary indicators and secondary indicators, wherein the primary indicators are service quality indicators and security management indicators;

[0014] All secondary indicators in the service quality indicators are used as subjective evaluation indicators, including delivery punctuality index, cargo integrity index, transportation time reliability index, communication timeliness index, customer satisfaction index and route deviation index, and the weights of all subjective evaluation indicators are obtained by expert opinion method;

[0015] All secondary indicators in the safety management index are used as objective evaluation indicators. The subjective evaluation indicators include traffic violation rate index, traffic accident rate index, fatigue driving index, truck modification rate index, truck assembly rate index, overload rate index and overspeed rate index. The entropy weight method is used to obtain the weights of all objective evaluation indicators.

[0016] Furthermore, the ranking of service quality indicators and safety management indicators is set, and the weight of each subjective evaluation indicator or objective evaluation indicator is normalized based on the ranking of service quality indicators and safety management indicators to obtain the final weight of each subjective evaluation indicator or objective evaluation indicator.

[0017] Furthermore, the TOPSIS method is used to score the comprehensive performance of all bidding trucks, and the bidding trucks are ranked from high to low according to the scores. The specific method is as follows:

[0018] A weighted decision matrix is ​​constructed to obtain the index value of each bidding truck relative to each subjective evaluation index or objective evaluation index according to the final weight of each subjective evaluation index and objective evaluation index;

[0019] Obtain the distance from each bidding truck to the ideal solution and the distance to the negative ideal solution;

[0020] The relative proximity of each bidding truck is obtained, and all bidding trucks are sorted based on the relative proximity.

[0021] Furthermore, the improved genetic algorithm is used to select the of the bidding trucks The methods for bidding on trucks are:

[0022] Select the order before the bidding trucks The bidding trucks are set up to generate a certain length of Chromosomes, and binary encoding of chromosome genes;

[0023] Screening step: obtaining bidding trucks participating in the target freight task according to the gene coding, obtaining the total load weight of all bidding trucks and the distance between each bidding truck and the shipping place, setting a penalty function, and removing chromosomes whose total load weight does not meet the weight of the target freight task and / or whose distance does not meet the preset distance threshold;

[0024] Establish a multi-objective-based fitness function to obtain the fitness value of the chromosome;

[0025] Obtaining the first optimal chromosome output based on the fitness value and a certain amount of chromosomes other than the first optimal chromosome to form a new population;

[0026] Perform crossover or mutation operations on the chromosomes in the new population to generate new chromosomes;

[0027] Jump to the screening step until the chromosome reaches the maximum number of iterations, output the second best chromosome based on the fitness value, decode the second best chromosome to obtain the selected Bidding truck.

[0028] Furthermore, the multi-objective-based fitness function is composed of the weighted total fuel cost and total time cost, specifically:

[0029] ,

[0030] in, is the weight of the total fuel cost, is the weight of the total time cost, For the Bidding trucks, For the The fuel cost per kilometer of the bidding trucks, For the The distance between the bidding truck and the place of shipment, For the The time cost per kilometer of a bidding truck.

[0031] Furthermore, a method for obtaining a certain amount of chromosomes other than the first optimal chromosome is:

[0032] Get the top ranking The distance between the bidding truck and the shipping place is used to rearrange the gene positions of all chromosomes in descending order based on the distance between the bidding truck and the shipping place;

[0033] Establish multiple concentric circles with equidistant increasing radii with the place of shipment as the center, and obtain one circular area and multiple annular areas;

[0034] The chromosomes other than the first optimal chromosome are grouped into a chromosome set, and the chromosomes in the chromosome set are sorted from small to large according to the binary code value;

[0035] Divide the chromosome group into multiple sub-populations according to the sorting, the number of sub-populations is consistent with the number of regions, so that each sub-population corresponds to each region in turn;

[0036] Chromosomes according to The area where the bidding trucks are located is divided into blocks, and each chromosome is composed of multiple chromosome segments, so that each chromosome segment corresponds to each area in turn;

[0037] Based on the greedy strategy, the subpopulations corresponding to the regions and the chromosome segments corresponding to the regions are compared, and the chromosome segment corresponding to the bidding truck with the largest load capacity is selected, and the chromosome corresponding to the chromosome segment is selected into the new population;

[0038] Among all the bidding trucks corresponding to the chromosomes selected into the new population, the load weight of each bidding truck is greater than a preset load weight threshold of the area where the bidding truck is located.

[0039] A logistics platform freight vehicle optimization device, comprising:

[0040] A collection module is used to decompose the type and weight of the target freight task, obtain the body structure and load capacity of the required truck, publish bidding information based on the body structure and load capacity of the required truck, and collect bidding trucks based on the bidding information;

[0041] A hierarchical structure model building module is used to build a hierarchical structure model to obtain all evaluation indicators;

[0042] A subjective evaluation index weight acquisition module is used to select the first part of all evaluation indexes as subjective evaluation indexes and obtain the weights of all subjective evaluation indexes;

[0043] An objective evaluation index weight acquisition module is used to select the second part of all evaluation indexes as objective evaluation indexes and obtain the weights of all objective evaluation indexes;

[0044] The sorting module is used to normalize the weights of all subjective and objective evaluation indicators, score the comprehensive performance of all bidding trucks, and sort the bidding trucks from high to low according to the scores;

[0045] The selection module is used to select the bidding truck based on the ranking. When the weight of the target freight task is less than or equal to the first weight threshold, the first truck is selected according to the ranking. When bidding for trucks, the weight of the target freight task is greater than the first weight threshold, the improved genetic algorithm is used to select the first truck in the sorting order. of the bidding trucks Bidding truck.

