Fresh flower harvesting green cold chain path optimization system and method based on improved wasp optimizer

Abstract

The invention discloses a fresh flower harvesting green cold chain path optimization system and method based on an improved wasp optimizer, and belongs to the field of operation planning optimization. The problems of high energy consumption, large carbon emission pollution amount, low customer satisfaction and the like in the fresh flower cold chain harvesting process are solved. Fresh flowers have the characteristics of being easy to rot and wither and difficult to preserve, fresh flower harvesting is an important link in a fresh flower logistics chain, and the harvesting efficiency and cost have obvious influences on the fresh flower industry. Firstly, a diversified and high-quality initial solution is generated by using random generation and a nearest neighbor algorithm, global search is performed by using a crossover operator of a genetic algorithm in a starting stage to expand a solution space, then a better solution is further searched by using six neighborhood search operators of a wasp spider optimizer, and in order to prevent the algorithm from falling into local optimum, an optimal solution is obtained. A new solution is generated by introducing mutation operation and copulation behaviors of the spider wasp optimizer. The total cost and the carbon dioxide emission of the fresh flowers in the cold-chain transportation process are effectively reduced, and the customer satisfaction is greatly improved.

Description

Technical Field

[0001] The invention relates to a flower harvesting green cold chain path optimization system and method based on an improved spider wasp optimizer, belonging to the field of operations research and optimization. Background Art

[0002] The flower industry is a characteristic industry in Yunnan, creating a large number of employment opportunities and economic income. Flowers are perishable and difficult to preserve, and usually need to be kept fresh in an environment of 2℃-8℃. Due to its particularity, the loss rate of flowers in ordinary logistics is as high as 30%, and the logistics cost accounts for 60% of the total cost. Therefore, the flowers after picking need to be transported back to the refrigerated warehouse by refrigerated vehicles as soon as possible. Cold chain logistics consumes more energy and emits more carbon than ordinary logistics. Therefore, it has become a general trend for cold chain logistics companies to accelerate the transformation to a low-carbon economy. This puts high demands on the path planning of cold chain logistics, which not only requires strict control of the total transportation cost, but also requires effective reduction of carbon emissions. More and more scholars have taken environmental factors such as carbon emissions into consideration when studying the cold chain vehicle path optimization problem. The green cold chain vehicle path optimization problem came into being. On the basis of the traditional cold chain vehicle path optimization problem, it further considers the impact of energy consumption on carbon emissions, and pays more attention to choosing a path solution that balances the economic benefits of the enterprise and is environmentally friendly. The green cold chain path optimization problem is actually a variant of the vehicle routing problem (VRP). VRP was first proposed by George Dantzig and John Ramser in 1959. It is a combinatorial optimization problem that has been widely studied in the field of operations research and optimization. Designing optimization models close to real applications and proposing algorithms with high accuracy and high speed have always been the focus of scholars' research.

[0003] In the cold chain harvesting process, there are problems such as high energy consumption, large carbon emission pollution, and low customer satisfaction. Therefore, many scholars have conducted in-depth research on the cold chain path optimization problem, but the consideration of cost structure is still insufficient. Studies have shown that the VRP problem is an NP-hard problem, and the scale of the solution space is very large. The optimization system based on real scenarios considers more and more factors, from the previous single objective and a small number of constraints to multiple objectives and multiple constraints, etc., and the complexity of the system has been greatly improved. The green cold chain path optimization problem involves more variables and constraints. For the solution algorithm, the exact algorithm cannot avoid the exponential explosion problem, and the classic heuristic algorithm is often trapped in the local optimal value. Therefore, it is necessary to find a system that meets practical applications and an efficient solution method. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a green cold chain path optimization system and method for flower harvesting based on an improved spider wasp optimizer, which can effectively solve the problems existing in the prior art.

