A Low-Carbon Logistics Distribution Route Planning Method Based on Multi-Objective Optimization

Abstract

This invention relates to the field of logistics distribution route planning, and more particularly to a low-carbon logistics distribution route planning method based on multi-objective optimization. The method includes: constructing a directed graph based on the logistics distribution network and obtaining the travel distance, travel time, and transportation economic cost of each route segment; calculating the carbon emissions of each route segment based on its travel distance and travel time; constructing a multi-objective function optimization model based on the travel time, carbon emissions, and transportation economic cost of each route segment; and using an improved non-dominated sorting genetic algorithm based on the multi-objective function optimization model to generate preliminary distribution route schemes, and introducing a service time window penalty mechanism to identify locally optimized candidate route segments and perform route optimization to obtain the optimal distribution route scheme. This method solves the problems of traditional logistics distribution route planning methods, such as insufficient integration of actual traffic conditions, unreasonable carbon emission structure design, lack of customer service time window constraints, and inadequate objective optimization strategies.

Description

Technical Field

[0001] This invention relates to the field of logistics distribution route planning, and in particular to a low-carbon logistics distribution route planning method based on multi-objective optimization. Background Technology

[0002] With the rapid development of e-commerce, instant delivery, and urban logistics, the importance of logistics and distribution in modern urban operations is increasing. At the same time, urban traffic congestion is becoming increasingly severe, and energy consumption and carbon emissions during transportation are becoming more prominent, becoming key factors restricting the development of green logistics. According to relevant research data, urban road traffic has become one of the main sources of urban carbon emissions. Delivery vehicles, due to their high frequency, short distances, and frequent stops, have a much higher carbon emission intensity per unit time than ordinary vehicles. Therefore, how to effectively reduce carbon emissions during logistics and distribution while meeting user service needs has become one of the core issues in current research on green logistics route optimization.

[0003] Traditional logistics route planning methods mostly focus on minimizing distance or cost, neglecting the nonlinear carbon emission behavior of vehicles under different traffic conditions. For example, during peak hours or in congested areas, vehicles may frequently start and stop, idling and waiting, which significantly increases carbon emissions per unit distance. However, existing models typically simplify carbon emissions to the product of distance and a fixed emission factor, failing to accurately reflect real-world emission levels under complex urban road conditions. This leads to low-carbon optimization results deviating from reality and weakening the practical feasibility of the proposed solutions.

[0004] On the other hand, although multi-objective optimization algorithms have been widely applied to path planning problems, such as Particle Swarm Optimization (PSO), Genetic Algorithm (GA), and Ant Colony Optimization (ACO), their adaptability in low-carbon delivery scenarios remains insufficient. Specifically: First, traditional algorithms often focus on single-objective or bi-objective optimization, making it difficult to achieve an effective balance between delivery timeliness, carbon emissions, and economic costs; second, they lack the comprehensive perception of dynamic changes in urban traffic and customer time window constraints, easily leading to infeasible paths such as time window breaches and service delays.

[0005] In recent years, in response to the demand for green and low-carbon delivery, some studies have attempted to introduce carbon emission models and heuristic algorithms to jointly optimize route selection. However, most of these studies have failed to effectively integrate traffic condition changes, service timeliness requirements, and multi-objective collaborative optimization mechanisms. For example, some algorithms only correct carbon emission assessments after route selection or simply add carbon emission term weights to the objective function, failing to dynamically perceive the spatiotemporal distribution characteristics of carbon emission structure.

[0006] Therefore, there is an urgent need for a route planning method that can fully integrate actual traffic conditions, carbon emission structure, customer service time windows and multi-objective optimization strategies to improve the overall rationality of route scheduling and the level of local service refinement, and help build a truly "low-carbon, efficient and intelligent" urban logistics and distribution system. Summary of the Invention

[0007] This invention provides a low-carbon logistics distribution route planning method based on multi-objective optimization to solve the technical problems of traditional logistics distribution route planning methods, such as insufficient integration of actual traffic conditions, unreasonable carbon emission structure design, lack of customer service time window constraints, and inadequate objective optimization strategies.

