Cold-chain logistics path optimization method and device and storage medium

Abstract

The invention provides a cold-chain logistics path optimization method and device and a storage medium, and belongs to the technical field of logistics paths. The method comprises the steps of collecting cold-chain logistics data, constructing an urban road network diagram according to urban traffic data, obtaining time required for a distribution vehicle to pass through a road section in different time periods based on the urban traffic data and the urban road network diagram, obtaining distribution time of the distribution vehicle in the road section through a time-varying network model, and obtaining the distribution time of the distribution vehicle in the road section through the time-varying network model. Constructing a cold-chain logistics path optimization model according to the cold-chain logistics data, determining a plurality of path nodes according to the actual distribution task, planning the plurality of path nodes through the time-varying network model and the cold-chain logistics path optimization model, and generating a plurality of distribution paths; and solving the distribution path through a particle swarm algorithm, and outputting an optimal distribution path. According to the method, the cold-chain logistics transportation path is dynamically adjusted based on the time-varying network environment, traffic condition changes in different time periods are fully considered, and the driving time and fuel consumption of vehicles are reduced.

Description

Technical Field

[0001] The present invention mainly relates to the technical field of logistics paths, and specifically to a cold chain logistics path optimization method, device and storage medium. Background Art

[0002] As the global climate change problem becomes increasingly serious, governments and enterprises around the world are paying more and more attention to the control and reduction of carbon emissions. The logistics industry, especially cold chain logistics, consumes huge amounts of energy due to its special temperature control requirements and is one of the main sources of carbon emissions. Most of the existing cold chain logistics path optimization methods focus on reducing transportation costs and improving distribution efficiency, but rarely consider the optimization of carbon emission costs.

[0003] Traditional logistics path planning methods are usually based on a static network environment, ignoring the dynamic changes in road traffic conditions. In actual applications, the traffic network is time-varying, and traffic flow and road conditions in different time periods will significantly affect the driving speed and fuel consumption of logistics vehicles. Therefore, path optimization methods that rely solely on a static network environment cannot accurately reflect the energy consumption and carbon emissions in the actual transportation process. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a cold chain logistics path optimization method, device and storage medium in view of the deficiencies in the prior art.

[0005] The technical solution of the present invention to solve the above technical problem is as follows: A cold chain logistics path optimization method comprises the following steps:

[0006] S1. Collect cold chain logistics data, including urban traffic data, delivery vehicle data, data on cold chain goods to be delivered, and carbon emission data;

[0007] S2. constructing a city road network diagram according to the city traffic data, obtaining the time required for a delivery vehicle to pass through a road segment at different time periods based on the city traffic data and the city road network diagram, dividing the required time into multiple time periods, setting a delivery task hypothesis according to the multiple time periods and the vehicle driving speeds corresponding to each time period and whether the delivery is across time periods, constructing the time-varying network model according to the delivery task hypothesis, and obtaining the delivery time of the delivery vehicle on the road segment through the time-varying network model;

[0008] S3, constructing a cold chain logistics path optimization model according to the delivery vehicle data, the cold chain goods data to be delivered, the carbon emission data and the delivery time;

[0009] S4. Determine multiple path nodes according to the actual delivery task, and plan the multiple path nodes through the time-varying network model and the cold chain logistics path optimization model to generate multiple delivery paths;

[0010] S5. The plurality of delivery paths are regarded as a plurality of particles, and the plurality of particles are solved by a particle swarm algorithm to output an optimal delivery path.

[0011] Another technical solution of the present invention to solve the above technical problem is as follows: a cold chain logistics path optimization device, comprising:

[0012] A data collection module is used to collect cold chain logistics data, wherein the cold chain logistics data includes urban traffic data, delivery vehicle data, data on cold chain goods to be delivered, and carbon emission data;

[0013] A task hypothesis module is used to construct an urban road network diagram according to the urban traffic data, obtain the time required for the delivery vehicle to pass through the road segment at different time periods based on the urban traffic data and the urban road network diagram, divide the required time into multiple time periods, set a delivery task hypothesis according to the multiple time periods and the vehicle driving speed corresponding to each time period and whether the delivery is across time periods, construct the time-varying network model according to the delivery task hypothesis, and obtain the delivery time of the delivery vehicle on the road segment through the time-varying network model;

[0014] A model building module, used to build a cold chain logistics path optimization model according to the delivery vehicle data, the cold chain goods to be delivered data, the carbon emission data and the delivery time;

[0015] A path generation module, used to determine multiple path nodes according to the actual delivery task, and plan the multiple path nodes through the time-varying network model and the cold chain logistics path optimization model to generate multiple delivery paths;

[0016] The path optimization module is used to take the multiple delivery paths as multiple particles, solve the multiple particles through a particle swarm algorithm, and output an optimal delivery path.

