Community group buying path planning method for various fresh produce products that consider the needs of heterogeneous customers

Abstract

This invention discloses a multi-category fresh produce community group-buying route planning method considering heterogeneous customer needs, comprising the following steps: S1, modeling three penalty costs based on the freshness decline curve of fresh produce and customer satisfaction; S2, constructing a multi-category fresh produce community group-buying delivery route optimization model considering heterogeneous customer needs; S3, using a large-scale nearest neighbor search algorithm to solve the multi-category fresh produce community group-buying route planning scheme considering heterogeneous customer needs. This invention's method can achieve efficient optimization schemes for the delivery of various types of fresh produce based on the pickup time window requirements of different types of customers, ensuring the freshness requirements of delivered fresh produce, improving customer satisfaction with the freshness of delivered fresh produce, and reducing the probability of returns. Simultaneously, using a large-scale nearest neighbor search algorithm as the delivery route optimization algorithm demonstrates strong applicability, high scalability, insensitivity to real-time parameters, and the ability to achieve good solution quality.

Description

Technical Field

[0001] This invention relates to the field of logistics technology, specifically to a method for planning the community group-buying routes for various types of fresh produce that takes into account the diverse needs of customers. Background Technology

[0002] With the continuous development of community group buying, online platforms such as community group buying have become one of the main ways for people to obtain daily necessities. People's lifestyles and consumption habits are also gradually changing, and this significant social transformation has led to a surge in online shopping demand. As a form of online shopping, community group buying for fresh produce has also experienced rapid development. Therefore, with the gradual increase in online shopping demand, the vehicle routing challenge with time windows in the context of the community group buying model has become a hot topic in academic research. In fact, because the freshness of fresh produce deteriorates rapidly, community residents are likely to receive fresh produce with lower freshness, leading to a decrease in customer satisfaction.

[0003] Traditional delivery route planning solutions struggle to meet the challenges of considering the diverse pickup needs of varied customers and the perishable nature of fresh produce in cold chain logistics. Therefore, systematically designing delivery plans to ensure timely delivery of fresh produce and maximize resident satisfaction is crucial in community group buying. Consequently, there is an urgent need for a multi-variety fresh produce community group buying route planning solution that can accommodate the diverse needs of customers. Summary of the Invention

[0004] The purpose of this invention is to provide a community group-buying path planning method for various types of fresh produce that takes into account the needs of heterogeneous customers, so as to solve the above-mentioned defects.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for planning the community group-buying path for various fresh produce categories that considers the needs of heterogeneous customers includes the following steps:

[0007] S1. Model three types of penalty costs based on the freshness decline curve of fresh produce and customer satisfaction.

[0008] S2. Construct a community group-buying delivery route optimization model for various types of fresh produce that takes into account the needs of heterogeneous customers.

[0009] S3. Use a large-scale nearest neighbor search algorithm to solve a community group-buying route planning scheme for various types of fresh produce that takes into account the heterogeneous needs of customers.

[0010] Preferably, in step S1, three penalty costs are modeled based on the freshness decline curve of the perishable goods and customer satisfaction. The specific steps are as follows:

[0011] S101. Construct the freshness degradation function for fresh produce stored in refrigerated trucks and at the group leader's location:

[0012]

[0013] In the formula, α p and β p These represent the spoilage coefficients related to the transportation process and the storage process in the shop, respectively, where t represents time. This represents the function that describes the decrease in freshness of perishable goods during transportation. This represents a function that indicates the decrease in the freshness of perishable goods when stored in a shop.

