Goods distribution path optimization method and device, computer equipment and medium

Abstract

The invention provides a cargo distribution path optimization method and device, computer equipment and a medium, and belongs to the field of path optimizing.The method comprises the steps that a vehicle path optimization model with the purpose of minimizing the total distribution cost is constructed; performing iterative solution on the vehicle path optimization model by adopting a genetic algorithm, dynamically calculating and applying a crossover probability and a mutation probability according to fitness distribution of a current population in the iterative process of the genetic algorithm, and periodically exchanging elite individuals among the populations by introducing an immigrant operator; and when the number of iterations reaches a preset number of times, the genetic algorithm stops iteration, and an optimal path is output. Therefore, when the path optimization is carried out, the search strategy can be automatically adjusted according to the search process, the population diversity is increased, the search space is further expanded, the algorithm stability is enhanced, and a high-quality fresh food logistics distribution scheme is stably obtained within reasonable time.

Description

Technical Field

[0001] This invention belongs to the field of route optimization, specifically relating to a method, apparatus, computer equipment, and medium for optimizing cargo delivery routes. Background Technology

[0002] With the rapid development of China's logistics industry, carbon emissions from road transportation have attracted widespread attention worldwide. As a crucial branch of the logistics industry, controlling and reducing carbon emissions from fresh produce distribution has become key to the industry's development. Chain convenience stores, as a major channel for fresh produce sales, play a vital role in optimizing their fresh produce delivery routes to improve efficiency, reduce logistics costs, and minimize environmental impact. Fresh produce cold chain logistics, as an important branch of the logistics industry, is still in its exploratory stage in my country. With the continuous growth in consumer demand for fresh produce, the fresh produce cold chain logistics industry is experiencing a golden age of rapid development. However, this process is also accompanied by new challenges: fresh produce cold chain transportation involves not only fuel consumption and carbon emissions from vehicle operation but also the additional energy consumption required to maintain a low-temperature environment during transportation. These factors collectively contribute to increased carbon emissions, exacerbating environmental pressure. As the fresh produce retail market continues to expand, the efficiency and cost control of fresh produce distribution for chain convenience stores, as important sales terminals, are becoming increasingly critical. Meanwhile, the carbon emissions from road transportation have received widespread attention, and due to its energy-intensive nature, carbon emission control during the cold chain logistics distribution process is particularly important. Therefore, optimizing delivery routes to reduce operating costs and carbon emissions while meeting the high-frequency, small-batch, and time-sensitive delivery needs of convenience stores has become an important research topic.

[0003] Existing research on route optimization for cold chain logistics vehicles, when extended to practical applications, falls into the NP-hard category for route optimization problems involving multiple distribution centers, multiple vehicle types (heterogeneous fleets), time windows, and complex constraints (such as urban traffic restrictions during certain hours). Traditional exact algorithms struggle to solve large-scale instances, while basic heuristic algorithms, such as standard genetic algorithms, are prone to getting trapped in local optima, exhibiting slow convergence, or premature convergence when solving such complex problems. These methods fail to guarantee obtaining high-quality solutions within a reasonable timeframe, resulting in low-quality optimized delivery routes.

[0004] Therefore, there is an urgent need for a fresh food logistics route optimization method that can reliably obtain high-quality delivery solutions. Summary of the Invention

[0005] To address the aforementioned problems, this invention provides a method, apparatus, computer equipment, and medium for optimizing cargo delivery routes.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A method for optimizing cargo delivery routes, the method comprising: Obtain the location information of the target distribution center and its corresponding multiple convenience stores; A vehicle routing optimization model is constructed to minimize the total delivery cost, starting from a target distribution center, reaching multiple convenience stores, and then returning to the target distribution center. The total delivery cost includes vehicle fixed costs, vehicle transportation costs, refrigeration costs, time window penalty costs for reaching convenience stores, and carbon emission costs. The model must satisfy constraints including route constraints, load constraints, and time constraints, with the time constraints including traffic restriction constraints during peak hours. The vehicle path optimization model is solved iteratively using a genetic algorithm. During the iteration process of the genetic algorithm, the crossover probability and mutation probability are dynamically calculated and applied based on a preset formula according to the fitness distribution of the current population. An immigration operator is introduced to periodically exchange elite individuals among the population. When the number of iterations reaches a preset number, the genetic algorithm stops iterating and outputs the optimal path.