[0046] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for selecting freight vehicles for a logistics platform when executing the program.

[0047] A non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the above-mentioned method for selecting freight vehicles for a logistics platform.

[0048] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0049] In view of the low matching degree between target freight tasks and freight vehicles in existing network freight logistics platforms, the present invention proposes a method for optimizing freight vehicles for logistics platforms. On the basis that the bidding trucks meet the required body structure and load-carrying capacity of the trucks, a hierarchical structure model is established based on the analytic hierarchy process, all evaluation indicators are stratified, the first part of all evaluation indicators is selected as subjective evaluation indicators, the weights of all subjective evaluation indicators are obtained, the second part of all evaluation indicators is selected as objective evaluation indicators, the weights of all objective evaluation indicators are obtained, the rationality of the weights of various indicators is ensured, and then the comprehensive performance of all bidding trucks is scored, and the ranking of the bidding trucks is obtained, which can improve the matching degree between the target freight tasks and freight vehicles, reduce work efficiency, and at the same time waste the waiting time of the consignee, and can also provide the consignee with a better receiving experience and improve service quality.

[0050] In the process of selecting the bidding truck based on the ranking, when the weight of the target freight task is less than or equal to the first weight threshold, it indicates that the weight of the target freight task is small, and the first truck is directly selected according to the ranking. The bidding truck is simple and fast. When the weight of the target freight task is greater than the first weight threshold, it indicates that the weight of the target freight task is large. The improved genetic algorithm is used to select the first of the bidding trucks The bidding trucks are eliminated with long-distance and small-load bidding trucks, which further improves the transportation time and reduces the use of bidding trucks. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The illustrative embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0052] Figure 1 A flow chart of a method for optimizing a freight vehicle for a logistics platform of the present invention;

[0053] Figure 2 In a specific embodiment of the present invention, the top Schematic diagram of the bidding truck and the location of the shipment;

[0054] Figure 3 It is a schematic diagram of the preferred device for the logistics platform freight vehicle of the present invention. DETAILED DESCRIPTION

[0055] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0056] In the following description, a large number of specific details are provided to provide a more thorough understanding of the present invention. However, it is apparent to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, in order to avoid confusion with the present invention, some technical features well known in the art are not described.

[0057] It should be understood that the present invention can be implemented in different forms and should not be interpreted as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to make the disclosure thorough and complete and to fully convey the scope of the present invention to those skilled in the art.

[0058] In one embodiment, Figure 1 As shown, a method for selecting a freight vehicle for a logistics platform is provided, comprising the following steps:

[0059] Step S1: Decompose the type and weight of the target freight task, obtain the body structure and load capacity of the required truck, publish bidding information based on the body structure and load capacity of the required truck, and collect bidding trucks according to the bidding information;

[0060] Step S2: Establish a hierarchical structure model to obtain all evaluation indicators;

[0061] Step S3: Select the first part of all evaluation indicators as subjective evaluation indicators, and obtain the weights of all subjective evaluation indicators;

[0062] Step S4: Select the second part of all evaluation indicators as objective evaluation indicators, and obtain the weights of all objective evaluation indicators;

[0063] Step S5: After the weights of all subjective evaluation indicators and objective evaluation indicators are normalized, the comprehensive performance of all bidding trucks is scored, and the bidding trucks are sorted from high to low according to the scores;

[0064] Step S6: Select bidding trucks based on the ranking. When the weight of the target freight task is less than or equal to the first weight threshold, select the top trucks based on the ranking. When bidding for trucks, the weight of the target freight task is greater than the first weight threshold, the improved genetic algorithm is used to select the first truck in the sorting order. of the bidding trucks Bidding truck.

[0065] In view of the low matching degree between target freight tasks and freight vehicles in existing network freight logistics platforms, the present invention proposes a method for optimizing freight vehicles for logistics platforms. On the basis that the bidding trucks meet the required body structure and load-carrying capacity of the trucks, a hierarchical structure model is established based on the analytic hierarchy process, all evaluation indicators are stratified, the first part of all evaluation indicators is selected as subjective evaluation indicators, the weights of all subjective evaluation indicators are obtained, the second part of all evaluation indicators is selected as objective evaluation indicators, the weights of all objective evaluation indicators are obtained, the rationality of the weights of various indicators is ensured, and then the comprehensive performance of all bidding trucks is scored, and the ranking of the bidding trucks is obtained, which can improve the matching degree between the target freight tasks and freight vehicles, reduce work efficiency, and at the same time waste the waiting time of the consignee, and can also provide the consignee with a better receiving experience and improve service quality.

[0066] In the process of selecting the bidding truck based on the ranking, when the weight of the target freight task is less than or equal to the first weight threshold, it indicates that the weight of the target freight task is small, and the first truck is directly selected according to the ranking. The bidding truck is simple and fast. When the weight of the target freight task is greater than the first weight threshold, it indicates that the weight of the target freight task is large. The improved genetic algorithm is used to select the first of the bidding trucks The bidding trucks are eliminated with long-distance and small-load bidding trucks, which further improves the transportation time and reduces the use of bidding trucks.

[0067] In step S1 of this embodiment, since the body structure and load capacity of the truck required for different types of materials are different, it is necessary to set the corresponding body structure and load capacity of the truck required for different target freight tasks. Based on the accurate body structure and load capacity, the accuracy of cargo transportation can be ensured, and trucks that do not meet the specifications can be prevented from transporting cargo, thereby improving the safety of the target freight tasks.