[0014] S102. Based on the relationship between the freshness degradation function and time, model the costs incurred due to freshness degradation inside the refrigerated truck. Costs incurred due to decreased freshness at the group leader's location Penalty costs for delivering goods later than the customer's time window They are respectively:

[0015]

[0016]

[0017] In the formula, ω represents the coefficient between the decrease in freshness and the resulting loss. This indicates the delivery time of the goods, specifically the time when the refrigerated truck unloads. Indicates the customer's earliest pickup time. Indicates the latest time the customer can pick up their order;

[0018] S103. Based on the relationship between delivery time and different customers' pickup time windows, the total penalty cost function is calculated as follows:

[0019]

[0020] Preferably, in step S2, a multi-category fresh produce community group-buying delivery route optimization model considering heterogeneous customer needs is constructed. The specific steps are as follows:

[0021] S201. By comprehensively utilizing the different customer pickup time window needs and product needs of each community node, establish a multi-category fresh food community group buying delivery route optimization model;

[0022] S202. Set the objective function as follows:

[0023]

[0024] In the formula, c ij x ijk This represents the basic transportation cost, where c ijLet x represent the cost from the i-th to the j-th community. ijk This indicates whether car k departs from location i and arrives at location j. Indicates the cost of punishment. q represents the total penalty. ilp This represents the total demand for product p from customer type l in community i.

[0025] First objective function c ij x ijk The second term represents minimizing all basic costs in the delivery plan. This represents minimizing the penalty costs caused by decreased product freshness and late delivery;

[0026] Set flow balance constraints as follows:

[0027]

[0028] In the formula, x 0jk Let x indicate whether car k departs from the warehouse and arrives at the j-th community. Then constraint (1) indicates that car k departs from the warehouse. j，|N|+1，k If k indicates whether vehicle j returns to the warehouse, then constraint (2) indicates that vehicle k eventually returns to the warehouse. Constraint (3) indicates that each vehicle will leave after completing its service at a certain community node, i.e., flow balancing constraint. Constraint (4) indicates that each community node has at least one vehicle for delivery. ipk Let k indicate whether vehicle k provides goods of type p to the i-th community. Constraint (5) indicates that a vehicle only serves a community when it passes through it. The initial and final positions of each vehicle are at the same depot.

[0029] Set the load capacity constraint as follows:

[0030]

[0031]

[0032] In the formula, ρ k Let represent the maximum number of cargo types that the k-th vehicle can carry. Then, constraint (6) means that the actual number of cargo types carried by the vehicle cannot exceed the upper limit; constraint (7) means that the p-th type of goods in the i-th community must be delivered by at least one vehicle; constraint (8) means that the k-th vehicle can only deliver goods to the i-th community when it passes through the i-th community; constraint (9) means that the total load capacity remains unchanged; constraint (10) means that each community is visited in chronological order; constraint (11) means that the time when departing from the warehouse is 0.

[0033] The following time window constraints are set to ensure that the delivery time of each vehicle meets the time window requirements of the community leader:

[0034]

[0035] In the formula, s i Let e ​​represent the service time of the i-th community. i and l i This indicates the earliest and latest time when the community receives the goods.

[0036] Preferably, in step S3, a large-scale nearest neighbor search algorithm is used to solve a community group-buying route planning scheme for various types of fresh produce that considers the heterogeneous needs of customers. The specific steps are as follows:

[0037] S301. Population initialization is performed using a heuristic greedy algorithm and random generation.

[0038] S302. Use the Split algorithm to assign a delivery scheme with a separator to each fixed-order node sequence;

[0039] S303. Use an elite selection strategy to select outstanding individuals from the population and perform crossover to obtain new sequences;

[0040] S304. The Split algorithm is used again to assign a delivery scheme with a separator to each fixed-order node sequence.

[0041] S305. Optimize the delivery route scheme using a large neighborhood search algorithm that includes deletion, insertion, and local neighborhood search.

[0042] Preferably, in step S301, a heuristic greedy algorithm and random generation are used for population initialization, and the specific steps are as follows:

[0043] In each generation, the population is divided into two subpopulations based on the feasibility of the solutions. The number of solutions in each subpopulation satisfies a lower bound μ and an upper bound μ+λ, where μ = 25 and λ = 40. During population initialization, 4μ individuals are generated. To generate each individual, 100 sets of random sequences are generated. These sequences are divided into initial solutions using the Split algorithm. The fitness of each initial solution is calculated, and the best one is added to the population as an initial individual. All individuals are trained. If an infeasible individual is generated, a correction operation is performed. At the end of initialization, the individuals in the population are divided into two feasibility-based subpopulations. Then, a selection operation is performed on the individuals in these two subpopulations, and the two selected subpopulations are used as the first generation of individuals.