[0007] Optionally, the time window penalty cost for arriving at the convenience store is calculated based on a preset static time window structure, which predefines an acceptable service time window, a time period for allowing early arrival penalties, a time period for allowing late arrival penalties, and a hard final time boundary for when service is unavailable for each convenience store.

[0008] Optionally, the vehicle type in the vehicle route optimization model includes ambient temperature vehicles for delivering ambient temperature goods and refrigerated vehicles for delivering refrigerated goods. The carbon emission cost includes the carbon emission cost generated by vehicle fuel consumption during transportation, and the carbon emission cost generated by the operation of the refrigeration equipment in the refrigerated vehicle.

[0009] Optionally, the formula for calculating the crossover probability is: ; The formula for calculating the mutation probability is: ; in, The initial crossover probability is set as follows. The initial mutation probability is set. For fitness, and This represents the maximum and average fitness values ​​in the contemporary population. For population index labeling, The total population. For the number of iterations, It is a very small positive number.

[0010] Optionally, the periodic exchange of elite individuals among populations includes: Replace the individual with the highest fitness in the current population with the individual with the lowest fitness in another independently generated or specified population.

[0011] A cargo delivery route optimization device, the device comprising: The acquisition module is used to acquire the location information of the target distribution center and its multiple corresponding convenience stores; The construction module is used to build a vehicle route optimization model that minimizes the total delivery cost from the target distribution center, through multiple convenience stores, and back to the target distribution center. The total delivery cost includes vehicle fixed cost, vehicle transportation cost, refrigeration cost, time window penalty cost for reaching the convenience store, and carbon emission cost. The constraints that the model must satisfy include route constraints, load constraints, and time constraints, with the time constraints including traffic restriction constraints during peak hours. The optimization module is used to iteratively solve the vehicle path optimization model using a genetic algorithm. During the iteration of the genetic algorithm, the crossover probability and mutation probability are dynamically calculated and applied based on a preset formula according to the fitness distribution of the current population. An immigration operator is introduced to periodically exchange elite individuals among the population. The output module is used to stop the genetic algorithm from iterating and output the optimal path when the number of iterations reaches a preset number.

[0012] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned method for optimizing cargo delivery routes.

[0013] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the aforementioned method for optimizing a goods delivery route.

[0014] The cargo delivery route optimization method provided by this invention has the following beneficial effects: The vehicle routing optimization model constructed in this invention aims to minimize the total delivery cost, comprehensively covering all aspects of vehicle costs. It accurately weighs the impact of various cost factors on the delivery plan, ensuring a good balance between economy and environmental friendliness, thus laying the foundation for high-quality delivery. When using a genetic algorithm for iterative solution, the crossover and mutation probabilities are dynamically calculated based on the fitness distribution of the current population. This allows for automatic adjustment of the search strategy according to the search progress, avoiding getting trapped in local optima and improving search efficiency and quality. Introducing a migration operator to periodically exchange elite individuals increases population diversity, further expands the search space, and enhances algorithm stability. Finally, after a preset number of iterations, the optimal path is output, stably obtaining a high-quality fresh produce logistics delivery plan within a reasonable time, effectively meeting practical needs. Attached Figure Description

[0015] To more clearly illustrate the embodiments and design schemes of the present invention, the accompanying drawings required for this embodiment will be briefly described below. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart illustrating a goods delivery route optimization method according to an exemplary embodiment of the present invention.

[0017] Figure 2 This is a schematic diagram of a goods delivery route problem provided by the present invention according to an exemplary embodiment.

