Cold chain low-carbon transport vehicle scheduling method under fuzzy driving time

Abstract

The invention discloses a cold chain low-carbon transport vehicle scheduling method under fuzzy driving time. The method comprises the following steps: initializing a main population and an auxiliary population; fusing the main population and the auxiliary population into a temporary population; according to a plurality of optimization target values in the scheduling model, niche reservation operation is carried out on individuals in the temporary population, ps elite individuals reserved in the temporary population are adopted to replace individuals in the main population, and an updated main population is obtained; determining a non-inferior solution set for the updated main population by using a Pareto non-dominated relationship according to a plurality of optimization target values in the scheduling model; updating a pheromone matrix by using the updated individuals in the non-inferior solution set of the main population, and sampling the pheromone matrix to update the auxiliary population; interacting individuals in the updated primary and auxiliary populations; performing local search on individuals in a non-inferior solution set of the main population and the auxiliary population after interaction; and judging a termination condition. By means of the method, the high-quality non-inferior solution set of the cold-chain low-carbon transport vehicle scheduling problem under the fuzzy driving time can be obtained within a short time.

Description

Technical Field

[0001] The invention relates to a cold chain low-carbon transport vehicle scheduling method under fuzzy driving time, belonging to the field of cold chain low-carbon transport vehicle scheduling. Background Art

[0002] With the rapid development of fresh food e-commerce, food processing and other industries, people's demand for cold chain food is increasing. In order to achieve the simultaneous delivery of goods at different temperatures, multi-cabin vehicles with multiple independent temperature-controlled compartments have come into being. Reasonable optimization of the driving routes of these vehicles can effectively improve the efficiency of cold chain transportation. Considering that in reality, due to the uncertainties of weather, road conditions, vehicles and the status of drivers, vehicles often cannot deliver goods at a certain time, so it is of great practical significance to reasonably dispatch multi-cabin vehicles under fuzzy driving time.

[0003] Nowadays, green development has become a global consensus, and all countries have actively introduced relevant policies and measures to promote the development and transformation of green industries. In the field of transportation, according to statistics from the United Nations and the International Energy Agency (IEA), greenhouse gases generated by transportation account for about 15%-20% of global greenhouse gas emissions, so it is imperative to optimize the carbon emission costs by incorporating them into the total transportation costs. In addition, in order to enhance the competitiveness and sustainable development of enterprises, on the one hand, we must always adhere to the principle of "customer first" and effectively improve customer satisfaction, and on the other hand, we must be considerate of employees, balance the workload of drivers as much as possible, and strive to improve the happiness of employees. Therefore, in addition to the goal of transportation cost, the balance of customer satisfaction and driver workload should also be optimized as the daily goals of the enterprise. However, when optimizing the above three goals at the same time, how to better maintain the diversity of solutions becomes a problem that needs to be solved. Summary of the invention

[0004] The present invention provides a cold-chain low-carbon transport vehicle scheduling method under fuzzy driving time. On the one hand, a scheduling model is established with minimizing total transportation cost, maximizing customer satisfaction and minimizing driver workload imbalance as optimization goals. On the other hand, a main-auxiliary population collaborative search mechanism is used to perform a global search. A local search operator combining the characteristics of the problem is further constructed and a variable neighborhood descent algorithm is used to perform a local search operation. The method of the present invention can obtain a high-quality non-inferior solution set of the cold-chain low-carbon transport vehicle scheduling problem under fuzzy driving time in a short time, providing decision-making assistance for cold-chain transportation and distribution companies.

[0005] The technical solution of the present invention is:

[0006] According to a first aspect of the present invention, a method for scheduling cold chain low-carbon transportation vehicles under fuzzy travel time is provided, comprising the following steps:

[0007] Step 1. Decimal coding is used for the cargo collection center and the customer number. Each vehicle starts from the cargo collection center and returns to the cargo collection center after the service is completed. The load of each compartment of the vehicle meets the capacity constraint.

