Logistics circulation sorting information management system

Abstract

The invention relates to the technical field of logistics information management systems, in particular to a logistics circulation sorting information management system, which is characterized in that priority calculation and sorting are carried out on all goods on the current day of a city sorting station, and target sorting point information is acquired from the goods on the current day; the priority of each sorting point is calculated by combining the specific characteristics of each sorting point, and a plurality of schemes are calculated through a sorting management comprehensive calculation module according to the number of current transportation personnel, the priority of the transportation sorting point of the goods needing to be delivered on the current day and the priority of all the goods needing to be delivered on the current day; the optimal scheme is obtained from the cost factor and the transportation efficiency factor, and the cargoes are sorted through the calculated optimal scheme, so that the cost is reduced as much as possible under the condition that the cargo transportation efficiency is ensured.

Description

Technical Field

[0001] This invention relates to the field of logistics information management system technology, specifically a logistics flow and sorting information management system. Background Technology

[0002] With the development of information technology, online shopping has gradually replaced offline shopping as the mainstream shopping method. To ensure accurate and fast delivery of goods to users, dedicated logistics transit stations have been established in each city. Goods are sorted at these transit stations, and then further sorted and delivered through designated transit stations in various districts and streets within the city. During the daily sorting and management process at these city transit stations, each item needs to be carefully managed to ensure efficient transportation. Due to the increase in online shopping promotions and technological advancements in recent years, the volume of goods requiring transportation has multiplied. Maintaining the original high efficiency of transportation has led to continuously increasing transportation costs, which have a significant impact on sorting centers. Therefore, reducing costs while ensuring fast delivery has become a crucial issue.

[0003] Therefore, a logistics flow and sorting information management system is proposed. Summary of the Invention

[0004] The purpose of this invention is to provide a logistics flow sorting information management system. By calculating and sorting the priority of all goods that need to be delivered to the city sorting station on the same day, and combining the priority of each sorting point calculated at the next level, the sorting management comprehensive calculation module calculates the optimal solution based on the current number of transportation personnel, the sorting points of goods to be delivered on the same day, and the priority of all goods to be delivered on the same day. The goods are then sorted according to the calculated optimal solution.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] The cargo information acquisition module acquires first feature information and second feature information of all cargoes that need to be delivered on the same day; the first feature information of all cargoes that need to be delivered on the same day is cargo feature information, and the second feature information of all cargoes that need to be delivered on the same day is cargo transportation feature information; the first feature information and the second feature information are used to calculate the priority of all cargoes that need to be delivered on the same day.

[0007] The cargo information acquisition module acquires the first feature information and the second feature information of all the cargoes that need to be delivered on the same day through the cargo information acquisition model;

[0008] The cargo information acquisition model includes a cargo first feature acquisition model and a cargo second feature acquisition model;

[0009] The cargo first feature acquisition model is used to automatically acquire the cargo first feature information;

[0010] Furthermore, the cargo first feature acquisition model includes an input module, a cargo first feature recognition module, and an output module;

[0011] The input module is used to input cargo information into the cargo first feature acquisition model to obtain the cargo's first feature information;

[0012] The cargo first feature recognition module includes a shape feature recognition module and a size feature recognition module;

[0013] The shape feature recognition module is used to identify the shape features of the goods. The shape feature recognition module includes three convolutional layers and one fully connected layer. Each convolutional layer is a 3*3 Sobel convolutional kernel.

[0014] The size feature recognition module is used to identify the size features of goods. The size feature recognition module includes three sets of convolutional layers. Each set of convolutional layers includes four convolutional layers and one fully connected layer. Each convolutional layer is a 5*5 Sobel convolutional kernel.

[0015] The output module is used to output the first feature information;

[0016] Furthermore, the recognition process of the first feature acquisition model is as follows:

[0017] The input module obtains input data H1 for all goods that need to be delivered on the same day.

[0018] The input data H1 is input into the shape feature recognition module to analyze the cargo shape feature information and obtain cargo shape feature H2;

[0019] The input data H2 is input into the size feature recognition module to analyze the cargo size feature information, and the cargo size feature H3 is obtained;

[0020] The cargo shape feature H2 and the cargo size feature H3 are input to the output module to obtain the first feature information H4;

[0021] The cargo second feature acquisition model is used to automatically acquire the cargo second feature information;

[0022] The cargo second feature acquisition model includes an input module, a cargo second feature recognition module, and an output module;

[0023] The input module is used to input cargo information to facilitate the cargo second feature acquisition model in acquiring the cargo's second feature information;

[0024] The cargo second feature recognition module is used to acquire and analyze the second feature of the cargo;

[0025] The cargo second feature acquisition model includes an input module, a cargo second feature recognition module, and an output module;

[0026] The input module is used to input cargo information to facilitate the cargo second feature acquisition model in acquiring the cargo's second feature information;

[0027] The cargo second feature recognition module is used to acquire and analyze the second feature of the cargo. The second feature recognition module includes a second feature encoder, a second feature decoder, and a second feature recognizer. The second feature encoder includes four encoding layers, each of which includes two convolutional layers. The first convolutional layer includes a 3*3 convolutional layer, and the second convolutional layer includes a 1*1 convolutional layer. The feature recognizer includes a Transform model.

[0028] The output module is used to output the second characteristic information of the goods;

[0029] Furthermore, the identification process of the cargo second feature acquisition model is as follows:

[0030] On the same day, all the aforementioned goods information input modules need to be delivered to obtain input data G1;

[0031] The input data G1 is input into the second feature encoder of the cargo second feature recognition module to analyze the second feature of the cargo and obtain feature information G2.

[0032] The feature information G2 is input into the second feature decoder for feature reconstruction to obtain feature information G3;

[0033] The feature information G3 is input to the second feature recognizer to identify features and obtain feature information G4.

[0034] The feature information G4 is input to the output module and is used to output the second feature information of all the goods that need to be delivered on the same day.

[0035] Furthermore, the cargo priority ranking module inputs the first feature information and the second feature information of all cargoes that need to be delivered on the same day into the cargo priority calculation model of the cargo priority module, and calculates and sorts the priority order of all cargoes that need to be delivered on the same day through the cargo priority calculation model;

[0036] The cargo priority ranking of the cargo priority ranking module is obtained through the cargo priority calculation model;

[0037] The cargo priority calculation model includes a feature information input module, a priority calculation module, and a sorting module;

[0038] After the cargo feature information is input into the feature information input module, the priority of the cargo is calculated by the priority calculation module, and the cargo is sorted by the sorting module according to the priority result.

[0039] The feature information input module is used to input the first feature information and the second feature information of the goods;

[0040] Furthermore, the priority calculation module is used to determine the priority level of the goods based on the first feature information and the second feature information of the goods using a priority calculation formula;

[0041] The priority calculation formula is as follows:

[0042]

[0043] Wherein, GDOP represents the priority level of the goods, and GX represents the priority level of the goods. i For the first feature information, λ i GY represents the information weight of the first feature information, m represents the total number of the first feature information, and GY represents the information weight of the first feature information. j For the second feature information, λ j is the information weight of the second feature information, and n is the total number of the second feature information;

[0044] The sorting module is used to sort the goods according to their priority level.

[0045] Furthermore, the sorting point priority ranking module obtains information about each sorting point in the transportation based on the second feature information of the goods. Based on the sorting point information, the module obtains the first sorting feature information and the second sorting feature information of each sorting point. The first sorting feature information and the second sorting feature information of each sorting point are input into the sorting point priority calculation model of the sorting point priority ranking module to calculate the priority order of each sorting point and sort them.