[0005] The technical solution adopted by the present invention is: combining the characteristics of flower cold chain path optimization, based on the hard time window constraint, comprehensively considering fixed cost, refrigeration cost, transportation cost, carbon emission cost, flower loss cost, and time penalty cost, a flower harvesting green cold chain path optimization model with the minimum comprehensive cost is constructed. A genetic neighborhood comprehensive spider wasp algorithm is proposed to solve the model. The algorithm first uses random generation and nearest neighbor algorithms to generate diverse and high-quality initial solutions. In the initial search stage, the crossover operator of the genetic algorithm is used to perform a global search to expand the solution space. Then, the six neighborhood search operators of the discretized spider wasp optimizer are used to further search for a better solution. In order to prevent the algorithm from falling into a local optimum, a reversal mutation operation and a discretized spider wasp optimizer mating behavior are introduced to generate a new solution, which reduces the total cost and carbon dioxide emissions of flowers during the cold chain transportation process, and greatly improves customer satisfaction.

[0006] A flower harvesting green cold chain path optimization system based on an improved spider wasp optimizer, comprising:

[0007] Flowers, goods that need to be picked up by refrigerated vehicles;

[0008] The warehouse center, the departure point for refrigerated vehicles and the final location to which the vehicles need to return after harvesting flowers;

[0009] Customers, customers need to wait at a designated location for a refrigerated vehicle to pick up the flowers within the time window;

[0010] Refrigerated vehicles are responsible for picking flowers from the customer's designated location and transporting them back to the refrigerated warehouse.

[0011] Preferably, the flowers are easily rotten and withered and difficult to preserve, and the types and methods of harvesting the flowers are similar.

[0012] Preferably, the customer location, the weight of flowers to be harvested at each location, the customer time window, and the customer service time are already known before the refrigerated vehicle departs. The customer time window is a hard time window. The vehicle can arrive earlier than the customer's earliest service time, but must wait until the customer's earliest service time to start harvesting flowers. If the vehicle arrives later than the latest time specified by the customer, the customer will refuse to accept the service.

[0013] Preferably, each location where flowers are placed can only be visited once by a refrigerated vehicle, and the sum of the weights of flowers to be harvested on the path of each vehicle is less than the vehicle capacity.

[0014] Preferably, the number of vehicles available at the warehouse center, the vehicle loads, and the vehicle travel speeds are known.

[0015] Preferably, the vehicle generates fuel consumption and carbon emissions when transporting flowers, and the engine is turned off when harvesting flowers, so as not to generate fuel consumption and carbon emissions.

[0016] A method for optimizing the green cold chain path of flower harvesting based on an improved spider wasp optimizer, the specific steps are as follows:

[0017] Step 1: Chromosomes are encoded using natural numbers. 0 represents the warehouse center, and natural numbers 1, 2, ..., n represent n customers. The chromosome is a sorted sequence of integers, and the order of arrangement represents the order in which vehicles visit customers. When decoding, the warehouse center is inserted into the sequence consisting of all customers according to the maximum load of the refrigerated vehicle and the customer's time window constraint, and decomposed into multiple different sub-routes. One chromosome represents a complete vehicle path plan.

[0018] Step 2: Generate the initial chromosome using two methods: random generation and nearest neighbor algorithm. The random generation method randomly generates an integer sequence without repeated numbers within the total number of customers as the newly generated solution. The nearest neighbor algorithm generation method first randomly selects the first customer, and then starts from the current customer to find the next closest customer who has not been visited and meets the load constraint and time window constraint. Repeat this operation until all customers are traversed.

[0019] Step 3: Considering the various costs of fixation, refrigeration, transportation, carbon emissions, flower loss and time penalty, an optimization objective function with the minimum total cost is constructed, and the inverse of the objective function is used as the fitness function of the solution;

[0020] Step 4: Three selection strategies: Roulette selection, the individual's fitness value determines the probability of being selected, and the higher the fitness value, the greater the probability of being selected for crossover and mutation; elite selection strategy, select the 10% of individuals with the highest fitness value for crossover and mutation; best individual retention strategy, retain the individuals with the largest fitness value to enter the next generation, and do not participate in crossover and mutation;

[0021] Step 5: At the beginning of the search, the partial mapping crossover operation in the genetic algorithm is used for preliminary search to expand the solution space. The specific operation is to first randomly select two crossover points, then retain the gene fragments between the two crossover points in parent 1 and parent 2, then exchange the gene fragments in the crossover region of parent 1 and parent 2 to generate two new offspring individuals, and finally the newly generated offspring needs to deal with duplicate and missing genes to ensure that each generated offspring is a legal and valid solution;