[0008] Step 2. Set the main parameters;

[0009] Step 3, initialization of main and auxiliary populations;

[0010] Step 4, merge the main population and the auxiliary population into a temporary population; according to the multiple optimization target values ​​in the scheduling model, perform microhabitat preservation operations on the individuals in the temporary population, and use the ps elite individuals retained in the temporary population to replace the individuals in the main population to obtain an updated main population; for the updated main population, use the Pareto non-dominated relationship to determine the non-inferior solution set according to the multiple optimization target values ​​in the scheduling model; use the individuals in the non-inferior solution set of the updated main population to update the pheromone matrix, sample the pheromone matrix to update the auxiliary population; interact with the individuals in the updated main and auxiliary populations;

[0011] Step 5: Perform local search on the individuals in the non-inferior solution set of the main population and auxiliary population after interaction;

[0012] Step 6. Determine the termination condition: If the termination condition is met, output the non-inferior solution set of the main population; otherwise, go to Step 4 to continue execution.

[0013] Furthermore, minimizing the total transportation cost, maximizing customer satisfaction and minimizing the imbalance of driver working hours are taken as optimization objectives, and constraints are constructed.

[0014] Furthermore, the optimization goal is specifically:

[0015] The first optimization goal is to minimize the total transportation cost Z1, which is calculated by the following formula:

[0016] MinimizeZ1=Z economy_cost +Z emission_cost

[0017] In the formula, Z1 represents the total transportation cost, Z economy_cost represents the driving cost, Z environment_cost represents the cost of carbon emissions;

[0018] The second optimization goal: maximize the customer satisfaction function, which is expressed as follows:

[0019]

[0020] In the formula, Z2 represents customer satisfaction, n c Represents the number of customers, μ i Represents customer i's satisfaction;

[0021] The third optimization goal is to minimize the imbalance of drivers’ working time. The expression is as follows:

[0022]

[0023] In the formula, Z3 represents the imbalance value of driver working time, represents the fuzzy travel time between customers, represents the decision variable, w i represents the waiting time that the vehicle spends serving customer i, s i represents the service time required for customer i, K = {1,2,...,n k} represents the vehicle set, n k Represents the total number of vehicles.

[0024] Furthermore, the initialization of the main and auxiliary populations is specifically as follows: individuals in the main population are initialized by a greedy rule and a random rule, and individuals in the auxiliary population are initialized by a random rule.

[0025] Furthermore, individuals in the main population are initialized using a greedy rule, and the specific steps are as follows:

[0026] 1) Construct a customer set to be served based on the customers who need services; use the cargo collection center as the starting point for each vehicle;

[0027] 2) Select the next customer to be served by the vehicle from the set of customers to be served based on the current greedy goal;

[0028] 3) Determine whether the product required by the selected customer exceeds the rated load of each compartment of the vehicle; if not, serve the customer and remove him from the set of customers to be served; otherwise, send the next vehicle;

[0029] 4) Determine whether the set of customers to be served is empty. If it is empty, it means that all customer services have been completed; otherwise, return to 2) and continue execution.

[0030] Furthermore, minimizing the total transportation cost and maximizing customer satisfaction are taken as greedy objectives.

[0031] Furthermore, the local search is performed on the individuals in the non-inferior solution sets of the main population and the auxiliary population after the interaction, specifically: for the obtained non-inferior solution set individuals in the main and auxiliary populations, the "Insert", "Exchange", "2-Opt" and "Special-Insert" neighborhood operations are executed in sequence, and a variable neighborhood descent strategy is adopted during the execution process.

[0032] Furthermore, the Special-Insert neighborhood operation is specifically as follows: for the current individual, a customer is randomly selected from the sub-path sequence with the longest driver working time among the individuals and inserted into any position of the sub-path sequence with the shortest driver working time; wherein the driver working time refers to the sum of the fuzzy driving time, the waiting time, and the service time.

[0033] According to a second aspect of the present invention, a processor is provided, which is used to run a program, wherein when the program is running, the cold chain low-carbon transportation vehicle scheduling method under fuzzy driving time described in any one of the above is executed.