[0046] The first sorting feature information of each sorting point is distance feature information, and the second sorting feature information of each sorting point is cargo transportation volume feature information; the first sorting feature information and the second sorting feature information are used to calculate the priority of each sorting point;

[0047] The sorting point priority module calculates the priority level of each sorting point through the sorting point priority calculation model, and sorts each sorting point according to the calculated priority level.

[0048] The sorting point priority calculation model includes a sorting point priority analysis module and a sorting point priority calculation module;

[0049] The sorting point priority analysis module is used to analyze the first sorting feature information and the second sorting feature information;

[0050] The sorting point priority analysis module includes a sorting feature information input module, a sorting feature information prediction module, and a sorting feature information evaluation module;

[0051] The feature information input module is used to input the first sorting feature information and the second sorting feature information;

[0052] The sorting feature information prediction module is used to acquire historical data of the first sorting feature information and the second sorting feature information, and predict first evaluation data of multiple first sorting feature information and second sorting feature information based on the historical data. The sorting feature information prediction module includes three long short-term memory neural networks, and each long short-term memory neural network includes four long short-term memory neural network units.

[0053] The sorting feature information evaluation module is used to calculate the first evaluation data and output the second evaluation data.

[0054] After the first sorting feature information and the second sorting feature information are input into the module, the first evaluation data is obtained through the sorting feature information prediction module, and the second evaluation data is obtained by inputting multiple reference evaluation data into the sorting feature information evaluation module.

[0055] Furthermore, the sorting point priority calculation module is used to calculate the analysis results of the sorting point priority analysis module to obtain the priority level of each sorting point;

[0056] The sorting point priority ranking module includes a sorting point priority calculation module:

[0057] The sorting point priority calculation module calculates the priority level of each sorting point using the sorting point priority calculation formula.

[0058] The formula for calculating the priority of sorting points is as follows:

[0059]

[0060] Wherein, WDOP represents the sorting point priority, and DIS... t The first sorting feature information is k, where k is the number of the first sorting feature information, and GIO is the number of sorting feature information. c For the second sorting feature information, λ cThe weight of the second sorting feature information is l, and the quantity of the second sorting feature information is l.

[0061] Furthermore, the sorting management comprehensive calculation module obtains the first logistics characteristic information of the current city sorting station, and inputs the first logistics characteristic information, the priority order of all goods to be delivered on the same day, and the priority order of each sorting point into the sorting management comprehensive calculation model of the sorting management comprehensive calculation module to obtain the optimal sorting scheme;

[0062] The first logistics characteristic information of the current city sorting station is transportation resource characteristic information;

[0063] The first logistics characteristic information, the priority of all the goods to be delivered on the same day, and the priority of each sorting point are used to calculate the optimal logistics sorting plan;

[0064] The sorting management integrated calculation module includes a sorting management integrated calculation model;

[0065] The sorting management integrated calculation model includes a data receiving module, a first scheme generation module, and a second scheme evaluation module.

[0066] The data receiving module is used to receive the first logistics characteristic information of the current city sorting station, the priority level of the sorting point, and the priority level of the goods;

[0067] Furthermore, the first scheme generation module is used to generate a first scheme using the data obtained by the receiving module;

[0068] The calculation formula for generating the first scheme is as follows:

[0069]

[0070] Where PLANSCH is the generated first scheme, GP ig For the goods that need to be transported on the same day, WDOP ig GDOP is the priority of the transport sorting point. ig Let g be the priority of the goods, g be the quantity of the goods to be transported that day, and TRANS be the priority of the goods. n The first logistics feature information; num is the quantity of the first logistics feature information;

[0071] Furthermore, the second scheme evaluation module is used to derive a second scheme by generating the content of the first scheme evaluation scheme;

[0072] The calculation formula for the second scheme evaluation module is as follows:

[0073]

[0074] Where P is the total cost calculation result of the scheme, θ RES Here, NRES represents the cost weight of the first scheme, and PLANSCH represents the total cost calculation. RES The first scheme generated;

[0075]

[0076] Wherein, PLAN represents the second scheme, P1, P2, ..., P n For the first scheme after calculating the cost, ω TRES As the time cost weight of the first scheme, PLANSCH TRES Let n be the number of schemes, and n be the first scheme.

[0077] Goods are dispatched according to the optimal sorting scheme.

[0078] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0079] 1. A cargo priority calculation formula is used to obtain the size, shape, and type of cargo in the city sorting station. Based on the obtained cargo feature information, the corresponding weight parameters of each cargo are obtained. Combining the corresponding weight parameters of the cargo with the volume and load data of the transport vehicle, the priority level of each cargo is calculated. By calculating the priority level of each cargo through the feature information of the cargo itself and the feature information of the transport vehicle, the cargo can be delivered within the specified time while reducing costs.

[0080] 2. A sorting point conveying priority calculation formula is adopted. The transportation weight parameters of each responsible sorting point are calculated by using information such as the distance between the city sorting station and the responsible sorting points in each district and street, as well as historical congestion information. The shipping weight parameters of each responsible sorting point are calculated by using information such as the historical shipment volume, capacity of goods, and number of transportation personnel. The conveying priority of each responsible sorting point is calculated by combining the shipping weight parameters and the transportation weight parameters, so as to ensure that the goods can be delivered in a timely manner while minimizing costs.

[0081] 3. Combining the calculated cargo priority level and the delivery priority of each responsible sorting point, a comprehensive sorting management calculation model is used. The volume of cargo delivered from the city sorting station to each sorting point on that day, the priority level of each sorting station, and the priority level of the cargo are input into the comprehensive sorting management calculation model to obtain the optimal sorting plan. This ensures that the city sorting station maximizes the use of current transportation resources and minimizes transportation costs, and ensures that the cargo is delivered to each responsible sorting point in the shortest possible time within a suitable cost range. Attached Figure Description

[0082] Figure 1 This is a schematic diagram of the process of the present invention;

[0083] Figure 2 This is a schematic diagram of the cargo first feature acquisition model of the present invention;

[0084] Figure 3 This is a schematic diagram of the cargo second feature acquisition model of the present invention;

[0085] Figure 4 This is a schematic diagram of the sorting point priority analysis module of the present invention. Detailed Implementation

[0086] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0087] As online shopping gradually becomes the mainstream shopping method, the logistics and transportation of goods have also developed alongside its rise. Currently, numerous cargo sorting centers have been built across my country, distributing goods according to their origin and destination to ensure efficient and safe transport. Once goods reach the destination city, they are sent from the city's sorting centers to designated regional sorting points, and then from these points to the recipients. With the increase in online shopping activities in recent years, the daily volume of goods handled by these sorting centers has multiplied. Maintaining the original transportation speed while processing this increasing volume, ensuring timely delivery, requires a large workforce working around the clock. This leads to continuously rising costs for the sorting centers. However, the capacity of each sorting center is limited, and costs cannot be increased indefinitely. Furthermore, the rising costs are impacting the centers. To reduce costs while maintaining transportation efficiency, this invention provides a logistics flow sorting information management system, the technical solution of which is as follows:

[0088] Reference Figure 1 In step one, the cargo information acquisition module automatically identifies and acquires the item characteristic information and transportation characteristic information of all cargoes that need to be delivered on the same day; the cargo priority ranking module inputs the item characteristic information and transportation characteristic information of all cargoes that need to be delivered on the same day into the cargo priority calculation model of the cargo priority module, and calculates and ranks the priority order of all cargoes that need to be delivered on the same day through the cargo priority calculation model.