[0068] According to different load capacities, trucks are divided into four types: heavy, medium, light, and mini trucks. Specifically, heavy trucks refer to vehicles with a gross mass of more than 12 tons and a length of more than 6 meters; medium trucks refer to vehicles with a gross mass of more than 4.5 tons and less than 12 tons and a length of less than 6 meters; light trucks refer to vehicles with a gross mass of less than or equal to 4.5 tons and a length of less than 6 meters; mini trucks refer to trucks with a gross mass of less than 1.8 tons and a length of less than or equal to 3.5 meters. Trucks with different load capacities can transport goods of different weights. When the amount of goods is small, the number of trucks required is 1, and the load capacity of the trucks required needs to be greater than the weight of the goods; when the amount of goods is large, the number of trucks required is multiple, and the total load capacity of the multiple trucks needs to be greater than the weight of the goods.

[0069] Different materials also require different trucks for transportation. At present, trucks are divided into flat cars, vans, special vehicles, container trucks, high guardrails, refrigerated trucks, tractors, tank trucks, ordinary trucks, trailers, supervision trucks, open trucks, micro trucks, and covered trucks.

[0070] In step S2 of this embodiment, the cargo transportation is usually evaluated in two aspects, specifically including service quality indicators and safety management indicators, to ensure low cost, customer satisfaction and transportation safety. Therefore, when establishing a hierarchical structure model, the method of stratifying all secondary indicators is as follows:

[0071] Service quality indicators and safety management indicators are selected as first-level indicators, each of the first-level indicators includes at least one second-level indicator, and all second-level indicators are used as second-level indicators.

[0072] Extract the service quality indicators and safety management indicators of all bidding trucks. As shown in Table 1, it is the indicator system of the freight vehicle optimization evaluation system of the network freight logistics platform, among which the freight vehicle optimization evaluation system of the network freight logistics platform includes service quality indicators and safety management indicators. The service quality indicators include on-time delivery rate indicators, cargo integrity rate indicators, transportation time reliability indicators, communication timeliness indicators, customer satisfaction indicators and route deviation indicators. The safety management indicators include traffic violation rate indicators, traffic accident rate indicators, fatigue driving indicators, truck modification rate indicators, truck assembly rate indicators, overload rate indicators and overspeed rate indicators.

[0073] Table 1 Index system of freight vehicle optimization evaluation system for online freight logistics platform

[0074]

[0075] Specifically, on-time delivery rate refers to the proportion of goods that are successfully delivered to the destination within the agreed time.

[0076] The cargo integrity rate index refers to the delivery of the products ordered by the consignee to the consignee without damage within the specified time according to the consignee's requirements.

[0077] Transportation time reliability refers to the degree to which the transportation of goods is carried out according to the agreed schedule. Generally, if the transportation time of a target freight task is 3 days, the transportation time of the truck should be close to 3 days with little fluctuation.

[0078] The communication timeliness metric measures how long it takes a customer service representative to respond to and resolve a customer inquiry, complaint, or issue.

[0079] Customer satisfaction indicators measure customer satisfaction with logistics services through customer surveys or complaint rates. The goal can be to achieve a certain satisfaction score or reduce the complaint rate.

[0080] The route deviation index refers to the degree of deviation between the truck position and the preset transportation route during the truck transportation process. In this embodiment, the deviation between the actual arrival position of the truck and the preset arrival position obtained by the GPS installed on the truck after each shipment is transported can be used to make a judgment. If the deviation is higher than the deviation threshold, it means that the truck has deviated. If the deviation is lower than the deviation threshold, it means that the truck has not deviated. The reliability of the transportation time can be judged based on the ratio of the number of deviations of the truck and the total number of monitoring times in the time period T.

[0081] The traffic violation rate index refers to the number of violations committed by trucks within a certain period of time.

[0082] The traffic accident rate index refers to the number of accidents involving trucks within a certain period of time.

[0083] The fatigue driving index refers to the number of fatigue driving incidents during a truck's transportation journey within a certain period of time.

[0084] The truck modification rate index refers to the number of times a truck is modified during its transportation journey within a certain period of time.

[0085] The truck assembly rate index refers to the number of assemblies in a truck's transportation journey within a certain period of time.

[0086] The overload rate index refers to the number of times a truck is overloaded during its transportation journey within a certain period of time.

[0087] The speed exceeding index refers to the number of times a truck exceeds the speed limit during its transportation journey within a certain period of time.

[0088] In step S3 of this embodiment, in the freight vehicle optimization evaluation system of the network freight logistics platform, the selection of bidding trucks is essentially a multi-objective optimization problem. The consignee and the consignor of the target freight task are concerned about transportation safety, time, cost and service quality. The network freight logistics platform needs to reduce transportation costs and increase transportation speed as much as possible under the premise of ensuring the safety of the goods. Therefore, based on the given network freight logistics platform freight vehicle optimization evaluation system, it is necessary to determine the weight of the subjective evaluation index.

[0089] In this embodiment, the expert opinion method is used to determine the weights of subjective evaluation indicators. Usually, the parameters of each indicator of the service quality indicator are mostly subjectively confirmed by the shipper and consignee of the target freight task. Therefore, in the indicator system of the freight vehicle optimization evaluation system of the network freight logistics platform, all secondary indicators in the service quality indicator are used as subjective evaluation indicators.