[0044] Preferably, in step S302, the Split algorithm is used to assign a delivery scheme with a separator to each fixed-order node sequence. The specific steps are as follows:

[0045] The Split algorithm requires the delivery order of all nodes and the cost between any arc arc(i,j), i,j∈V, where each node represents the demand for a certain fresh produce in the community.

[0046] The cost of arc(i,j), where i,j∈V, is expressed as:

[0047]

[0048] In the formula, f ij This represents the total cost of a single delivery visit from community i to community j. Each arc arc(i, j), where i, j ∈ V, guarantees that a vehicle can only deliver to one community once. 0，i+1 c represents the basic delivery cost from the warehouse to the first community i; j，0 c represents the cost from the last community j to the warehouse; h，h+1 Let c represent the basic cost from community h to community h+1. If h and h+1 represent different demands for various types of fresh produce within the same community, then c... h，h+1 =0; Let τ represent the total cost of vehicle delivery between communities, where arc(i,j), i,j∈V, also includes penalty costs for different customers in different communities; hk Let q represent the arrival time of vehicle k. hlp If customer l in community h has a demand for fresh produce p, then H lp (τ hk )q hlp Let represent the penalty cost for customer l in community h for fresh produce p. Therefore, the total penalty cost is expressed as: M is a sufficiently large positive integer.

[0049] After calculating all costs of arc arc(i,j), i,j∈V, we can obtain the optimal solution in a fixed sequence of nodes.

[0050] Preferably, in step S303, an elite selection strategy is used to select superior individuals from the population and perform crossover to obtain a new sequence. The specific steps are as follows:

[0051] First, randomly select two parents from all individuals; then, randomly select two cutoff points i and j in Parent1, and copy the substrings P1(i)...P1(j) to the corresponding positions in Child1; finally, starting from node j+1 in Parent2, if the node is not found in C1, insert the node into C1 in turn. Another child C2 can be obtained in the same way by swapping P1 and P2.

[0052] Preferably, in step S305, a large neighborhood search algorithm that includes deletion, insertion, and local neighborhood search is used to optimize the delivery route scheme. The specific steps are as follows:

[0053] Each operation traverses all possible pairs (u,v) in the route. Each pair can contain one or more nodes. All nodes in a node pair may come from the same route or different routes. Each node contains information about the community order number and product type. x and y represent the next node of u and v on their paths, respectively. T(u) represents the route of node u.

[0054] M1. If u is a client node, remove u and then insert it after v;

[0055] M2. If u and x are clients, remove them and then insert (u, x) after v;

[0056] M3. If u and x are clients, remove them and then insert (x, u) after v;

[0057] M4. If u and x are clients, swap u and x;

[0058] M5. If u,x and v are clients, swap (u,x) and v;

[0059] M6. If (u,x) and (v,y) are clients, swap (u,x) and (v,y);

[0060] M7. If T(u) = T(v), replace (u,x) and (v,y) with (u,v) and (x,y);

[0061] M8. If T(u)≠T(v), replace (u,x) and (v,y) with (u,v) and (x,y);

[0062] M9. If T(u)≠T(v), replace (u,y) and (x,v) with (u,v) and (x,y).