[0022] Step 6: Based on the solution after searching in Step 5, the discrete spider wasp optimizer further conducts a global search of the solution space by weighing the six neighborhood search operators in the three stages of search phase, follow and escape phase, and nesting behavior phase; in the search phase, the neighborhood search is guided by the common information of the temporal distance matrix and the excellent genes in the elite population, in the follow and escape phase, the neighborhood search is based on the genes in the same route, and in the nesting behavior phase, the neighborhood search is based on the genes in different routes;

[0023] Step 7: Two mutation operations introduce new solutions, reversal mutation, randomly select two positions of the chromosome, determine the reversal operation interval, and reverse the genes in the selected interval; the mating behavior of the discrete spider wasp optimizer, based on the elite population, constructs a temporary set of three consecutive customers with the highest frequency of occurrence, randomly selects two parents 1 and 2, and reserves the same gene fragments in parents 1 and 2 as in the temporary set to offspring 1 and offspring 2 respectively, and the vacant positions in offspring 1 are filled by non-repeated genes in parent 2, and the vacant positions in offspring 2 are filled by non-repeated genes in parent 1.

[0024] The beneficial effects of the present invention are: based on the hard time window constraint, the fixed cost, refrigeration cost, transportation cost, carbon emission cost, flower loss cost, and time penalty cost are comprehensively considered to construct a green cold chain path optimization system for flower harvesting with the lowest comprehensive cost. The introduction of the hard time window can force the system to strictly abide by the time plan to serve customers and maximize customer satisfaction. Punctuality is necessary for cold chain transportation of perishable goods or time-sensitive goods. While ensuring the quality of flowers, because the flowers can be picked up from customers on time, customers can trust the logistics company and expand the market influence of the logistics company. Due to the large number of factors to be considered, the complexity of the problem is greatly increased, and a better algorithm is needed to solve it. Therefore, we propose a genetic neighborhood integrated spider wasp algorithm to solve it. The algorithm has a high convergence speed and good stability. It can find a flower harvesting path plan with a smaller total cost and carbon emissions, effectively reducing the total cost and carbon dioxide emissions of flowers during cold chain transportation, and greatly improving customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the system model of the present invention;

[0026] Figure 2 It is a schematic diagram of chromosome encoding and decoding of the present invention;

[0027] Figure 3 It is a schematic diagram of the operation of partial mapping crossover of the genetic algorithm of the present invention;

[0028] Figure 4It is a schematic diagram of the operation of the search operator 1 in the follow and escape phases in the discretized spider wasp optimizer of the present invention;

[0029] Figure 5 It is a schematic diagram of the operation of the search operator 2 in the follow and escape phases in the discretized spider wasp optimizer of the present invention;

[0030] Figure 6 It is a schematic diagram of the operation of the search operator 1 in the nesting behavior stage in the discretized spider wasp optimizer of the present invention;

[0031] Figure 7 It is a schematic diagram of the operation of the search operator 2 in the nesting behavior stage in the discretized spider wasp optimizer of the present invention;

[0032] Figure 8 is a schematic diagram of the reverse mutation operation of the genetic algorithm of the present invention;

[0033] Fig. 9 is a schematic diagram of the mating behavior mutation operation in the discretized spider wasp optimizer of the present invention;

[0034] Fig.10 It is a schematic diagram of the solution algorithm flow of the present invention. DETAILED DESCRIPTION

[0035] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods.

[0036] Example 1: Figure 1 As shown, a flower harvesting green cold chain path optimization system based on an improved spider wasp optimizer comprises:

[0037] Flowers, goods that need to be picked up by refrigerated vehicles;

[0038] The warehouse center, the departure point for refrigerated vehicles and the final location to which the vehicles need to return after harvesting flowers;

[0039] Customers, customers need to wait at a designated location for a refrigerated vehicle to pick up the flowers within the time window;

[0040] Refrigerated vehicles are responsible for picking flowers from the customer's designated location and transporting them back to the refrigerated warehouse.

[0041] Furthermore, the flowers are easy to rot and wither, and difficult to preserve. The types of flowers that need to be harvested by the refrigerated vehicles are the same, such as fresh cut flowers (roses), and the way of harvesting flowers when the vehicles arrive at the customer's location is the same.