[0090] In this embodiment, the expert opinion method is used to determine the weights of subjective evaluation indicators. Usually, the parameters of each indicator of the service quality indicator are mostly subjectively confirmed by the shipper and consignee of the target freight task. Therefore, in the indicator system of the freight vehicle optimization evaluation system of the network freight logistics platform, all secondary indicators in the service quality indicator are used as subjective evaluation indicators.

[0091] In this embodiment, it is not appropriate to have too many or too few experts. Too few experts will limit the representativeness; too many experts will be difficult to organize and the workload will be heavy. The number of experts should be appropriate to achieve the required credibility. According to research, the credibility increases less with the increase of the number of experts after the number of experts exceeds 15. Generally speaking, 10 to 50 experts are appropriate.

[0092] In step S3 of this embodiment, a certain amount of secondary indicators are selected as subjective evaluation indicators, and the method of using the expert opinion method to determine the weights of all subjective evaluation indicators is as follows:

[0093] Step S301: All secondary indicators in the service quality indicators are used as subjective evaluation indicators, and the subjective evaluation indicators include delivery punctuality rate indicator, cargo integrity rate indicator, transportation time reliability indicator, communication timeliness indicator, customer satisfaction indicator and route deviation indicator;

[0094] Step S302: select multiple experts, ask for their opinions through a questionnaire survey, and use a proportional scale to convert the qualitative results of all subjective evaluation indicators into quantitative ranking results for processing;

[0095] Step S3403: constructing a first judgment matrix A1 based on the ranking results of each expert;

[0096] Step S304: using the matrix geometric mean method to process the first judgment matrix of all experts to obtain the second judgment matrix A2;

[0097] Step S305: Obtaining the weight of each subjective evaluation index based on the second judgment matrix.

[0098] In this embodiment, after obtaining the first judgment matrix A1 and the second judgment matrix A2, it is necessary to perform consistency ratio analysis on each first judgment matrix or the second judgment matrix one by one. When , it is considered that the consistency of the first judgment matrix or the second judgment matrix is ​​acceptable. When , the judgment matrix needs to be adjusted until the matrix satisfies CR<0.1.

[0099] In a specific implementation of this embodiment, 20 experts are selected to score the delivery punctuality index, the cargo integrity index, the transportation time reliability index, the communication timeliness index, the customer satisfaction index and the route deviation index to obtain the first judgment matrix A1.

[0100] The first judgment matrices of all experts are processed by the matrix geometric mean method to obtain the second judgment matrix A2, as shown in Table 2, which is a second judgment matrix table obtained by the matrix geometric mean method in a specific implementation of this embodiment.

[0101] Table 2 The second judgment matrix obtained by matrix geometric mean method

[0102]

[0103] The weights of the various subjective evaluation indicators are obtained based on the second judgment matrix, as shown in Table 3, which is a weight table of various subjective evaluation indicators in a specific implementation of this embodiment.

[0104] Table 3 Weight table of each subjective evaluation index

[0105]

[0106] And get , , , indicating that the consistency check is satisfied.

[0107] In step S4 of this embodiment, the entropy weight method is used to determine the objective weight. Usually, the safety performance index is an inherent parameter in the transportation journey of a truck within a certain period of time. Therefore, in the index system of the freight vehicle optimization evaluation system of the network freight logistics platform, all secondary indicators in the safety management index are used as objective evaluation indicators, and an objective evaluation index set is established for all objective evaluation indicators.

[0108] The entropy weight method is used to determine the objective weight of each objective evaluation indicator. Historical data is obtained from the online freight platform. The transportation itineraries of a certain number of trucks within a certain period of time are taken as samples to obtain the probability of occurrence of traffic violation rate indicators, traffic accident rate indicators, fatigue driving indicators, truck modification rate indicators, truck assembly rate indicators, overload rate indicators and overspeed rate indicators in the bidding trucks.

[0109] Specifically, the method of using the entropy weight method to determine the objective weight of each objective evaluation indicator is as follows:

[0110] Step S401: All secondary indicators in the safety performance index are used as objective evaluation indicators, and the objective evaluation indicators include traffic violation rate index, traffic accident rate index, fatigue driving index, truck modification rate index, truck assembly rate index, overload rate index and overspeed rate index.

[0111] Step S402: historical data is obtained in the online freight logistics platform, the transportation itineraries of a certain number of trucks within a certain period of time are taken as samples, and the occurrence probability of each objective evaluation indicator of each truck is taken as the objective indicator value.

[0112] Step S403: performing standardization processing on all objective indicator values ​​respectively.

[0113] Step S404: Obtain information entropy of all objective evaluation indicators.

[0114] Specifically, the method for obtaining information entropy is: for objective evaluation indicators, calculate their information entropy. The information entropy calculation formula is as follows:

[0115]

[0116] in, For the The probability of an objective evaluation index appearing in all samples is is the total number of objective evaluation indicators, For the The information entropy of an objective evaluation index.

[0117] Step S404: Obtain the weight of each objective evaluation index.

[0118] Specifically, the method for obtaining the weight is: multiply the information entropy of the objective evaluation index by its frequency of occurrence in the sample, then sum it, and finally divide it by the total information entropy of the sample to obtain the weight of each objective evaluation index. The weight calculation formula is as follows:

[0119]

[0120] in, For the The weight of an objective evaluation indicator, is the total information entropy of all samples.