[0063] The beneficial effects of this invention are as follows:

[0064] This invention presents a multi-category fresh produce community group-buying route planning method that considers the diverse needs of customers. While taking into account the requirements of different customer groups, it comprehensively considers the pickup time window needs and satisfaction with the freshness of fresh produce among different types of customers within the community. It integrates the different degrees of freshness degradation of various fresh produce items to construct a multi-category fresh produce delivery model that considers the pickup needs of diverse customers. The system design relies on a route planning solution based on a large nearest neighbor search algorithm. This enables the system to design efficient optimization schemes for the delivery of various types of fresh produce based on the pickup time window needs of different customer groups, ensuring the freshness requirements of delivered fresh produce, improving customer satisfaction with the freshness of delivered fresh produce, and reducing the probability of returns. Furthermore, the use of a large nearest neighbor search algorithm as the delivery route optimization algorithm demonstrates strong applicability, high scalability, insensitivity to real-time parameters, and the ability to achieve good solution quality. Attached Figure Description

[0065] Figure 1 : A schematic flowchart of the method of the present invention;

[0066] Figure 2 : Flowchart of the large-scale nearest neighbor search algorithm in the method of this invention. Detailed Implementation

[0067] The present invention will be further described below with reference to the embodiments. It should be noted that these are merely examples and descriptions of the inventive concept. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the inventive concept or exceed the scope defined in the claims, they should all be considered to fall within the protection scope of the present invention.

[0068] Example 1:

[0069] Figure 1 This is a schematic diagram of the process of the method of the present invention, as shown below. Figure 1 As shown, the community group-buying path planning method for various types of fresh produce, considering the heterogeneous needs of customers, includes the following steps:

[0070] S1. Model three types of penalty costs based on the freshness decline curve of fresh produce and customer satisfaction. The specific steps are as follows:

[0071] S101. Construct the freshness degradation function for fresh produce stored in refrigerated trucks and at the group leader's location:

[0072]

[0073] In the formula, α p and β p These represent the spoilage coefficients related to the transportation process and the storage process in the shop, respectively, where t represents time. This represents the function that describes the decrease in freshness of perishable goods during transportation. This represents the function that indicates the decrease in freshness of perishable goods when stored in a shop.

[0074] S102. Based on the relationship between the freshness degradation function and time, model the costs incurred due to freshness degradation inside the refrigerated truck. Costs incurred due to decreased freshness at the group leader's location Penalty costs for delivering goods later than the customer's time window They are respectively:

[0075]

[0076] In the formula, ω represents the coefficient between the decrease in freshness and the resulting loss. This indicates the delivery time of the goods, specifically the time when the refrigerated truck unloads. Indicates the customer's earliest pickup time. This indicates the latest time a customer can pick up their order.

[0077] S103. Based on the relationship between delivery time and different customers' pickup time windows, the total penalty cost function is calculated as follows:

[0078]

[0079] S2. Construct a community group-buying delivery route optimization model for various types of fresh produce that considers the needs of heterogeneous customers. The specific steps are as follows:

[0080] S201. By comprehensively utilizing the different customer pickup time window needs and product needs of each community node, establish a multi-category fresh food community group buying delivery route optimization model;

[0081] S202. Set the objective function as follows:

[0082]

[0083] In the formula, c ij x ijk This represents the basic transportation cost, where c ij Let x represent the cost from the i-th to the j-th community. ijk Indicate whether car k departs from point i and arrives at point j; Indicates the cost of punishment. q represents the total penalty. ilp This represents the total demand for product p from customer type l in community i.

[0084] First objective function c ij x ijk The second term represents minimizing all basic costs in the delivery plan. This indicates minimizing the penalty costs caused by decreased product freshness and late delivery.

[0085] S203. Construct a delivery model that includes flow balance constraints, load capacity constraints, and time window constraints.

[0086] Set flow balance constraints as follows:

[0087]

[0088] In the formula, x 0jk Let x indicate whether car k departs from the warehouse and arrives at the j-th community. Then constraint (1) indicates that car k departs from the warehouse. j，|N|+1，k If k indicates whether vehicle j returns to the warehouse, then constraint (2) indicates that vehicle k eventually returns to the warehouse. Constraint (3) indicates that each vehicle will leave after completing its service at a certain community node, i.e., flow balancing constraint. Constraint (4) indicates that each community node has at least one vehicle for delivery. ipk Let k indicate whether vehicle k provides goods of type p to the i-th community. Constraint (5) indicates that a vehicle only serves a community when it passes through it. The initial and final positions of each vehicle are at the same depot.

