Production and transportation collaborative optimization method based on two-stage differential evolution four-layer supply chain network

Abstract

The invention discloses a two-stage differential evolution four-layer supply chain network-based production and transportation collaborative optimization method. The method comprises the steps of 1, constructing a four-layer supply chain network production and transportation collaborative optimization model; 2, solving an optimal production scheme set of the first stage; and 3, solving an optimal transportation scheme set of two-stage optimization. According to the method, the optimal production and transportation scheme with small CO2 emission can be obtained, effective control over carbon emission of the whole link and efficient allocation of resources are achieved, and an efficient and feasible important decision support is provided for green design and optimization of the supply chain.

Description

Technical Field

[0001] The present invention relates to the field of supply chain network management, and more specifically to the coordinated optimization of supply chain network transportation and production based on differential evolution to reduce the overall CO2 emissions of the network. , used in green design management of supply chain networks. Background Art

[0002] In modern supply chain management, a large amount of waste is generated throughout the entire process from raw material acquisition to product manufacturing, transportation, and use, causing serious environmental pollution and damage to the ecosystem. As the global economic and social development has increasingly higher requirements for green and sustainable development, building a green supply chain has become an inevitable choice for corporate development. The supply chain network design with carbon emissions (CO2) as the main optimization target aims to achieve low-carbon goals by optimizing production and transportation plans. However, the application of existing technologies in this field still has significant problems and limitations.

[0003] Many existing optimization methods usually regard suppliers, manufacturers, warehouses, and retailers in the supply chain network as independent units, and fail to fully consider the multi-level, cross-link complex coupling relationships between them. This approach easily leads to the dependencies and interactivity of each link in the supply chain not being accurately reflected, making it difficult for the optimization results to truly match the actual supply chain design requirements. Secondly, when dealing with multiple constraints in the supply chain network, existing technologies often adopt an isolated approach, optimizing the constraints of each link such as the supplier's delivery time, the factory's production capacity, and the warehouse's storage restrictions separately, ignoring the coupling effects between different links. This localized approach may lead to only local optimization, rather than the global optimality of the entire supply chain, and it is impossible to obtain the optimal production and transportation solutions.

[0004] Therefore, in the context of green design of supply chain networks, a single global optimization strategy is difficult to fully cope with the multi-dimensional and multi-constrained green design and optimization of supply chains, and it is difficult to provide enterprises with comprehensive and effective decision-making support for green supply chain optimization solutions. Summary of the invention

[0005] In order to overcome the shortcomings of the prior art, the present invention proposes a production and transportation collaborative optimization method based on a two-stage differential evolution four-layer supply chain network, in order to reduce the total CO2 emissions, achieve effective control of carbon emissions in the entire link and efficient allocation of resources, thereby providing efficient and feasible important decision-making support for the green design and optimization of the supply chain, and better solve the problem of collaborative optimization of production and transportation in the supply chain network.

[0006] The present invention adopts the following technical solutions to solve the technical problems:

[0007] The production and transportation collaborative optimization method based on a two-stage differential evolution four-layer supply chain network of the present invention is characterized in that the supplier set Raw materials collection provided Transport to the factory for assembly Produce and obtain product collection After that, it will be transported to the warehouse center for collection. Store and finally transport to retailers for collection In the four-layer supply chain network, let Any supplier, Any raw material, Represents any factory, Any product, represents any warehouse center, Any retailer; will produce the product Raw materials required The consumption is recorded as ,supplier Raw materials The maximum supply is recorded as ,supplier Get unit raw materials The CO2 emissions generated when ,factory Production products The upper limit of production capacity is ,factory Storing surplus units of raw materials The amount of CO2 emissions generated is recorded as ,factory Production unit products The amount of CO2 emissions generated is recorded as , warehouse center About Products The upper limit of storage capacity is recorded as , warehouse center Storage Unit Products The amount of CO2 emissions generated is recorded as , Retailer About Products The demand for ;supplier Transport unit raw materials To the factory The resulting CO2 emissions are recorded as ,factory Unit Product To the warehouse center The amount of CO2 emissions generated is recorded as , warehouse center Transport unit product To retailers The amount of CO2 emissions generated is recorded as The production and transportation collaborative optimization method is carried out in the following steps:

[0008] Step 1: Build a production and transportation collaborative optimization model:

