Cold-chain logistics path intelligent method and system based on chaos genetic algorithm

Abstract

The invention discloses a cold-chain logistics path intelligent method and system based on a chaos genetic algorithm, and relates to the technical field of intelligent path planning. According to the method, the carbon emission dynamic cost and the dynamic freshness evaluation coefficient are introduced in model construction, meanwhile, the chaos genetic algorithm is introduced into cold-chain logistics path optimization, the process model is continuously improved, and finally the ideal cold-chain logistics optimal path with the shortest distribution distance, the lowest total distribution cost and the lowest low-carbon emission is obtained. Aiming at the defects that a traditional genetic algorithm is slow in convergence speed, a calculation result is easy to fall into a local optimal solution and an evolution effect is influenced by invariant evolution parameters in the aspect of processing big data, the invention provides a chaos adaptive genetic algorithm, chaos disturbance parameters are added through Circle chaos mapping, the problem that the calculation result is easy to fall into the local optimal solution is solved, and the robustness of the algorithm is improved. Through a chaos roulette selection algorithm, a self-adaptive crossover operator and a self-adaptive mutation operator, the influence of invariant evolution parameters on the evolution effect is solved.

Description

Technical Field

[0001] This invention relates to the field of intelligent route planning technology, specifically to an intelligent method and system for cold chain logistics routes based on chaotic genetic algorithms. Background Technology

[0002] In recent years, with the rise of fresh food e-commerce platforms, the delivery of fresh products has experienced explosive growth, making cold chain logistics an important research area in modern logistics management. In the cold chain delivery process, in addition to effectively solving traditional vehicle routing problems, various factors such as temperature control and time window constraints must be comprehensively considered. To improve transportation efficiency and optimize delivery routes and logistics networks, it is urgent to introduce new technologies for refined scheduling. It is worth noting that cold chain delivery should not only focus on delivery costs but also consider social and environmental aspects, balancing corporate interests and social benefits to promote a sustainable logistics model.

[0003] The technical solution described in Chinese patent "CN116502989B A method for optimizing the route of cold chain logistics vehicles based on a hybrid balance optimization algorithm" fails to consider two important issues when constructing the cold chain logistics vehicle route optimization model: dynamic balance of carbon emissions and dynamic freshness of cold chain goods. Furthermore, it has poor adaptability to the overtaking strategy problem and its breadth of solutions to the cold chain logistics route optimization problem is slightly limited.

[0004] In March 2025, Zhang Tianrui et al. published an article in the Journal of Chongqing Normal University (Natural Science Edition) entitled "Research on Cold Chain Logistics Route Optimization of Fresh Products Based on Improved Harris Eagle Algorithm". Considering traffic congestion, they constructed an objective function model that comprehensively considers fixed costs, transportation costs, damage costs, refrigeration costs, and penalty costs, and improved the Harris Eagle optimization algorithm to enhance solution accuracy. However, the article did not consider the constraint of dynamic carbon emission balance, a crucial contemporary environmental indicator, and the single distribution center limited the application scenarios and applicability of the optimization method.

[0005] Analysis of domestic and international research on route optimization reveals that most studies focus on conventional delivery or transportation, with limited research on cold chain logistics considering carbon emissions and insufficient dynamic response mechanisms. Existing algorithms largely rely on historical data modeling, exhibiting weak dynamic adjustment capabilities for real-time traffic congestion, extreme weather, and other emergencies, resulting in a relatively limited research scope. The following key issues urgently need to be addressed: 1) Insufficient responsiveness to dynamic scenarios. Most research models are based on static or simplified dynamic assumptions, leading to slow responses to complex dynamic scenarios such as sudden traffic congestion and extreme weather. Dynamic adjustment mechanisms based on real-time data are still incomplete. 2) Inadequate multi-cost coordination mechanisms. There are clear trade-offs between cost control and carbon emission reduction objectives. Existing research lacks dynamic balance criteria, making it difficult to adapt to complex and ever-changing real-world scenarios, and it rarely considers carbon emission factors.

[0006] In summary, the cold chain logistics vehicle routing problem model suffers from several issues, including incomplete construction, limited consideration of factors, excessive constraints from distribution centers, insufficient responsiveness to dynamic scenarios, and a lack of dynamic balance in carbon emissions. Furthermore, traditional genetic algorithms exhibit technical problems when handling optimization problems under multi-objective and multi-constraint conditions, such as slow convergence, susceptibility to local optima, and the influence of invariant evolutionary parameters on evolutionary outcomes. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention proposes an intelligent method and system for cold chain logistics routes based on chaotic genetic algorithms. This addresses the technical problems of incomplete cold chain logistics vehicle route models and poor evolutionary performance of traditional genetic algorithms in big data application scenarios.

[0008] The technical solution of this invention is as follows:

[0009] On the one hand, this invention provides an intelligent method for cold chain logistics paths based on chaotic genetic algorithms, comprising the following steps:

[0010] Considering vehicle fixed costs, vehicle transportation costs, cargo damage costs, vehicle cooling costs, dynamic carbon emission costs, and vehicle time penalty costs, construct a path optimization function model that minimizes the total cost;

[0011] An improved chaotic genetic algorithm is used to iteratively solve the path optimization function model with the minimum total cost to obtain the optimal solution; the improved chaotic genetic algorithm incorporates chaotic perturbation parameters through Circle chaotic mapping;

[0012] The optimal solution is decoded to form the final optimized cold chain logistics distribution route.

[0013] Furthermore, considering vehicle fixed costs, vehicle transportation costs, cargo damage costs, vehicle cooling costs, dynamic carbon emission costs, and vehicle time penalty costs, a path optimization function model with minimum total cost is constructed, specifically including:

[0014] First, determine the location information of the distribution center and each customer; determine the quantity of products required by the customer; set the number of refrigerated trucks and the fixed load capacity of each refrigerated truck; set that all refrigerated trucks depart from the distribution center and return to the distribution center after completing the delivery task; set that customers can only choose delivery service once within a time period; set that the total distance traveled by all refrigerated trucks cannot exceed the set total mileage, and the loading capacity of each refrigerated truck cannot exceed the set maximum load capacity; the number of refrigerated trucks carrying out delivery tasks at the same time cannot exceed the number of refrigerated trucks owned by the distribution center; products must meet minimum freshness standards upon delivery; delivery service time is strictly subject to time window restrictions, and violations of time window restrictions will incur penalty costs;

[0015] The fixed cost of the vehicle is expressed as:

[0016] (1);

[0017] In the formula, Fixed costs for vehicles, For fixed costs per vehicle unit, , The nodes are numbered, and these nodes include customers and distribution centers. For the number of customers; Number the distribution center The number of distribution centers; Number the refrigerated truck. The number of refrigerated trucks; As a decision variable, the distribution center Refrigerated truck From node Driving towards the node The value is 1 if the condition is met, and 0 otherwise.