[0018] Figure 3 This is a schematic diagram of a time window provided by the present invention according to an exemplary embodiment.

[0019] Figure 4 This is a schematic diagram of a PMX crossover method provided by the present invention according to an exemplary embodiment.

[0020] Figure 5 This is a schematic flowchart of a mutation operation provided by the present invention according to an exemplary embodiment.

[0021] Figure 6 This is a schematic diagram of a genetic algorithm flow provided by the present invention according to an exemplary embodiment.

[0022] Figure 7 This is a block diagram of a goods delivery route optimization device provided by the present invention according to an exemplary embodiment. Detailed Implementation

[0023] To enable those skilled in the art to better understand and implement the technical solutions of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0024] The technical solutions provided by the various embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0025] First, this invention provides a method for optimizing cargo delivery routes, specifically as follows: Figure 1 As shown, it includes the following steps: S101. Obtain the location information of the target distribution center and its corresponding multiple convenience stores, and construct a vehicle route optimization model with the goal of minimizing the total delivery cost.

[0026] In this invention, delivery vehicles need to depart from a distribution center, deliver goods to all convenience stores, and then return to the same distribution center. Therefore, this invention requires constructing a vehicle routing optimization model that minimizes the total delivery cost, starting from a target distribution center, delivering goods to multiple convenience stores, and then returning to the target distribution center. This total delivery cost includes vehicle fixed costs, vehicle transportation costs, refrigeration costs, time window penalty costs, and carbon emission costs. The model must satisfy constraints including route constraints, load constraints, and time constraints, with the time constraints including traffic restrictions during peak hours.

[0027] In this step, the model constructed is a multi-distribution-center, multi-vehicle chain convenience store fresh food route optimization problem (CETR-MDMVRPTW) considering carbon emissions, and taking into account factors such as time windows and traffic restrictions. The CETR-MDMVRPTW problem studied in this invention can be specifically described as follows: there are multiple distribution centers, the number of convenience stores is known, and the geographical locations of both are known. The order demand of each convenience store is known. Each distribution center, in cooperation with a leasing company, rents one set of refrigerated trucks and one set of ambient temperature trucks, with different and known load capacities. Different vehicle types have different energy consumption. Items with the same transport temperature should be placed in the same vehicle type. The vehicle load must not exceed the maximum load limit of the vehicle. The vehicle speed is fixed. At certain times, certain roads are closed to traffic due to road maintenance, traffic control, traffic congestion, etc. Under these conditions, the optimal delivery route should be selected to provide services to the convenience stores within their required time windows as much as possible. Otherwise, a certain waiting or penalty cost will be charged. The goal is to rationally plan vehicle routes to minimize the overall delivery cost while considering carbon emissions. A schematic diagram of this problem is shown below. Figure 2 As shown.

[0028] To simplify the calculations, the model includes the following assumptions: (1) The transported products are divided into two categories: the first category is room temperature goods, which are transported by room temperature vehicles, and the second category is refrigerated goods, which are transported by refrigerated vehicles.

[0029] (2) All delivery vehicles depart from the distribution center and return to the original distribution center.

[0030] (3) The distribution center has enough delivery vehicles to complete the delivery tasks.

[0031] (4) This invention only considers the simple delivery situation and does not include the pickup task.

[0032] (5) All relevant information about convenience stores is known, including demand, location, product cost and time window.

[0033] (6) The model does not consider the influence of road surface type, including road type, road surface slope, vehicle acceleration and deceleration, etc.

[0034] (7) The vehicle travels at a fixed speed during the transportation process.

[0035] During vehicle use, certain fixed costs are incurred, such as worker wages and rental fees. For ease of calculation, this invention assumes that these costs are only related to the number of vehicles used in the transportation task, and that vehicle fixed costs... It can be represented as:

[0036] .

