A low-carbon logistics distribution path planning module and method for fresh products based on adaptive dynamic search ant colony algorithm

Abstract

A low-carbon logistics path planning system based on an adaptive dynamic search ant colony algorithm includes: a data input module, a path planning module, when a vehicle selects a path, the vehicle gives priority to delivering to the customer with the largest transfer probability, when the random number is greater than a fixed value, the vehicle gives priority to delivering to the customer with the largest transfer factor; when the random number is less than a fixed value, the roulette method is used to select the next customer point; a data calculation module, after updating the pheromone, a new round of vehicle path planning continues. This application outputs the path planning results of logistics distribution vehicles, outputs the various target parameters of the final planned path, and draws the change curve of the total target of each iteration and the vehicle path schematic diagram. The system and method can quickly plan paths for logistics distribution vehicles, while also taking into account economic costs and environmental costs, which is conducive to logistics distribution companies to better plan vehicle paths.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of path planning, and more specifically, to a low-carbon logistics path planning method based on an adaptive dynamic search ant colony algorithm. Background Art

[0002] As people's living standards improve, food safety and quality are receiving more and more attention from consumers. Fresh products such as fruits and vegetables account for an increasing proportion of people's daily consumption needs, and more and more consumers are willing to pay a higher cost for better quality fresh products. In order to meet market demand, the cold chain logistics industry has developed rapidly. While controlling transportation costs, companies have increased customer stickiness and enhanced their competitiveness by improving the freshness of fresh products, while also reducing food waste.

[0003] There are two main problems in the existing research on vehicle routing optimization for cold chain logistics distribution:

[0004] (1) Regarding the cold chain logistics path optimization problem, on the one hand, many scholars have studied the vehicle path optimization under static networks, and on the other hand, many scholars have focused on the optimization of carbon emissions during the distribution process. In theoretical scenarios, they have studied how to reasonably arrange the vehicle distribution routes and loading sequences with the goal of minimizing the total distribution cost. They have used heuristic algorithms, exact algorithms, meta-heuristic algorithms and other intelligent algorithms to solve similar path optimization problems.

[0005] (2) Ant colony algorithm is widely used to solve vehicle routing problems due to its good robustness and parallel characteristics. However, the global search ability of the ant colony algorithm is poor. In the initial stage of the algorithm, it is easy to fall into the local optimum due to the lack of path pheromones. Its solution to the VRP problem can be further optimized and improved. Summary of the invention

[0006] The present invention provides a low-carbon logistics path planning module and method based on an adaptive dynamic search ant colony algorithm. Based on the perspectives of economic cost and environmental cost, chaos theory is introduced, and a new dynamic probability selection strategy and pheromone update strategy are designed and integrated into the traditional ant colony algorithm, and the effective information recorded in the iteration is used to guide subsequent operations.

[0007] The present invention is achieved through the following technical solutions:

[0008] A low-carbon logistics path planning system based on adaptive dynamic search ant colony algorithm includes a data input module, a path planning module, a data calculation module, an optimization adjustment module, and a data output module.

[0009] The data input module is used to read various basic data of the distribution center, the number of delivery vehicles, and customers, including: the location coordinates of the distribution center; the number of delivery vehicles and load limits; the number of customers, location coordinates, demand, service time for opening the door to unload goods, and delivery time window;

[0010] The path planning module is used to adaptively plan the vehicle path. The unit heuristic factor of the ant colony algorithm is an important factor affecting the transfer of ants from one point to another, and is an important component of the transfer probability. Different adaptive functions are assigned to the heuristic factors so that the weight ratio between the three factors changes continuously, and the transfer probability is adaptively adjusted.

[0011] When the vehicle selects a route, it will give priority to delivering to the customer with the highest transfer probability. When the random number is greater than a fixed value, the vehicle will give priority to delivering to the customer with the largest transfer factor. When the random number is less than a fixed value, the roulette method is used to select the next customer point.

[0012] The data calculation module is used to calculate the total transportation distance, total transportation cost, fixed cost and other different costs, carbon emissions, number of vehicles used, etc., so as to prepare for subsequent optimization and adjustment;

[0013] The optimization and adjustment module is used to update pheromones using the ant-week model, and the pheromone change amount is the pheromone update constant divided by the total distance when the optimal goal is achieved; the pheromone on the optimal path is volatilized and the pheromone change amount is added, and other paths that are not the optimal path only volatilize pheromones without adding other forms of pheromone supplements, and a new round of vehicle path planning is continued after the pheromones are updated;

[0014] The data output module is used to output various target parameters of the final planned path, including total transportation distance, total transportation cost, fixed cost and other different costs, carbon emissions, number of vehicles used, etc.; logistics distribution vehicle path planning results, including the path of each vehicle; change curves of the total target of vehicle paths in previous iterations; and a schematic diagram of the vehicle path of the final optimization result.