[0018] The vehicle transportation cost is expressed as follows:

[0019] (2);

[0020] In the formula, For vehicle transportation costs, For the unit transportation cost of vehicles, For refrigerated trucks from nodes To the node The distance;

[0021] Cargo damage costs consist of two components: ongoing cargo damage costs during transportation and instantaneous cargo damage costs during loading and unloading operations. The ongoing cargo damage costs during transportation are as follows:

[0022] (3);

[0023] In the formula, For ongoing cargo damage costs during transportation, Number the customer; Indicates distribution center Refrigerated truck Does it meet customer needs? To meet the demand for fresh produce and satisfy the needs of the times =1, otherwise =0; The unit price for delivering fresh produce. Reflecting on customers Order size This refers to the dynamic rate of quality degradation of fresh produce during transportation. This refers to the average speed of the refrigerated truck.

[0024] The instantaneous cargo damage cost during loading and unloading operations is:

[0025] (4);

[0026] In the formula, The cost of instantaneous cargo damage during loading and unloading operations, Complete customer service for refrigerated trucks The remaining cargo capacity after the delivery service The instantaneous degradation rate of fresh produce quality during unloading. For customers Required unloading time;

[0027] The total cost of cargo damage is:

[0028] (5);

[0029] in, Cost of damaged goods;

[0030] The vehicle's refrigeration cost is:

[0031] (6);

[0032] In the formula, For vehicle cooling costs, This refers to the refrigerant consumption coefficient during delivery. This refers to the refrigerant consumption coefficient during loading and unloading. For nodes To the node Delivery time;

[0033] The dynamic cost of carbon emissions is:

[0034] (7);

[0035] In the formula, For refrigerated trucks from nodes To the node Load capacity, This refers to the carbon emissions per unit distance of refrigerated trucks. This represents the carbon emissions per unit distance for a refrigerated truck when its load capacity is zero. This refers to the carbon emissions per unit distance for a fully loaded refrigerated truck. This refers to the maximum load capacity of the refrigerated truck. For unit fuel cost, The carbon tax cost per unit distance traveled by refrigerated trucks. For carbon tax, Carbon emission factor For the dynamic cost of carbon emissions;

[0036] The vehicle time penalty cost is:

[0037] (8);

[0038] In the formula, The cost of penalties for vehicle time windows. For refrigerated trucks to reach customers Time, [ , For customers Delivery service time window, The earliest arrival time, The latest arrival time, Early arrival penalty coefficient, The penalty coefficient for being late; As a decision variable, the distribution center Refrigerated truck From node Driving towards customers The value is 1 if the condition is met, and 0 otherwise.

[0039] The path optimization function model that minimizes total cost is:

[0040] (9);

[0041] st

[0042] (10);

[0043] (11);

[0044] (12);

[0045] (13);

[0046] (14);

[0047] (15);

[0048] (16);

[0049] in, For total cost, Indicates refrigerated truck Does it meet customer needs? To meet the demand for fresh produce and satisfy the needs of the times =1, otherwise =0; For customers; , and For decision variables; For the customer's number; The node number; For refrigerated trucks, For distribution center collection, This refers to the maximum driving range of a refrigerated truck. For customers The cargo capacity.

[0050] Furthermore, the step of using an improved chaotic genetic algorithm to iteratively solve the path optimization function model that minimizes the total cost to obtain the optimal solution specifically includes:

[0051] A1: Construct an initial population; the initial population includes several chromosomes, each chromosome representing the delivery route of the refrigerated truck to complete the delivery task, and each gene in the chromosome representing a customer or distribution center passed through the delivery route;

[0052] A2: Calculate the fitness function value for each chromosome in the current population;

[0053] The fitness function is expressed as follows:

[0054] (17);

[0055] In the formula, Chromosomes This is the improved fitness function; Let be the objective function. For specific parameters.

[0056] A3: Based on the fitness function value, perform chaotic roulette wheel selection on the chromosomes in the current population;

[0057] The chaotic roulette wheel selection specifically involves: performing a roulette wheel selection operation first, and then adding a chaotic perturbation to the chromosomes replicated during the roulette wheel selection operation after the roulette wheel selection operation.

[0058] The process of increasing the chaotic perturbation of the replicated chromosomes in the roulette wheel selection operation is as follows:

[0059] (18);

[0060] In the formula, For the number of iterations, The new chromosomes selected for the chaotic roulette wheel game. To select the chromosome to replicate during the roulette wheel selection operation, For chaotic perturbation operators;

[0061] A4: Perform adaptive crossover on the new chromosomes obtained through chaotic roulette wheel selection;

[0062] A5: Perform adaptive mutation operations on chromosomes that have undergone adaptive crossover;

[0063] A6: The chromosomes after the adaptive mutation operation are divided into two parts according to their fitness function values. The chromosomes with fitness function values ​​higher than or equal to the set value are retained, and the chromosomes with fitness function values ​​lower than the set value are further processed by chaotic mapping.

[0064] A7: Integrate the chromosomes with high fitness function values ​​and the chromosomes obtained from chaotic mapping into the current population. Determine if the current iteration count has reached the set maximum iteration count. If yes, output the chromosome with the highest current fitness function value as the optimal solution. Otherwise, increment the iteration count by 1 and return to A3.

[0065] Furthermore, A1 specifically includes:

[0066] A1.1: Set the total number of customers in the current delivery task. A customer is randomly selected from the customer set for the initial visit. ;

[0067] A1.2: Set the value of the first gene on the chromosome to "0", then use the initial access client. As the starting point of the path, visit the remaining customers in a circular order. If the remaining cargo capacity of the refrigerated truck can meet the needs of the customer, add the customer to the current path. Otherwise, close the current path and add a gene value "0" to the chromosome. Continue to visit the remaining customers in the above way until all customers have been traversed. Add a gene value "0" to the end of the chromosome to complete the generation of a chromosome.