[0017] Another technical solution of the present invention to solve the above-mentioned technical problem is as follows: a cold chain logistics path optimization device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the cold chain logistics path optimization method as described above is implemented.

[0018] Another technical solution of the present invention to solve the above technical problem is as follows: a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the cold chain logistics path optimization method as described above is implemented.

[0019] The beneficial effects of the present invention are: based on the time-varying network environment, the cold chain logistics transportation route is dynamically adjusted, the changes in traffic conditions in different time periods are fully considered, the vehicle's driving time and fuel consumption are reduced, and the energy consumption and carbon emissions in the actual transportation process are accurately reflected. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A schematic diagram of a cold chain logistics path optimization method provided by an embodiment of the present invention;

[0021] Figure 2 A module block diagram of a cold chain logistics path optimization device provided by an embodiment of the present invention;

[0022] Figure 3 A schematic diagram of speed-time variation provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0024] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0025] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] like Figure 1 As shown, a cold chain logistics path optimization method includes the following steps:

[0027] S1. Collect cold chain logistics data, including urban traffic data, delivery vehicle data, data on cold chain goods to be delivered, and carbon emission data;

[0028] S2. constructing a city road network diagram according to the city traffic data, obtaining the time required for a delivery vehicle to pass through a road segment at different time periods based on the city traffic data and the city road network diagram, dividing the required time into multiple time periods, setting a delivery task hypothesis according to the multiple time periods and the vehicle driving speeds corresponding to each time period and whether the delivery is across time periods, constructing the time-varying network model according to the delivery task hypothesis, and obtaining the delivery time of the delivery vehicle on the road segment through the time-varying network model;

[0029] S3, constructing a cold chain logistics path optimization model according to the delivery vehicle data, the cold chain goods data to be delivered, the carbon emission data and the delivery time;

[0030] S4. Determine multiple path nodes according to the actual delivery task, and plan the multiple path nodes through the time-varying network model and the cold chain logistics path optimization model to generate multiple delivery paths;

[0031] S5. The plurality of delivery paths are regarded as a plurality of particles, and the plurality of particles are solved by a particle swarm algorithm to output an optimal delivery path.

[0032] In the above embodiments, based on the time-varying network environment, the cold chain logistics transportation route is dynamically adjusted, fully considering the changes in traffic conditions in different time periods, reducing the vehicle's driving time and fuel consumption, and accurately reflecting the energy consumption and carbon emissions in the actual transportation process.

[0033] Preferably, after the cold chain logistics data is collected, a step S10 of preprocessing the cold chain logistics data is further included, specifically:

[0034] The cold chain logistics data is cleaned to remove noise and outliers, and the cleaned cold chain logistics data is format converted according to a set data format.

[0035] In the above embodiment, cold chain logistics data that does not meet the requirements can be filtered out, making data analysis more accurate.

[0036] Preferably, in S2, an urban road network is constructed according to the urban traffic data, and the time required for a delivery vehicle to pass through a road segment at different time periods is obtained based on the urban traffic data and the urban road network diagram, and the required time is divided into multiple time periods, specifically:

[0037] Generate an urban road network graph through the road structure data in the urban traffic data, wherein the nodes in the urban road network graph represent intersections or logistics distribution centers, and the edges between two nodes represent road segments;

[0038] The time required for a vehicle to pass through a road segment at different time periods is determined according to the speed variation pattern caused by actual urban traffic congestion in the urban traffic data and the distance of the road segment, and the required time is divided into multiple time periods, and each time period corresponds to a different vehicle driving speed, and the vehicle driving speed is constant in each time period.

[0039] like Figure 3 As shown, in the above embodiment, for the time-varying network situation, the time (the time required for a vehicle to pass through the road section at different time periods) is divided into w time periods according to the speed variation law caused by actual urban traffic congestion. This can serve as a basis for analyzing whether cross-time distribution is necessary, ensuring that logistics transportation is both economical and efficient, and meeting customers' requirements for timeliness and cost.