[0089] Set the load capacity constraint as follows:

[0090]

[0091] In the formula, ρ k Let represent the maximum number of cargo types that the k-th vehicle can carry. Then constraint (6) means that the actual number of cargo types carried by the vehicle cannot exceed the upper limit. Constraint (7) means that the p-th type of goods in the i-th community must be delivered by at least one vehicle. Constraint (8) means that the k-th vehicle can only deliver goods to the i-th community when it passes through the i-th community. Constraint (9) means that the total load capacity remains unchanged; constraint (10) means that each community is visited in chronological order; constraint (11) means that the time when departing from the warehouse is 0.

[0092] The following time window constraints are set to ensure that the delivery time of each vehicle meets the time window requirements of the community leader:

[0093]

[0094] In the formula, s i Let e ​​represent the service time of the i-th community. i and l i This indicates the earliest and latest time when the community receives the goods.

[0095] S3 Figure 2 The flowchart of the large-scale nearest neighbor search algorithm in the method of this invention is as follows: Figure 2As shown, the large-scale nearest neighbor search algorithm is used to solve the community group-buying route planning scheme for various types of fresh produce that consider the heterogeneous needs of customers. The specific steps are as follows:

[0096] S301. Population initialization is performed using a heuristic greedy algorithm and random generation.

[0097] In each generation, the population is divided into two subpopulations based on the feasibility of solutions. The number of solutions in each subpopulation satisfies a lower bound μ and an upper bound μ+λ, where μ = 25 and λ = 40. During population initialization, 4μ individuals are generated. To generate each individual, 100 random sequences need to be generated. Then... Figure 2 The "sub-process" in the process is as follows: the sequences are divided into initial solutions using the Split algorithm, the fitness of each initial solution is calculated, and the best one is placed into the population as the initial individual; all individuals are trained, and if an infeasible individual is generated, a correction operation is performed. At the end of the initialization, the individuals in the population are divided into two subpopulations based on whether they are feasible. Then, a selection operation is performed on the individuals in these two subpopulations, and the two selected subpopulations are used as the first generation of individuals.

[0098] S302. Use the Split algorithm to assign a delivery scheme with a separator to each fixed-order node sequence.

[0099] The Split algorithm requires the delivery order of all nodes and the cost between any arc arc(i,j), i,j∈V, where each node represents the demand for a certain fresh produce in the community.

[0100] The cost of arc(i,j), where i,j∈V, is expressed as:

[0101]

[0102] In the formula, f ， This represents the total cost of a single delivery visit from community i to community j. Each arc arc(i, j), where i, j ∈ V, guarantees that a vehicle can only deliver to one community once. 0，i+1 c represents the basic delivery cost from the warehouse to the first community i; j，0 c represents the cost from the last community j to the warehouse; h，h+1 Let c represent the basic cost from community h to community h+1. If h and h+1 represent different demands for various types of fresh produce within the same community, then c... h，h+1 =0; Let τ represent the total cost of vehicle delivery between communities, where arc(i,j), i,j∈V, also includes penalty costs for different customers in different communities; hk Let q represent the arrival time of vehicle k. hlp If customer l in community h has a demand for fresh produce p, then Hlp (τ hk )q hlp Let represent the penalty cost for customer l in community h for fresh produce p. Therefore, the total penalty cost is expressed as: M represents a sufficiently large positive integer.

[0103] After calculating all costs of arc arc(i,j), i,j∈V, we can obtain the optimal solution in a fixed sequence of nodes.