[0008] The present invention provides a low-carbon logistics distribution route planning method based on multi-objective optimization, which specifically includes the following technical solutions:

[0009] A low-carbon logistics distribution route planning method based on multi-objective optimization includes the following steps:

[0010] S1. Construct a directed graph based on the logistics distribution network and obtain the travel distance, travel time and transportation economic cost of the path segments; based on the travel distance and travel time of the path segments, construct a traffic state perception-based carbon emission model that integrates travel distance and idling time, and calculate the carbon emissions of the path segments; based on the travel time, carbon emissions and transportation economic cost of the path segments, construct a multi-objective function optimization model.

[0011] S2. Based on the multi-objective function optimization model, an improved non-dominated sorting genetic algorithm is used to generate a preliminary delivery route plan. Based on the preliminary delivery route plan, a service time window penalty mechanism is introduced to identify local optimization candidate path segments and perform path optimization to obtain the optimal delivery route plan.

[0012] Preferably, S1 specifically includes:

[0013] Based on the travel time of a route segment, an idling ratio coefficient is introduced to obtain the idling time of the route segment. Based on the idling time and travel distance of the route segment, and combined with the carbon emission structure ratio of travel distance and idling time of urban delivery vehicles in actual operation, a traffic state perception carbon emission model based on the fusion of travel distance and idling time is constructed, and the carbon emission of the route segment is obtained.

[0014] Preferably, S1 specifically includes:

[0015] Based on the directed graph, candidate delivery route schemes are obtained; the total delivery time, total carbon emissions, and total transportation economic cost of the candidate delivery route schemes are calculated and minimized respectively, and a multi-objective function optimization model is constructed to obtain the objective functions of total delivery time, total carbon emissions, and total transportation economic cost.

[0016] Preferably, S2 specifically includes:

[0017] In the implementation of the improved non-dominated sorting genetic algorithm, each individual corresponds to a candidate delivery route scheme. The multi-objective function optimization model is combined with the weight coefficients of each objective function to calculate the crowding distance of the individual. Based on the sorted crowding distance, the preliminary delivery route scheme is obtained.

[0018] Preferably, S2 specifically includes:

[0019] Based on the preliminary delivery route plan, the actual arrival time of the nodes is calculated. Based on the actual arrival time of the nodes and combined with the preset service time window, the path segments in the preliminary delivery route plan that violate the service time window are identified: when there is at least one individual whose complete delivery route in the preliminary delivery route plan meets the service time window, the individual with the largest congestion distance is selected from the individuals in the preliminary delivery route plan that meet the service time window as the optimal delivery route plan; otherwise, the path segments that violate the service time window are identified within each individual in the preliminary delivery route plan, and the path segments are marked as local optimization candidate path segments for the corresponding individual.

[0020] Preferably, S2 specifically includes:

[0021] The service time window penalty mechanism calculates the time difference based on the actual arrival time of the node and the preset service time window; based on the time difference, a path segment selection probability function is introduced to optimize local optimization candidate path segments.

[0022] Preferably, S2 specifically includes:

[0023] The path segment selection probability function is based on the travel time and carbon emissions of local optimization candidate path segments. It introduces guiding weights and time window penalty factors to calculate the probability from the upstream node of the local optimization candidate path segment to each passable adjacent node, and selects the node with the highest probability to replace the downstream node of the original path segment. Path segments that violate the service time window are adjusted to generate new individuals.

[0024] Preferably, S2 specifically includes:

[0025] Calculate the updated congestion distance based on the new individual, and select the individual with the largest updated congestion distance as the optimal delivery route solution output.

[0026] The beneficial effects of the technical solution of the present invention are:

[0027] 1. The traffic state perception carbon emission model constructed in this invention, which integrates travel distance and idling time, effectively reflects the real carbon emission level of delivery routes under different traffic conditions by combining travel distance and idling time factors of route segments. Compared with the traditional method of estimating carbon emissions based solely on distance, it further considers the contribution of idling emissions under congestion conditions, making carbon emission calculations closer to actual operating scenarios. This improves the accuracy and practicality of low-carbon optimization goals and helps to achieve precise and controllable green logistics route planning.