[0040] Preferably, in S2, a delivery task hypothesis is set according to the multiple time periods and the vehicle driving speeds corresponding to each time period and whether the delivery is across time periods, and the time-varying network model is constructed according to the delivery task hypothesis. The delivery time of the delivery vehicle on the road segment is obtained through the time-varying network model, specifically:

[0041] Based on the different lengths of the edges between any two nodes, three delivery task assumptions are set according to the multiple time periods and the vehicle driving speeds corresponding to each time period during delivery, including completing the delivery task in a single time period, completing the delivery task across one time period, and completing the delivery task across two or more time periods, where:

[0042] The delivery task completed in a single time period is: the delivery task between two nodes in a single time period needs to satisfy the requirement that the distance traveled by vehicle k in the remaining time within a certain time period is greater than or equal to the actual distance between node i and node j. The delivery time of the road segment is determined based on the delivery task completed in a single time period as follows:

[0043]

[0044] Among them, t ij is the time required from node i to node j, d ij is the distance between node i and node j, v w is the driving speed in the wth time period;

[0045] The delivery task completed across a time period is: the delivery task between two nodes completed across a time period needs to satisfy that the distance traveled by vehicle k in the remaining time of a time period is less than the actual distance between node i and node j and the distance traveled by vehicle k in the next time period is greater than or equal to the remaining distance traveled between node i and node j. Based on the delivery task completed across a time period, the delivery time of the road segment is determined as follows:

[0046]

[0047] Among them, t ij is the time required from node i to node j, t i is the time when the vehicle departs, t w is the time t at which the vehicle departs i The time period, d ij is the distance between node i and node j, v w is the driving speed in the wth time period;

[0048] The delivery task completed across two or more time periods is: if the distance traveled by vehicle k in the remaining time of a time period is less than the actual distance between node i and node j and the distance traveled by vehicle k in the next time period is less than the remaining distance traveled between node i and node j, then vehicle k needs to cross a new time period again and repeat until the maximum distance traveled by vehicle k in the new time period is greater than or equal to the remaining distance traveled between node i and node j. Based on the delivery task completed across two or more time periods, the delivery time of the road segment is determined as follows:

[0049]

[0050] Among them, t ij is the time required from node i to node j, t i is the time when the vehicle departs, t w is the time point t at which the vehicle departs i The time period, d ij is the distance between node i and node j, v w is the driving speed in the wth time period.

[0051] In the above embodiments, while optimizing the path, the transportation route and time schedule are dynamically adjusted to fully consider the time-varying characteristics of the transportation network, reduce the vehicle's driving time and fuel consumption, and improve the overall transportation efficiency of cold chain logistics.

[0052] Preferably, in S3, a cold chain logistics path optimization model is constructed according to the delivery vehicle data, the cold chain goods data to be delivered, the carbon emission data and the delivery time, specifically:

[0053] A fixed cost calculation formula is established based on the number of delivery vehicles, vehicle purchase costs, daily maintenance costs, and driver wages in the delivery vehicle data. The fixed cost expression is:

[0054]

[0055] Where C0 is the fixed cost of the vehicle; xijk is a binary decision variable, taking values ​​of 0 or 1.

[0056] A transportation cost calculation formula is established based on the change in fuel consumption in the delivery vehicle data. The transportation cost expression is:

[0057]

[0058] Among them, E fue is the fuel consumption expression, E fue =E bas ·(1+a(v t -v bas ) 2 ), d ij is the distance from node i to j, P fue is the fuel price, x ijk is a binary decision variable, taking values ​​of 0 or 1.

[0059] Based on the change of the value of the goods in the cold chain goods data to be delivered, a cargo damage cost expression is established, and the cargo damage cost expression is:

[0060]

[0061] Among them, λ is the quality decay rate, p0 is the unit price of cold chain goods, q i is the demand of customer point i, t ij is the time required to go from node i to j, y ik is a binary decision variable, taking values ​​of 0 or 1.

[0062] Establish a time window for customers to receive service based on the delivery time and pickup appointment time in the cold chain goods data to be delivered. t ,l t ], and based on the time window [e t ,l t ] Establish a time window penalty cost expression, the time window penalty cost expression is:

[0063]

[0064] Among them, f1 is the waiting cost per unit time when the vehicle arrives early; f2 is the delay cost per unit time when the vehicle arrives late; [e t ,l t ] is the customer's expected delivery time window; [E t ,L t ] is the delivery time acceptable to customers.