[0015] A method for a low-carbon logistics path planning system based on an adaptive dynamic search ant colony algorithm comprises the following steps:

[0016] Read basic data of distribution centers, delivery vehicles, and customers;

[0017] The initial vehicle path planning is performed based on the adaptive transfer probability of the improved ant colony algorithm and the basic data of the distribution center, the number of delivery vehicles, and the customers;

[0018] The total distance, total cost, carbon emissions and other target parameters of the initial vehicle path planning are calculated through formulas, and the optimal value of each target parameter is taken;

[0019] Based on the principle of chaos, after obtaining the optimal solution through global solution, local search is performed, and the pheromone is updated according to the newly obtained optimal solution and then optimized and adjusted;

[0020] Output the logistics distribution vehicle path planning results, output the various target parameters of the final planned path, and draw the change curve of the total goal of each iteration and the vehicle path schematic diagram.

[0021] As the best embodiment of the present invention, the basic data of the distribution center, the number of distribution vehicles, and the customers specifically include:

[0022] The location coordinates of the distribution center;

[0023] The number and load limits of delivery vehicles;

[0024] Data such as the number of customers, location coordinates, demand, service time for opening the door and unloading, and delivery time window.

[0025] As the optimal embodiment of the present invention, the adaptive transition probability includes different heuristic factors, and the heuristic factors have different adaptive importance levels. The transition probability of the traditional ant colony algorithm easily causes the algorithm to produce a local optimum. To solve this problem, a random number is introduced to select the transition probability, and a selection is made from two transition modes by generating a random number.

[0026] As the optimal embodiment of the present invention, the heuristic factor is an important influencing factor for ants to transfer from one point to another, and is an important component of the transfer probability. As the number of iterations increases, the importance of the heuristic factor also changes. If its importance is a fixed constant, it is not conducive to getting rid of the drawbacks of the original algorithm. Therefore, different adaptive functions are assigned to make the weight ratio between the three constantly change, and the transfer probability is adjusted adaptively.

[0027] As the optimal embodiment of the present invention, after the random number is selected, it is compared with another fixed value. When the random number is greater than the fixed value, the vehicle gives priority to delivering to the customer with the largest transfer factor; when the random number is less than the fixed value, the roulette method is used to select the next customer point; at the same time, the vehicle gives priority to delivering to the customer with the largest transfer probability.

[0028] As the best embodiment of the present invention, after obtaining a vehicle path, calculating its target parameters specifically includes:

[0029] Total transport distance, total transport cost, various costs such as fixed costs, carbon emissions, number of vehicles used, etc.;

[0030] After calculating the results, compare them to get the optimal goal.

[0031] As the best embodiment of the present invention, the chaotic disturbance mechanism of pheromone is updated. Specifically,

[0032] Chaotic perturbation uses the random characteristics of a pure system to generate a pseudo-random variable within a range. Adding this random variable to the update of pheromones can help escape the local optimum when searching for the optimal solution, thereby improving the search performance of the algorithm. The pheromone update adopts the ant-week model, which is specifically expressed as follows: the pheromone change is the pheromone update constant divided by the total distance to achieve the optimal goal.

[0033] As the optimal embodiment of the present invention, a chaotic variable is added, and an adjustment change coefficient is added thereto. After the pheromone on the optimal path evaporates, the pheromone change amount is added. Other paths that are not the optimal path only evaporate the pheromone without adding other forms of pheromone supplementation.

[0034] As the best embodiment of the present invention, the vehicle path result and the image output are finally optimized, specifically including:

[0035] The target parameters of the final planned route include total transportation distance, total transportation cost, fixed cost and other different costs, carbon emissions, number of vehicles used, etc.;

[0036] Logistics distribution vehicle path planning results, including the path of each vehicle;

[0037] The change curve of the overall target of the vehicle path in each iteration;

[0038] Schematic diagram of the vehicle path of the final optimization result.