[0089] Step 2: The sorting point priority ranking module obtains information about each sorting point in the transportation based on the second feature information of the goods. Based on the sorting point information, the module obtains the first sorting feature information and the second sorting feature information of each sorting point. The first sorting feature information and the second sorting feature information of each sorting point are input into the sorting point priority calculation model of the sorting point priority ranking module to calculate the priority order of each sorting point and sort them.

[0090] Step 3: The sorting management comprehensive calculation module obtains the transportation resource characteristic information of the current city sorting station, and inputs the transportation resource characteristic information, the priority order of all goods to be delivered on the same day, and the priority order of each sorting point into the sorting management comprehensive calculation model of the sorting management comprehensive calculation module. Through comprehensive calculation of cost and transportation time, the optimal sorting plan is obtained.

[0091] Dispatch goods according to the optimal sorting plan;

[0092] The cargo information acquisition module automatically obtains the item and transportation characteristics of all goods to be delivered that day. Based on this information, it calculates the priority level for each item and sorts them according to the calculated priority levels. The priority of the transported goods is then determined based on the sorting results. The sorting point priority ranking module obtains information about each sorting point that needs to be transported that day based on the transportation characteristics of all goods to be delivered. It then obtains the distance and cargo volume characteristics of each sorting point and inputs these characteristics into the sorting point priority ranking model. The model calculates the priority of each sorting point... The sorting points are prioritized and sorted to obtain the sorting results. Based on the sorting results, the priority of each sorting point is obtained. The calculated priority of goods, the priority of each sorting point, and the transportation resource characteristics of the current city sorting stations are input into the sorting management comprehensive calculation module. The sorting management comprehensive calculation module calculates multiple sorting schemes. Then, by calculating the transportation cost and comparing the transportation time of each sorting scheme, the optimal sorting scheme is obtained. Goods are dispatched according to the optimal sorting scheme. By utilizing the resources currently available at each sorting station, the calculated optimal scheme can sort and dispatch goods in a reasonable manner, saving costs without affecting transportation efficiency, and delivering goods on time with the highest efficiency.

[0093] Example 1

[0094] As a first-tier city, N City experiences a surge in cargo transportation during the Double 11 shopping festival. The numerous sorting points also contribute to exceptionally high transportation costs. The cargo information acquisition module retrieves the primary and secondary characteristics of all goods requiring delivery to N City's sorting stations that day. The primary characteristics include packaging size and shape, while the secondary characteristics include transportation details such as destination, expedited processing, and presence of fragile labels. These characteristics are used to calculate the priority of all goods requiring delivery to N City's sorting stations that day. Given the large volume and diverse needs of the goods (e.g., cold chain transport for goods with a limited shelf life, bulky items, and fragile goods), the module combines information obtained during scanning with image recognition technology to automatically acquire all the cargo's characteristics. These characteristics represent the cargo's priority, facilitating subsequent priority calculations.

[0095] The cargo information acquisition module at the N City sorting station obtains the first and second characteristic information of all cargoes requiring delivery that day through a cargo information acquisition model. This model includes a first characteristic acquisition model and a second characteristic acquisition model for cargoes at the N City sorting station. Since the first characteristic information is cargo feature information, and the second characteristic information is cargo transportation feature information, the information contained within them differs. A single model would be insufficient for analyzing their characteristics. Therefore, based on the characteristics of the feature information, two separate models are used for analysis. Different models output different results for different feature information, facilitating subsequent cargo priority calculations.

[0096] The first feature acquisition model for goods at the N City sorting station is used to automatically acquire the feature information of all goods that need to be delivered to the N City sorting station on the same day.

[0097] Reference Figure 2 As shown, taking cuboid cargo as an example, the cargo first feature acquisition model of the city sorting station in N city includes an input module, a cargo first feature recognition module, and an output module;

[0098] The input module is used to input the information of the goods so that the first feature acquisition model of the goods can obtain the feature information of the goods;

[0099] The first feature recognition module for goods at the N City urban sorting station includes the shape feature recognition module and the size feature recognition module for the N City urban sorting station.

[0100] The shape feature recognition module is used to identify the shape features of goods. The shape feature recognition module includes three convolutional layers and one fully connected layer. Each convolutional layer is a 3*3 Sobel convolutional kernel.

[0101] The size feature recognition module is used to identify the size features of goods. The size feature recognition module includes three sets of convolutional layers. Each set of convolutional layers includes four convolutional layers and one fully connected layer. Each convolutional layer has a 5*5 Sobel convolutional kernel.

[0102] The output module is used to output cargo characteristic information;

[0103] The identification process of the first feature acquisition model for the N City urban sorting station is as follows:

[0104] On that day, all goods at the N City sorting station passed through the input module, resulting in input data H1.

[0105] Input data H1 is input into the shape feature recognition module of the city sorting station in N city to analyze the shape feature information of the goods and obtain the shape feature of the goods H2.

[0106] Input data H2 is fed into the size feature recognition module of the city sorting station in N city to analyze the size feature information of goods and obtain the size feature of goods H3.

[0107] The cargo shape feature H2 and cargo size feature H3 are input to the output module to obtain cargo feature information H4;

[0108] Image recognition technology is used to obtain the shape outline of each item at the city sorting station in N city. The shape features are obtained based on the shape outline, and the size of the item is calculated based on the key points of the outline. Since truck volume is prioritized, the size of the item also determines its transportation priority level. Larger items need to be put in in advance, while smaller items can be put in after the larger items are put in. Some non-cubic or cubic items are also classified and sorted to allow trucks to carry as many items as possible within the specified requirements, while also reducing the workload of transportation personnel.

[0109] The second feature acquisition model for goods at the N City urban sorting station is used to automatically acquire the cargo transportation feature information of goods at the N City urban sorting station;

[0110] Reference Figure 3 As shown, the cargo second feature acquisition model for the N City urban sorting station includes an input module, a cargo second feature recognition module, and an output module;

[0111] The input module is used to input cargo information to facilitate the cargo transportation feature information of the cargo second feature acquisition model at the N City urban sorting station;

[0112] The cargo second feature recognition module is used to acquire and analyze the second feature of the cargo. The second feature recognition module includes a second feature encoder, a second feature decoder, and a second feature recognizer. The second feature encoder includes four encoding layers, each encoding layer includes two convolutional layers. The first convolutional layer includes a 3*3 convolutional layer, and the second convolutional layer includes a 1*1 convolutional layer. The feature recognizer includes a Transform model.

[0113] The output module is used to output the cargo transportation characteristic information of the goods;

[0114] The identification process of the cargo second feature acquisition model is as follows:

[0115] The input data G1 is obtained from the module that inputs all goods that need to be delivered on the same day.

[0116] Input data G1 is input into the second feature encoder of the cargo second feature recognition module to analyze the second feature of the cargo and obtain feature information G2;

[0117] Feature information G2 is input into the second feature decoder for feature reconstruction to obtain feature information G3;

[0118] Feature information G3 is input to the second feature recognizer to identify features and obtain feature information G4.