[0068] A1.3: Repeat A1.1-A1.2 until an initial population of the specified size is generated.

[0069] Furthermore, the adaptive crossover operation specifically includes:

[0070] The first step is to randomly select a pair of parent chromosomes, generate a random number for them and compare it with the crossover probability. When the random number is not greater than the crossover probability, the genes are swapped at the same starting and ending positions of the pair of parent chromosomes. The swapped offspring chromosomes are then subjected to conflict detection. A mapping table is established based on the two swapped gene sets. The conflicting genes in the newly generated chromosomes are then transformed according to the mapping table. Gene conflict detection is completed, and the crossover-transformed offspring chromosomes are formed.

[0071] The crossover probability is:

[0072] (19);

[0073] In the formula, For crossover probability, The initial crossover probability; For the maximum crossover probability, To minimize the crossover probability, The maximum fitness function value. The minimum fitness function value. This is the current fitness function value. This represents the current average fitness function value of the population.

[0074] Furthermore, the adaptive mutation operation specifically includes:

[0075] A pair of parent chromosomes is randomly selected, a random number is generated for it and compared with the mutation probability. When the random number is not greater than the mutation probability, two positions on the chromosome are randomly selected as the starting points of the inversion interval. The genes in the selected inversion interval are reversed to obtain new offspring chromosomes. However, when a gene value of "0" is encountered, it is not involved in the inversion operation.

[0076] The mutation probability is:

[0077] (20);

[0078] In the formula, The mutation probability, The initial mutation probability; The maximum mutation probability, This represents the minimum mutation probability.

[0079] Furthermore, A6 specifically includes:

[0080] The formula for the Circle chaotic mapping is as follows:

[0081] (twenty one);

[0082] In the formula, The chaotic perturbation operator at the l-th iteration. The average rotational speed; To control the strength of nonlinear interactions;

[0083] The chaotic perturbation is added as shown in the following formula:

[0084] (twenty two);

[0085] In the formula, The chromosome obtained by chaotic mapping. , These are the chromosomes with the largest and smallest fitness function values ​​in the population during the l-th iteration, respectively.

[0086] On the other hand, this invention also proposes an intelligent system for cold chain logistics routes based on chaotic genetic algorithms, used to implement an intelligent method for cold chain logistics routes based on chaotic genetic algorithms, including:

[0087] The model building module is used to construct a path optimization function model with the minimum total cost, taking into account vehicle fixed costs, vehicle transportation costs, cargo damage costs, vehicle cooling costs, dynamic carbon emission costs, and vehicle time penalty costs.

[0088] The solution module is used to iteratively solve the path optimization function model with the minimum total cost using an improved chaotic genetic algorithm to obtain the optimal solution;

[0089] The solution generation module is used to decode the optimal solution and form the final cold chain logistics distribution route optimization solution.

[0090] Thirdly, this application proposes an electronic device, including: one or more processors, and a memory for storing instructions, which, when executed by the one or more processors, cause the one or more processors to execute the aforementioned intelligent method for cold chain logistics paths based on chaotic genetic algorithms.

[0091] Fourthly, this application proposes a computer-readable storage medium storing executable instructions that, when executed, cause a processor to perform the aforementioned intelligent method for cold chain logistics paths based on a chaotic genetic algorithm.

[0092] Fifthly, this application proposes a computer program product, including a computer program or instructions, which, when executed by a processor, implements the aforementioned intelligent method for cold chain logistics paths based on a chaotic genetic algorithm.

[0093] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0094] This invention introduces dynamic carbon emission costs and dynamic freshness evaluation coefficients into the model construction to build a multi-distribution center cold chain logistics delivery route optimization model (the path optimization function model with the minimum total cost). This invention also introduces a chaotic genetic algorithm into cold chain logistics route optimization, utilizing the survival-of-the-fittest principle in genetic algorithms to continuously improve the process model, ultimately obtaining the ideal optimal cold chain logistics route with the shortest delivery distance, lowest total delivery cost, and lowest carbon emissions.

[0095] This invention addresses the shortcomings of traditional genetic algorithms in handling large datasets, such as slow convergence speed, susceptibility to local optima, and the impact of invariant evolutionary parameters on evolutionary performance. It proposes a chaotic adaptive genetic algorithm that incorporates chaotic perturbation parameters through a Circle chaotic mapping, thus resolving the problem of getting trapped in local optima. Furthermore, it utilizes a chaotic roulette wheel selection algorithm, adaptive crossover operators, and adaptive mutation operators to mitigate the impact of invariant evolutionary parameters on evolutionary performance.

[0096] The optimization method provided by this invention shortens the delivery distance, reduces the total delivery cost and carbon emissions, improves the level of logistics management and operational efficiency, and meets the application requirements. Attached Figure Description

[0097] Figure 1 This is a flowchart of an intelligent cold chain logistics path method based on a chaotic genetic algorithm in an embodiment of the present invention;

[0098] Figure 2 This is a schematic diagram of the parent chromosome in an embodiment of the present invention;

[0099] Figure 3 This is a schematic diagram of the chromosomes after crossover in an embodiment of the present invention;

[0100] Figure 4 This is a mapping table for the gene detection process in this embodiment of the invention;

[0101] Figure 5 This is a schematic diagram of the offspring chromosomes after gene conflict detection in an embodiment of the present invention;

[0102] Figure 6 This is a schematic diagram of inversion variation in an embodiment of the present invention. Detailed Implementation

[0103] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0104] Example 1:

[0105] A smart method for cold chain logistics routes based on chaotic genetic algorithms, such as Figure 1 As shown, it includes the following steps:

[0106] S1: Considering vehicle fixed costs, vehicle transportation costs, cargo damage costs, vehicle cooling costs, dynamic carbon emission costs, and vehicle time penalty costs, construct a path optimization function model with the minimum total cost;

[0107] This embodiment first determines the location information of the distribution center and each customer; determines the quantity of products required by the customer; sets the number of refrigerated trucks and the fixed load capacity of each refrigerated truck; sets that all refrigerated trucks depart from the distribution center and return to the distribution center after completing the delivery task; sets that customers can only choose delivery service once within a time period; sets that the total distance traveled by all refrigerated trucks cannot exceed the set total mileage, and that the loading capacity of each refrigerated truck cannot exceed the set maximum load capacity; the number of refrigerated trucks performing delivery tasks at the same time cannot exceed the number of refrigerated trucks owned by the distribution center; products must meet minimum freshness standards upon delivery; delivery service time is strictly subject to time window restrictions, refrigerated trucks can arrive at customers in advance, but cannot deliver in advance; and sets that violating the time window restrictions will incur penalty costs.