[0034] According to a third aspect of the present invention, a computer-readable storage medium is provided, which includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned cold chain low-carbon transportation vehicle scheduling methods under fuzzy driving time.

[0035] The beneficial effects of the present invention are as follows: first, the present invention uses greedy rules and random rules to initialize the population, thereby improving the convergence speed of the algorithm; second, the mechanism of collaborative search between the main and auxiliary populations is adopted to improve the global search capability of the algorithm; finally, a variable neighborhood descent algorithm with 4 neighborhood operators is used for local search, thereby enhancing the local search capability of the algorithm. This method can obtain a high-quality non-inferior solution set for the vehicle scheduling problem in the low-carbon transportation process of cold-chain food under fuzzy driving time in a short time, providing a high-quality decision-making basis for transportation companies. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is the overall algorithm flow chart of the present invention;

[0037] Figure 2 A schematic diagram of vehicle optimization scheduling in the low-carbon transportation process of cold chain food of the present invention;

[0038] Figure 3 It is a schematic diagram of the operation of PMX of the present invention;

[0039] Figure 4 Schematic diagram of the neighborhood operations "Insert", "Exchange", "2-Opt" and "Special-Insert" of the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention clearer, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. It should be noted that the embodiments in this application and the features in the embodiments can be combined with each other arbitrarily without conflict.

[0041] Example 1: Figure 1-4 As shown, according to a first aspect of an embodiment of the present invention, a cold chain low-carbon transportation vehicle scheduling method under fuzzy driving time is provided, comprising:

[0042] Step 1. Decimal coding is used for the cargo collection center and the delivery customer number. Each vehicle starts from the cargo collection center and returns to the cargo collection center after the service is completed. The load of each vehicle compartment meets the capacity constraint. For example, Figure 2 As shown in the figure, the vehicle in each sub-path starts from the cargo collection center. Due to the constraints of the vehicle cabin capacity, it must serve customers and return to the cargo collection center under the premise of satisfying the constraints. Therefore, the driving path of each vehicle constitutes a sub-path, with the cargo collection center number 0 as the boundary. For example, the solution (individual) formed by 10 customers is π m =[0,5,2,7,3,0,6,4,8,0,10,1,9,0], where customers 5, 2, 7 and 3 are served by the first car, and their sequence [0,5,2,7,3,0] is the first subpath; customers 6, 4 and 8 are served by the second car, and their sequence [0,6,4,8,0] is the second subpath; customers 10, 1 and 9 are served by the third car, and their sequence [0,10,1,9,0] is the third subpath.

[0043] Step 2, set the main parameters: the main parameters include the main and auxiliary population sizes ps, pheromone importance factor α, heuristic function importance factor β and pheromone volatility coefficient ρ;

[0044] Step 3, initialization of main and auxiliary populations;

[0045] Step 4, merge the main population and the auxiliary population into a temporary population; according to the multiple optimization target values ​​in the scheduling model, perform microhabitat preservation operations on the individuals in the temporary population, and use the ps elite individuals retained in the temporary population to replace the individuals in the main population to obtain an updated main population; for the updated main population, use the Pareto non-dominated relationship to determine the non-inferior solution set according to the multiple optimization target values ​​in the scheduling model; use the individuals in the non-inferior solution set of the updated main population to update the pheromone matrix, sample the pheromone matrix to update the auxiliary population; interact with the individuals in the updated main and auxiliary populations;

[0046] Step 5: Perform local search on the individuals in the non-inferior solution set of the main population and auxiliary population after interaction;

[0047] Step 6. Determine the termination condition: If the termination condition is met, output the non-inferior solution set of the main population; otherwise, go to Step 4 to continue execution.

[0048] Furthermore, minimizing the total transportation cost, maximizing customer satisfaction and minimizing the imbalance of driver working hours are taken as optimization objectives, and constraints are constructed.