[0039] Beneficial effects: The present invention is dedicated to solving the green vehicle routing problem with a soft time window in the distribution process of fresh products, and takes carbon emissions and freshness into consideration on the basis of the traditional vehicle routing problem. The total distance, total cost, carbon emissions and other target parameters of the initial vehicle routing planning are calculated through formulas, and the optimal value of each target parameter is taken; based on the principle of chaos, after the global solution is obtained, a local search is performed, and the pheromone is updated according to the newly obtained optimal solution for optimization and adjustment; the logistics distribution vehicle routing planning results are output, and the target parameters of the final planned path are output, and the change curves of the total target of each iteration and the vehicle routing diagram are drawn. The system and method can quickly plan routes for logistics distribution vehicles, while also taking into account economic costs and environmental costs, which is conducive to logistics distribution companies to better plan vehicle routes. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 This is a schematic diagram of the process of this application.

[0042] Figure 2 This is a schematic diagram of the module structure of this application. DETAILED DESCRIPTION

[0043] The following is a detailed description of an embodiment of the present invention in conjunction with the accompanying drawings: This embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation method and a specific operation process are given, but the protection scope of the present invention is not limited to the following embodiments.

[0044] like Figure 1 , 2 As shown, a low-carbon logistics path planning system based on adaptive dynamic search ant colony algorithm includes a data input module, a path planning module, a data calculation module, an optimization adjustment module, and a data output module.

[0045] The data input module is used to read various basic data of the distribution center, the number of delivery vehicles, and customers, including: the location coordinates of the distribution center; the number of delivery vehicles and load limits; the number of customers, location coordinates, demand, service time for opening the door to unload goods, and delivery time window;

[0046] The path planning module is used to adaptively plan the vehicle path. The unit heuristic factor of the ant colony algorithm is an important factor affecting the transfer of ants from one point to another, and is an important component of the transfer probability. Different adaptive functions are assigned to the heuristic factors so that the weight ratio between the three factors changes continuously, and the transfer probability is adaptively adjusted.

[0047] When the vehicle selects a route, it will give priority to delivering to the customer with the highest transfer probability. When the random number is greater than a fixed value, the vehicle will give priority to delivering to the customer with the largest transfer factor. When the random number is less than a fixed value, the roulette method is used to select the next customer point.

[0048] The data calculation module is used to calculate the total transportation distance, total transportation cost, fixed cost and other different costs, carbon emissions, number of vehicles used, etc., so as to prepare for subsequent optimization and adjustment;

[0049] The optimization and adjustment module is used to update pheromones using the ant-week model, and the pheromone change amount is the pheromone update constant divided by the total distance when the optimal goal is achieved; the pheromone on the optimal path is volatilized and the pheromone change amount is added, and other paths that are not the optimal path only volatilize pheromones without adding other forms of pheromone supplements, and a new round of vehicle path planning is continued after the pheromones are updated;

[0050] The data output module is used to output various target parameters of the final planned path, including total transportation distance, total transportation cost, fixed cost and other different costs, carbon emissions, number of vehicles used, etc.; logistics distribution vehicle path planning results, including the path of each vehicle; change curves of the total target of vehicle paths in previous iterations; and a schematic diagram of the vehicle path of the final optimization result.

[0051] A low-carbon logistics path planning method based on an adaptive dynamic search ant colony algorithm comprises the following steps:

[0052] Read the location coordinates of the distribution center;

[0053] Read the number of delivery vehicles and load limits;

[0054] Read data such as the number of customers, location coordinates, demand, service time for opening the door to unload goods, and delivery time window;

[0055] Record all coordinate points in an array.

[0056] The first ant starts from the coordinates of the distribution center.

[0057] Delete the coordinate points of the distribution center from the array.

[0058] The ant searches for the next customer point that needs to be served based on the service time window and demand of each service point.

[0059] Every time the ant passes a client point, it deletes the coordinates of that point from the array;

[0060] When an ant starts from a point, the next point it can move to must meet three conditions: the customer has not been served yet,

[0061] The service time window is reached.

[0062] Meet the vehicle's load limit,

[0063] So the next point that can be moved to is a set;

[0064] When the set is empty, the ant returns to the distribution center and starts again from the distribution center. At this time, the number of required vehicles is increased by 1, and the subsequent path is the path of the next new vehicle;

[0065] When the set is not empty, the ants can move according to the transition probability;

[0066] The transition probability should be composed of the pheromone concentration and the heuristic factor,

[0067] The pheromone concentration is specified at the beginning,

[0068] The heuristic factor is an important factor affecting the transfer of ants from one point to another, so it is determined according to the goal of the algorithm.

[0069] This embodiment uses two heuristic factors, including

[0070] The inverse of the total delivery cost of a vehicle from customer i to customer j is η ij

[0071] The inverse of the product quality loss from customer i to customer j, θ ij

[0072] The importance of the heuristic factor changes as the number of iterations increases.