[0119] The feature information G4 input / output module is used to output the cargo transportation feature information of all goods that need to be delivered on the same day;

[0120] In addition to the characteristics of the goods themselves, other information about the object is also scanned, such as the receiving point of the delivery note, expedited orders, and characteristics of flammable, explosive, and fragile items. By scanning this textual information of the goods, the textual features are extracted and identified through the second feature acquisition model of the goods at the N City sorting station. The recognition results are used to obtain another layer of priority feature information of the goods. The priority level of the goods is considered more comprehensively based on the textual feature information and the characteristics of the goods themselves.

[0121] The cargo priority ranking module will input the cargo characteristic information and cargo transportation characteristic information of all goods that need to be delivered to the city sorting station of N city on the same day into the cargo priority calculation model of the cargo priority module of the city sorting station of N city. The priority order of all goods that need to be delivered on the same day will be calculated and sorted through the cargo priority calculation model.

[0122] The cargo priority ranking of the cargo priority ranking module at the N City urban sorting station is obtained through a cargo priority calculation model.

[0123] The cargo priority calculation model includes a feature information input module, a priority calculation module, and a sorting module;

[0124] After the cargo feature information is input into the module, the priority of the cargo is calculated by the priority calculation module, and the cargo is sorted by the sorting module based on the priority results.

[0125] The feature information input module is used to input the cargo feature information and cargo transportation feature information;

[0126] The priority calculation module is used to determine the priority level of goods based on their characteristics and transportation characteristics using a priority calculation formula.

[0127] The priority calculation formula is as follows:

[0128]

[0129] Where GDOP represents the priority level of the goods, and GX represents the priority level of the goods. i For cargo characteristic information, λ i GY represents the information weight of cargo feature information, m represents the total number of cargo feature information, and GY represents the information weight of cargo feature information. j For cargo transportation characteristic information, λ j represents the information weight of cargo transportation feature information, and n is the total number of cargo transportation feature information;

[0130] The sorting module is used to sort goods according to their priority level.

[0131] This table only shows some key characteristics of the goods. The sorting priority of some goods at the N City sorting station is shown in Table 1:

[0132] Table 1. Ranking of Priority for Selected Goods at City Sorting Stations in City N

[0133]

[0134] Goods XXXXXXXX23 are large rectangular goods containing batteries and require expedited delivery; Goods XXXXXXXX46 are small to medium-sized cubic goods requiring expedited delivery only; Goods XXXXXXXX84 are medium-sized rectangular goods containing fresh produce and have a limited shelf life. Based on the calculated priority of each characteristic, Goods XXXXXXXX84 has the highest priority (1), followed by Goods XXXXXXXX23 (priority 2), and then Goods XXXXXXXX46 (priority 3). After calculating the priority of each good based on its characteristics, Goods XXXXXXXX84 is sent first, followed by Goods XXXXXXXX23, and finally Goods XXXXXXXX46. By comprehensively considering the textual characteristics and shape information of each good, and using the priority calculation module of the goods priority sorting module at the N City sorting station, the priority of each good is calculated and sorted according to the calculation results. This comprehensive approach considers the transportation of each good, ensuring efficient transportation without affecting the cost of the N City sorting station.

[0135] As shown in step two, the sorting point priority ranking module obtains information on each sorting point in N City based on the cargo transportation characteristics information of the goods at the city sorting station in N City. Based on the sorting point information in N City, the module obtains the first sorting characteristic information and the second sorting characteristic information of each sorting point in N City. The first sorting characteristic information and the second sorting characteristic information of each sorting point in N City are input into the sorting point priority calculation model of the sorting point priority ranking module in N City to calculate the priority order of each sorting point in N City and sort them.

[0136] The first sorting characteristic information for each sorting point in City N is distance characteristic information, and the second sorting characteristic information for each sorting point in City N is cargo transportation volume characteristic information; the distance characteristic information and cargo transportation volume characteristic information are used to calculate the priority of each sorting point in City N;

[0137] The N City sorting point priority module calculates the priority level of each sorting point in N City using a sorting point priority calculation model, and sorts the sorting points in N City according to the calculated priority level.

[0138] The sorting point priority calculation model includes a sorting point priority analysis module and a sorting point priority calculation module;

[0139] The sorting point priority analysis module is used to analyze distance characteristic information and cargo transportation volume characteristic information;

[0140] Reference Figure 4 As shown, the sorting point priority analysis module includes a sorting feature information input module, a sorting feature information prediction module, and a sorting feature information evaluation module;

[0141] The feature information input module is used to input distance feature information and cargo transportation volume feature information;

[0142] The sorting feature information prediction module is used to acquire historical data of distance feature information and cargo transportation volume feature information, and predict the first evaluation data of multiple distance feature information and cargo transportation volume feature information based on the historical data. The sorting feature information prediction module includes three long short-term memory neural networks, and each long short-term memory neural network includes four long short-term memory neural network units.

[0143] The sorting feature information evaluation module is used to calculate the first evaluation data and output the second evaluation data.

[0144] After the distance feature information and cargo transportation volume feature information are input into the module, the first evaluation data is obtained through the sorting feature information prediction module, and multiple reference evaluation data are input into the sorting feature information evaluation module to obtain the second evaluation data.

[0145] The sorting point priority calculation module is used to calculate the analysis results of the sorting point priority analysis module and obtain the priority level of each sorting point in N City.

[0146] Since goods need to be sent to sorting points in various regions for further distribution, such as the sorting point in a district of city N or the sorting point in a street of city N, the process of transporting goods from the city sorting station in city N to each sorting point will directly or indirectly affect transportation costs and efficiency. Therefore, the sorting points to which goods need to be transported are included in the priority consideration factors. By analyzing the factors that directly or indirectly affect the cost and transportation efficiency of each sorting point, the sorting points are prioritized to further reduce costs and improve efficiency. Rational sorting and transportation of goods can also reduce the pressure on each sorting point and avoid the accumulation of goods.

[0147] The sorting point priority ranking module includes a sorting point priority calculation module:

[0148] The sorting point priority calculation module uses the sorting point priority calculation formula to determine the priority level of each sorting point in City N.

[0149] The formula for calculating the priority of sorting points is as follows:

[0150]

[0151] Where WDOP represents the priority of the sorting point in City N, and DIS... t For distance feature information, k is the number of distance feature information, GIO c For cargo transportation volume characteristic information, λ c is the weight of the freight transport volume feature information, and l is the number of freight transport volume feature information;

[0152] The table below only shows some key feature information. The sorting priority of each sorting point in City N is shown in Table 2:

[0153] Table 2. Priority Ranking of Sorting Points in City N

[0154]

[0155] The sorting point in City X is 10.3km from the city sorting station in City N. There is no historical congestion. The assessed throughput for this sorting point is high, and the assessed cargo capacity is medium. The sorting point in City Y is 19.7km from the city sorting station in City N. There is historical congestion. The assessed throughput for this sorting point is medium, and the assessed cargo capacity is high. The sorting point in Street Z is 28.4km from the city sorting station in City N. There is no historical congestion. The assessed throughput for this sorting point is low, and the assessed cargo capacity is high. According to the sorting point priority calculation module, the priority of the sorting point in City X is 1 (the highest priority), and the priority of the sorting point in Street Z is 2. The first priority is 3, which is the second priority. The second priority is 3, which is the third priority. According to the sorting results, the goods are transported in the order of X, Z and Y. The distance of each sorting point from the city sorting station in city N is different. The transportation mode and traffic conditions are different. The cargo throughput of each sorting point is also different. These comprehensive factors will indirectly affect the transportation cost and transportation efficiency. Therefore, based on the cargo of the city sorting station in city N, the sorting points in city N need to be transported. According to the transportation distance, transportation mode and cargo throughput of each sorting point in city N, the priority level of the city is obtained through the sorting point priority calculation formula to improve the accuracy of sorting each sorting point.