[0037] Vehicle transportation costs primarily refer to the expenses that vary depending on the travel conditions during transportation. These costs are typically related to the vehicle's load capacity, the distance traveled, and the toll fees involved. To simplify calculations and focus on core factors, this invention will primarily explore the impact of travel distance on vehicle transportation costs, while temporarily ignoring other factors. It can be represented as:

[0038] .

[0039] Fresh produce, due to its high perishability, requires stringent temperature control during transportation, necessitating the use of refrigerated trucks for delivery. However, during operation, refrigerated trucks must continuously consume energy to maintain the low-temperature environment due to temperature differences and heat conduction between the inside and outside of the truck, thus incurring refrigeration costs. It can be represented as:

[0040] .

[0041] In actual logistics and delivery processes, delivery vehicles often struggle to arrive on time due to uncontrollable factors such as traffic control, road congestion, and unpredictable weather. Especially when delivery times exceed the convenience store's designated time window, time window penalty costs are incurred, becoming a significant cost factor in the delivery process. This invention employs a soft time window to penalize vehicles; if a vehicle is outside the designated time window... Arriving at the convenience store within the specified time frame will not incur any penalty costs if the vehicle is within the specified time frame. and Arriving at the convenience store within the designated timeframe will incur corresponding penalty costs, as illustrated in the time window diagram. Figure 3 As shown.

[0042] Time window penalty cost It can be represented as: ; .

[0043] During vehicle delivery, carbon dioxide emissions primarily originate from the combustion of engine fuel. However, numerous factors influence vehicle fuel consumption, such as vehicle design, road conditions, driving distance, climate conditions, and load. To simplify the study, this invention focuses on the impact of three key factors—driving distance, load, and driving speed—on fuel consumption and carbon emissions, temporarily ignoring other secondary factors. Assume the vehicle's maximum load capacity is... The vehicle's current cargo capacity is The fuel consumption per unit distance when the vehicle is fully loaded is The fuel consumption per unit distance when the vehicle is unloaded is .

[0044] Therefore, fuel consumption per unit distance It can be represented as: ; Therefore, during the delivery process, if from a convenience store transport Goods delivered to convenience stores Therefore, the carbon emissions generated during this journey can be expressed as: ; in, This represents the carbon dioxide emission coefficient. For convenience stores to convenience store The distance.

[0045] The carbon dioxide emissions from refrigeration equipment during delivery are also related to the vehicle's travel distance and load capacity. During delivery, if the vehicle is moving from a convenience store... transport Goods delivered to convenience stores Therefore, the carbon emissions generated during this journey can be expressed as:

[0046] ; in, This represents the carbon dioxide emission coefficient generated by refrigeration per unit weight of goods transported per unit distance. For convenience stores to convenience store The distance.

[0047] Carbon emission costs It can be represented as: .

[0048] Based on the above analysis, this invention constructs a mathematical model for the CETR-MDMVRPTW problem, with the objective of minimizing the sum of vehicle fixed costs, vehicle transportation costs, refrigeration costs, time window penalty costs, and carbon emission costs in the convenience store fresh produce delivery process. The specific model is as follows: (1) The objective function of the model includes: .

[0049] (2) Model constraints, including: 1) Line-related constraints: ; This constraint means that each convenience store can only be served by a room-temperature vehicle once.

[0050] ; This constraint means that each convenience store can only be served by a refrigerated truck once.

[0051] ; ; This constraint means that the vehicle must proceed to the next node after completing its service at a convenience store.

[0052] ; This constraint means that all vehicles leaving the distribution center eventually return to the distribution center.

[0053] ; This constraint means that vehicles are not allowed to return to the distribution center without serving a convenience store after they have departed.

[0054] ; This constraint means eliminating sub-loops in the vehicle delivery route.

[0055] ; ; This constraint means that each convenience store can only be served by one distribution center.

[0056] 2) Load-related constraints: ; This constraint represents the load limit for vehicles operating at normal temperatures.

[0057] ; This constraint represents the load limit for refrigerated trucks.