[0028] 2. This invention introduces an improved non-dominated sorting genetic algorithm, which performs a global search by calculating congestion distance to obtain preliminary delivery route schemes. This ensures a high-quality trade-off between travel time, carbon emissions, and transportation economic costs, improving the overall optimization level of the route schemes. Simultaneously, a service time window penalty mechanism is introduced, and a path segment selection probability function is designed to fine-tune local optimization candidate path segments, effectively improving the time window satisfaction rate of service nodes. This two-layer collaborative optimization strategy enhances the convergence performance of the logistics delivery route planning method and the output quality of the delivery route schemes, significantly improving the practicality and timeliness of low-carbon delivery. Attached Figure Description

[0029] Figure 1 This is a flowchart of a low-carbon logistics distribution route planning method based on multi-objective optimization as described in this invention. Detailed Implementation

[0030] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0032] The following description, in conjunction with the accompanying drawings, details a specific scheme for a low-carbon logistics distribution route planning method based on multi-objective optimization provided by the present invention.

[0033] See attached document Figure 1 The diagram illustrates a flowchart of a low-carbon logistics distribution route planning method based on multi-objective optimization, provided by an embodiment of the present invention. The method includes the following steps:

[0034] S1. Construct a directed graph based on the logistics distribution network and obtain the travel distance, travel time and transportation economic cost of the path segments; based on the travel distance and travel time of the path segments, construct a traffic state perception-based carbon emission model that integrates travel distance and idling time, and calculate the carbon emissions of the path segments; based on the travel time, carbon emissions and transportation economic cost of the path segments, construct a multi-objective function optimization model.

[0035] Construct a directed graph based on the existing logistics and distribution network. ,in This represents the set of nodes in a logistics and distribution network, including factories, warehouses, and transfer centers. Represents a set of path segments, consisting of nodes. To the node path segment Includes the following parameters: Represents path segment The travel distance, in kilometers (km), is obtained from existing traffic databases; Indicates that the vehicle is on the route segment Travel time, in minutes (min), is obtained from existing traffic databases; Represents path segment The transportation economic cost, in yuan (CNY), is calculated by weighting factors such as vehicle unit energy consumption, labor unit price, and travel distance from the existing logistics enterprise operation database. This is a technical method well-known to those skilled in the art and will not be elaborated here. Represents path segment The carbon emissions, in grams (g), were calculated using a traffic state perception-based carbon emission model that integrates travel distance and idling time.

[0036] The traffic state perception-based carbon emission model, which integrates travel distance and idling time, is constructed based on the IPCC National Greenhouse Gas Inventory Guidelines and incorporates the carbon emission structure ratio of travel distance and idling time for urban delivery vehicles in actual operation. This allows for a more accurate calculation of carbon emissions for route segments. The formula is as follows:

[0037]

[0038] in, Represents path segment Carbon emissions, in grams (g); Represents path segment travel distance The carbon emission percentage weight is set based on expert experience, and its value range is [value range missing]. ; Represents path segment The travel distance, in kilometers (km); Carbon emission intensity per unit distance represents the carbon emissions generated by a vehicle per kilometer, measured in grams per kilometer (g / km). The values ​​were obtained by referencing the IPCC National Greenhouse Gas Inventory Guidelines and actual operational data from delivery vehicles. The range for carbon emission intensity per unit distance for electric vehicles is as follows: The range of carbon emission intensity per unit distance for gasoline vehicles is as follows: ; Represents path segment The carbon emission percentage of idling time ; Indicates that the vehicle is on the route segment Idle time, in minutes (min); Indicates that the vehicle is on the route segment Travel time on the road, in minutes (min); This is the idle speed proportional coefficient, representing the vehicle's speed on the road segment. Upward travel time The percentage of time spent in an idling state (including waiting at traffic lights, slow-moving traffic, stop-and-go traffic, and other unstable states) is determined by analyzing route segments in the existing traffic database. The ratio of historical idling time to travel time is used to calculate the value, which ranges from [value missing]. ; The carbon emission intensity at idle time is expressed as the amount of carbon emitted per minute when a vehicle is idling, measured in grams per minute (g / min). The values ​​were obtained by referring to the IPCC National Greenhouse Gas Inventory Guidelines and actual operational data from delivery vehicles. The range for the carbon emission intensity at idle time for electric vehicles is as follows: The range of values ​​for the carbon emission intensity per unit time at idle speed of gasoline vehicles is as follows: ;