[0108] The vehicle's fixed costs mainly cover labor costs and additional expenses incurred by the vehicle. Labor costs include employee salaries and performance bonuses, while additional expenses include depreciation and maintenance costs. The vehicle's fixed costs are expressed as follows:

[0109] (1);

[0110] In the formula, Fixed costs for vehicles, For fixed costs per vehicle unit, , The nodes are numbered, and these nodes include customers and distribution centers. For the number of customers; Number the distribution center The number of distribution centers; Number the refrigerated truck. The number of refrigerated trucks; As a decision variable, the distribution center Refrigerated truck From node Driving towards the node The value is 1 if the condition is met, and 0 otherwise.

[0111] The transportation costs for fresh produce delivery vehicles mainly include vehicle costs and fuel costs. Vehicle costs include vehicle maintenance, repairs, and insurance. Fuel costs vary depending on the type of vehicle. The vehicle transportation cost is expressed as follows:

[0112] (2);

[0113] In the formula, For vehicle transportation costs, For the unit transportation cost of vehicles, For refrigerated trucks from nodes To the node The distance;

[0114] To quantify the spoilage patterns of fresh produce during distribution, a dynamic freshness assessment model is proposed, incorporating an exponential function. This reflects the relationship between the quality of fresh produce and its timeliness. For time, This is a freshness assessment coefficient, and its value is highly correlated with fluctuations in ambient temperature, humidity, and oxygen content. Increased temperature or oxygen levels accelerate the deterioration rate of fresh produce. During loading and unloading operations, frequent opening of cargo doors causes significant changes in temperature, humidity, and gas composition within the storage space due to heat exchange. The preservation conditions during this stage are more severe than during transportation. Therefore, this invention subdivides quality loss into two components: continuous damage costs during transportation and instantaneous damage costs during loading and unloading operations.

[0115] The ongoing cost of cargo damage during transportation is:

[0116] (3);

[0117] In the formula, For ongoing cargo damage costs during transportation, Number the customer; Indicates distribution center Refrigerated truck Does it meet customer needs? To meet the demand for fresh produce and satisfy the needs of the times =1, otherwise =0; The unit price for delivering fresh produce. Reflecting on customers Order size This refers to the dynamic rate of quality degradation of fresh produce during transportation. This refers to the average speed of the refrigerated truck.

[0118] The instantaneous cargo damage cost during loading and unloading operations is:

[0119] (4);

[0120] In the formula, The cost of instantaneous cargo damage during loading and unloading operations, Complete customer service for refrigerated trucks The remaining cargo capacity after the delivery service The instantaneous degradation rate of fresh produce quality during unloading. For customers Required unloading time;

[0121] Therefore, the total cost of cargo damage is:

[0122] (5);

[0123] in, Cost of damaged goods;

[0124] The refrigeration costs involved in the cold chain logistics of fresh produce during delivery mainly include two aspects: first, the cost incurred by the refrigeration equipment used to maintain the freshness of the produce during delivery; and second, the refrigeration costs incurred when external air enters the vehicle compartment and activates the refrigeration equipment during loading and unloading. The vehicle refrigeration cost is as follows:

[0125] (6);

[0126] In the formula, For vehicle cooling costs, This refers to the refrigerant consumption coefficient during delivery. This refers to the refrigerant consumption coefficient during loading and unloading. For nodes To the node Delivery time;

[0127] In the cold chain logistics and distribution of fresh produce, the carbon emission costs mainly consist of two parts: first, the costs incurred by the vehicles during delivery; and second, the CO2 emissions generated by the vehicles during delivery. 2 The resulting carbon emission costs. Therefore, the dynamic cost of carbon emissions is the sum of these two parts, and the dynamic cost of carbon emissions is:

[0128] (7);

[0129] In the formula, For refrigerated trucks from nodes To the node Load capacity, This refers to the carbon emissions per unit distance of refrigerated trucks. This represents the carbon emissions per unit distance for a refrigerated truck when its load capacity is zero. This refers to the carbon emissions per unit distance for a fully loaded refrigerated truck. This refers to the maximum load capacity of the refrigerated truck. For unit fuel cost, The carbon tax cost per unit distance traveled by refrigerated trucks. For carbon tax, Carbon emission factor For the dynamic cost of carbon emissions;

[0130] If a refrigerated truck in a cold chain logistics system exceeds the customer's required delivery time window, a time window penalty will be implemented to ensure the freshness and quality of the perishable goods. The vehicle time penalty cost is:

[0131] (8);

[0132] In the formula, The cost of penalties for vehicle time windows. For refrigerated trucks to reach customers Time, [ , For customers Delivery service time window, The earliest arrival time, The latest arrival time, Early arrival penalty coefficient, The penalty coefficient for being late; As a decision variable, the distribution center Refrigerated truck From node Driving towards customers The value is 1 if the condition is met, and 0 otherwise.

[0133] The path optimization function model that minimizes total cost is:

[0134] (9);

[0135] st

[0136] (10);

[0137] (11);

[0138] (12);

[0139] (13);

[0140] (14);

[0141] (15);

[0142] (16);

[0143] in, For total cost, Indicates refrigerated truck Does it meet customer needs? To meet the demand for fresh produce and satisfy the needs of the times =1, otherwise =0; For customer groups (including) (1 customer) , and For decision variables; For the customer's number; The node number; For refrigerated trucks (including) (a refrigerated truck) For distribution center collection (including (one distribution center) This refers to the maximum driving range of a refrigerated truck. For customers The cargo capacity;

[0144] In summary, the above is the path optimization function model with the minimum total cost. Equation (9) represents the overall objective function of the path optimization function model with the minimum total cost in the cold chain logistics distribution of fresh products. Equation (10) indicates that only one refrigerated truck can travel from the distribution center to each customer. Equation (11) indicates that when the distribution center carries out cold chain logistics activities to distribute fresh products, only one refrigerated truck can travel on the delivery path between each two nodes. Equation (12) indicates that the delivery path of the refrigerated truck must not exceed the set maximum mileage of the refrigerated truck. Equation (13) indicates that the refrigerated trucks of the distribution center... The load capacity cannot exceed the maximum load capacity (rated load capacity) of the refrigerated truck itself; Equation (14) indicates that the number of refrigerated trucks performing delivery tasks in a complete delivery cycle should not exceed the total number of refrigerated trucks owned by the distribution center; Equation (15) indicates that the refrigerated truck starts from the designated distribution center and returns to the distribution center after completing the delivery task, and all refrigerated trucks have the same initial position; Equation (16) indicates the customer's time window requirements.