[0058] 3) Time constraints: ; This constraint represents a time window constraint.

[0059] ; This constraint represents the time constraint for road traffic restrictions.

[0060] ; This constraint indicates that the delivery process is continuous.

[0061] S102. The genetic algorithm is used to iteratively solve the vehicle path optimization model. During the iteration process of the genetic algorithm, the crossover probability and mutation probability are dynamically calculated and applied based on the fitness distribution of the current population and a preset formula. An immigration operator is introduced to periodically exchange elite individuals among the population.

[0062] This step uses a genetic algorithm (GA) combined with adaptive mechanisms to optimize the vehicle routing optimization model constructed in the previous steps.

[0063] Based on the needs of model construction, the set of models and their corresponding descriptions are shown in Table 1 below: Table 1. Set of models and their corresponding descriptions The model parameters and their corresponding descriptions are shown in Table 2 below: Table 2. Model parameters and corresponding descriptions In genetic algorithms, the fitness function plays a crucial role. It is not only a quantitative standard for measuring an individual's survival ability in a specific environment, but also a key indicator for evaluating the quality of potential solutions. A higher fitness function value generally means that the individual is more competitive in the environment, and thus has a greater chance of being selected for the next generation. The fitness function is often constructed by cleverly transforming the objective function to ensure that the algorithm accurately reflects the essence of the problem and the search direction. The fitness function of this invention is shown below:

[0064] ; in, Represents the fitness function. Representing the The target value for each chromosome, It is a positive number in the interval (0, 1), mainly to prevent the extreme case of the denominator being 0, which would cause the search direction to deviate and improve the search efficiency of the algorithm.

[0065] This invention uses Stochastic Universal Sampling (SUS) as the selection mechanism to select a certain proportion of individuals from the current population based on their fitness values ​​to enter the next generation. Its core idea is to allocate a roulette wheel partition proportional to each individual's fitness. Then, several fixed points are placed at even intervals, with each fixed point corresponding to one selected individual. This method avoids some problems in traditional roulette wheel selection, such as over-selection of high-fitness individuals or complete neglect of low-fitness individuals. The specific operational process of stochastic universal sampling is as follows:

[0066] Step 1: Calculate the population size and represent it as... ; Step 2: Calculate the cumulative sum of the population fitness values. ; Step 3: Calculate the distance between equally spaced selected points. The number of individuals to be selected; Step 4: Determine the location of each equally spaced selection point. A random number between 0 and 1; Step 5: Determine the selected individuals, if... , Individual Selected; Step 6: Randomly shuffle the selected individuals. To ensure the randomness and fairness of the selection, the original order of individuals selected based on fitness is shuffled by randomly reordering them.

[0067] This invention selects the classic PMX crossover operator and optimizes and improves it. By combining an adaptive mechanism, the algorithm can dynamically adjust the parameters of the crossover operator to better adapt to the problem-solving process. The specific crossover steps are as follows:

[0068] Step 1: Adaptive Crossover Probability Calculation: This invention employs adaptive crossover probability... Select the individuals to be cross-linked. and This represents the maximum and average fitness values ​​in the contemporary population. It is a very small positive number, to prevent the group from and Equal values ​​ensure that the crossover probability is not zero, guaranteeing the possibility of changes in superior individuals in the early stages. This allows the crossover probability to adjust as fitness changes, preventing it from getting trapped in a local optimum. The calculation formula is as follows:

[0069] .

[0070] Step 2: Select the crossover point: If the calculated crossover probability is... If the result is greater than or equal to a random number, then crossover is performed on this pair of chromosomes. Two random positions are selected as crossover points, determining a crossover segment. If the two crossover segments are equal, then crossover is not performed.

[0071] Step 3: Exchange crossover segments: In two parent individuals, exchange the genes in these two segments according to the crossover point; Step 4: Handling Duplicate Genes: After gene swapping, duplicate genes may be generated in some individuals. To solve this problem, the function checks each individual for duplicate genes after the gene swap. If a duplicate gene is found, the function will find the duplicate gene at a location other than the swap point and replace it with the original gene at that location, ensuring that the genes in each individual are unique.