[0065] Based on the driving carbon emission data and the refrigeration carbon emission data in the carbon emission data, a driving carbon emission cost expression and a refrigeration carbon emission cost expression are respectively constructed. The driving carbon emission cost expression is:

[0066]

[0067] Among them, d ij is the distance from node i to j; E fue For vehicles in v t speed; c is the carbon emission cost per kilogram; β is the carbon emission generated per unit fuel consumption; x ijk is a binary decision variable, taking values ​​of 0 or 1.

[0068] The refrigeration carbon emission cost expression is:

[0069]

[0070] Among them, α is the carbon emission per unit cooling time; t ij is the time required from node i to node j; c is the cost per kilogram of carbon emissions; x ijk is a binary decision variable, taking values ​​of 0 or 1.

[0071] In the above embodiment, the introduction of carbon emission costs and time-varying network factors into path optimization not only achieves environmentally friendly and efficient cold chain logistics management, but also improves the dynamic adaptability and service quality of the logistics system.

[0072] Preferably, in S5, the multiple delivery paths are used as multiple particles, and the multiple particles are solved by a particle swarm algorithm to output an optimal delivery path, specifically:

[0073] The plurality of delivery paths are regarded as a plurality of particles, and a chaotic initialization process is performed on the plurality of particles through a Logistic chaos model;

[0074] Perform adaptive inertia weight processing on the initialized particles to obtain a particle swarm carrying an inertia weight factor;

[0075] A first learning factor and a second learning factor are set, wherein the first learning factor represents the component learned by the particle from its own experience, and the second learning factor represents the degree to which the particle learns toward the global optimal position. The first learning factor and the second learning factor are used to search for the optimal value from a particle group carrying an inertia weight factor, thereby outputting an optimal delivery path according to the optimal value.

[0076] In the above embodiment, the particle swarm algorithm can quickly find the optimal delivery path from multiple delivery paths.

[0077] Specifically, the improved particle swarm algorithm is introduced in detail below.

[0078] The particle swarm optimization algorithm has the advantages of strong global search capability, dynamic adjustment strategy, applicability to multi-objective optimization, simple implementation and strong adaptability when dealing with logistics path optimization problems. It can effectively solve optimization problems in the field of logistics. The values ​​of parameters such as inertia weight and learning factor in the particle swarm optimization algorithm play a very important role in the convergence performance of the algorithm. However, researchers mostly use conventional values ​​for these parameters, which will inevitably affect the convergence and convergence speed of the algorithm. Similarly, initialization also has a great impact on its accuracy and speed. The uneven distribution of the solution space often falls into the local optimum.

[0079] First, chaos initialization is performed.

[0080] Traditional particle swarm algorithms generally use random initialization to determine the position distribution of the initial population. A random number is generated by a computer, and then the initial position of each particle is randomly generated according to the following formula.

[0081] Positions=rand*(ub-lb)+lb

[0082] Among them, Positions represents the generated particle position; rand represents the generated random number, and its value range is [0, 1]; ub and lb are the upper and lower bounds of the solution space respectively.

[0083] Usually, random initialization can generate different initial populations each time, which is convenient to use. However, this method has a disadvantage, that is, the initial particles are unevenly distributed in the solution space, resulting in particles in some areas being too dense and particles in other areas being too sparse. This situation is very unfavorable for the early convergence of the optimization algorithm, especially for the swarm optimization algorithm that is prone to fall into the local optimum, which may cause the convergence speed to decrease or even fail to converge.

[0084] In contrast, chaos initialization can effectively avoid these problems. Chaos initialization has the characteristics of randomness, ergodicity and regularity. It continuously traverses the search space within a certain range according to its own rules, so the generated initial population has obvious improvements in solution accuracy and convergence speed.

[0085] In this embodiment, the Logistic chaos model is used for initialization. The formula of the Logistic chaos model is as follows:

[0086] x k+1 =λ*x k (1-x k ),λ∈(0,4),

[0087] Where λ is the chaotic model mapping parameter. If λ∈[3.57,4], the system is in a chaotic state. If λ=4, the system is in a completely chaotic state. Generally, λ is taken as 4.

[0088] Next, adaptive inertia weight processing is performed.

[0089] The inertia weight factor has a great impact on the performance of PSO. When the value of the inertia weight factor is large, it is conducive to global search and helps the algorithm find the potential optimal solution. On the contrary, when the value of the inertia weight factor is small, it can speed up the convergence of the algorithm and promote local meticulous optimization, which is of great help in finding a more accurate optimal solution.