[0104] S303. Based on the fitness function (which in practical problems is related to the freshness of perishable goods, customer satisfaction, and basic costs), an elite selection strategy is used to select superior individuals from the population, and crossover is performed to obtain a new sequence. The crossover steps are as follows:

[0105] First, randomly select two parents from all individuals; then, randomly select two cutoff points i and j in Parent1 (P1) and copy the substrings P1(i)...P1(j) to the corresponding positions in Child1 (C1); finally, starting from j+1 in Parent2 (P2), if the node is not found in C1, insert the node into C1 in turn. Another child C2 can be obtained in the same way by swapping P1 and P2.

[0106] S304, proceed again Figure 2 The “subprocess” in the process: The Split algorithm is used to assign a delivery scheme with a separator to each fixed-order node sequence, see S301 for details.

[0107] S305. Optimize the delivery route using a large neighborhood search algorithm that includes deletion, insertion, and local neighborhood search. The local search operators are M1-M9. After iterating through M1-M9, assign delivery routes with separators to the "sub-processes," thus completing the local search step. If the optimal solution is improved, return to the previous steps and repeat until the maximum number of iterations or the maximum running time is reached. If there is no improvement, perform population diversification, that is, add the diversity ranking based on Hamming distance to the objective function with a certain weighted coefficient to select more diverse individuals, increase the possibility of mutation, and then return to the previous steps and repeat.

[0108] Each operation traverses all possible pairs (u,v) in the route. Each pair can contain one or more nodes. All nodes in a node pair may come from the same route or different routes. Each node contains information about the community order number and product type. x and y represent the next node of u and v on their paths, respectively. T(u) represents the route of node u.

[0109] M1. If u is a client node, remove u and then insert it after v;

[0110] M2. If u and x are clients, remove them and then insert (u, x) after v;

[0111] M3. If u and x are clients, remove them and then insert (x, u) after v;

[0112] M4. If u and x are clients, swap u and x;

[0113] M5. If u,x and v are clients, swap (u,x) and v;

[0114] M6. If (u,x) and (v,y) are clients, swap (u,x) and (v,y);

[0115] M7. If T(u) = T(v), replace (u,x) and (v,y) with (u,v) and (x,y);

[0116] M8. If T(u)≠T(v), replace (u,x) and (v,y) with (u,v) and (x,y);

[0117] M9. If T(u)≠T(v), replace (u,y) and (x,v) with (u,v) and (x,y).

[0118] This invention presents a multi-category fresh produce community group-buying route planning method that considers the diverse needs of customers. While taking into account the requirements of different customer groups, it comprehensively considers the pickup time window needs and satisfaction with the freshness of fresh produce among different types of customers within the community. It integrates the different degrees of freshness degradation of various fresh produce items to construct a multi-category fresh produce delivery model that considers the pickup needs of diverse customers. The system design relies on a route planning solution based on a large nearest neighbor search algorithm. This enables the system to design efficient optimization schemes for the delivery of various types of fresh produce based on the pickup time window needs of different customer groups, ensuring the freshness requirements of delivered fresh produce, improving customer satisfaction with the freshness of delivered fresh produce, and reducing the probability of returns. Furthermore, the use of a large nearest neighbor search algorithm as the delivery route optimization algorithm demonstrates strong applicability, high scalability, insensitivity to real-time parameters, and the ability to achieve good solution quality.

[0119] The above is an exemplary description of the invention. Obviously, the specific implementation of the invention is not limited to the above-described manner. Any non-substantial improvement made using the inventive concept and technical solution of the invention, or the direct application of the inventive concept and technical solution to other situations without modification, is within the protection scope of the invention.