[0121] In a specific implementation of this embodiment, the transportation itineraries of 12 trucks over three years are taken as samples, the occurrence probability of each objective evaluation indicator of each truck is taken as the objective indicator value, and the method provided in this embodiment is used to obtain a weight table of each objective evaluation indicator as shown in Table 4.

[0122] Table 4 Weight table of each objective evaluation index

[0123]

[0124] After obtaining the weights of all subjective evaluation indicators and the weights of all objective evaluation indicators, the service quality indicators and safety management indicators are sorted according to the requirements of the consignee, and then the weights of all subjective evaluation indicators and the weights of all objective evaluation indicators are normalized by setting evaluation weights, so as to obtain the weight of each subjective evaluation indicator or objective evaluation indicator.

[0125] In the specific implementation of this embodiment, both the shipper and the consignee of the target freight task believe that the safety management index is more important. Therefore, the evaluation weight of the safety management index is set to 0.7, and the evaluation weight of the service quality index is set to 0.3. The final weights of all secondary indicators are shown in Table 5.

[0126] Table 5 Final weights of all secondary indicators

[0127]

[0128] The TOPSIS method is an evaluation method that compares and selects schemes based on relative proximity by calculating the distance between each scheme and the ideal scheme and the negative ideal scheme. The TOPSIS method makes full use of the original data and has no strict restrictions on data distribution, sample size, and indicators. The calculation method is simple and has a wide range of applications. Therefore, the TOPSIS method is selected for the comparison and selection of bidding trucks.

[0129] Specifically include:

[0130] Constructing a weighted decision matrix , the matrix elements ,in For the The weight of a subjective evaluation index or an objective evaluation index, For the The bidding trucks are relative to The weight of a subjective evaluation index or an objective evaluation index. Let the ideal solution be , , the negative ideal solution is , , No. The distance from the bidding truck to the ideal solution and the distance to the negative ideal solution According to the following formula:

[0131]

[0132]

[0133] Among them, The closer the bidding truck is to the ideal solution and the farther it is from the negative ideal solution, the better it is. However, it is difficult to meet both requirements in the actual decision-making process. Therefore, relative proximity is introduced. The concept of relative closeness is greater, indicating the The closer the distance of the bidding truck to the ideal solution is, the farther the distance to the negative ideal solution is. The better the bidding trucks, the closer the Calculated by formula.

[0134] The closer the evaluated solution is to the ideal solution and the farther it is from the negative ideal solution, the better it is. However, it is difficult to meet both requirements in the actual decision-making process. Therefore, relative proximity is introduced. The concept of relative proximity is that the closer the solution is to the ideal solution and the farther it is from the negative ideal solution, that is, the better the solution is. The relative proximity of the bidding trucks The calculation formula is:

[0135]

[0136] In this embodiment, when the freight volume is large, it is necessary to select a certain number of bidding trucks as a fleet to jointly undertake the target freight task. If a selected bidding truck is far away from the shipping place, even if the bidding truck is ranked higher, it cannot be ensured that the bidding truck can reach the shipping place quickly. This greatly increases the waiting time of the entire fleet, and also increases the waiting time of the shipper and the consignee. Therefore, it will greatly affect the efficiency of the fleet in completing the target freight task, and also reduce the shipping efficiency of the shipper and the receiving efficiency of the consignee.

[0137] In order to solve this problem, this embodiment needs to select a certain number of bidding trucks from some bidding trucks that are ranked higher. These bidding trucks are relatively close to the shipping place, so that these bidding trucks can quickly reach the shipping place, thereby saving time costs. In addition, fuel costs also need to be considered when selecting bidding trucks, thereby saving fuel costs. The traditional solution is to directly undertake freight tasks through a fleet, but this method is usually not only expensive, but also has a high vacancy rate.

[0138] Genetic algorithm is a random search method. Although it may not be able to find the optimal solution to a problem, the probability of obtaining the optimal solution can be increased through improved genetic algorithms.

[0139] The shipper of the target freight task can publish the target freight task on the online freight logistics platform, and the bidding trucks that meet the required truck body structure and load capacity can bid, which makes it easier for the online freight logistics platform to select.

[0140] When a batch of goods is divided into multiple parts, the load capacity of each bidding truck participating in the target freight task must be reflected in the chromosome, and then the total load capacity of all bidding trucks can be obtained. The distance between the bidding trucks participating in the target freight task and the shipping place must also be reflected.

[0141] In step S6 of this embodiment, when the weight of the target freight task is greater than the first weight threshold, the improved genetic algorithm is used to select the first of the bidding trucks The specific method for bidding for a truck is as follows:

[0142] Step S601: Select the first order according to the order of the bidding trucks The bidding trucks are set up to generate a certain length of Chromosomes, and binary encoding of chromosome genes;

[0143] In step S601 of this embodiment, the chromosome is initialized, and the first order is selected according to the order of the bidding trucks. Bidding trucks, establishing populations, generating chromosomes, the gene number of chromosomes is 1~ , corresponding to 1 to bidding trucks, and the code corresponding to each gene is 0 or 1, 0 means that the bidding truck does not participate in the task, and 1 means that the bidding truck participates in the task. This coding is used to obtain all bidding trucks participating in the task.

[0144] Each gene in the chromosome contains three pieces of information: the bidding truck number, the load capacity of the bidding truck pair, and the distance between the bidding truck and the shipping place.

[0145] Step S602: Screening step: Obtain bidding trucks participating in the target freight task according to the gene encoding, obtain the total load weight of all bidding trucks and the distance between each bidding truck and the shipping place, set a penalty function based on the total load weight and distance, and remove chromosomes whose total load weight does not meet the weight of the target freight task and / or whose distance does not meet the preset distance threshold.