[0090] In order to balance the global search capability and local improvement capability of PSO, researchers proposed a nonlinear dynamic inertia weight coefficient formula. This formula can dynamically adjust the value of the inertia weight according to the current state of the algorithm, so that the algorithm can better balance the needs of global exploration and local optimization at different stages.

[0091] Specifically, the nonlinear dynamic inertia weight coefficient formula can adjust the size of the inertia weight according to the number of iterations of the algorithm or other state variables to adapt to the optimization requirements of different stages. Such a design can effectively improve the performance and convergence speed of PSO, making it more suitable for different types of optimization problems. Its expression is:

[0092]

[0093] Among them, f represents the real-time objective function value of the particle, f avg and f min They represent the average value and minimum target value of all particles at present. As can be seen from the above formula, the inertia weight changes with the change of the particle objective function value. When the particle target values ​​are dispersed, the inertia weight is reduced; when the particle target values ​​are consistent, the inertia weight is increased.

[0094] Next, dynamic learning factor processing is performed.

[0095] In the standard particle swarm algorithm, the learning factors c1 and c2 are key parameters, which usually represent the characteristics of individual cognition and social learning of particles in the particle swarm algorithm. Among them, the learning factor c1 represents the particle's attention to its own historical optimal position, that is, the component of the particle's learning from its own experience; while the learning factor c2 reflects the degree to which the particle learns from the global optimal position.

[0096] During the execution of the particle swarm algorithm, in order to achieve extensive spatial search and increase the diversity of particles in the initial stage, a larger c1 value and a smaller c2 value should be set so that particles are more inclined to search in local areas, which helps to explore the overall search space.

[0097] However, as the algorithm progresses, in order to enable particles to converge to the global optimal solution quickly and accurately, the learning factor value strategy needs to be adjusted. Therefore, usually in the later stages of the algorithm, c1 should be set to a smaller value and c2 should be set to a larger value to encourage particles to move more actively to the global optimal position, speed up the convergence speed of the algorithm and improve the accuracy.

[0098] In order to implement this dynamic adjustment strategy, c1 is set as a monotonically decreasing function, and c2 is set as a monotonically increasing function. This design allows the learning factor to change with the number of iterations of the algorithm and can be adaptively adjusted at different stages, thereby more effectively balancing the exploration and utilization capabilities of the algorithm and improving the performance and convergence effect of the particle swarm algorithm.

[0099]

[0100] Where t is the current iteration number. max is the maximum number of iterations of the particle swarm. From the above, we can see that by dynamically adjusting the value of the learning factor, the group can quickly search for the optimal value in a short time in the early stage of evolution, and can quickly and accurately converge to the optimal solution in the later stage of evolution.

[0101] The optimization algorithm process is:

[0102] Initialization: Initialize the path planning parameters according to the starting point and the end point.

[0103] Path search: Perform path search in a time-varying network graph and calculate the comprehensive cost of each path.

[0104] Dynamic adjustment: Dynamically adjust path selection based on real-time traffic conditions and carbon emissions to select the path with the lowest overall cost.

[0105] The logistics path of the present invention is simulated and optimized below.

[0106] Simulation verification:

[0107] Simulation software is used to simulate and verify the optimized route and analyze the improvements in carbon emissions and transportation efficiency.

[0108] The effects of different path planning schemes are compared to verify the superiority of the present invention.

[0109] Optimization adjustments:

[0110] According to the simulation results, the path planning algorithm and model parameters are further optimized to improve the applicability and effect in practical applications.

[0111] The method of the present invention can be integrated into system implementation and application.

[0112] System Development:

[0113] Embed optimization algorithms and models into cold chain logistics management systems and develop corresponding software tools.

[0114] Provides a friendly user interface for easy operation and real-time monitoring.

[0115] Practical Application:

[0116] The method of the present invention is applied in actual cold chain logistics operations to verify and adjust the optimization plan.

[0117] Through actual data analysis, the effectiveness of the present invention in reducing carbon emissions and improving transportation efficiency is evaluated.

[0118] Through the above embodiments, the present invention realizes a path optimization method that comprehensively considers carbon emission costs and special needs of cold chain logistics in a time-varying network environment, thereby significantly improving the environmental protection and operational efficiency of cold chain logistics.