[0156] As shown in step three, the sorting management comprehensive calculation module obtains the first logistics characteristic information of the city sorting station in N City. It inputs the first logistics characteristic information, the priority order of all goods to be delivered on the same day, and the priority order of each sorting point into the sorting management comprehensive calculation model of the city sorting management comprehensive calculation module in N City to obtain the optimal sorting plan. The priority ranking level of each goods in N City and the ranking level of each sorting point are calculated by the N City goods priority ranking module and the N City sorting point priority ranking module. Combined with the number of personnel and transport trucks that the city sorting station in N City can transport at present, multiple sorting plans are calculated through the above comprehensive factors. The optimal sorting plan is selected from the plans. Transporting goods according to the optimal sorting plan can effectively ensure the efficiency of goods transportation. At the same time, it can maximize the utilization of the current transportation resources of the city sorting station in N City and reduce costs.

[0157] The primary logistics characteristic information of the current city sorting station is the transportation resource characteristic information of City N, such as the number of transport personnel or transport vehicles;

[0158] Transportation resource characteristics, priority of all goods to be delivered to City N on the same day, and priority of each sorting point are used to calculate the optimal logistics sorting plan for City N.

[0159] The N City sorting management integrated calculation module includes a sorting management integrated calculation model;

[0160] The comprehensive calculation model for sorting management includes a data receiving module, a first scheme generation module, and a second scheme evaluation module.

[0161] The first option is the initial option, and the second option is the final option;

[0162] The data receiving module is used to receive the transportation resource characteristics, sorting point priority level, and cargo priority level of the current city sorting station;

[0163] The first scheme generation module is used to generate the initial scheme for City N based on the data obtained by the receiving module.

[0164] The initial calculation formula for the N City plan is as follows:

[0165]

[0166] Where PLANSCH is the initial scheme for city N generated, GP ig For goods that need to be transported on the same day, WDOP ig GDOP prioritizes transportation sorting points. ig For cargo priority, g represents the quantity of cargo to be transported that day, TRANS n This represents the characteristic information of transportation resources; num represents the number of characteristic information items of transportation resources.

[0167] The second scheme evaluation module is used to evaluate the content of the initial scheme for City N and derive the final scheme for City N.

[0168] The formula for calculating the final solution evaluation module is as follows:

[0169]

[0170] Where P is the total cost calculation result of the scheme, θ RES Here, NRES represents the cost weight of the first scheme, and PLANSCH represents the total cost calculation. RES The first scheme generated;

[0171]

[0172] Where PLAN is the final plan for City N, P1, P2, ..., P n For the initial plan for City N after cost calculation, ω TRESAssigning time cost weights to the initial plan for City N, PLANSCH TRES Let n be the initial plan for city N, and n be the number of plans.

[0173] This table only shows partial data. The optimal sorting scheme for City N is shown in Table 3.

[0174] Table 3 Selection of the Optimal Sorting Scheme in City N

[0175] plan Cost (ten thousand yuan / day) Transportation time is less than average time Priority Option A1 130 Greater than Not considering Option A2 156 Less than 2 Option A3 143 Less than 1

[0176] The first solution generation module generated several solutions, including solutions A1, A2, and A3. The estimated cost of solution A1 was 1.3 million yuan, that of solution A2 was 1.56 million yuan, and that of solution A3 was 1.43 million yuan. From a cost perspective, solution A1 was the most cost-effective. However, considering the transportation efficiency of the goods, the estimated transportation time for each solution also needed to be calculated. Based on the transportation time of all solutions, the average transportation time of each solution was calculated. The transportation time of solution A1 was longer than the average time, that of solution A2 was shorter than the average time, and that of solution A3 was shorter than the average time. After comprehensive consideration, solution A3 was the most effective in terms of both cost and transportation efficiency, and was therefore the optimal sorting solution. Goods were then dispatched to the N City transfer station based on solution A3.

[0177] The solution calculation is based on the current transportation resources in City N, the priority of all goods in City N, and the priority of the sorting points that need to transport all goods in City N. Multiple sorting solutions are calculated, taking cost factors into account. The total cost is calculated based on the transportation cost required for each solution, and the solutions are ranked from lowest to highest total cost. Considering that lower-cost solutions may incur higher transportation time and affect transportation efficiency, the average transportation time for all solutions is calculated. The solution with the lowest cost and transportation time less than the average transportation time is identified as the optimal solution. This also ensures that each sorting point can receive and distribute goods in a timely manner. This solution guarantees lower costs and faster transportation speed, while also improving the work efficiency of staff at the City N sorting stations and reducing the workload of sorting goods.

[0178] Example 2

[0179] As a second- or third-tier city, City C has become a logistics hub for various regions due to the development of logistics in recent years and its geographical location. It also has its own urban transit station. However, City C's road network is not as extensive as that of first-tier cities, and its workforce is insufficient. With the increase in online shopping activities in recent years, the amount of goods transported has increased. Whether it is seafood, fruit, and meat that require cold chain transportation, or some large items such as refrigerators, color TVs, and washing machines, due to the city's economic conditions and manpower issues, it is necessary to minimize transportation costs. This makes it necessary for City C's transit station to allocate transportation more scientifically. By scanning the information of these goods and combining it with image recognition technology, priority information can be obtained to facilitate subsequent priority calculations.

[0180] The cargo information acquisition module at the City C sorting station obtains the first and second characteristic information of all cargoes that need to be delivered that day through a cargo information acquisition model. The cargo information acquisition model includes a first characteristic acquisition model and a second characteristic acquisition model for cargoes at the City C sorting station. The first characteristic information is the cargo characteristic information of City C, and the second characteristic information is the cargo transportation characteristic information of City C. Different models are used for feature analysis based on their different information characteristics. Compared with the same model, different models can obtain more accurate analysis results based on different information characteristics. Therefore, two models are used to analyze the two types of characteristic information separately, and different results are output based on the analysis of different characteristic information by the models, so as to facilitate the subsequent calculation of cargo priority.

[0181] The first feature acquisition model for goods at the City C sorting station is used to automatically acquire the feature information of all goods that need to be delivered to the City C sorting station on the same day;

[0182] The cargo first feature acquisition model for the City C sorting station includes an input module, a cargo first feature recognition module, and an output module. The recognition process of the City C sorting station first feature acquisition model is as follows:

[0183] On that day, all goods at the City C sorting station passed through the input module, resulting in input data H1.

[0184] Input data H1 is input into the shape feature recognition module of the city sorting station in City C to analyze the shape feature information of the goods and obtain the shape feature of the goods H2.

[0185] Input data H2 is fed into the size feature recognition module of the city sorting station in City C to analyze the size feature information of the goods and obtain the size feature of the goods H3.

[0186] The cargo shape feature H2 and cargo size feature H3 are input to the output module to obtain cargo feature information H4;

[0187] The shape and outline of goods include the shape features of the goods themselves, which affect the priority level of goods transportation. By using image recognition technology to obtain the shape and outline of each goods at the city sorting station in City C, the outline features of the goods are automatically obtained, and the priority of goods transportation is determined according to their shape and size, so that the goods can be loaded into trucks under the specified requirements, maximizing the load capacity of trucks, reducing the transportation time of trucks, and also reducing the workload of transportation personnel in City C.