[0145] S2: The improved chaotic genetic algorithm is used to iteratively solve the path optimization function model with the minimum total cost to obtain the optimal solution; the improved chaotic genetic algorithm incorporates chaotic perturbation parameters through Circle chaotic mapping;

[0146] Specifically, the following steps are included:

[0147] S2.1: Construct an initial population; the initial population includes several chromosomes, each chromosome representing the delivery route of the refrigerated truck to complete the delivery task, and each gene representing a customer or distribution center passed through the delivery route;

[0148] Specifically, the following steps are included:

[0149] S2.1.1: Set the total number of customers in the current delivery task. A customer is randomly selected from the customer set for the initial visit. ;

[0150] S2.1.2: Set the first gene value of the chromosome to "0", representing the start of delivery from the distribution center, and then starting with the initial customer. As the starting point of the path, visit the remaining customers in a circular sequence. +1、 +2、┈、 1, 2, ... -1) If the remaining cargo capacity of the refrigerated truck can meet the customer's needs, add the customer to the current path; otherwise, close the current path and add a gene value "0" to the chromosome. Continue to visit the remaining customers in the same way until all customers have been traversed. Add a gene value "0" to the end of the chromosome, which means that the vehicle has returned to the distribution center after the delivery is completed. This completes the generation of a chromosome.

[0151] In this embodiment, integer encoding is used for chromosome encoding. Assuming the distribution center has K refrigerated trucks, it needs to encode... Delivery services are provided to 1 customer, where 0 represents a distribution center, and 1~ Indicates refrigerated truck, 1~ Let's say we represent the customers who need delivery. Assuming the chromosome number is (0,2,4,6,8,0,3,5,7,1,0), then there are 2 refrigerated trucks that need to deliver to 8 customers. Specifically, the first truck departs from the distribution center, delivers fresh produce to customers 2-4-6-8 in that order, and then returns to the distribution center. The second truck departs from the distribution center, delivers fresh produce to customers 3-5-7-1 in that order, and then returns to the distribution center.

[0152] S2.1.3: Repeat S2.1.1-S2.1.2 until an initial population of the specified size is generated;

[0153] S2.2: Calculate the fitness function value for each chromosome in the current population;

[0154] In this embodiment, an exponential transformation method is selected to adjust the relative proportions between fitness function values, thereby controlling the selection and guiding the algorithm's global search. The fitness function is expressed as:

[0155] (17);

[0156] In the formula, Chromosomes This is the improved fitness function; Let be the objective function. These are specific parameters used to increase the diversity of a population;

[0157] S2.3: Based on the fitness function value, perform chaotic roulette wheel selection on the chromosomes in the current population;

[0158] Roulette wheel selection is a classic selection algorithm, but it is prone to getting stuck in local optima, reducing search efficiency. This invention adds a low-probability chaotic perturbation to the chromosomes replicated during the roulette wheel selection operation after the initial selection operation. This reduces premature convergence caused by the selection operation itself and also lowers the probability of crossover between identical chromosomes.

[0159] The process of adding a small-probability chaotic perturbation to the chromosomes replicated in the roulette wheel selection operation is as follows:

[0160] (18);

[0161] In the formula, For the number of iterations, The new chromosomes selected for the chaotic roulette wheel game. To select the chromosome to replicate during the roulette wheel selection operation, For chaotic perturbation operators; It can make the perturbation uniformly traverse the points centered on the gene with an interval length of 2cd;

[0162] S2.4: Perform adaptive crossover on the new chromosomes obtained through chaotic roulette wheel selection;

[0163] In the early stages of evolution, a higher crossover probability is needed to improve the algorithm's global search capability. However, as evolution progresses, the number of individuals with high fitness function values ​​gradually increases, and the need for rapid convergence grows. At this point, the crossover probability should be lower. Therefore, the improved adaptive crossover probability is related to both the evolutionary process and the fitness function value during evolution; it is necessary to increase chromosome diversity while simultaneously improving search speed to find the optimal solution.

[0164] (19);

[0165] In the formula, For crossover probability, The initial crossover probability is set to 0.8 in this embodiment; For the maximum crossover probability, To minimize the crossover probability, The maximum fitness function value. The minimum fitness function value. This is the current fitness function value. This represents the current average fitness function value of the population.

[0166] The crossover operator is a partial matching crossover. The first step involves randomly selecting a pair of parent chromosomes, generating a random number for each, and comparing it to the crossover probability. If the random number is not greater than the crossover probability, gene exchange occurs at the points where the start and end positions of the parent chromosomes are the same. Figure 2 As shown. The offspring chromosomes after the exchange are as follows. Figure 3 As shown, then conflict detection is performed, and a mapping table is established based on the two exchanged gene sets, as follows: Figure 4 As shown. The conflicting genes in the newly generated chromosome are transformed according to the mapping table to ensure that all genes in the resulting offspring chromosome are unique. At this point, gene conflict detection is complete, and the offspring chromosome after crossover is formed, as shown. Figure 5 As shown.

[0167] S2.5: Perform adaptive mutation operations on chromosomes that have undergone adaptive crossover;

[0168] Similar to the crossover probability, the mutation probability is related to both the number of generations and the individual's fitness function value. In the later stages of evolution, to prevent the algorithm from getting trapped in local optima, the mutation probability needs to be increased to improve its ability to escape local optima. However, when the individual's fitness function value exceeds the average fitness function value, search fatigue occurs, and at this point, it is necessary to increase the mutation probability to enhance individual diversity.

[0169] (20);

[0170] In the formula, The mutation probability, The initial mutation probability is set to 0.001 in this embodiment; The maximum mutation probability, The minimum mutation probability;

[0171] The mutation operator is the inversion mutation method, specifically as follows: Figure 6 As shown, a pair of parent chromosomes are randomly selected, a random number is generated for them and compared with the mutation probability. When the random number is not greater than the mutation probability, two positions on the chromosome are randomly selected as the starting points of the inversion interval. The genes in the selected inversion interval are reversed to obtain new offspring chromosomes. However, when a gene value of "0" is encountered, it does not participate in the inversion operation.