[0119] like Figure 2 As shown, the present invention also provides a cold chain logistics path optimization device, comprising:

[0120] A data collection module is used to collect cold chain logistics data, wherein the cold chain logistics data includes urban traffic data, delivery vehicle data, data on cold chain goods to be delivered, and carbon emission data;

[0121] A task hypothesis module is used to construct an urban road network diagram according to the urban traffic data, obtain the time required for the delivery vehicle to pass through the road segment at different time periods based on the urban traffic data and the urban road network diagram, divide the required time into multiple time periods, set a delivery task hypothesis according to the multiple time periods and the vehicle driving speed corresponding to each time period and whether the delivery is across time periods, construct the time-varying network model according to the delivery task hypothesis, and obtain the delivery time of the delivery vehicle on the road segment through the time-varying network model;

[0122] A model building module, used to build a cold chain logistics path optimization model according to the delivery vehicle data, the cold chain goods to be delivered data, the carbon emission data and the delivery time;

[0123] A path generation module, used to determine multiple path nodes according to the actual delivery task, and plan the multiple path nodes through the time-varying network model and the cold chain logistics path optimization model to generate multiple delivery paths;

[0124] The path optimization module is used to take the multiple delivery paths as multiple particles, solve the multiple particles through a particle swarm algorithm, and output an optimal delivery path.

[0125] Preferably, it also includes a preprocessing module, which is used to remove noise and outliers from the cold chain logistics data after the cold chain logistics data is collected, and convert the format of the cleaned cold chain logistics data according to a set data format.

[0126] The present invention also provides a cold chain logistics path optimization device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the cold chain logistics path optimization method described above is implemented.

[0127] The present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the cold chain logistics path optimization method as described above is implemented.

[0128] The cold chain logistics path optimization method considering carbon emission costs under time-varying networks of the present invention has the following beneficial effects:

[0129] 1. Reduce carbon emissions: By incorporating carbon emission costs into the route optimization model and selecting transportation routes with lower carbon emissions, greenhouse gas emissions in the cold chain logistics process can be effectively reduced, promoting environmental protection and sustainable development.

[0130] 2. Improve transportation efficiency: While optimizing the path, dynamically adjust the transportation route and time schedule, fully consider the time-varying characteristics of the transportation network, reduce vehicle driving time and fuel consumption, and improve the overall transportation efficiency of cold chain logistics.

[0131] 3. Take into account both transportation costs and timeliness: When optimizing route selection, comprehensively consider transportation costs, carbon emission costs and transportation timeliness, and provide multi-objective optimization solutions to ensure that logistics transportation is both economical and efficient, and meet customers' requirements for timeliness and cost.

[0132] 4. Optimization of temperature control management: In view of the special needs of cold chain logistics, temperature control factors are considered to avoid long-term stops and frequent opening of carriages, ensure that the goods maintain the best temperature control state during transportation, and improve the quality of cold chain logistics services.

[0133] 5. Dynamic response capability: By using the time-varying network model, it can respond to changes in traffic conditions in real time, flexibly adjust transportation routes, enhance the dynamic adaptability of the cold chain logistics system, and improve the stability and reliability of logistics operations.

[0134] 6. Practical application value: The path optimization method and system provided by the present invention can be widely used in cold chain logistics management. Through practical application verification and adjustment, it can ensure that carbon emissions can be effectively reduced and transportation efficiency can be improved in different scenarios, and it has significant economic and social benefits.

[0135] In summary, the present invention introduces carbon emission costs and time-varying network factors into path optimization, which not only achieves environmentally friendly and efficient cold chain logistics management, but also improves the dynamic adaptability and service quality of the logistics system, and has important practical application value and promotion prospects.

[0136] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0137] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0138] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0139] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.

[0140] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0141] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. For such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0142] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A cold chain logistics path optimization method, It is characterized in that The steps include: S1. Collect cold chain logistics data, including urban traffic data, delivery vehicle data, data on cold chain goods to be delivered, and carbon emission data; S2. constructing a city road network diagram according to the city traffic data, obtaining the time required for a delivery vehicle to pass through a road segment at different time periods based on the city traffic data and the city road network diagram, dividing the required time into multiple time periods, setting a delivery task hypothesis according to the multiple time periods and the vehicle driving speeds corresponding to each time period and whether the delivery is across time periods, constructing the time-varying network model according to the delivery task hypothesis, and obtaining the delivery time of the delivery vehicle on the road segment through the time-varying network model; S3, constructing a cold chain logistics path optimization model according to the delivery vehicle data, the cold chain goods data to be delivered, the carbon emission data and the delivery time; S4. Determine multiple path nodes according to the actual delivery task, and plan the multiple path nodes through the time-varying network model and the cold chain logistics path optimization model to generate multiple delivery paths; S5. The plurality of delivery paths are regarded as a plurality of particles, and the plurality of particles are solved by a particle swarm algorithm to output an optimal delivery path.