[0188] The second feature acquisition model for goods at the City C sorting station is used to automatically acquire the cargo transportation feature information of goods at the City C sorting station;

[0189] The cargo second feature acquisition model for the city sorting station in City C includes an input module, a cargo second feature recognition module, and an output module; the recognition process of the cargo second feature acquisition model is as follows:

[0190] The input data G1 is obtained from the module that inputs all goods that need to be delivered on the same day.

[0191] Input data G1 is input into the second feature encoder of the cargo second feature recognition module to analyze the second feature of the cargo and obtain feature information G2;

[0192] Feature information G2 is input into the second feature decoder for feature reconstruction to obtain feature information G3;

[0193] Feature information G3 is input to the second feature recognizer to identify features and obtain feature information G4.

[0194] The feature information G4 input / output module is used to output the cargo transportation feature information of all goods that need to be delivered on the same day;

[0195] The textual features of goods are extracted by the second feature acquisition model of the city sorting station in City C. These textual features also contain a large number of factors that determine the priority level of goods. These factors are identified by the second feature acquisition model to obtain another layer of priority feature information of the goods. The priority level of goods is considered more comprehensively based on the textual feature information and the feature information of the goods themselves.

[0196] The cargo priority ranking module will input the cargo characteristic information and cargo transportation characteristic information of all goods that need to be delivered to the city sorting station of City C on the same day into the cargo priority calculation model of the cargo priority module of City C sorting station. The priority order of all goods that need to be delivered on the same day will be calculated and sorted through the cargo priority calculation model.

[0197] The cargo priority ranking module of the city sorting station in City C obtains the cargo priority ranking through a cargo priority calculation model;

[0198] The cargo priority calculation model includes a feature information input module, a priority calculation module, and a sorting module;

[0199] After the cargo feature information is input into the module, the priority of the cargo is calculated by the priority calculation module, and the cargo is sorted by the sorting module based on the priority results.

[0200] The feature information input module is used to input the cargo feature information and cargo transportation feature information;

[0201] The priority calculation module is used to determine the priority level of goods based on their characteristics and transportation characteristics using a priority calculation formula.

[0202] The priority calculation formula is as follows:

[0203]

[0204] Where GDOP represents the priority level of the goods, and GX represents the priority level of the goods. i For cargo characteristic information, λ i GY represents the information weight of cargo feature information, m represents the total number of cargo feature information, and GY represents the information weight of cargo feature information. j For cargo transportation characteristic information, λ j represents the information weight of cargo transportation feature information, and n is the total number of cargo transportation feature information;

[0205] The sorting module is used to sort goods according to their priority level.

[0206] This table only shows some key characteristics of the goods. The sorting priority of some goods at the City C sorting station is shown in Table 4:

[0207] Table 4. Ranking of Priority for Selected Goods at City C's Urban Sorting Station

[0208]

[0209] Goods XXXXXXXX14 is a large rectangular cargo, including fragile items; Goods XXXXXXXX32 is a small rectangular cargo, urgently needed and an important item; Goods XXXXXXXX70 is a medium-sized rectangular cargo with no additional transport characteristics. Based on the calculated priorities, Goods XXXXXXXX32 has the highest priority (1), followed by Goods XXXXXXXX14 (priority 2), then Goods XXXXXXXX70 (priority 3). After calculating priorities based on their characteristics, Goods XXXXXXXX32 is sent first, followed by Goods XXXXXXXX14, and finally Goods XXXXXXXX70. According to the priority ranking module of the City C sorting station, the textual characteristics and shape information of each cargo are input into the priority ranking module to calculate the priority of each cargo and rank them accordingly. This reduces the cost of the City C sorting station while ensuring timely transport and maintaining high efficiency, while also improving the efficiency of sorting and staff work.

[0210] As shown in step two, the sorting point priority ranking module obtains information on each sorting point in City C based on the cargo transportation characteristics information of the goods at the City C sorting station. Based on the sorting point information in City C, the module obtains the first sorting characteristic information and the second sorting characteristic information of each sorting point in City C. The first sorting characteristic information and the second sorting characteristic information of each sorting point in City C are input into the sorting point priority calculation model of the City C sorting point priority ranking module to calculate the priority order of each sorting point in City C and sort them.

[0211] The first sorting characteristic information for each sorting point in City C is distance characteristic information, and the second sorting characteristic information for each sorting point in City C is cargo transportation volume characteristic information; the distance characteristic information and cargo transportation volume characteristic information are used to calculate the priority of each sorting point in City C;

[0212] The sorting point priority module in City C calculates the priority level of each sorting point in City C using a sorting point priority calculation model, and sorts the sorting points in City C according to the calculated priority level.

[0213] The sorting point priority calculation model includes a sorting point priority analysis module and a sorting point priority calculation module;

[0214] The sorting point priority analysis module is used to analyze distance feature information and cargo transportation volume feature information; the sorting point priority analysis module includes a sorting feature information input module, a sorting feature information prediction module, and a sorting feature information evaluation module; the feature information input module is used to input distance feature information and cargo transportation volume feature information;

[0215] The sorting feature information prediction module is used to acquire historical data of distance feature information and cargo transportation volume feature information, and predict the first evaluation data of multiple distance feature information and cargo transportation volume feature information based on the historical data. The sorting feature information prediction module includes three long short-term memory neural networks, and each long short-term memory neural network includes four long short-term memory neural network units.

[0216] The sorting feature information evaluation module is used to calculate the first evaluation data and output the second evaluation data; after the distance feature information and cargo transportation volume feature information input modules are used, the first evaluation data is obtained through the sorting feature information prediction module, and multiple reference evaluation data are input into the sorting feature information evaluation module to obtain the second evaluation data; the sorting point priority calculation module is used to calculate the analysis results of the sorting point priority analysis module to obtain the priority level of each sorting point in City C;

[0217] Due to the limited number of sorting personnel and less favorable road transportation conditions compared to first-tier cities, and the need for goods to be sent to sorting points in various regions for further distribution, a significant amount of time is required during transportation. Therefore, the characteristics of sorting points are taken into consideration, such as the distance between the sorting points and the city's sorting station, and the cargo storage capacity of each sorting point in City C. By analyzing the factors that directly or indirectly affect the cost and transportation efficiency of each sorting point, the sorting points are prioritized to improve the transportation efficiency of the sorting station in City C.

[0218] The sorting point priority ranking module includes a sorting point priority calculation module:

[0219] The sorting point priority calculation module uses the sorting point priority calculation formula to determine the priority level of each sorting point in City C.