Claims

1. A method for planning the community group-buying route for various types of fresh produce that considers the needs of heterogeneous customers, characterized in that, Includes the following steps: S1. Model three types of penalty costs based on the freshness decline curve of fresh produce and customer satisfaction. S2. Construct a community group-buying delivery route optimization model for various types of fresh produce that takes into account the needs of heterogeneous customers. S3. Use a large-scale nearest neighbor search algorithm to solve the community group buying route planning scheme for various fresh produce that takes into account the heterogeneous needs of customers. In step S1, three types of penalty costs are modeled based on the freshness decline curve of fresh produce and customer satisfaction. The specific steps are as follows: S101. Construct the freshness degradation function for fresh produce stored in refrigerated trucks and at the group leader's location: ， ， In the formula, and These represent the spoilage coefficients related to the transportation process and the storage process in the store, respectively. Indicates time, This represents the function that describes the decrease in freshness of perishable goods during transportation. This represents the function that describes the decrease in freshness of perishable goods when stored in a shop. S102. Based on the relationship between the freshness degradation function and time, model the costs incurred due to freshness degradation inside the refrigerated truck. The cost of decreased freshness at the group leader's location Penalty costs for delivering goods later than the customer's time window They are respectively: ， ， ， In the formula, This represents the coefficient between the decrease in freshness and the resulting loss. This indicates the delivery time of the goods, specifically the time when the refrigerated truck unloads. Indicates the customer's earliest pickup time. Indicates the latest time the customer can pick up their order; S103. Based on the relationship between delivery time and different customers' pickup time windows, the total penalty cost function is calculated as follows: ； In step S2, a multi-category fresh produce community group-buying delivery route optimization model considering heterogeneous customer needs is constructed. The specific steps are as follows: S201. By comprehensively utilizing the different customer pickup time window needs and product needs of each community node, establish a multi-category fresh food community group buying delivery route optimization model; S202. Set the objective function as follows: ； In the formula, This represents the basic transportation cost, of which Indicates from the first arrive The cost of a community express Did the car come from Departure from the place to arrive land, Indicates the cost of punishment. Indicates the total penalty. Indicates the first The first in the community Type of customer for the first Total demand for a certain commodity; First objective function The second term represents minimizing all basic costs in the delivery plan. This represents minimizing the penalty costs caused by decreased product freshness and late delivery; S203. Construct a delivery model that includes flow balance constraints, load capacity constraints, and time window constraints; First, let's explain the model variables: Set flow balance constraints as follows: ； In the formula, express Did the car depart from the warehouse and arrive at the destination? Then constraint (1) represents the first community. The vehicle departed from the warehouse; express Did the car come from If the land returns to the warehouse, then constraint (2) means that the first... The vehicle was eventually returned to the warehouse; Constraint (3) indicates that each vehicle will leave after completing its service in a certain community, which is a flow balance constraint; Constraint (4) indicates that each community node has at least one vehicle for delivery. express Did the car give the first The community provides the first For goods of this type, constraint (5) indicates that the vehicle only serves the community when it passes through the community; where the initial and final locations of each vehicle are at the same depot; Set the load capacity constraint as follows: ； In the formula, Indicates the first If the maximum number of cargo types a vehicle can carry is specified, then constraint (6) means that the actual number of cargo types a vehicle can carry cannot exceed the upper limit. Constraint (7) indicates the first The first community Each type of commodity must be delivered by at least one vehicle; constraint (8) indicates that the first type of commodity is delivered by at least one vehicle. The vehicle can only pass the first Delivery to a community can only be made when there is a community; Constraint (9) indicates that the total load capacity remains constant; Constraint (10) indicates that each community is visited in chronological order; constraint (11) indicates that the time is 0 when starting from the warehouse; Set the following time window constraints to ensure that the delivery time of each vehicle meets the time window requirements of the community leader: ； In the formula, Indicates the first Service hours for each community and Indicates the earliest and latest time the community receives the goods; In step S3, a large-scale nearest neighbor search algorithm is used to solve a community group-buying route planning scheme for various types of fresh produce that takes into account the heterogeneous needs of customers. The specific steps are as follows: S301. Population initialization is performed using a heuristic greedy algorithm and random generation. S302. Use the Split algorithm to assign a delivery scheme with a separator to each fixed-order node sequence; S303. Use an elite selection strategy to select outstanding individuals from the population and perform crossover to obtain new sequences; S304. The Split algorithm is used again to assign a delivery scheme with a separator to each fixed-order node sequence. S305. Optimize the delivery route scheme using a large neighborhood search algorithm that includes deletion, insertion, and local neighborhood search.