2. The cold chain logistics path optimization method according to claim 1, It is characterized in that After the cold chain logistics data is collected, the method further includes a step S10 of pre-processing the cold chain logistics data, specifically: The cold chain logistics data is cleaned to remove noise and outliers, and the cleaned cold chain logistics data is format converted according to a set data format.

3. The cold chain logistics path optimization method according to claim 1, It is characterized in that In S2, an urban road network is constructed according to the urban traffic data, and the time required for a delivery vehicle to pass through a road segment at different time periods is obtained based on the urban traffic data and the urban road network diagram, and the required time is divided into multiple time periods, specifically: Generate an urban road network graph through the road structure data in the urban traffic data, wherein the nodes in the urban road network graph represent intersections or logistics distribution centers, and the edges between two nodes represent road segments; The time required for a vehicle to pass through a road segment at different time periods is determined according to the speed variation pattern caused by actual urban traffic congestion in the urban traffic data and the distance of the road segment, and the required time is divided into multiple time periods, and each time period corresponds to a different vehicle driving speed, and the vehicle driving speed is constant in each time period.

4. The cold chain logistics path optimization method according to claim 3, It is characterized in that In S2, a delivery task hypothesis is set according to the multiple time periods and the vehicle driving speeds corresponding to each time period and whether the delivery is across time periods, and the time-varying network model is constructed according to the delivery task hypothesis. The delivery time of the delivery vehicle on the road segment is obtained through the time-varying network model, which is specifically: Based on the different lengths of the edges between any two nodes, three delivery task assumptions are set according to the multiple time periods and the vehicle driving speeds corresponding to each time period during delivery, including completing the delivery task in a single time period, completing the delivery task across one time period, and completing the delivery task across two or more time periods, where: The delivery task completed in a single time period is: the delivery task between two nodes in a single time period needs to satisfy that the distance traveled by vehicle k in the remaining time within a certain time period is greater than or equal to the actual distance between node i and node j; the delivery time of the road segment is determined based on the delivery task completed in a single time period: Among them, t ij is the time required from node i to node j, d ij is the distance between node i and node j, v w is the driving speed in the wth time period; The delivery task completed across a time period is: the delivery task between two nodes completed across a time period needs to satisfy that the distance traveled by vehicle k in the remaining time of a certain time period is less than the actual distance between node i and node j and the distance traveled by vehicle k in the next time period is greater than or equal to the remaining distance traveled between node i and node j; the delivery time of the road segment is determined based on the delivery task completed across a time period as follows: Among them, t ij is the time required from node i to node j, t i is the time point of vehicle departure, t w is the time point t at which the vehicle departs i The time period, d ij is the distance between node i and node j, v w is the driving speed in the wth time period; The delivery task completed across two or more time periods is: if the distance traveled by vehicle k in the remaining time of a time period is less than the actual distance between node i and node j and the distance traveled by vehicle k in the next time period is less than the remaining distance traveled between node i and node j, then vehicle k needs to cross a new time period again and repeat until the maximum distance traveled by vehicle k in the new time period is greater than or equal to the remaining distance traveled between node i and node j; the delivery time of the road segment is determined based on the delivery task completed across two or more time periods as follows: Among them, t ij is the time required from node i to node j, t i is the time point of vehicle departure, t w is the time point t at which the vehicle departs i The time period, d ij is the distance between node i and node j, v w is the driving speed in the wth time period.