[0072] Step 5: Looping Operation: Iterate through the selected chromosomes, processing one pair at a time, until all chromosomes have been processed or only one remains (if the total number of chromosomes is odd). If the total number of chromosomes is odd, the last chromosome does not participate in the crossover operation;

[0073] like Figure 4 As shown in the figure, this is an example of the PMX crossover method, which effectively maintains the uniqueness and order of genes, helps to maintain population diversity, and avoids the generation of illegal solutions.

[0074] The specific process and schematic diagram of the mutation operation in this invention are as follows: Figure 5 As shown: Step 1: Adaptive Mutation Probability Calculation: This invention employs adaptive mutation probability... Select the individuals to be mutated. and This represents the maximum and average fitness values ​​in the contemporary population. It is a very small positive number, to prevent the group from and Equal values ​​ensure that the crossover probability is not zero, guaranteeing the possibility of changes in superior individuals in the early stages. This allows the mutation probability to adjust with changes in fitness, preventing it from getting trapped in local optima. The number of iterations is factored into the selection of the mutation probability. The mutation probability is related to the number of mutations, and the calculation formula is as follows:

[0075] .

[0076] Step 2: Randomly select two non-zero mutation points on the parent's chromosome; Step 3: Swap the positions of these two mutation points to obtain the mutated offspring.

[0077] In genetic algorithms, the migration operator is an operation used to maintain population diversity. During the algorithm's iteration process, the migration operator randomly selects a subset of individuals and replaces them with new individuals to ensure population diversity and prevent the algorithm from getting trapped in local optima.

[0078] In this invention, an elite retention strategy is employed. At the end of each evolutionary cycle, it identifies and records the individuals with the highest and lowest fitness in each population; these individuals are typically referred to as "elites." The core of the elite retention strategy lies in the fact that by preserving these superior individuals, we can replace the worst-performing chromosome in the target population with the best chromosome from the source population during subsequent migration processes. This strategy not only maximizes the preservation of high-quality genes within the population but also promotes the exchange and fusion of superior genes between populations. In this way, the elite retention strategy ensures that the population always contains high-quality solutions, thereby significantly improving the algorithm's convergence speed and global search performance.

[0079] Based on the above description, the genetic algorithm process used in this invention is as follows: Figure 6As shown, after parameter initialization, the algorithm enters a multi-stage parallel computation phase, performing calculations for populations 1, 2, and up to N. For each population, the fitness of each individual is first calculated. Then, the SUS (Random Universal Sampling) method is used for individual selection, followed by adaptive crossover, where selected individuals exchange genetic information with a certain probability. Adaptive mutation is then performed, randomly altering individual genes to increase diversity. The processed individuals are then reinserted into the population to form a new generation. After each independent population completes one generation of evolution, a migration operator is used to replace the worst individual in one population with the best individual in another, achieving information exchange and co-evolution between populations. Finally, the best individuals from all populations are selected to form an elite population, and it is determined whether preset termination conditions, such as the maximum number of iterations or a fitness threshold, have been met. If not, the algorithm returns to continue the above evolutionary operations; if the termination conditions are met, the final result is output, and the algorithm terminates.

[0080] S103. When the number of iterations reaches the preset number, the genetic algorithm stops iterating and outputs the optimal path.

[0081] The preset number of iterations is the maximum number of iterations. When the maximum number of iterations is reached, the output path scheme will be taken as the optimal path.