[0172] S2.6: The chromosomes after the adaptive mutation operation are divided into two parts according to their fitness function values. The chromosomes with fitness function values ​​higher than or equal to the set value are retained, and the chromosomes with fitness function values ​​lower than the set value are further subjected to chaotic mapping processing to further improve the population diversity and make it less likely to fall into local optima.

[0173] This invention, after genetic operations, retains individuals with high fitness functions in the population, while further subjecting individuals with low fitness functions to chaotic optimization. This increases the genetic diversity of the population, making it less prone to getting trapped in local optima. Simultaneously, adding chaotic optimization may bring individuals closer to the optimal solution, thus reducing the number of generations required for evolution.

[0174] Currently, common chaotic mapping methods include Logistic chaotic mapping, Sine chaotic mapping, and Circle chaotic mapping. Among them, Logistic and Sine chaotic mappings exhibit strong nonlinear dynamic characteristics, especially within specific parameter ranges, where the generated chaotic sequences may show abrupt changes. Circle chaotic mapping, on the other hand, exhibits circular motion characteristics, possessing strong periodicity and continuity. These properties make the chaotic sequences generated by Circle mapping more uniform and coherent in terms of cyclicity and smooth transitions compared to mappings based on Logistic or Sine functions. These advantages of Circle mapping have played a significant role in improving algorithm performance. By introducing Circle chaotic mapping, the algorithm can traverse the solution space more uniformly during the search process, reducing the possibility of getting trapped in local optima, thereby improving global search capabilities. Simultaneously, its smooth transition characteristics also help the algorithm maintain stable convergence speed and accuracy in continuous optimization problems.

[0175] The formula for the Circle chaotic mapping is as follows:

[0176] (twenty one);

[0177] In the formula, Let be the chaotic perturbation operator (angle) at the l-th iteration. The average rotational speed is [0,1]. To control the strength of the nonlinear interaction, the value is taken as [0, 1].

[0178] The chaotic perturbation is added as shown in the following formula:

[0179] (twenty two);

[0180] In the formula, The chromosome obtained by chaotic mapping. , These are the chromosomes with the largest and smallest fitness function values ​​in the population during the l-th iteration, respectively.

[0181] S2.7: Integrate the chromosomes with high fitness function values ​​and the chromosomes obtained from chaotic mapping into the current population. Determine if the current iteration count has reached the set maximum iteration count. If yes, output the chromosome with the highest fitness function value as the optimal solution. Otherwise, increment the iteration count by 1 and return to S2.3.

[0182] S3: Decode the optimal solution to form the final cold chain logistics distribution route optimization scheme;

[0183] In this embodiment, a company in Hunnan District, Shenyang City is used as a case study. The company has 4 distribution centers and 34 customer points. Relevant data is obtained based on on-site investigation and reference to data released by the Shenyang Transportation Department.

[0184] Refrigerated truck specifications: Overall dimensions 6000×2200×3300mm, container dimensions 4200×2100×550mm, gross vehicle weight 4500 kg, rated load capacity 1800 kg, average vehicle speed 60 km / h, diesel fuel, constant speed no-load fuel consumption 10.2L / 100 km, combined fuel consumption 16.3L / 100 km.

[0185] Model basic parameters: unit fixed cost 180 yuan, unit transportation cost 3 yuan / km, unit refrigeration cost 1.5 yuan / h, average unit price of fresh agricultural products 3000 yuan / t, dynamic decay rate of product quality during transportation 0.002, instantaneous decay of product quality during unloading 0.003, penalty cost for early arrival per unit time 30 yuan / h, penalty cost for delayed arrival per unit time 45 yuan / h, carbon emission coefficient 2.2 kg / L, unit carbon tax price 0.35 yuan / kg.

[0186] Genetic algorithm parameters: population size 100, maximum number of iterations 300, initial crossover probability 0.8, initial mutation probability 0.001.

[0187] The traditional genetic algorithm and the improved chaotic genetic algorithm were each run 10 times. The best result was selected as the optimal solution for the genetic algorithm by comparison. The cost comparison is shown in Table 1.

[0188] Table 1. Cost comparison of the two algorithms before and after improvement;

[0189]

[0190] Apart from fixed costs, all other costs have been improved after optimization, achieving the goal of maximizing economic and environmental benefits, which proves the scientific validity and feasibility of the improved chaotic genetic algorithm in solving such complex optimization problems.

[0191] To consider whether carbon emission constraints affect path selection and optimal solution value based on the improved chaotic genetic algorithm, the solution results of the objective function without carbon emission dynamic cost and with carbon emission dynamic cost are compared. The algorithm is solved 10 times, and the best result among the 10 results is taken as the optimal solution, as shown in Table 2.

[0192] Table 2 Data Comparison;

[0193]

[0194] Whether considering various costs, vehicle load rate, or total vehicle mileage, the optimal solution that takes carbon emission constraints into account is superior to the optimal solution that does not.

[0195] Example 2:

[0196] A cold chain logistics path intelligent system based on chaotic genetic algorithm, used to implement an intelligent method for cold chain logistics paths based on chaotic genetic algorithm, including:

[0197] The model building module is used to construct a path optimization function model with the minimum total cost, taking into account vehicle fixed costs, vehicle transportation costs, cargo damage costs, vehicle cooling costs, dynamic carbon emission costs, and vehicle time penalty costs.

[0198] The solution module is used to iteratively solve the path optimization function model with the minimum total cost using an improved chaotic genetic algorithm to obtain the optimal solution;

[0199] The solution generation module is used to decode the optimal solution and form the final cold chain logistics distribution route optimization solution.

[0200] Example 3:

[0201] This embodiment proposes an electronic device, including: one or more processors, and a memory, wherein the memory is used to store instructions, and when the instructions are executed by the one or more processors, the one or more processors execute the aforementioned intelligent method for cold chain logistics paths based on chaotic genetic algorithms.

[0202] The electronic device can be a mobile phone, computer, or tablet computer, etc., and includes a memory and a processor. The memory stores a computer program, which, when executed by the processor, implements a smart method for cold chain logistics paths based on a chaotic genetic algorithm as described in the embodiment. It is understood that the electronic device may also include input / output (I / O) interfaces and communication components.