[0049] Furthermore, the three optimization target calculation methods are:

[0050] (1) Total transportation cost: The travel cost Z economy_cost and carbon emission cost Z environment_cost The meaning of each cost is described in detail below.

[0051] Driving cost refers to the variable costs incurred during transportation, mainly including various fuel costs, tire wear costs, and vehicle repair and maintenance costs incurred during vehicle use. The calculation formula for the driving cost is as follows:

[0052]

[0053] In the formula, C1 represents the driving cost per unit distance, and in this example, C1 = 5. ij represents the distance between customer i and customer j, represents the decision variable (takes 0, 1; 1 means vehicle k goes from node i to node j; otherwise takes 0), K = {1, 2, ..., n k} represents the vehicle set, n k Represents the total number of vehicles; node set V = V c ∪{0}, 0 represents the collection center, V c ={1,2,...,n c} represents the customer set; n c Represents the total number of customers.

[0054] The carbon emission cost calculation formula is as follows:

[0055]

[0056] Where C2 represents the carbon emission cost per liter of fuel. In this example, C2 = 12.6. Represents the fuzzy fuel consumption of the vehicle traveling from node i to node j.

[0057] Furthermore, the is the fuzzy fuel consumption of the vehicle traveling from customer i to customer j, and the specific expression is:

[0058]

[0059] Among them, λ=ξ / κψ, γ=1 / 1000n tf η,α=τ+gsinθ+gC r cosθ,β=0.5C d ρ1A. Specific parameter descriptions and values ​​are shown in Table 1. i is the cargo volume after the vehicle leaves customer i, is the fuzzy speed of the vehicle traveling between customer i and customer j (calculated based on the distance when the fuzzy travel time is known).

[0060] Table 1 Symbols, definitions and values ​​of parameters related to carbon emissions

[0061] symbol Interpretation Value h Engine friction coefficient (kJ / rev / L) 0.2 M Engine speed (rev / s) 33 <![CDATA[V 发 ]]> Engine displacement(L) 5 ξ Fuel to air mass ratio 1 κ Diesel calorific value (kJ / g) 44 ψ Conversion factor (g / s to L / s) 737 <![CDATA[n tf ]]> Vehicle drivetrain efficiency 0.4 η Engine efficiency parameters 0.9 τ <![CDATA[Acceleration (m / s 2 )]]> 0 g <![CDATA[Acceleration due to gravity (m / s 2 )]]> 9.81 θ Road slope 0 <![CDATA[C r ]]> Rolling resistance coefficient 0.01 w Vehicle curb weight (kg) 6350 <![CDATA[C d ]]> Air resistance coefficient 0.7 <![CDATA[ρ1]]> <![CDATA[Air density (kg / m 3 )]]> 1.2014 A <![CDATA[Frontal area (m 2 )]]> 3.912

[0062] In summary, the minimum total transportation cost Z1 is calculated by the following formula:

[0063] MinimizeZ1=Z economy_cost +Z emission_cost (4)

[0064] (2) The function of maximizing customer satisfaction is expressed as follows:

[0065]

[0066] In the formula, Z2 represents customer satisfaction, n c Represents the number of customers, μ i Represents customer i's satisfaction.

[0067] Exemplarily, the satisfaction function may be adopted as follows:

[0068]

[0069] In the formula, The fuzzy start time of serving customer i for vehicle, e i , l i The lower and upper limits of the service time window for customer i, E i , L i are the lower and upper limits of the service time window that can be extended for customer i.

[0070] (3) The expression of minimizing the imbalance value of driver working time Z3 is as follows:

[0071]

[0072] In the formula, represents the fuzzy travel time between customers, w irepresents the waiting time that the vehicle spends serving customer i, s i The service time required on behalf of customer i.

[0073] Furthermore, the fuzzy travel time is expressed using triangular fuzzy numbers, and the expression of triangular fuzzy numbers is as follows:

[0074]

[0075] In the formula, c l is the lower limit of the triangular fuzzy number, c m is the most likely value of the triangular fuzzy number, that is, the kernel of the triangular fuzzy number, c u is the upper limit of the triangular fuzzy number.