[0042] Furthermore, the customer location, the weight of flowers to be harvested at each location, the customer time window, and the customer service time are already known before the refrigerated vehicle departs. The customer time window is a hard time window. The vehicle can arrive earlier than the customer's earliest service time, but must wait until the customer's earliest service time to start harvesting flowers. If the vehicle arrives later than the latest time specified by the customer, the customer will refuse to accept the service.

[0043] Furthermore, each location where flowers are placed can only be visited once by a refrigerated truck, and the sum of the weights of flowers that need to be harvested on the path of each vehicle must be less than the vehicle capacity.

[0044] Furthermore, the number of available vehicles, vehicle loads, and vehicle travel speeds at the warehouse center are known.

[0045] Furthermore, the vehicle consumes fuel and generates carbon emissions when transporting flowers, but when harvesting flowers, the engine is turned off and no fuel consumption and carbon emissions are generated.

[0046] like Figure 2 As shown, the chromosome is encoded using natural numbers, 0 represents the warehouse center, and natural numbers 1, 2, ..., 8 represent 8 customers. The chromosome encoding is a string of integer sequences. When decoding, 0 is inserted into the integer array composed of customers according to the maximum load of the vehicle and the customer's time window constraint. The decoded chromosome represents a complete path, and the interval between 0 and 0 represents a route.

[0047] like Figure 3 As shown, two crossover points are randomly selected, the gene fragments between the crossover points in parent 1 and parent 2 are retained, and then the gene fragments in the crossover region of parent 1 and parent 2 are exchanged, and finally the duplicated genes in the chromosome are deleted and the missing genes are filled, thus generating offspring 1 and offspring 2.

[0048] like Figure 4 As shown, two adjacent edges in the route are randomly selected to be disconnected and reconnected. If the total cost of the newly formed path is reduced, the operation is performed, otherwise the status quo is maintained.

[0049] like Figure 5 As shown, a small segment in the route is randomly selected, which is 2 consecutive nodes, and moved to another position in the route. If the cost of the new path after the move is lower than the cost of the original path, the move is accepted.

[0050] like Figure 6 As shown in the figure, a customer in the current route is randomly deleted and inserted into other different routes. If the total cost of the new path formed is less than the total cost of the original path, the operation is performed, otherwise it is not performed.

[0051] like Figure 7As shown, two different routes are randomly selected and one of their customers is exchanged. If the total cost of the newly formed path is less than the total cost of the original path, the exchange operation is performed, otherwise it is not performed.

[0052] like Figure 8 As shown, two positions of the chromosome are randomly selected, the reversal operation interval is determined, and the genes in the selected interval are reversed.

[0053] like Fig. 9 As shown, based on the elite population, a temporary set consisting of three consecutive customers with the highest occurrence frequency is constructed, and two parents 1 and 2 are randomly selected. The gene fragments in parents 1 and 2 that are the same as those in the temporary set are reserved for offspring 1 and offspring 2 respectively, and the vacant positions in offspring 1 are filled by non-repeated genes in parent 2, and the vacant positions in offspring 2 are filled by non-repeated genes in parent 1.

[0054] like Fig.10 As shown in the figure, firstly, two methods, random generation and nearest neighbor algorithm, are used to generate initial solutions with diversified and high quality; the crossover operation of genetic algorithm is performed in the initial search phase to conduct preliminary search; then the six neighborhood search operators in the search phase, follow-up and escape phase, and nesting behavior phase of the discretized spider wasp optimizer are weighed to further search the solution space globally. Finally, the mutation operation of genetic algorithm and the mating operation of discretized spider wasp optimizer are used to introduce new solutions. The search operator 1 in the search phase of the discretized spider wasp optimizer is: select the 10% chromosomes with the best fitness in the population to form an elite population, count the customers with the highest frequency at each position of the chromosome in the elite population, construct a temporary path, retain the customers with the same temporary path, and delete the rest of the customers. Finally, the connection probability between two customers is calculated based on the common information of the spatiotemporal distance matrix and the elite population, and the deleted customers are inserted back into the path, where the larger the connection probability, the smaller the incremental cost of the path when the customer is inserted into the position. The search operator 2 in the search phase is: retain the longest common subsequence of the individual and the elite chromosome, delete the remaining customers, and finally calculate the connection probability between two customers based on the space-time distance matrix and the common information of the elite population, and insert the deleted customers back into the path.