[0203] The processor is used to execute all or part of the steps in the intelligent cold chain logistics path method based on chaotic genetic algorithm as described in the above embodiments. The memory is used to store various types of data, which may include, for example, instructions for any application or method in the electronic device, as well as application-related data.

[0204] The processor can be implemented as an Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor, or other electronic components, and is used to execute the intelligent cold chain logistics path method based on chaotic genetic algorithm described in the above embodiments.

[0205] Example 4:

[0206] This embodiment proposes a computer-readable storage medium that stores executable instructions. When these instructions are executed, if they are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0207] The computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the intelligent cold chain logistics path method based on chaotic genetic algorithm described in various embodiments of this application.

[0208] The aforementioned storage media include: flash memory, hard disks, multimedia cards, card-type memory (e.g., SD (Secure Digital Memory Card) or DX (Memory Data Register, MDR) memory), random access memory (RAM), static random-access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic storage, disks, optical discs, servers, APP (Application) application stores, and other media capable of storing program verification codes. These media store computer programs, which, when executed by a processor, can implement the various steps of the aforementioned intelligent cold chain logistics path method based on chaotic genetic algorithms.

[0209] Example 5:

[0210] This embodiment proposes a computer program product, including a computer program or instructions, which, when executed by a processor, implements the aforementioned intelligent method for cold chain logistics paths based on a chaotic genetic algorithm.

[0211] Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a computer program product.

[0212] The various embodiments in this application are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0213] The scope of protection of this application is not limited to the embodiments described above. Obviously, those skilled in the art can make various modifications and variations to this disclosure without departing from the scope and spirit of this disclosure. If such modifications and variations fall within the scope of this disclosure and its equivalents, then the intent of this disclosure also includes these modifications and variations.

Claims

1. A smart method for cold chain logistics paths based on chaotic genetic algorithms, characterized in that, Includes the following steps: Considering vehicle fixed costs, vehicle transportation costs, cargo damage costs, vehicle cooling costs, dynamic carbon emission costs, and vehicle time penalty costs, construct a path optimization function model that minimizes the total cost; An improved chaotic genetic algorithm is used to iteratively solve the path optimization function model with the minimum total cost to obtain the optimal solution; the improved chaotic genetic algorithm incorporates chaotic perturbation parameters through Circle chaotic mapping; The optimal solution is decoded to form the final optimized cold chain logistics distribution route.

2. The intelligent method for cold chain logistics paths based on chaotic genetic algorithm according to claim 1, characterized in that, The process considers vehicle fixed costs, vehicle transportation costs, cargo damage costs, vehicle cooling costs, dynamic carbon emission costs, and vehicle time penalty costs to construct a path optimization function model that minimizes the total cost. Specifically, this includes: First, determine the location information of the distribution center and each customer; determine the quantity of products required by the customer; set the number of refrigerated trucks and the fixed load capacity of each refrigerated truck; set that all refrigerated trucks depart from the distribution center and return to the distribution center after completing the delivery task; set that customers can only choose delivery service once within a time period; set that the total distance traveled by all refrigerated trucks cannot exceed the set total mileage, and the loading capacity of each refrigerated truck cannot exceed the set maximum load capacity; the number of refrigerated trucks carrying out delivery tasks at the same time cannot exceed the number of refrigerated trucks owned by the distribution center; products must meet minimum freshness standards upon delivery; delivery service time is strictly subject to time window restrictions, and violations of time window restrictions will incur penalty costs; The fixed cost of the vehicle is expressed as: （1）； In the formula, Fixed costs for vehicles, For fixed costs per vehicle unit, , The nodes are numbered, and these nodes include customers and distribution centers. For the number of customers; Number the distribution center The number of distribution centers; Number the refrigerated truck. The number of refrigerated trucks; As a decision variable, the distribution center Refrigerated truck From node Driving towards the node The value is 1 if the condition is met, and 0 otherwise. The vehicle transportation cost is expressed as follows: （2）； In the formula, For vehicle transportation costs, For the unit transportation cost of vehicles, For refrigerated trucks from nodes To the node The distance; Cargo damage costs consist of two components: ongoing cargo damage costs during transportation and instantaneous cargo damage costs during loading and unloading operations. The ongoing cargo damage costs during transportation are as follows: （3）； In the formula, For ongoing cargo damage costs during transportation, Number the customer; Indicates distribution center Refrigerated truck Does it meet customer needs? To meet the demand for fresh produce and satisfy the needs of the times =1, otherwise =0; The unit price for delivering fresh produce. Reflecting on customers Order size This refers to the dynamic rate of quality degradation of fresh produce during transportation. This refers to the average speed of the refrigerated truck. The instantaneous cargo damage cost during loading and unloading operations is: （4）； In the formula, The cost of instantaneous cargo damage during loading and unloading operations, Complete customer service for refrigerated trucks The remaining cargo capacity after the delivery service The instantaneous degradation rate of fresh produce quality during unloading. For customers Required unloading time; The total cost of cargo damage is: （5）； in, Cost of damaged goods; The vehicle's refrigeration cost is: （6）； In the formula, For vehicle cooling costs, This refers to the refrigerant consumption coefficient during delivery. This refers to the refrigerant consumption coefficient during loading and unloading. For nodes To the node Delivery time; The dynamic cost of carbon emissions is: （7）； In the formula, For refrigerated trucks from nodes To the node Load capacity, This refers to the carbon emissions per unit distance of refrigerated trucks. This represents the carbon emissions per unit distance for a refrigerated truck when its load capacity is zero. This refers to the carbon emissions per unit distance for a fully loaded refrigerated truck. This refers to the maximum load capacity of the refrigerated truck. For unit fuel cost, The carbon tax cost per unit distance traveled by refrigerated trucks. For carbon tax, Carbon emission factor For the dynamic cost of carbon emissions; The vehicle time penalty cost is: （8）； In the formula, The cost of penalties for vehicle time windows. For refrigerated trucks to reach customers Time, [ , For customers Delivery service time window, The earliest arrival time, The latest arrival time, Early arrival penalty coefficient, The penalty coefficient for being late; As a decision variable, the distribution center Refrigerated truck From node Driving towards customers The value is 1 if the condition is met, and 0 otherwise. The path optimization function model that minimizes total cost is: （9）； st （10）； （11）； （12）； （13）； （14）； （15）； （16）； in, For total cost, Indicates refrigerated truck Does it meet customer needs? To meet the demand for fresh produce and satisfy the needs of the times =1, otherwise =0; For customers; , and For decision variables; For the customer's number; The node number; For refrigerated trucks, For distribution center collection, This refers to the maximum driving range of a refrigerated truck. For customers The cargo capacity.