[0146] In step S602 of this embodiment, setting a penalty function specifically includes:

[0147] The total load first penalty function uses the constraint condition that the total load of all bidding trucks is greater than the weight of the target freight task, ensuring that the formed fleet can take away all the target freight tasks at one time, reducing the possibility of round-trip transportation. The expression is:

[0148]

[0149] in, For the The load capacity of the bidding trucks, is the total weight of the cargo, is the total number of trucks bidding.

[0150] The second penalty function of total load is to select the total load of all bidding trucks to be less than or equal to a certain multiple of the weight of the target freight task as a constraint, to prevent the empty rate of bidding trucks from being too high and increase freight costs. The expression is:

[0151]

[0152] in, A multiple of the target cargo mission weight, usually Take 1.2-1.4.

[0153] Distance constraint: The distance between each bidding truck and the shipping place is less than the preset distance threshold as a constraint condition, to prevent a single bidding truck from being too far away from the shipping place and affecting the time of the entire fleet. The expression is:

[0154]

[0155] in, is the preset distance threshold, For the The distance between the bidding truck and the shipping location.

[0156] Step S603: Establish a multi-objective-based fitness function to obtain the fitness value of the chromosome.

[0157] The vehicle performance indicators are considered based on the multi-objective fitness function, and the vehicle performance indicators include fuel consumption indicators and labor cost indicators.

[0158] The fuel cost index refers to the amount of fuel consumed per unit time, unit mileage or unit transportation volume during the use of a truck.

[0159] The labor cost indicator refers to the hourly labor fee or salary of the driver during the use of the truck.

[0160] Specifically, the multi-objective-based fitness function is composed of the total fuel cost and the total time cost weighted as follows:

[0161] ,

[0162] in, is the weight of the total fuel cost, is the weight of the total time cost, For the Bidding trucks, For the The fuel cost per kilometer of the bidding trucks, For the The distance between the bidding truck and the place of shipment, For the The time cost per kilometer of a bidding truck.

[0163] In step S603 of this embodiment, in order to save time and cost, a fitness function is set, which is composed of the weighted total fuel cost and the total time cost, and a bidding fleet consisting of a certain number of bidding trucks with the lowest fitness value is selected as the optimal solution.

[0164] Step S604: Obtain the first optimal chromosome output based on the fitness value and a certain amount of chromosomes other than the first optimal chromosome to form a new population.

[0165] Step S605: Perform a crossover or mutation operation on the chromosomes in the new population to generate new chromosomes.

[0166] Step S606: Jump to the screening step until the chromosome reaches the maximum number of iterations, output the second best chromosome based on the fitness value, decode the second best chromosome to obtain the selected Bidding truck.

[0167] Specifically, the method for obtaining a certain amount of chromosomes other than the first optimal chromosome is:

[0168] Step S6041: Obtain the top ranking The distance from the bidding truck to the shipping place is used to rearrange the gene positions of all chromosomes in descending order based on the distance from the bidding truck to the shipping place.

[0169] In this embodiment, the gene positions of all chromosomes are rearranged in descending order based on the distance between the bidding truck and the shipping place, so that the bidding truck corresponding to the gene positions arranged in sequence is closer and closer to the shipping place. The closer the bidding truck is to the shipping place, the lower the time cost required for the bidding truck to reach the shipping place. The top The bidding trucks can meet the service quality indicators and safety management indicators to a certain extent. Therefore, it is necessary to select bidding trucks that are closer to the shipping place and have a larger load capacity to complete the target freight task, as shown in Table 6 below, which is a chromosome gene rearrangement table. The bidding trucks are getting closer and closer to the shipping place from left to right.

[0170] Table 6 Chromosome gene rearrangement table

[0171]

[0172] in, , ,..., ,..., For the 1st, 2nd, ...., n ,..., N The load capacity of the bidding trucks, , ,... ,..., For the 1st, 2nd, ...., n ,..., N The distance between the bidding truck and the place of shipment, the 1st, 2nd, ...., n ,..., N The distance between the bidding truck and the shipping location is getting closer and closer.

[0173] In the specific implementation of this embodiment, the top 12 bidding trucks are selected, and the If there are 10 bidding trucks, For example, if a chromosome is 010111101101, it means that the selected bidding trucks are the 1st bidding truck, the 3rd bidding truck, the 4th bidding truck, the 6th bidding truck, the 7th bidding truck, the 8th bidding truck, the 9th bidding truck and the 3rd bidding truck.

[0174] Step S6042: Establish multiple concentric circles with equidistantly increasing radii with the shipping location as the center, to obtain one circular area and multiple annular areas.

[0175] like Figure 2 As shown, the top ranking in a specific embodiment of the present invention is Schematic diagram of the bidding trucks and the locations of the shipping places. In the specific implementation manner of this embodiment, there are a total of 4 areas, namely 1 circular area and 3 annular areas. Taking the shipping place as the starting point, the 4 areas are set from near to far as the first area, the second area, the third area and the fourth area. The 12 bidding trucks are arranged according to the distance in a plurality of concentric circles with equidistant increasing radius established with the shipping place as the center.

[0176] Step S6043: Group the chromosomes other than the first optimal chromosome into a chromosome group, and sort the chromosomes in the chromosome group from small to large according to the binary code value.

[0177] Step S6044: Divide the chromosome group into multiple sub-populations according to the sorting, and the number of sub-populations is consistent with the number of regions, so that each sub-population corresponds to each region in turn.