[0076] Furthermore, the constraints of the scheduling model include: first constraint: each customer can only be served by one vehicle and can only be served once (i.e., equations (9) and (10)); second constraint: each vehicle departs from the collection center and returns to the collection center after serving the customer (i.e., equation (11)); third constraint: flow balance, that is, after a vehicle enters a customer's location and completes the service, it must leave the location to visit another customer (i.e., equation (12)); fourth constraint: the current cabin load is equal to the sum of the corresponding product demand of all customers served by the vehicle plus the corresponding product demand of the current service customer (i.e., equation (13)); fifth constraint: the cargo load of each vehicle cabin cannot exceed the rated load of each cabin (i.e., equation (14)); sixth constraint: the collection center There is no waiting time and service time (i.e., formula (15)); the seventh constraint condition: the fuzzy arrival time of the vehicle to the current customer is equal to the fuzzy arrival time to the previous customer plus the waiting service time and service time of the previous customer, plus the fuzzy driving time from the previous customer to the current customer (i.e., formula (16)); the eighth constraint condition: the fuzzy start service time of the vehicle serving the current customer is equal to the fuzzy start service time of the vehicle to the current customer plus the waiting time (i.e., formula (17)); the ninth constraint condition: the value of the waiting time (i.e., formula (18)); the tenth constraint condition: the time window constraint (i.e., formula (19)); the eleventh constraint condition: used for sub-loop elimination (i.e., formula (20)); the twelfth constraint condition: the value of the decision variable (i.e., formula (21)).

[0077] Specifically, the expressions involved in the constraints are as follows:

[0078]

[0079] w0=s0=0(15)

[0080]

[0081] In the formula, represents the load of product p after vehicle k serves customer i, q ip represents the quantity of product p that customer i needs to deliver, and the product set P = {1,2,...,n p},n p represents the maximum number of product types required by customers (i.e. the number of cabins used in the vehicle); Q p Represents the maximum load of each cabin, w i represents the waiting time that a vehicle spends serving customer i, s i represents the service time required by customer i, w0 and s0 represent the waiting time and service time required by the vehicle at the collection center; represents the fuzzy time when the vehicle arrives at customer i, Respectively The lower limit, most likely value, and upper limit of represents the fuzzy travel time between customers; represents the fuzzy service start time at customer i; E i Represents the lower limit of the time window for customer i.

[0082] Exemplarily, the initialization of the main parameters includes: obtaining the main parameter values ​​by an orthogonal experimental design method. In the example of the present invention, the values ​​of the four main parameters are: ps=30, α=2, β=3, ρ=0.4.

[0083] Furthermore, the initialization of the main and auxiliary populations is specifically as follows: individuals in the main population are initialized by a greedy rule and a random rule, and individuals in the auxiliary population are initialized by a random rule.

[0084] Furthermore, the main population is initialized by a greedy rule and a random rule, specifically: the first preset number of individuals is initialized by a greedy rule, and the second preset number of individuals is initialized by a random rule; wherein the sum of the first preset number and the second preset number is the main population size ps. Exemplarily, the first preset number is set to 2, and when ps=30, the second preset number is set to 28.

[0085] Furthermore, individuals in the main population are initialized using a greedy rule, which is achieved through the following steps:

[0086] 1) Construct a customer set to be served based on the customers who need services; use the cargo collection center as the starting point for each vehicle;

[0087] 2) Select the next customer to be served by the vehicle from the set of customers to be served based on the current greedy goal;

[0088] 3) Determine whether the product required by the selected customer exceeds the rated load of each compartment of the vehicle; if not, serve the customer and remove him from the set of customers to be served; otherwise, send the next vehicle;

[0089] 4) Determine whether the set of customers to be served is empty. If it is empty, it means that all customer services have been completed; otherwise, return to 2) and continue execution.