[0055] A method for optimizing the green cold chain path of flower harvesting based on an improved spider wasp optimizer, the specific steps are as follows:

[0056] Step 1: Chromosomes are encoded using natural numbers. 0 represents the warehouse center, and natural numbers 1, 2, ..., n represent n customers. The chromosome is a sorted sequence of integers, and the order of arrangement represents the order in which vehicles visit customers. When decoding, the warehouse center is inserted into the sequence consisting of all customers according to the maximum load of the refrigerated vehicle and the customer's time window constraint, and decomposed into multiple different sub-routes. One chromosome represents a complete vehicle path plan.

[0057] Step 2: Generate the initial chromosome using two methods: random generation and nearest neighbor algorithm. The random generation method randomly generates an integer sequence without repeated numbers within the total number of customers as the newly generated solution. The nearest neighbor algorithm generation method first randomly selects the first customer, and then starts from the current customer to find the next closest customer who has not been visited and meets the load constraint and time window constraint. Repeat this operation until all customers are traversed.

[0058] Step 3: Considering the various costs of fixation, refrigeration, transportation, carbon emissions, flower loss and time penalty, an optimization objective function with the minimum total cost is constructed, and the inverse of the objective function is used as the fitness function of the solution;

[0059] Step 4: Three selection strategies: Roulette selection, the individual's fitness value determines the probability of being selected, and the higher the fitness value, the greater the probability of being selected for crossover and mutation; elite selection strategy, select the 10% of individuals with the highest fitness value for crossover and mutation; best individual retention strategy, retain the individuals with the largest fitness value to enter the next generation, and do not participate in crossover and mutation;

[0060] Step 5: At the beginning of the search, the partial mapping crossover operation in the genetic algorithm is used for preliminary search to expand the solution space. The specific operation is to first randomly select two crossover points, then retain the gene fragments between the two crossover points in parent 1 and parent 2, then exchange the gene fragments in the crossover region of parent 1 and parent 2 to generate two new offspring individuals, and finally the newly generated offspring needs to deal with duplicate and missing genes to ensure that each generated offspring is a legal and valid solution;

[0061] Step 6: Based on the solution after searching in Step 5, the discrete spider wasp optimizer further conducts a global search of the solution space by weighing the six neighborhood search operators in the three stages of search phase, follow and escape phase, and nesting behavior phase; in the search phase, the neighborhood search is guided by the common information of the temporal distance matrix and the excellent genes in the elite population, in the follow and escape phase, the neighborhood search is based on the genes in the same route, and in the nesting behavior phase, the neighborhood search is based on the genes in different routes;

[0062] Step 7: Two mutation operations introduce new solutions, reversal mutation, randomly select two positions of the chromosome, determine the reversal operation interval, and reverse the genes in the selected interval; the mating behavior of the discrete spider wasp optimizer, based on the elite population, constructs a temporary set of three consecutive customers with the highest frequency of occurrence, randomly selects two parents 1 and 2, and reserves the same gene fragments in parents 1 and 2 as in the temporary set to offspring 1 and offspring 2 respectively, and the vacant positions in offspring 1 are filled by non-repeated genes in parent 2, and the vacant positions in offspring 2 are filled by non-repeated genes in parent 1.

[0063] 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 flower harvesting green cold chain path optimization system based on an improved spider wasp optimizer, characterized in that: include: Flowers, goods that need to be picked up by refrigerated vehicles; The warehouse center, the departure point for refrigerated vehicles and the final location to which the vehicles need to return after harvesting flowers; Customers, customers need to wait at a designated location for a refrigerated vehicle to pick up the flowers within the time window; Refrigerated vehicles are responsible for picking flowers from the customer's designated location and transporting them back to the refrigerated warehouse.

2. A flower harvesting green cold chain path optimization system based on an improved spider wasp optimizer according to claim 1, characterized in that: The flowers are easily perishable and withered and difficult to preserve. The types of flowers harvested by the refrigerated vehicles are the same as the way the flowers are harvested when arriving at the customer's location.

3. The flower harvesting green cold chain path optimization system based on the improved spider wasp optimizer according to claim 1, characterized in that: The customer location, the weight of flowers to be harvested at each location, the customer time window, and the customer service time are already known before the refrigerated vehicle departs. The customer time window is a hard time window. The vehicle can arrive earlier than the customer's earliest service time, but must wait until the customer's earliest service time to start harvesting flowers. If the vehicle arrives later than the latest time specified by the customer, the customer will refuse to accept the service.