[0178] In the specific implementation of this embodiment, since there are 4 regions in total, the chromosome set is divided into 4 sub-populations, namely the first sub-population, the second sub-population, the third sub-population and the fourth sub-population.

[0179] Step S6045: sort the chromosomes according to The area where the bidding trucks are located is divided into blocks, and each chromosome is composed of multiple chromosome fragments, so that each chromosome fragment corresponds to each area in turn.

[0180] For example, chromosome 010111101101 is divided into blocks according to the regions where the bidding trucks are located. After the division, the chromosome is: 01, 011, 1101, 101. The chromosome is divided into 4 chromosome segments, namely the first chromosome segment, the second chromosome segment, the third chromosome segment and the fourth chromosome segment. The first chromosome segment corresponds to the first region, in which there are the 1st bidding truck, the 2nd bidding truck and the 3rd bidding truck; the second chromosome segment corresponds to the second region, in which there are the 4th bidding truck, the 5th bidding truck, the 6th bidding truck and the 7th bidding truck; the third chromosome segment corresponds to the third region, in which there are the 8th bidding truck, the 9th bidding truck and the 10th bidding truck; the fourth chromosome segment corresponds to the fourth region, in which there are the 11th bidding truck and the 12th bidding truck.

[0181] Step S6046: Based on the greedy strategy, the sub-populations corresponding to the regions and the chromosome segments corresponding to the regions are compared, the chromosome segment corresponding to the bidding truck with the largest load capacity is selected, and the chromosome corresponding to the chromosome segment is selected into the new population.

[0182] For example, there are two chromosomes in the first subpopulation, namely, the first chromosome 010111101011 and the second chromosome 010111101100. The first chromosome 010111101011 and the second chromosome 010111101100 are divided into blocks according to the areas where the bidding trucks are located.

[0183] Then the first chromosome is divided into blocks: 01, 011, 1101, 011;

[0184] The second chromosome is divided into blocks: 01, 011, 1101, 100;

[0185] Based on the greedy strategy, the first subpopulation corresponding to the first region and the chromosome fragments corresponding to the region are compared. Since the load capacity of the third bidding truck is greater than the load capacity of the first bidding truck and the load capacity of the second bidding truck, the second chromosome is selected. This can reduce the use of trucks and reduce the vacancy rate, thereby reducing costs.

[0186] Example 2

[0187] Embodiment 2 provides a logistics platform freight vehicle optimization device, such as Figure 3 As shown, including:

[0188] A collection module is used to decompose the type and weight of the target freight task, obtain the body structure and load capacity of the required truck, publish bidding information based on the body structure and load capacity of the required truck, and collect bidding trucks based on the bidding information;

[0189] A hierarchical structure model building module is used to build a hierarchical structure model to obtain all evaluation indicators;

[0190] A subjective evaluation index weight acquisition module is used to select the first part of all evaluation indexes as subjective evaluation indexes and obtain the weights of all subjective evaluation indexes;

[0191] An objective evaluation index weight acquisition module is used to select the second part of all evaluation indexes as objective evaluation indexes and obtain the weights of all objective evaluation indexes;

[0192] The sorting module is used to normalize the weights of all subjective and objective evaluation indicators, score the comprehensive performance of all bidding trucks, and sort the bidding trucks from high to low according to the scores;

[0193] The selection module is used to select the bidding truck based on the ranking. When the weight of the target freight task is less than or equal to the first weight threshold, the first truck is selected according to the ranking. When bidding for trucks, the weight of the target freight task is greater than the first weight threshold, the improved genetic algorithm is used to select the first truck in the sorting order. of the bidding trucks Bidding truck.

[0194] Example 3

[0195] Embodiment 3 provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for selecting freight vehicles for a logistics platform when executing the program.

[0196] Example 4

[0197] Embodiment 3 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 above-mentioned method for selecting freight vehicles for a logistics platform is implemented.

[0198] Those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including several instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. 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 and other media that can store program codes.

[0199] Those skilled in the art will appreciate that the above-mentioned embodiments are specific examples for implementing the present invention, and in actual applications, various changes may be made thereto in form and detail without departing from the spirit and scope of the present invention.