[0090] Furthermore, minimizing the total transportation cost and maximizing customer satisfaction are taken as greedy objectives.

[0091] Furthermore, the random rule initialization is achieved through the following steps:

[0092] S1. Build a customer set to be served based on the customers who need services; take the cargo collection center as the starting point of each vehicle;

[0093] S2. Randomly select a customer from the set of customers to be served as the next service customer of the current vehicle;

[0094] S3, determining whether the product required by the selected customer exceeds the rated load of each compartment of the vehicle; if not, serving the customer and removing him from the set of customers to be served; otherwise, dispatching the next vehicle;

[0095] S4. Determine whether the set of customers to be served is empty. If it is empty, it means that all services for customers have been completed; otherwise, return to S2 to continue execution.

[0096] Furthermore, the update of the pheromone matrix is ​​as follows:

[0097] Let fpFrontSet=(π1,...,π spf ) is the non-inferior solution set in the main population, and the updating method of the element “pheromone concentration” in the pheromone matrix is ​​shown in formula (22).

[0098] τ gen+1 (c i ,c j )=ρ×τ gen (c i ,c j )+Δ(c i ,c j )(twenty two)

[0099] In the formula, τ gen (c i ,c j ) and τ gen+1 (c i ,c j ) represents the pheromone concentration on the current section and the updated section, and the endpoint c of the section i 、c j For customers or collection centers; Δ(ci ,c j ) is c i to c j The newly accumulated pheromone concentration on the road segment is calculated as follows:

[0100]

[0101] In the formula, as well as Represents individual π m The total transportation cost, customer satisfaction and driver working time imbalance value; edge(c i ,c j )∈π m Indicates road segment (c i ,c j ) belongs to the solution of π m ;

[0102] Furthermore, the pheromone matrix is ​​sampled to update the auxiliary population. During the auxiliary population update process, it is necessary to continuously calculate the transfer probability between customers based on the pheromone matrix, and continuously select the next service customer based on the transfer probability to obtain the sampled individual, and update and replace the auxiliary population individuals based on the sampled individual. i ,c j ) is calculated according to the following formula:

[0103]

[0104] Where N j represents the set of customers to be served, η(c i ,c j ) represents the heuristic function value, which is calculated by the following formula:

[0105]

[0106] In the formula, Z1(c i ,c j ) indicates that the vehicle is from c i Go to service c j The transportation cost incurred on the road section, Z2(c i ,c j ) indicates that the vehicle is from c i Go to service c j When customer c j satisfaction.

[0107] Furthermore, the individuals in the updated main and auxiliary populations interact with each other, specifically: the individuals in the auxiliary population are cross-pollinated with the individuals in the non-inferior solution set in the main population, and the sub-individuals generated after the PMX cross-pollination are updated (replaced) with the corresponding individuals in the auxiliary population participating in the cross-pollination. Figure 3In the PMX crossover process shown in the figure, a random portion of individuals in the main population are selected and retained, and the rest are supplemented by individuals in the auxiliary population to complete the crossover operation. It should be noted that during the program writing process, the array subscript starts from 1, so for Figure 3-Figure 4 In the code, 1 is used as the collection center number and 2 is used as the starting number for the customer number.

[0108] Furthermore, the local search of the individuals in the non-inferior solution set of the main population and the auxiliary population after the interaction is specifically as follows: for the non-inferior solution set individuals in the main and auxiliary populations obtained, the following operations are performed in sequence: Figure 4 The “Insert”, “Exchange”, “2-Opt”, and “Special-Insert” neighborhood operations shown adopt a variable neighborhood descent strategy during execution, that is, when an individual becomes better when executing a certain neighborhood operation, the individual will re-execute from the “Insert” neighborhood operation.