[0082] Using the above method, the first constructed vehicle routing optimization model aims to minimize the total delivery cost, comprehensively covering all aspects of vehicle costs. It accurately weighs the impact of various cost factors on the delivery plan, ensuring that the generated plan achieves a good balance between economy and environmental friendliness, laying the foundation for high-quality delivery. When using a genetic algorithm for iterative solution, the crossover and mutation probabilities are dynamically calculated based on the fitness distribution of the current population. This allows for automatic adjustment of the search strategy according to the search progress, avoiding getting trapped in local optima and improving search efficiency and quality. Introducing a migration operator to periodically exchange elite individuals increases population diversity, further expands the search space, and enhances algorithm stability. Finally, after a preset number of iterations, the optimal path is output, stably obtaining a high-quality fresh produce logistics delivery plan within a reasonable time, effectively meeting practical needs.

[0083] In one embodiment, the effectiveness of the above algorithm was also verified. The first set of examples was a small-scale example, containing 2 distribution centers and 12 chain convenience stores; the second set of examples contained 2 distribution centers and 24 chain convenience stores; the third set of examples contained 2 distribution centers and 48 chain convenience stores; and the fourth set of examples contained 3 distribution centers and 90 chain convenience stores. The location information, product demand, service time windows, and other information of all convenience stores were known.

[0084] When using CPLEX software, if results cannot be output within 2 hours, the algorithm is considered unusable for solving that example. When using MATLAB software, the values ​​of the parameters in the genetic algorithm are as follows: The maximum number of iterations was set to 100, and each case was repeated 5 times. The average value was taken as the final result. The solution results are shown in Table 3.

[0085] Table 3 Comparison of results for cases of different sizes Where h1 represents the number of vehicles with normal temperature, h2 represents the number of vehicles with refrigerated conditions, GAP1 = (average total cost of the basic genetic algorithm - total cost of CPLEX) / total cost of CPLEX * 100%, GAP2 = (average total cost of the improved genetic algorithm - total cost of CPLEX) / total cost of CPLEX * 100%.

[0086] Based on the above results, it can be seen that the genetic algorithm proposed in this invention is necessary and efficient.

[0087] Secondly, the present invention also provides a cargo delivery route optimization device, such as... Figure 7 As shown, it includes: The acquisition module 201 is used to acquire the location information of the target distribution center and its corresponding multiple convenience stores.

[0088] Module 202 is used to construct a vehicle route optimization model that minimizes the total delivery cost from the target distribution center, through multiple convenience stores, and back to the target distribution center. The total delivery cost includes vehicle fixed cost, vehicle transportation cost, refrigeration cost, time window penalty cost for reaching the convenience store, and carbon emission cost. The constraints that the model needs to satisfy include route constraints, load constraints, and time constraints, with the time constraints including traffic restriction constraints during peak hours.

[0089] The optimization module 203 is used to iteratively solve the vehicle path optimization model using a genetic algorithm. During the iteration process of the genetic algorithm, the crossover probability and mutation probability are dynamically calculated and applied based on a preset formula according to the fitness distribution of the current population. An immigration operator is introduced to periodically exchange elite individuals among the population.

[0090] The output module 204 is used to stop the genetic algorithm from iterating and output the optimal path when the number of iterations reaches a preset number.

[0091] By employing the above method, a vehicle routing optimization model is constructed with the goal of minimizing total delivery costs. This model comprehensively covers all aspects of vehicle costs, accurately weighing the impact of various cost factors on the delivery plan. This ensures that the generated plan achieves a good balance between economy and environmental friendliness, laying the foundation for high-quality delivery. When using a genetic algorithm for iterative solution, the crossover and mutation probabilities are dynamically calculated based on the fitness distribution of the current population. This allows for automatic adjustment of the search strategy according to the search progress, avoiding getting trapped in local optima and improving search efficiency and quality. Introducing a migration operator to periodically exchange elite individuals increases population diversity, further expands the search space, and enhances algorithm stability. Finally, after a preset number of iterations, the optimal path is output, stably obtaining a high-quality fresh produce logistics delivery plan within a reasonable timeframe, effectively meeting practical needs.

[0092] The present invention also provides a computer-readable storage medium storing a computer program that can be used to execute the above-described... Figure 1 The steps of the provided method for optimizing cargo delivery routes.