3. The intelligent method for cold chain logistics paths based on chaotic genetic algorithm according to claim 1, characterized in that, The process of using an improved chaotic genetic algorithm to iteratively solve the path optimization function model that minimizes the total cost to obtain the optimal solution specifically includes: A1: Construct an initial population; the initial population includes several chromosomes, each chromosome representing the delivery route of the refrigerated truck to complete the delivery task, and each gene in the chromosome representing a customer or distribution center passed through the delivery route; A2: Calculate the fitness function value for each chromosome in the current population; The fitness function is expressed as follows: （17）； In the formula, Chromosomes This is the improved fitness function; Let be the objective function. For specific parameters; A3: Based on the fitness function value, perform chaotic roulette wheel selection on the chromosomes in the current population; The chaotic roulette wheel selection specifically involves: performing a roulette wheel selection operation first, and then adding a chaotic perturbation to the chromosomes replicated during the roulette wheel selection operation after the roulette wheel selection operation. The process of increasing the chaotic perturbation of the replicated chromosomes in the roulette wheel selection operation is as follows: （18）； In the formula, For the number of iterations, The new chromosomes selected for the chaotic roulette wheel game. To select the chromosome to replicate during the roulette wheel selection operation, For chaotic perturbation operators; A4: Perform adaptive crossover on the new chromosomes obtained through chaotic roulette wheel selection; A5: Perform adaptive mutation operations on chromosomes that have undergone adaptive crossover; A6: Divide the chromosomes after the adaptive mutation operation into two parts according to their fitness function values. The part of the chromosomes with fitness function values ​​higher than or equal to the set value is retained, and the part of the chromosomes with fitness function values ​​lower than the set value is subjected to chaotic mapping. A7: Integrate the chromosomes with high fitness function values ​​and the chromosomes obtained from chaotic mapping into the current population. Determine if the current iteration count has reached the set maximum iteration count. If yes, output the chromosome with the highest current fitness function value as the optimal solution. Otherwise, increment the iteration count by 1 and return to A3.

4. The intelligent method for cold chain logistics paths based on chaotic genetic algorithm according to claim 3, characterized in that, A1 specifically includes: A1.1: Set the total number of customers in the current delivery task. A customer is randomly selected from the customer set for the initial visit. ; A1.2: Set the value of the first gene on the chromosome to "0", then use the initial access client. As the starting point of the path, visit the remaining customers in a circular order. If the remaining cargo capacity of the refrigerated truck can meet the needs of the customer, add the customer to the current path; otherwise, close the current path and add a gene value "0" to the chromosome. Continue to visit the remaining customers in the same way until all customers have been traversed and add a gene value "0" to the end of the chromosome to complete the generation of a chromosome. A1.3: Repeat A1.1-A1.2 until an initial population of the specified size is generated.

5. The intelligent method for cold chain logistics paths based on chaotic genetic algorithm according to claim 3, characterized in that, The adaptive crossover operation is specifically as follows: The first step is to randomly select a pair of parent chromosomes, generate a random number for them and compare it with the crossover probability. When the random number is not greater than the crossover probability, the genes are swapped at the same starting and ending positions of the pair of parent chromosomes. The swapped offspring chromosomes are then subjected to conflict detection. A mapping table is established based on the two swapped gene sets. The conflicting genes in the newly generated chromosomes are then transformed according to the mapping table. Gene conflict detection is completed, and the crossover-transformed offspring chromosomes are formed. The crossover probability is: （19）； In the formula, For crossover probability, The initial crossover probability; For the maximum crossover probability, To minimize the crossover probability, The maximum fitness function value. The minimum fitness function value. This is the current fitness function value. This represents the current average fitness function value of the population.

6. The intelligent method for cold chain logistics paths based on chaotic genetic algorithm according to claim 3, characterized in that, The adaptive mutation operation specifically includes: A pair of parent chromosomes is randomly selected, a random number is generated for it and compared with the mutation probability. When the random number is not greater than the mutation probability, two positions on the chromosome are randomly selected as the starting points of the inversion interval. The genes in the selected inversion interval are reversed to obtain new offspring chromosomes. However, when a gene value of "0" is encountered, it is not involved in the inversion operation. The mutation probability is: （20）； In the formula, The mutation probability, The initial mutation probability; The maximum mutation probability, This represents the minimum mutation probability.

7. The intelligent method for cold chain logistics paths based on chaotic genetic algorithm according to claim 3, characterized in that, A6 specifically includes: The formula for the Circle chaotic mapping is as follows: （21）； In the formula, The chaotic perturbation operator at the l-th iteration. The average rotational speed; To control the strength of nonlinear interactions; The chaotic perturbation is added as shown in the following formula: （22）； In the formula, The chromosome obtained by chaotic mapping. , These are the chromosomes with the largest and smallest fitness function values ​​in the population during the l-th iteration, respectively.

8. A cold chain logistics path intelligent system based on chaotic genetic algorithm, used to implement the cold chain logistics path intelligent method based on chaotic genetic algorithm as described in any one of claims 1-7, characterized in that, include: The model building module is used to construct a path optimization function model with the minimum total cost, taking into account vehicle fixed costs, vehicle transportation costs, cargo damage costs, vehicle cooling costs, dynamic carbon emission costs, and vehicle time penalty costs. The solution module is used to iteratively solve the path optimization function model with the minimum total cost using an improved chaotic genetic algorithm to obtain the optimal solution; The solution generation module is used to decode the optimal solution and form the final cold chain logistics distribution route optimization solution.

9. An electronic device, characterized in that, include: One or more processors, and a memory for storing instructions that, when executed by the one or more processors, cause the one or more processors to perform the intelligent cold chain logistics path method based on a chaotic genetic algorithm as described in any one of claims 1-7.

10. A computer program product, characterized in that, Includes a computer program or instructions that, when executed by a processor, implement the intelligent cold chain logistics path method based on a chaotic genetic algorithm as described in any one of claims 1-7.

Images