[0109] Furthermore, the Special-Insert neighborhood operation is specifically as follows: for the current individual, a customer is randomly selected from the sub-path sequence with the longest driver working time and inserted into any position of the sub-path sequence with the shortest driver working time; wherein the driver working time refers to the fuzzy driving time Waiting time i , Service time i The sum of the three; m =[0,5,2,7,3,0,6,4,8,0,10,1,9,0] for example, there are three sub-paths, and three driver working times are obtained. It should be noted that if there are multiple sub-path sequences with the longest / shortest driver working time, one is randomly selected to participate in the neighborhood operation.

[0110] In order to verify the effectiveness of the scheduling method proposed in this invention, an experimental comparison is made between it and the classic mainstream multi-objective optimization algorithms MOEAD, NSGAII and SPEA2.

[0111] To ensure the fairness of the comparison, the above four algorithms all use the same running time as the termination condition to solve the RC series of examples in the VRPTW standard dataset. All algorithms use Intel i5 processor (2.5GHZ), 16GB memory, Win10 operating system, matlab2022b programming environment, and run 20 times independently when solving each example, and use two indicators such as (26) and (27) for evaluation:

[0112]

[0113] Among them, v Hrepresents the volume of the hypercube from the Hth individual to the reference point on the Pareto front surface obtained by each algorithm, and S is the set of non-inferior solutions. H The pfNum represents the distance between the Hth individual on the Pareto front obtained by each algorithm and the individual on the nearest approximate Pareto front, and pfNum represents the number of individuals on the approximate Pareto front. The approximate optimal Pareto front of the present invention is composed of the first front individual selected by combining the Pareto front individuals obtained by each algorithm.

[0114] The obtained results are shown in Table 2. The bold in the table indicates the optimal value of each example. The larger the HV index, the better, and the smaller the IGD index, the better. As can be seen from Table 2, the test results of the algorithm in all examples are better than those of the other three comparison algorithms, thus verifying the effectiveness of EACO.

[0115] Table 2

[0116]

[0117] In addition, in order to verify the effectiveness of the collaborative global search of the main and auxiliary populations of the key operations in the enhanced ant colony optimization scheduling method proposed in the present invention, it is compared with the variant EACO that does not use this collaborative evolution mechanism. dco A comparison was made. EACO dco The rest of the invention is the same as the present invention, but it has only one population that follows the traditional ant colony optimization evolution rules. Comparisons are made on different examples as shown in Table 3. As can be seen from Table 3, the main and auxiliary populations used in the present invention to coordinate global search have obvious comparative advantages, which verifies the effectiveness of the key operations of the present invention.

[0118] Table 3

[0119]

[0120] According to a second aspect of an embodiment of the present invention, a processor is provided, wherein the processor is used to run a program, wherein when the program is running, the cold chain low-carbon transportation vehicle scheduling method under fuzzy driving time described above is executed.

[0121] According to a third aspect of an embodiment of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned cold chain low-carbon transportation vehicle scheduling methods under fuzzy driving time.

[0122] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0123] The specific implementation modes of the present invention are described in detail above in conjunction with the accompanying drawings, but the present invention is not limited to the above implementation modes, and various changes can be made within the knowledge scope of ordinary technicians in this field without departing from the purpose of the present invention.

Claims

1. A cold chain low-carbon transportation vehicle scheduling method under fuzzy driving time, characterized in that: The following steps are involved: Step 1. Decimal coding is used for the cargo collection center and the customer number. Each vehicle starts from the cargo collection center and returns to the cargo collection center after the service is completed. The load of each compartment of the vehicle meets the capacity constraint. Step 2. Set the main parameters; Step 3, initialization of main and auxiliary populations; Step 4: Merge the main population and auxiliary population into a temporary population; According to multiple optimization target values ​​in the scheduling model, the individuals in the temporary population are subjected to microhabitat preservation operations, and the ps elite individuals retained in the temporary population are used to replace the individuals in the main population to obtain an updated main population; the Pareto non-dominated relationship is used to determine the non-inferior solution set for the updated main population according to multiple optimization target values ​​in the scheduling model; Use the individuals in the non-inferior solution set of the updated main population to update the pheromone matrix, and sample the pheromone matrix to update the auxiliary population; Interact with individuals in the updated main and auxiliary populations; Step 5: Perform local search on the individuals in the non-inferior solution set of the main population and auxiliary population after interaction; Step 6. Determine the termination condition: If the termination condition is met, output the non-inferior solution set of the main population; otherwise, go to Step 4 to continue execution.