4. The flower harvesting green cold chain path optimization system based on the improved spider wasp optimizer according to claim 1, characterized in that: Each flower placement location can only be visited once by a refrigerated truck, and the sum of the weights of flowers that need to be harvested on each vehicle's path must be less than the vehicle's capacity.

5. The flower harvesting green cold chain path optimization system based on the improved spider wasp optimizer according to claim 1, characterized in that: The number of available vehicles, vehicle loads, and vehicle speeds at the warehouse center are known.

6. The flower harvesting green cold chain path optimization system based on the improved spider wasp optimizer according to claim 1, characterized in that: The vehicle consumes fuel and generates carbon emissions when transporting flowers. When harvesting flowers, the engine is turned off and no fuel consumption and carbon emissions are generated.

7. A method for optimizing the green cold chain path of flower harvesting based on an improved spider wasp optimizer, characterized in that: The specific steps are as follows: Step 1: Chromosomes are encoded using natural numbers. 0 represents the warehouse center, and natural numbers 1, 2, ..., n represent n customers. The chromosome is a sorted sequence of integers, and the order of arrangement represents the order in which vehicles visit customers. When decoding, the warehouse center is inserted into the sequence consisting of all customers according to the maximum load of the refrigerated vehicle and the customer's time window constraint, and decomposed into multiple different sub-routes. One chromosome represents a complete vehicle path plan. Step 2: Generate the initial chromosome using two methods: random generation and nearest neighbor algorithm. The random generation method randomly generates an integer sequence without repeated numbers within the total number of customers as the newly generated solution. The nearest neighbor algorithm generation method first randomly selects the first customer, and then starts from the current customer to find the next closest customer who has not been visited and meets the load constraint and time window constraint. Repeat this operation until all customers are traversed. Step 3: Considering the various costs of fixation, refrigeration, transportation, carbon emissions, flower loss and time penalty, an optimization objective function with the minimum total cost is constructed, and the inverse of the objective function is used as the fitness function of the solution; Step 4: Three selection strategies: Roulette selection, the individual's fitness value determines the probability of being selected, and the higher the fitness value, the greater the probability of being selected for crossover and mutation; elite selection strategy, select a set proportion of individuals with the highest fitness value for crossover and mutation; best individual retention strategy, retain the individuals with the largest fitness value to enter the next generation, and do not participate in crossover and mutation; Step 5: At the beginning of the search, the partial mapping crossover operation in the genetic algorithm is used for preliminary search to expand the solution space. The specific operation is to first randomly select two crossover points, then retain the gene fragments between the two crossover points in parent 1 and parent 2, then exchange the gene fragments in the crossover region of parent 1 and parent 2 to generate two new offspring individuals, and finally the newly generated offspring needs to deal with duplicate and missing genes to ensure that each generated offspring is a legal and valid solution; Step 6: Based on the solution after searching in Step 5, the discrete spider wasp optimizer further conducts a global search of the solution space by weighing the six neighborhood search operators in the three stages of search phase, follow and escape phase, and nesting behavior phase; in the search phase, the neighborhood search is guided by the common information of the temporal distance matrix and the excellent genes in the elite population, in the follow and escape phase, the neighborhood search is based on the genes in the same route, and in the nesting behavior phase, the neighborhood search is based on the genes in different routes; Step 7: Two mutation operations introduce new solutions, reversal mutation, randomly select two positions of the chromosome, determine the reversal operation interval, and reverse the genes in the selected interval; the mating behavior of the discrete spider wasp optimizer, based on the elite population, constructs a temporary set of three consecutive customers with the highest frequency of occurrence, randomly selects two parents 1 and 2, and reserves the same gene fragments in parents 1 and 2 as in the temporary set to offspring 1 and offspring 2 respectively, and the vacant positions in offspring 1 are filled by non-repeated genes in parent 2, and the vacant positions in offspring 2 are filled by non-repeated genes in parent 1.

8. The flower harvesting green cold chain path optimization method based on the improved spider wasp optimizer according to claim 7, characterized in that: In the Step 4, the ratio is set to 10%.

Images