[0093] This invention also provides a computer device. At the hardware level, the computer device includes a processor, an internal bus, a network interface, memory, and non-volatile memory, and may also include other hardware required for various operations. The processor reads the corresponding computer program from the non-volatile memory into memory and then executes it to achieve the above-mentioned functions. Figure 1 The steps of the provided method for optimizing cargo delivery routes.

[0094] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0095] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0096] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0097] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0098] It should be noted that the specific embodiments described above enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. Therefore, although the present invention has been described in detail in this specification, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the present invention; and all technical solutions and improvements that do not depart from the spirit and scope of the present invention are covered within the protection scope of the patent of the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for optimizing cargo delivery routes, characterized in that, The method includes: Obtain the location information of the target distribution center and its corresponding multiple convenience stores; A vehicle routing optimization model is constructed to minimize the total delivery cost, starting from a target distribution center, reaching multiple convenience stores, and then returning to the target distribution center. The total delivery cost includes vehicle fixed costs, vehicle transportation costs, refrigeration costs, time window penalty costs for reaching convenience stores, and carbon emission costs. The model must satisfy constraints including route constraints, load constraints, and time constraints, with the time constraints including traffic restriction constraints during peak hours. The vehicle path optimization model is solved iteratively using a genetic algorithm. During the iteration process of the genetic algorithm, the crossover probability and mutation probability are dynamically calculated and applied based on a preset formula according to the fitness distribution of the current population. An immigration operator is introduced to periodically exchange elite individuals among the population. When the number of iterations reaches a preset number, the genetic algorithm stops iterating and outputs the optimal path.

2. The method according to claim 1, characterized in that, The time window penalty cost for arriving at the convenience store is calculated based on a preset static time window structure. The static time window structure predefines an acceptable service time window, a period of time during which early arrival penalty is allowed, a period of time during which late arrival penalty is allowed, and a final time boundary where service is not possible for each convenience store.

3. The method according to claim 1, characterized in that, The vehicle route optimization model includes vehicles for delivering ambient temperature goods and refrigerated vehicles for delivering refrigerated goods. The carbon emission cost includes the carbon emission cost generated by vehicle fuel consumption during transportation, as well as the carbon emission cost generated by the operation of refrigeration equipment in refrigerated vehicles.

4. The method according to claim 1, characterized in that, The formula for calculating the crossover probability is: ； The formula for calculating the mutation probability is: ； in, The initial crossover probability, The initial mutation probability is set. For fitness, and This represents the maximum and average fitness values ​​in the contemporary population. For population index labeling, The total population. For the number of iterations, It is a very small positive number.

5. The method according to claim 1, characterized in that, The periodic exchange of elite individuals among populations includes: Replace the individual with the highest fitness in the current population with the individual with the lowest fitness in another independently generated or specified population.

6. A cargo delivery route optimization device, characterized in that, The device includes: The acquisition module is used to acquire the location information of the target distribution center and its multiple corresponding convenience stores; The construction module is used to build a vehicle route optimization model that minimizes the total delivery cost from the target distribution center, through multiple convenience stores, and back to the target distribution center. The total delivery cost includes vehicle fixed cost, vehicle transportation cost, refrigeration cost, time window penalty cost for reaching the convenience store, and carbon emission cost. The constraints that the model must satisfy include route constraints, load constraints, and time constraints, with the time constraints including traffic restriction constraints during peak hours. The optimization module is used to iteratively solve the vehicle path optimization model using a genetic algorithm. During the iteration of the genetic algorithm, the crossover probability and mutation probability are dynamically calculated and applied based on a preset formula according to the fitness distribution of the current population. An immigration operator is introduced to periodically exchange elite individuals among the population. The output module is used to stop the genetic algorithm from iterating and output the optimal path when the number of iterations reaches a preset number.

7. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the method described in any one of claims 1 to 5.

8. A computer device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method described in any one of claims 1 to 5.

Images