2. The community group-buying route planning method for various fresh produce products considering heterogeneous customer needs as described in claim 1, characterized in that, In step S301, a heuristic greedy algorithm and random generation are used for population initialization. The specific steps are as follows: In each generation, the population is divided into two subpopulations based on the feasibility of solutions, with each subpopulation having a lower bound on the number of solutions. and upper limit ,in =25, =40; During population initialization, generate 4 To generate each individual, 100 random sequences need to be generated. These sequences are divided into initial solutions using the Split algorithm. The fitness of each initial solution is calculated, and the best one is placed into the population as the initial individual. All individuals are trained. If an infeasible individual is generated, a correction operation is performed. At the end of the initialization, the individuals in the population are divided into two feasibility-based subpopulations. Then, a selection operation is performed on the individuals in these two subpopulations, and the two selected subpopulations are used as the first generation of individuals.

3. The community group-buying route planning method for various types of fresh produce considering heterogeneous customer needs as described in claim 1, characterized in that, In step S302, the Split algorithm is used to assign a delivery scheme with a separator to each fixed-order node sequence. The specific steps are as follows: The Split algorithm requires the delivery order of all nodes and the order of delivery in any arc segment. The cost between nodes, where each node represents the demand for a certain type of fresh produce in the community; The cost is expressed as: In the formula, Indicates a delivery visit from arrive The total cost of the community, per arc Ensure that each vehicle can only deliver to one community once; Indicates the journey from the warehouse to the first community. Basic delivery costs; Indicates from the last community Costs to the warehouse; Indicates from the community To the community The basic cost, if and This indicates the different demands for various types of fresh produce within the same community. ; This indicates the total cost of vehicle delivery between communities. It also includes penalty costs for different communities and different customers; Indicates vehicle Arrival time, Indicates customer In the community For fresh produce The demand, Indicates customer In the community For fresh produce The penalty cost is therefore, the total penalty cost is expressed as ; It is a sufficiently large positive integer; Arc at the calculation point After accounting for all costs, we can obtain the optimal solution in a fixed sequence of nodes.

4. The community group-buying route planning method for various types of fresh produce considering heterogeneous customer needs as described in claim 1, characterized in that, In step S303, an elite selection strategy is used to select superior individuals from the population and perform crossover to obtain a new sequence. The specific steps are as follows: First, randomly select two parents from all individuals; then, randomly select two cut points from Parent1. and , substring ... Copy to the corresponding location in Child1; finally, from Parent2... Starting with a node, if the node is not found in C1, then insert the node into C1 in turn. Another child node C2 can be obtained in the same way by swapping P1 and P2.

5. The community group-buying route planning method for various types of fresh produce considering heterogeneous customer needs as described in claim 4, characterized in that, In step S305, a large neighborhood search algorithm that includes deletion, insertion, and local neighborhood search is used to optimize the delivery route scheme. The specific steps are as follows: Each operation traverses all possible pairs (u,v) in the route. Each pair can contain one or more nodes. All nodes in a node pair may come from the same route or different routes. Each node contains information about the community order number and product type. x and y represent the next node of u and v on their paths, respectively. T(u) represents the route of node u. Traversing M1-M9 in sequence constitutes a complete local search operation. M1. If u is a client node, remove u and then insert it after v; M2. If u and x are clients, remove them and then insert (u, x) after v; M3. If u and x are clients, remove them and then insert (x, u) after v; M4. If u and x are clients, swap u and x; M5. If u, x, and v are clients, swap (u, x) and v. M6. If (u, x) and (v, y) are clients, swap (u, x) and (v, y); M7. If T(u) = T(v), replace (u, x) and (v, y) with (u, v) and (x, y); M8. If T(u) ≠ T(v), replace (u, x) and (v, y) with (u, v) and (x, y); M9. If T(u) ≠ T(v), replace (u, y) and (x, v) with (u, v) and (x, y).

Images