[0039] Unlike traditional route planning optimization, the above formula constructs a traffic state-aware carbon emission model by integrating travel distance and idling time, in order to meet the requirements for dynamic carbon emission response characteristics in route planning optimization.

[0040] Furthermore, based on the directed graph, candidate delivery route schemes are obtained. In the route planning, a multi-objective function optimization model is constructed by minimizing the total delivery time, total carbon emissions, and total transportation economic cost of the candidate delivery route schemes. The multi-objective function formula is as follows:

[0041]

[0042] Indicates candidate delivery route options The objective function for total delivery time; Indicates candidate delivery route options The objective function for total carbon emissions; Indicates candidate delivery route options The objective function for the total economic cost of transportation; Candidate delivery route options Included path segments gather;

[0043] The formula for the multi-objective function optimization model is: ,in This is a set of all feasible delivery route options (i.e., candidate delivery route options), derived from existing logistics and delivery networks.

[0044] S2. Based on the multi-objective function optimization model, an improved non-dominated sorting genetic algorithm is used to generate a preliminary delivery route plan. Based on the preliminary delivery route plan, a service time window penalty mechanism is introduced to identify local optimization candidate path segments and perform path optimization to obtain the optimal delivery route plan.

[0045] Based on a multi-objective optimization model, an improved non-dominated sorting genetic algorithm (LC-NSGA-II) is used to calculate congestion distance by comprehensively considering total delivery time, total carbon emissions, and total transportation economic cost. The formula is as follows:

[0046]

[0047] in, This represents the first step of the LC-NSGA-II algorithm. Individual; and They represent the first The and the first Each individual has a corresponding delivery route plan (i.e., a candidate delivery route plan), which includes several complete delivery routes; Represents an individual The crowding distance is used for elite selection in the non-dominated ranking phase. Larger individuals are given higher priority for preservation. Entering the next generation of the population; In this invention, the number of objective functions is represented. That is, the objective function includes the total delivery time objective function, the total carbon emissions objective function, and the total transportation economic cost objective function; Indicates the index value of the objective function; and They represent the first time. Objective function In the middle, individuals ,individual The objective function value of the corresponding delivery route plan; and They represent the first Objective function The maximum and minimum values ​​are obtained by calculating the delivery route plan for all individuals on the 1st. Objective function The objective function value is obtained by selecting the maximum and minimum values, which are then used for normalization. Indicates the first The weight coefficients of each objective function are set according to expert experience and can take values ​​of [values ​​to be filled in]. This is used to strengthen the carbon emission optimization orientation;

[0048] The above congestion distance calculation formula comprehensively considers the total delivery time, total carbon emissions, and total transportation economic cost, and takes the goal of minimizing total carbon emissions as the core reference indicator for congestion distance. This makes the improved non-dominated sorting genetic algorithm more inclined to choose low-carbon paths, thus better meeting the optimization needs of green delivery systems.

[0049] The LC-NSGA-II algorithm will consider the crowding distance. Sort from largest to smallest and select the top... Individual components constitute a preliminary delivery route plan;

[0050] Furthermore, the actual arrival time of the nodes is calculated and compared with the service time window set based on the existing logistics order system to identify whether there are path segments in the delivery path of each individual in the preliminary delivery path plan that violate the service time window: If at least one individual's complete delivery path in the preliminary delivery path plan meets the service time window, then the individual with the largest congestion distance is selected from the individuals in the preliminary delivery path plan that meet the service time window as the optimal delivery path plan of the present invention; if no individual's complete delivery path in the preliminary delivery path plan meets the service time window, then the path segments that violate the service time window are identified within each individual in the preliminary delivery path plan, and the path segments are marked as local optimization candidate path segments for the corresponding individual; a service time window penalty mechanism is introduced to optimize the local optimization candidate path segments; the specific implementation process is as follows:

[0051] Based on nodes Actual arrival time Determine the node Valid time of arrival on the day ,in Minutes, or 24 hours, are used to indicate the actual arrival time. Align the nodes with the service time window. Valid arrival time and service window on the day of arrival Compare and calculate nodes. Valid time of arrival on the day Time difference with service time window If node valid time of arrival on the day During service hours Within the range, then If node Valid time of arrival on the day Earlier than the node Earliest permitted service time The node is not open at this time; you will have to wait until the earliest permitted service time of the day. Only those who can serve If node Valid time of arrival on the day Later than the node Latest permitted service time The node is currently closed and service will resume at the earliest permitted time the following day. The node Actual arrival time By following the corresponding individuals in sequence Travel time of the middle route segment The result is obtained by summing, in minutes (min); node Service Hours The data comes from the existing logistics order system. Represents a node The earliest permitted service time, Represents a node The latest allowed service time, in minutes (min). If the service time window spans multiple days, add 1440 minutes (24 hours) to the latest allowed service time of the service time window when setting it, so that it becomes a logically continuous interval. For example, if the service time is from 8:00 to 3:00 am, then set the service time window to (480 min, 1620 min).

[0052] In the implementation of the service time window penalty mechanism, based on each of the above individuals The local optimization candidate path segment is calculated using a path segment selection probability function. The probability of reaching each accessible adjacent node from the upstream node of the local optimization candidate path segment is calculated, and the node with the highest probability is selected to replace the downstream node of the original path segment. This allows for fine-tuning of path segments that violate the service time window. After fine-tuning, a new path segment is generated. A new individual Each new individual This corresponds to a delivery route plan that meets the service time window; finally, based on the new individual... Calculate the updated congestion distance Select the updated congestion distance The largest individual The optimal delivery route is output as the optimal delivery route plan; the formula for the route segment selection probability function is as follows:

[0053]

[0054] in, Indicates the upstream node of the candidate path segment from the local optimization. Select node The probability of being a downstream node, i.e., a path segment The probability of being selected; Path segment The guiding weight is determined according to the expert method, and the parameter value range is [range missing]. ; The sensitivity coefficient representing the guiding weight is set according to expert experience, and its value range is [value range missing]. ; Indicates that the vehicle is on the route segment The passage time on the road; Represents path segment Carbon emissions; This is a time sensitivity index, set based on expert experience, with a value range of [value range missing]. This is used to control the degree of emphasis placed on delivery timeliness; The carbon emission sensitivity index is set based on expert experience, and its value range is [value range missing]. This is used to control the level of emphasis placed on low carbon; The importance factor is set based on expert experience, and its value range is [value range missing]. ; Indicates the current node Passable adjacent nodes The set of nodes, that is, the set of nodes of the local optimization candidate path segment after excluding visited nodes or prohibited nodes. Visited nodes come from the delivery path record corresponding to the current individual, and prohibited nodes come from the existing logistics order system. Path segment The guiding weight; Indicates that the vehicle is on the route segment The passage time on the road; Represents path segment Carbon emissions; This is a time window penalty function used to penalize arriving nodes. The severity of early or late arrivals is suppressed, and the range of the time window penalty function value is [value missing]. ; The time window penalty factor is set based on expert experience, and its value range is [value range missing]. Unit: min -1 ; For nodes The time difference between the valid arrival time on the day of arrival and the service time window, in minutes (min).

[0055] The above formula dynamically incorporates a time window penalty in the path selection process, gradually guiding the path construction towards the direction of meeting the service time window constraints. This ensures that the path segment selection probability function can still efficiently meet the timeliness requirements of customer service under the premise of low carbon emissions, thus balancing the dual goals of green development and timeliness.

[0056] In summary, a low-carbon logistics distribution route planning method based on multi-objective optimization has been developed.

[0057] The order of the embodiments is for illustrative purposes only and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0058] The various embodiments in this specification 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.