[0220] The formula for calculating the priority of sorting points is as follows:

[0221]

[0222] Wherein, WDOP represents the priority of the sorting point in City C, and DIS... t For distance feature information, k is the number of distance feature information, GIO c For cargo transportation volume characteristic information, λ c is the weight of the freight transport volume feature information, and l is the number of freight transport volume feature information;

[0223] The table below only shows some key feature information. The sorting priority of each sorting point in City C is shown in Table 5:

[0224] Table 5. Priority Ranking of Sorting Points in City C

[0225]

[0226] Sorting point F is 21.4km from the city sorting station in City C. There are no historical congestion issues. Its assessed throughput and cargo capacity are both medium. Sorting point G is 17.3km from the city sorting station in City C. There are no congestion issues. Its assessed throughput and cargo capacity are both low. Sorting point H is 25.6km from the city sorting station in City C. There are no historical congestion issues. Its assessed throughput and cargo capacity are both low. According to the sorting point priority calculation module, sorting point F has a priority of 1 (highest priority), sorting point G has a priority of 2 (second highest priority), and sorting point H... The priority of the picking point is 3, which is the third priority. According to the ranking results, the goods are transported in the order of F, G, and H. Based on the cargo transportation volume characteristics and distance characteristics of each picking point, including the assessment of throughput, assessment of cargo capacity, distance, and congestion, multiple factors are input into the priority ranking module of picking points in City C. By comprehensively considering the situation of each picking point in City C from multiple perspectives, the priority level of each picking point in City C is obtained and ranked according to the analysis and calculation results of comprehensive factors. This provides another perspective on the problem of insufficient manpower in City C, which causes transportation costs and efficiency issues, and facilitates the solution of subsequent problems.

[0227] As shown in step three, the sorting management comprehensive calculation module obtains the first logistics characteristic information of the city sorting station in City C. It then inputs this first logistics characteristic information, the priority order of all goods to be delivered that day, and the priority order of each sorting point into the sorting management comprehensive calculation model of the City C sorting management comprehensive calculation module to derive the optimal sorting plan. Due to the limited transportation personnel and costs in City C, the priority ranking of each goods and the ranking of each sorting point in City C are calculated using the City C goods priority ranking module and the City C sorting point priority ranking module. Multiple sorting plans are calculated based on these factors, and the optimal sorting plan is selected from these plans. This optimal sorting plan maximizes the utilization of the City C sorting station's transportation resources to transport goods, effectively ensuring high efficiency in goods transportation while minimizing transportation costs. It also reduces the workload of the City C sorting station staff, providing a better working environment.

[0228] The primary logistics characteristic information of the current city sorting station is the transportation resource characteristic information of City C, such as the number of transport personnel or transport vehicles;

[0229] Transportation resource characteristics, priority of all goods to be delivered to City C on the same day, and priority of each sorting point are used to calculate the optimal logistics sorting plan for City C.

[0230] The sorting management comprehensive calculation module for City C includes a comprehensive calculation model for sorting management;

[0231] The comprehensive calculation model for sorting management includes a data receiving module, a first scheme generation module, and a second scheme evaluation module.

[0232] The first option is the initial option, and the second option is the final option;

[0233] The data receiving module is used to receive the transportation resource characteristics, sorting point priority level, and cargo priority level of the current city sorting station;

[0234] The first scheme generation module is used to generate the initial scheme for City C based on the data obtained by the receiving module.

[0235] The initial calculation formula for City C is as follows:

[0236]

[0237] Among them, PLANSCH is the initial scheme for City C generated, and GP ig For goods that need to be transported on the same day, WDOP ig GDOP prioritizes transportation sorting points. ig For cargo priority, g represents the quantity of cargo to be transported that day, TRANS n This represents the characteristic information of transportation resources; num represents the number of characteristic information items of transportation resources.

[0238] The second scheme evaluation module is used to evaluate the content of the initial scheme for City C and derive the final scheme for City C.

[0239] The formula for calculating the final solution evaluation module is as follows:

[0240]

[0241] Where P is the total cost calculation result of the scheme, θ RES Here, NRES represents the cost weight of the first scheme, and PLANSCH represents the total cost calculation. RES The first scheme generated;

[0242]

[0243] Wherein, PLAN is the final plan for City C, P1, P2, ..., P n For the initial plan for City C after cost calculation, ω TRES Assigning time cost weights to the initial plan for City C, PLANSCH TRES Let n be the initial plan for City C, and n be the number of plans.

[0244] This table only shows partial data. The optimal sorting solution for City C is shown in Table 6.

[0245] Table 6 Selection of the Optimal Sorting Scheme in City C

[0246] plan Cost (ten thousand yuan / day) Transportation time is less than average time Priority Option A1 84 Less than 3 Option A2 79 Less than 1 Option A3 82 Less than 2

[0247] Option A1, Option A2, and Option A3 are several options generated by the first option generation module. The estimated cost of Option A1 is 840,000 yuan, the estimated cost of Option A2 is 790,000 yuan, and the estimated cost of Option A3 is 820,000 yuan. From a cost perspective, Option A2 has the best cost. The average transportation time of the options is calculated based on the transportation time of all options. The transportation time of Option A2 is also less than the average time of the options. After comprehensive consideration, Option A2 has the best effect in terms of both cost and transportation efficiency, and is the optimal sorting option. The goods are dispatched to the transfer station in City C according to Option A2.

[0248] By prioritizing all goods in City C and the sorting points that need to transport all goods in City C, multiple sorting schemes are derived based on the calculated priorities and current transportation resources. By analyzing the cost and transportation efficiency factors of each sorting scheme, the optimal sorting scheme is obtained after comprehensive calculation. Using this sorting scheme for transportation can not only reduce costs while ensuring efficient transportation, but also reduce the workload of transportation personnel. The rational sorting scheme can also ensure the sustainable operation of the City C sorting station and provide room for gradual growth in its future development.

[0249] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A logistics flow and sorting information management system, characterized in that, include: The cargo information acquisition module acquires the first and second characteristic information of all cargoes that need to be delivered on the same day. The cargo priority ranking module inputs the first feature information and the second feature information of all cargoes that need to be delivered on the same day into the cargo priority calculation model of the cargo priority module, and calculates and sorts the priority order of all cargoes that need to be delivered on the same day through the cargo priority calculation model. The sorting point priority ranking module obtains information about each sorting point in the transportation based on the second feature information of the goods. Based on the sorting point information, the module obtains the first sorting feature information and the second sorting feature information of each sorting point. The first sorting feature information and the second sorting feature information of each sorting point are input into the sorting point priority calculation model of the sorting point priority ranking module to calculate the priority order of each sorting point and sort them. The sorting management comprehensive calculation module obtains the first logistics characteristic information of the current city sorting station, and inputs the first logistics characteristic information, the priority order of all goods to be delivered on the same day, and the priority order of each sorting point into the sorting management comprehensive calculation model of the sorting management comprehensive calculation module to obtain the optimal sorting scheme. Goods are dispatched according to the optimal sorting scheme.

2. The logistics flow sorting information management system according to claim 1, characterized in that, include: The first characteristic information of all goods that need to be delivered on the same day is the goods characteristic information, and the second characteristic information of all goods that need to be delivered on the same day is the goods transportation characteristic information; the first characteristic information and the second characteristic information are used to calculate the priority of all goods that need to be delivered on the same day; The first sorting feature information of each sorting point is distance feature information, and the second sorting feature information of each sorting point is cargo transportation volume feature information; the first sorting feature information and the second sorting feature information are used to calculate the priority of each sorting point. The first logistics characteristic information of the current city sorting station is transportation resource characteristic information; The first logistics characteristic information, the priority of all the goods to be delivered on the same day, and the priority of each sorting point are used to calculate the optimal logistics sorting plan.