[0200] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for selecting freight vehicles for a logistics platform, characterized in that: include: Decompose the type and weight of the target freight task, obtain the body structure and load capacity of the required truck, publish bidding information based on the body structure and load capacity of the required truck, and collect bidding trucks based on the bidding information; Establish a hierarchical structure model to obtain all evaluation indicators; Select the first part of all evaluation indicators as subjective evaluation indicators, and obtain the weights of all subjective evaluation indicators; Select the second part of all evaluation indicators as objective evaluation indicators, and obtain the weights of all objective evaluation indicators; After the weights of all subjective and objective evaluation indicators are normalized, the comprehensive performance of all bidding trucks is scored and the bidding trucks are ranked from high to low according to the scores; Based on the ranking, the bidding truck is selected. When the weight of the target freight task is less than or equal to the first weight threshold, the first truck is selected according to the ranking. When bidding for trucks, the weight of the target freight task is greater than the first weight threshold, the improved genetic algorithm is used to select the first truck in the sorting order. of the bidding trucks The specific method for bidding for a truck is as follows: Select the order before the bidding trucks The bidding trucks are set up to generate a certain length of Chromosomes, and binary encoding of chromosome genes; Screening step: obtaining bidding trucks participating in the target freight task according to the gene coding, obtaining the total load weight of all bidding trucks and the distance between each bidding truck and the shipping place, setting a penalty function, and removing chromosomes whose total load weight does not meet the weight of the target freight task and / or whose distance does not meet the preset distance threshold; Establish a multi-objective-based fitness function to obtain the fitness value of the chromosome; The first optimal chromosome output based on the fitness value and a certain amount of chromosomes other than the first optimal chromosome are obtained to form a new population. The specific method is: Get the top ranking The distance between the bidding truck and the shipping place is used to rearrange the gene positions of all chromosomes in descending order based on the distance between the bidding truck and the shipping place; Establish multiple concentric circles with equidistant increasing radii with the place of shipment as the center, and obtain one circular area and multiple annular areas; The chromosomes other than the first optimal chromosome are grouped into a chromosome set, and the chromosomes in the chromosome set are sorted from small to large according to the binary code value; Divide the chromosome group into multiple sub-populations according to the sorting, the number of sub-populations is consistent with the number of regions, so that each sub-population corresponds to each region in turn; Chromosomes according to The area where the bidding trucks are located is divided into blocks, and each chromosome is composed of multiple chromosome segments, so that each chromosome segment corresponds to each area in turn; Based on the greedy strategy, the subpopulation corresponding to the region is obtained, the chromosome fragments corresponding to the region are compared, the chromosome fragment corresponding to the bidding truck with the largest load capacity is selected, and the chromosome corresponding to the chromosome fragment is selected into the new population; Among all the bidding trucks corresponding to the chromosomes selected into the new population, the load weight of each bidding truck is greater than the preset load weight threshold of the area where the bidding truck is located; Perform crossover or mutation operations on the chromosomes in the new population to generate new chromosomes; Jump to the screening step until the chromosome reaches the maximum number of iterations, output the second best chromosome based on the fitness value, decode the second best chromosome to obtain the selected Bidding truck.

2. The method for selecting a freight vehicle for a logistics platform according to claim 1, characterized in that: The hierarchical structure model includes primary indicators and secondary indicators, wherein the primary indicators are service quality indicators and security management indicators; All secondary indicators in the service quality indicators are used as subjective evaluation indicators, including delivery punctuality index, cargo integrity index, transportation time reliability index, communication timeliness index, customer satisfaction index and route deviation index, and the weights of all subjective evaluation indicators are obtained by expert opinion method; All secondary indicators in the safety management index are used as objective evaluation indicators. The subjective evaluation indicators include traffic violation rate index, traffic accident rate index, fatigue driving index, truck modification rate index, truck assembly rate index, overload rate index and overspeed rate index. The entropy weight method is used to obtain the weights of all objective evaluation indicators.

3. The method for selecting a freight vehicle for a logistics platform according to claim 2, characterized in that: The ranking of service quality indicators and safety management indicators is set, and the weight of each subjective evaluation indicator or objective evaluation indicator is normalized based on the ranking of service quality indicators and safety management indicators to obtain the final weight of each subjective evaluation indicator or objective evaluation indicator.

4. The method for selecting a freight vehicle for a logistics platform according to claim 1, characterized in that: The TOPSIS method is used to score the comprehensive performance of all bidding trucks, and the bidding trucks are ranked from high to low according to the scores. The specific method is as follows: A weighted decision matrix is ​​constructed to obtain the index value of each bidding truck relative to each subjective evaluation index or objective evaluation index according to the final weight of each subjective evaluation index and objective evaluation index; Obtain the distance from each bidding truck to the ideal solution and the distance to the negative ideal solution; The relative proximity of each bidding truck is obtained, and all bidding trucks are sorted based on the relative proximity.

5. The method for selecting a freight vehicle for a logistics platform according to claim 1, characterized in that: The multi-objective-based fitness function is composed of the total fuel cost and the total time cost weighted as follows: ， in, is the weight of the total fuel cost, is the weight of the total time cost, For the Bidding trucks, For the The fuel cost per kilometer of the bidding trucks, For the The distance between the bidding truck and the place of shipment, For the The time cost per kilometer of a bidding truck.

6. A logistics platform freight vehicle optimization device, applied to the logistics platform freight vehicle optimization method according to any one of claims 1 to 5, characterized in that: include: A collection module is used to decompose the type and weight of the target freight task, obtain the body structure and load capacity of the required truck, publish bidding information based on the body structure and load capacity of the required truck, and collect bidding trucks based on the bidding information; A hierarchical structure model building module is used to build a hierarchical structure model to obtain all evaluation indicators; A subjective evaluation index weight acquisition module is used to select the first part of all evaluation indexes as subjective evaluation indexes and obtain the weights of all subjective evaluation indexes; An objective evaluation index weight acquisition module is used to select the second part of all evaluation indexes as objective evaluation indexes and obtain the weights of all objective evaluation indexes; The sorting module is used to normalize the weights of all subjective and objective evaluation indicators, score the comprehensive performance of all bidding trucks, and sort the bidding trucks from high to low according to the scores; The selection module is used to select the bidding truck based on the ranking. When the weight of the target freight task is less than or equal to the first weight threshold, the first truck is selected according to the ranking. When bidding for trucks, the weight of the target freight task is greater than the first weight threshold, the improved genetic algorithm is used to select the first truck in the sorting order. of the bidding trucks Bidding truck.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, it implements the logistics platform freight vehicle optimization method as described in any one of claims 1 to 5.

8. 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, it implements the logistics platform freight vehicle optimization method as described in any one of claims 1 to 5.

Images