[0073] Different adaptive functions are assigned to different heuristic factors so that the weight ratio among the three factors changes continuously and the transition probability is adjusted adaptively.

[0074] In this embodiment, an adaptive function is assigned to the importance of the pheromone concentration and the two heuristic factors, which is specifically expressed as follows:

[0075] Among them, iter is the number of iterations, maxiter is the maximum number of iterations,

[0076]

[0077] δ, μ and ε are the pheromone concentration and the importance of the two heuristic factors, respectively.

[0078] Introduce a random number r to determine the ant's transfer probability.

[0079] The pheromone concentration is multiplied by different heuristic factors, and their importance is raised to the square of each factor. This formula is called the transfer factor.

[0080] Choose a fixed number r 0 , compare it with the introduced random number. When the random number is greater than the fixed value, the vehicle will give priority to delivering to the customer with the largest transfer factor. When the random number is less than the fixed value, the roulette method is used to select the next customer point.

[0081] In this embodiment, the transition probability is

[0082]

[0083] Among them, τ ij is the pheromone concentration, η ij is the inverse of the total delivery cost, θ ij It is the inverse of the product quality loss. The set of other customer points that vehicle k can choose to serve after serving customer i.

[0084] This formula gives the transition probability of each point in the set:

[0085] The vehicle will prioritize delivery to customers with the highest probability of transfer.

[0086] When all customer points have been served, the ants return to the distribution center.

[0087] Every time the ant returns to the distribution center, it replaces a delivery vehicle.

[0088] When the ant returns to the distribution center after passing all customer points, all the trajectories of the ant's movement are recorded.

[0089] Calculate the number of vehicles after the ant has traversed all customers.

[0090] Calculate various targets such as fixed costs and freshness,

[0091] Calculate total distance travelled and carbon emissions.

[0092] The first vehicle path planning is the initial planning.

[0093] Each item in the initial plan is marked as the optimal target.

[0094] Release the second ant from the distribution center coordinates.

[0095] Let it continue to traverse all customer points according to the above transfer.

[0096] After the second ant traverses all customers, it also calculates various objectives.

[0097] Compare the calculated target value with the optimal target obtained last time.

[0098] If the current goal is not as suitable as the optimal goal obtained last time, the last vehicle path will be maintained and its goal will continue to be the optimal goal.

[0099] If the current goal is more suitable than the optimal goal obtained last time, the optimal path planning will be updated to the current vehicle path, and the current goal will be the optimal goal.

[0100] The pheromone update adopts the ant week model.

[0101] The pheromone change is the pheromone update constant divided by the total distance to achieve the optimal goal.

[0102] Added chaos perturbation mechanism for updating pheromones,

[0103] Add chaos variables to the pheromone update.

[0104] After the pheromone on the optimal path evaporates, the pheromone change is added.

[0105] Other paths that are not the optimal path only volatilize pheromones and do not add other forms of pheromone supplementation.

[0106] In this embodiment, the update strategy is

[0107] On the best path

[0108] τ on non-optimal path ij (t+n)=(1-ρ)τ ij (t)+λz ij

[0109] Among them, ρ is the pheromone volatility factor, λ is the coefficient for adjusting the mixed purity variable, z ij is the mixed pure variable generated by the mixed pure system.

[0110] After the pheromone is updated, the number of iterations is increased by 1, and the algorithm starts again from path planning until the maximum number of iterations is reached.

[0111] After the last iteration, the optimal goal is obtained.

[0112] List the path of each vehicle,

[0113] Draw the change curve of the total target of the vehicle path in each iteration,

[0114] Draw a schematic diagram of the vehicle path of the final optimization result.