3. The logistics flow sorting information management system according to claim 1, characterized in that, The cargo information acquisition module includes: The cargo information acquisition module acquires the first feature information and the second feature information of all the cargoes that need to be delivered on the same day through the cargo information acquisition model; The cargo information acquisition model includes a cargo first feature acquisition model and a cargo second feature acquisition model; The cargo first feature acquisition model is used to automatically acquire the cargo first feature information; The cargo second feature acquisition model is used to automatically acquire the cargo second feature information.

4. The logistics flow sorting information management system according to claim 1, characterized in that, The cargo information acquisition module includes: The cargo information acquisition module includes a cargo first feature acquisition model; The cargo first feature acquisition model includes an input module, a cargo first feature recognition module, and an output module; The input module is used to input cargo information into the cargo first feature acquisition model to obtain the cargo's first feature information; The cargo first feature recognition module includes a shape feature recognition module and a size feature recognition module; The shape feature recognition module is used to identify the shape features of the goods. The shape feature recognition module includes three convolutional layers and one fully connected layer. Each convolutional layer is a 3*3 Sobel convolutional kernel. The size feature recognition module is used to identify the size features of goods. The size feature recognition module includes three sets of convolutional layers. Each set of convolutional layers includes four convolutional layers and one fully connected layer. Each convolutional layer is a 5*5 Sobel convolutional kernel. The output module is used to output the first feature information; The recognition process of the first feature acquisition model is as follows: The input module obtains input data H1 for all goods that need to be delivered on the same day. The input data H1 is input into the shape feature recognition module to analyze the cargo shape feature information and obtain cargo shape feature H2; The input data H2 is input into the size feature recognition module to analyze the cargo size feature information, and the cargo size feature H3 is obtained; The cargo shape feature H2 and the cargo size feature H3 are input to the output module to obtain the first feature information H4.

5. The logistics flow sorting information management system according to claim 1, characterized in that, The cargo information acquisition module includes: The cargo information acquisition module includes a cargo second feature acquisition model; The cargo second feature acquisition model includes an input module, a cargo second feature recognition module, and an output module; The input module is used to input cargo information to facilitate the cargo second feature acquisition model in acquiring the cargo's second feature information; The cargo second feature recognition module is used to acquire and analyze the second feature of the cargo. The second feature recognition module includes a second feature encoder, a second feature decoder, and a second feature recognizer. The second feature encoder includes four encoding layers, each of which includes two convolutional layers. The first convolutional layer includes a 3*3 convolutional layer, and the second convolutional layer includes a 1*1 convolutional layer. The feature recognizer includes a Transform model. The output module is used to output the second characteristic information of the goods; The identification process of the cargo second feature acquisition model is as follows: On the same day, all the aforementioned goods information input modules need to be delivered to obtain input data G1; The input data G1 is input into the second feature encoder of the cargo second feature recognition module to analyze the second feature of the cargo and obtain feature information G2. The feature information G2 is input into the second feature decoder for feature reconstruction to obtain feature information G3; The feature information G3 is input to the second feature recognizer to identify features and obtain feature information G4. The feature information G4 is input to the output module and is used to output the second feature information of all the goods that need to be delivered on the same day.

6. The logistics flow sorting information management system according to claim 1, characterized in that, The cargo priority ranking module includes: The cargo priority ranking of the cargo priority ranking module is obtained through the cargo priority calculation model; The cargo priority calculation model includes a feature information input module, a priority calculation module, and a sorting module; After the cargo feature information is input into the feature information input module, the priority of the cargo is calculated by the priority calculation module, and the cargo is sorted by the sorting module according to the priority result. The feature information input module is used to input the first feature information and the second feature information of the goods; The priority calculation module is used to determine the priority level of the goods based on the first feature information and the second feature information of the goods, using a priority calculation formula. The priority calculation formula is as follows: Wherein, GDOP represents the priority level of the goods, and GX represents the priority level of the goods. i For the first feature information, λ i GY represents the information weight of the first feature information, m represents the total number of the first feature information, and GY represents the information weight of the first feature information. j For the second feature information, λ j is the information weight of the second feature information, and n is the total number of the second feature information; The sorting module is used to sort the goods according to their priority level.

7. The logistics flow sorting information management system according to claim 1, characterized in that, The sorting point priority ranking module includes: The sorting point priority module calculates the priority level of each sorting point through the sorting point priority calculation model, and sorts each sorting point according to the calculated priority level. The sorting point priority calculation model includes a sorting point priority analysis module and a sorting point priority calculation module; The sorting point priority analysis module is used to analyze the first sorting feature information and the second sorting feature information; The sorting point priority analysis module includes a sorting feature information input module, a sorting feature information prediction module, and a sorting feature information evaluation module; The feature information input module is used to input the first sorting feature information and the second sorting feature information; The sorting feature information prediction module is used to acquire historical data of the first sorting feature information and the second sorting feature information, and predict first evaluation data of multiple first sorting feature information and second sorting feature information based on the historical data. The sorting feature information prediction module includes three long short-term memory neural networks, and each long short-term memory neural network includes four long short-term memory neural network units. The sorting feature information evaluation module is used to calculate the first evaluation data and output the second evaluation data. After the first sorting feature information and the second sorting feature information are input into the module, the first evaluation data is obtained through the sorting feature information prediction module, and multiple reference evaluation data are input into the sorting feature information evaluation module to obtain the second evaluation data; The sorting point priority calculation module is used to calculate the analysis results of the sorting point priority analysis module to obtain the priority level of each sorting point.

8. The logistics flow sorting information management system according to claim 1, characterized in that, The sorting point priority ranking module includes: The sorting point priority ranking module includes a sorting point priority calculation module: The sorting point priority calculation module calculates the priority level of each sorting point using the sorting point priority calculation formula. The formula for calculating the priority of sorting points is as follows: Wherein, WDOP is the sorting point priority, and DIS t The first sorting feature information is k, where k is the number of the first sorting feature information, and GIO is the number of sorting feature information. c For the second sorting feature information, λ c is the weight of the second sorting feature information, and l is the number of the second sorting feature information.

9. A logistics flow sorting information management system according to claim 1, characterized in that, The sorting management integrated calculation module includes: The sorting management integrated calculation module includes a sorting management integrated calculation model; The sorting management integrated calculation model includes a data receiving module, a first scheme generation module, and a second scheme evaluation module. The data receiving module is used to receive the first logistics characteristic information of the current city sorting station, the priority level of the sorting point, and the priority level of the goods; The first scheme generation module is used to generate a first scheme using the data obtained by the receiving module; The second scheme evaluation module is used to generate the content of the first scheme evaluation scheme to derive the second scheme.

10. A logistics flow sorting information management system according to claim 9, characterized in that, The generation of the first scheme and the evaluation of the second scheme include: The calculation formula for generating the first scheme is as follows: Where PLANSCH is the generated first scheme, GP ig For the goods that need to be transported on the same day, WDOP ig GDOP is the priority of the transport sorting point. ig Let g be the priority of the goods, g be the quantity of the goods to be transported that day, and TRANS be the priority of the goods. n The first logistics feature information; num is the quantity of the first logistics feature information; The calculation formula for the second scheme evaluation module is as follows: Where P is the total cost calculation result of the scheme, θ RES Here, NRES represents the cost weight of the first scheme, and PLANSCH represents the total cost calculation. RES The first scheme generated; Wherein, PLAN represents the second scheme, P1, P2, ..., P n For the first scheme after calculating the cost, ω TRES As the time cost weight of the first scheme, PLANSCH TRES Let n be the first scheme, and n be the number of schemes.

Images