2. The cold chain low-carbon transportation vehicle scheduling method under fuzzy travel time according to claim 1 is characterized in that: The scheduling model is specifically as follows: minimizing the total transportation cost, maximizing customer satisfaction and minimizing the imbalance of driver working hours are the optimization objectives, and constraints are constructed.

3. The cold chain low-carbon transportation vehicle scheduling method under fuzzy driving time according to claim 2 is characterized in that: The optimization objectives are specifically: The first optimization goal is to minimize the total transportation cost Z1, which is calculated by the following formula: MinimizeZ1=Z economy_cost +Z emission_cost In the formula, Z1 represents the total transportation cost, Z economy_cost represents the driving cost, Z environment_cost represents the cost of carbon emissions; The second optimization goal: maximize the customer satisfaction function, which is expressed as follows: In the formula, Z2 represents customer satisfaction, n c Represents the number of customers, μ i Represents customer i's satisfaction; The third optimization goal is to minimize the imbalance of drivers’ working time. The expression is as follows: In the formula, Z3 represents the imbalance value of driver working time, represents the fuzzy travel time between customers, represents the decision variable, w i represents the waiting time that the vehicle spends serving customer i, s i represents the service time required for customer i, K = {1,2,...,n k } represents the vehicle set, n k Represents the total number of vehicles.

4. The cold chain low-carbon transportation vehicle scheduling method under fuzzy travel time according to claim 1 is characterized in that: The initialization of the main and auxiliary populations is specifically as follows: individuals in the main population are initialized by a greedy rule and a random rule, and individuals in the auxiliary population are initialized by a random rule.

5. The cold chain low-carbon transportation vehicle scheduling method under fuzzy travel time according to claim 4 is characterized in that: The individuals in the main population are initialized using the greedy rule. The specific steps are as follows: 1) Construct a customer set to be served based on the customers who need services; use the cargo collection center as the starting point for each vehicle; 2) Select the next customer to be served by the vehicle from the set of customers to be served based on the current greedy goal; 3) Determine whether the product required by the selected customer exceeds the rated load capacity of each compartment of the vehicle; If it is not exceeded, the customer is served and removed from the set of customers to be served; otherwise, the next car is sent; 4) Determine whether the set of customers to be served is empty. If it is empty, it means that all customer services have been completed; otherwise, return to 2) and continue execution.

6. The cold chain low-carbon transportation vehicle scheduling method under fuzzy travel time according to claim 5 is characterized in that: Minimizing the total transportation cost and maximizing customer satisfaction are taken as greedy goals.

7. The cold chain low-carbon transportation vehicle scheduling method under fuzzy travel time according to claim 1 is characterized in that: The local search for individuals in the non-inferior solution sets of the main population and the auxiliary population after the interaction is specifically as follows: for the obtained non-inferior solution set individuals in the main and auxiliary populations, "Insert", "Exchange", "2-Opt" and "Special-Insert" neighborhood operations are performed in sequence, and a variable neighborhood descent strategy is adopted during the execution process.

8. The cold chain low-carbon transportation vehicle scheduling method under fuzzy travel time according to claim 7 is characterized in that: The Special-Insert neighborhood operation is specifically as follows: for the current individual, a customer is randomly selected from the sub-path sequence with the longest driver working time among the individuals and inserted into any position of the sub-path sequence with the shortest driver working time; wherein the driver working time refers to the sum of the fuzzy driving time, the waiting time, and the service time.

9. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the cold chain low-carbon transportation vehicle scheduling method under fuzzy driving time described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the cold chain low-carbon transportation vehicle scheduling method under fuzzy driving time described in any one of claims 1-8.

Images