[0115] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A low-carbon logistics path planning system based on adaptive dynamic search ant colony algorithm, characterized in that: It includes data input module, path planning module, data calculation module, optimization adjustment module and data output module. The data input module is used to read various basic data of the distribution center, the number of delivery vehicles, and the customers, including: the location coordinates of the distribution center; the number of delivery vehicles and the load limit; the number of customers, location coordinates, demand, service time for opening the door to unload goods, and the time window data for delivery; The path planning module is used to adaptively plan the vehicle path. The unit heuristic factor of the ant colony algorithm is an important factor affecting the transfer of ants from one point to another, and is an important component of the transfer probability. Different adaptive functions are assigned to the heuristic factors so that the weight ratio between the three factors changes continuously, and the transfer probability is adjusted adaptively. The adaptive function is assigned to the pheromone concentration and the importance of the two heuristic factors, which is specifically expressed as: Among them, iter is the number of iterations, maxiter is the maximum number of iterations, δ, μ and ε are the importance of pheromone concentration and two heuristic factors respectively. The pheromone concentration and different heuristic factors are multiplied, and their importance is used as the square of each factor. This formula is called the transfer factor. A random number is introduced, and the transfer probability is calculated based on the random number and the transfer factor. When a vehicle selects a route, it selects a customer for delivery based on the transfer probability. When the random number is greater than a fixed value, the vehicle will give priority to delivering to the customer with the largest transfer factor. When the random number is less than a fixed value, the roulette method is used to select the next customer point. The data calculation module is used to calculate the total transportation distance, total transportation cost, various fixed costs, carbon emissions, and the number of vehicles used, so as to prepare for subsequent optimization and adjustment; The optimization and adjustment module is used to update pheromones using the ant-week model, and the pheromone change amount is the pheromone update constant divided by the total distance when the optimal goal is achieved; the pheromone on the optimal path is volatilized and the pheromone change amount is added, and other paths that are not the optimal path only volatilize pheromones without adding other forms of pheromone supplements, and a new round of vehicle path planning is continued after the pheromones are updated; The data output module is used to output various target parameters of the final planned path, including total transportation distance, total transportation cost, various fixed costs, carbon emissions, and the number of vehicles used; logistics distribution vehicle path planning results, including the path of each vehicle; the change curve of the total target of the vehicle path in previous iterations; and the vehicle path schematic diagram of the final optimization result.

2. A method for a low-carbon logistics path planning system based on an adaptive dynamic search ant colony algorithm according to claim 1, characterized in that: The steps include: Read basic data of distribution centers, delivery vehicles, and customers; The initial vehicle path planning is performed based on the adaptive transfer probability of the improved ant colony algorithm and the basic data of the distribution center, the number of delivery vehicles, and the customers; The total distance, total cost, and carbon emissions of the initial vehicle path planning are calculated through formulas, and the optimal value of each target parameter is taken; Based on the principle of chaos, after obtaining the optimal solution through global solution, local search is performed, and the pheromone is updated according to the newly obtained optimal solution and then optimized and adjusted; Output the logistics distribution vehicle path planning results, output the various target parameters of the final planned path, and draw the change curve of the total goal of each iteration and the vehicle path schematic diagram.

3. The method of a low-carbon logistics path planning system based on an adaptive dynamic search ant colony algorithm according to claim 1 is characterized in that: The basic data of the distribution center, the number of distribution vehicles, and the customers include: The location coordinates of the distribution center; The number and load limits of delivery vehicles; The number of customers, location coordinates, demand, service time for opening the door to unload, and delivery time window data.

4. The method of a low-carbon logistics path planning system based on an adaptive dynamic search ant colony algorithm according to claim 1 is characterized in that: After obtaining a vehicle path, calculate its target parameters, including: Total transport distance, total transport cost, various fixed costs, carbon emissions, number of vehicles used; After calculating the results, compare them to get the optimal goal.

5. The method of a low-carbon logistics path planning system based on an adaptive dynamic search ant colony algorithm according to claim 1 is characterized in that: Update the chaotic perturbation mechanism of pheromones. Specifically, Chaotic perturbation uses the random characteristics of a mixed pure system to generate a pseudo-random variable within a range. Adding this random variable to the update of pheromones can make it possible to jump out of the local optimum when looking for the optimal solution during the search process, thereby improving the search performance of the algorithm. The pheromone update adopts the ant cycle model, which is specifically expressed as follows: the pheromone change is the pheromone update constant divided by the total distance to achieve the optimal goal.

6. The method of a low-carbon logistics path planning system based on an adaptive dynamic search ant colony algorithm as described in claim 1 is characterized in that: Chaotic variables are added, and an adjustment coefficient is added to them. After the pheromone on the optimal path evaporates, the pheromone change amount is added. Other paths that are not the optimal path only evaporate pheromones, and no other forms of pheromone supplementation are added.

7. The method of a low-carbon logistics path planning system based on an adaptive dynamic search ant colony algorithm as described in claim 1 is characterized in that: The final optimized vehicle path results and image output include: The target parameters of the final planned route include total transportation distance, total transportation cost, various fixed costs, carbon emissions, and number of vehicles used; Logistics distribution vehicle path planning results, including the path of each vehicle; The change curve of the overall target of the vehicle path in each iteration; Schematic diagram of the vehicle path of the final optimization result.

Images