5. The cold chain logistics path optimization method according to claim 1, It is characterized in that In S3, a cold chain logistics path optimization model is constructed according to the delivery vehicle data, the cold chain goods data to be delivered, the carbon emission data and the delivery time, specifically: A fixed cost calculation formula is established based on the number of delivery vehicles, vehicle purchase costs, daily maintenance costs, and driver wages in the delivery vehicle data. The fixed cost expression is: Among them, C 0 is the fixed cost of the vehicle, x ijk is a binary decision variable, taking values ​​of 0 or 1; A transportation cost calculation formula is established based on the change in fuel consumption in the delivery vehicle data. The transportation cost expression is: Among them, E fue is the fuel consumption expression, E fue =E bas ·(1+a(v t -v bas ) 2 ), d ij is the distance from node i to j, P fue is the fuel price, x ijk is a binary decision variable, taking values ​​of 0 or 1; Based on the change of the value of the goods in the cold chain goods data to be delivered, a cargo damage cost expression is established, and the cargo damage cost expression is: Among them, λ is the quality decay rate, p 0 is the unit price of cold chain goods, q i is the demand of customer point i, t ij is the time required to go from node i to j, y ik is a binary decision variable, taking values ​​of 0 or 1; The customer's expected delivery time window is established based on the delivery time and the pickup appointment time in the cold chain goods data to be delivered, and a time window penalty cost expression is established based on the customer's expected delivery time window. The time window penalty cost expression is: Among them, f 11 is the waiting cost per unit time when the vehicle arrives early, f 2 is the delay cost per unit time when the vehicle arrives late, [e t , l t ] is the customer's expected delivery time window; Based on the driving carbon emission data and the refrigeration carbon emission data in the carbon emission data, a driving carbon emission cost expression and a refrigeration carbon emission cost expression are respectively constructed. The driving carbon emission cost expression is: Among them, d ij is the distance from node i to j, E fue For vehicles in v t The fuel consumption at the speed, c is the carbon emission cost per kilogram, β is the carbon emission generated by unit fuel consumption, x ijk is a binary decision variable, taking values ​​of 0 or 1; The refrigeration carbon emission cost expression is: Among them, α is the carbon emission per unit cooling time, t ij is the time required to travel from node i to node j, c is the cost of carbon emissions per kilogram, x ijk is a binary decision variable, taking values ​​of 0 or 1.

6. The cold chain logistics path optimization method according to claim 1, It is characterized in that In S5, the multiple delivery paths are used as multiple particles, and the multiple particles are solved by a particle swarm algorithm to output the optimal delivery path, which is specifically: The plurality of delivery paths are regarded as a plurality of particles, and a chaotic initialization process is performed on the plurality of particles through a Logistic chaos model; Perform adaptive inertia weight processing on the initialized particles to obtain a particle swarm carrying an inertia weight factor; A first learning factor and a second learning factor are set, wherein the first learning factor represents the component learned by the particle from its own experience, and the second learning factor represents the degree to which the particle learns toward the global optimal position. The first learning factor and the second learning factor are used to search for the optimal value from a particle group carrying an inertia weight factor, thereby outputting an optimal delivery path according to the optimal value.

7. A cold chain logistics path optimization device, It is characterized in that include: A data collection module is used to collect cold chain logistics data, wherein the cold chain logistics data includes urban traffic data, delivery vehicle data, data on cold chain goods to be delivered, and carbon emission data; A task hypothesis module is used to construct an urban road network diagram according to the urban traffic data, obtain the time required for the delivery vehicle to pass through the road segment at different time periods based on the urban traffic data and the urban road network diagram, divide the required time into multiple time periods, set a delivery task hypothesis according to the multiple time periods and the vehicle driving speed corresponding to each time period and whether the delivery is across time periods, construct the time-varying network model according to the delivery task hypothesis, and obtain the delivery time of the delivery vehicle on the road segment through the time-varying network model; A model building module, used to build a cold chain logistics path optimization model according to the delivery vehicle data, the cold chain goods to be delivered data, the carbon emission data and the delivery time; A path generation module, used to determine multiple path nodes according to the actual delivery task, and plan the multiple path nodes through the time-varying network model and the cold chain logistics path optimization model to generate multiple delivery paths; The path optimization module is used to take the multiple delivery paths as multiple particles, solve the multiple particles through a particle swarm algorithm, and output an optimal delivery path.

8. The cold chain logistics path optimization device according to claim 7, It is characterized in that It also includes a preprocessing module, which is used to remove noise and outliers from the cold chain logistics data after the cold chain logistics data is collected, and convert the format of the cleaned cold chain logistics data according to a set data format.

9. A cold chain logistics path optimization device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the cold chain logistics path optimization method according to any one of claims 1 to 6 is implemented.

10. A computer-readable storage medium storing a computer program. It is characterized in that When the computer program is executed by a processor, the cold chain logistics path optimization method as described in any one of claims 1 to 6 is implemented.

Images