[0059] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A low-carbon logistics distribution route planning method based on multi-objective optimization, characterized in that, Includes the following steps: S1. Construct a directed graph based on the logistics distribution network and obtain the travel distance, travel time, and transportation economic cost of the path segments; based on the travel distance and travel time of the path segments, construct a traffic state perception-based carbon emission model that integrates travel distance and idling time, and calculate the carbon emissions of the path segments; based on the directed graph, obtain candidate delivery route schemes; calculate the total delivery time, total carbon emissions, and total transportation economic cost of the candidate delivery route schemes, and minimize them respectively, construct a multi-objective function optimization model, and obtain the objective functions for total delivery time, total carbon emissions, and total transportation economic cost. S2. Based on a multi-objective function optimization model, an improved non-dominated sorting genetic algorithm is used to calculate the congestion distance based on the total delivery time, total carbon emissions, and total transportation economic cost. Based on the crowding distance sorted from largest to smallest, a preliminary delivery route plan is generated; Based on the preliminary delivery route plan, the actual arrival time of the nodes is calculated. Based on the actual arrival time of the nodes, and combined with a preset service time window, path segments in the preliminary delivery route plan that violate the service time window are identified: When at least one individual's complete delivery route in the preliminary delivery route plan satisfies the service time window, the individual with the largest congestion distance among those individuals in the preliminary delivery route plan that satisfies the service time window is selected as the optimal delivery route plan; otherwise, path segments violating the service time window are identified within each individual in the preliminary delivery route plan, and these path segments are marked as local optimization candidate path segments for the corresponding individual. A service time window penalty mechanism is introduced, and the local optimization candidate path segments are optimized through a path segment optimization probability function to obtain the optimal delivery route plan. The path segment optimization probability function is based on the travel time and carbon emissions of the local optimization candidate path segments, and incorporates guiding weights and time window penalty factors to calculate the probability from the upstream node of the local optimization candidate path segment to each passable adjacent node, as shown in the following formula: in, Indicates the upstream node of the candidate path segment from the local optimization. Select node The probability of being a downstream node; Path segment The guiding weight; The sensitivity coefficient representing the guiding weight; Indicates that the vehicle is on the route segment The passage time on the road; Represents path segment Carbon emissions; It is a time sensitivity index; Carbon emission sensitivity index; Indicating importance factor; Indicates the current node Passable adjacent nodes A set; Path segment The guiding weight; Indicates that the vehicle is on the route segment The passage time on the road; Represents path segment Carbon emissions; For time window penalty function; The time window penalty factor; For nodes The time difference between the valid arrival time on the day of arrival and the service time window; The node with the highest probability is selected to replace the downstream node of the original path segment. Path segments that violate the service time window are adjusted to generate new individuals.

2. The low-carbon logistics distribution route planning method based on multi-objective optimization according to claim 1, characterized in that, S1 specifically includes: Based on the travel time of a route segment, an idling ratio coefficient is introduced to obtain the idling time of the route segment. Based on the idling time and travel distance of the route segment, and combined with the carbon emission structure ratio of travel distance and idling time of urban delivery vehicles in actual operation, a traffic state perception carbon emission model based on the fusion of travel distance and idling time is constructed, and the carbon emission of the route segment is obtained.

3. The low-carbon logistics distribution route planning method based on multi-objective optimization according to claim 1, characterized in that, S2 specifically includes: In the implementation of the improved non-dominated sorting genetic algorithm, each individual corresponds to a candidate delivery path scheme. The multi-objective function optimization model is combined with the weight coefficients of each objective function to calculate the crowding distance of the individual.

4. The low-carbon logistics distribution route planning method based on multi-objective optimization according to claim 1, characterized in that, S2 specifically includes: The service time window penalty mechanism calculates the time difference based on the actual arrival time of the node and the preset service time window; based on the time difference, a path segment selection probability function is introduced to optimize local optimization candidate path segments.

5. The low-carbon logistics distribution route planning method based on multi-objective optimization according to claim 1, characterized in that, S2 specifically includes: Calculate the updated congestion distance based on the new individual, and select the individual with the largest updated congestion distance as the optimal delivery route solution output.

Images