[0009] According to the coupling relationship between production decision variables and transportation decision variables in the production and transportation process of the four-layer supply chain network, the production and transportation decision variables are divided into the decision variables of the factory output in the first stage and the transportation plan decision variables of the raw materials or final products in the second stage. The decision variables of the factory output include: the number of product f produced by factory k ; The decision variables for the transportation plan of the raw materials or final products include: supplier Transport of raw materials To the factory Number of ,factory Transport Products To the warehouse center Number of , warehouse center Transport Products To retailers Number of ;

[0010] Taking the minimum total CO2 emissions in the supply chain network as the goal, the objective function of the production and transportation collaborative optimization model is constructed using formula (1): :

[0011] (1)

[0012] The constraints of the production and transportation collaborative optimization model are constructed using equations (2) to (5):

[0013] (2)

[0014] (3)

[0015] (4)

[0016] (5)

[0017] (6)

[0018] (7)

[0019] Step 2: Based on the differential evolution algorithm, optimize the decision variables of the factory output in the first stage to obtain the optimal production plan set:

[0020] Step 2.1, initialization parameters, including: population size is , the current number of iterations is , the maximum number of iterations is ; The mutation operator is and crossover operator ; Initialize t=1;

[0021] Step 2.2: The tth generation population The encoding:

[0022] Step 2.2.1: Take the production plan consisting of the quantity of each product produced by each factory as a production individual, so that the t-generation population consists of N individuals. ,in, represents the tth generation population The i-th production individual in , and the encoding method of each production individual is integer encoding, the encoding length dim of each production individual is the product of the number of factories and the number of product types, and each bit of the encoding in each production individual represents the quantity of a certain product produced by a certain factory;

[0023] Step 2.2.2, let the upper bound of any a-th bit code be , the lower bound is ; a∈[1,dim]; and ∈ , represents the upper bound set, ∈ , represents the lower bound set;

[0024] Step 2.3: Use greedy selection to select the i-th production individual Perform a preliminary evaluation and obtain the i-th production individual Minimum CO2 emissions , thus obtaining the minimum CO2 emission set of the tth generation ;

[0025] Step 2.4: According to ,right After mutation and crossover operations, the t-th generation offspring population is obtained , and then use greedy selection to Calculate total CO2 emissions , and and Compare each production individual at the corresponding position in the t+1 generation, and select the production individual with the smallest total emission to join the t+1 generation population. :

[0026] Step 2.5, after assigning t+1 to t, if t> , then the final Generation population This is the optimal production plan set; otherwise, return to step 2.4 and execute sequentially;

[0027] Step 3: Based on the differential evolution algorithm, optimize the decision variables of the transportation plan for raw materials or final products in the first and second stages to obtain the optimal transportation plan set:

[0028] Step 3.1, initialization parameters, including: population size is , the current number of iterations is The maximum number of iterations is , initialize the number of iterations ;

[0029] Step 3.2: Itth generation population The encoding:

[0030] Step 3.2.1. Generation population The i-th individual The procurement plan, production allocation plan, storage and transshipment plan form a transportation plan and serve as the i-th transportation individual , thus obtaining the It generation population , and the encoding method of each transport individual is integer encoding, The length of the code for each transported individual; express No. Bit encoding;

[0031] Step 3.2.2, let any b-th bit be encoded The upper bound of , the lower bound is ; b∈[1,D]; and ∈ , represents the upper bound set, ∈ , represents the lower bound set;

[0032] Step 3.3: Use the repair strategy to Make corrections and get the corrected Transport units ; Thus, the revised It generation population is obtained ,in, express The Bit encoding;

[0033] Step 3.4: Use differential evolution algorithm to After crossover and mutation operations, the It-th generation offspring population is obtained. ;

[0034] Step 3.5: Use the repair strategy to Each individual in is corrected to obtain the corrected offspring population ;

[0035] Step 3.6: Use formula (1) to calculate Total CO2 emissions Z( )and Total CO2 emissions Z( ), and and Compare each transport individual at the corresponding position in the , and select the transport individual with the smallest total emission to join the It+1 generation transport population ;

[0036] Step 3.5: Assign to ,like , then the final Generation population This is the optimal transportation solution set. Otherwise, return to step 3.4 and execute sequentially.

[0037] The production and transportation collaborative optimization method based on a two-stage differential evolution four-layer supply chain network described in the present invention is also characterized in that step 2.3 includes:

[0038] Step 2.3.1, with the goal of minimizing the sum of the CO2 emissions generated when suppliers obtain unit raw materials and the CO2 emissions generated when suppliers transport unit raw materials to the factory, sort all suppliers in ascending order, and then judge each supplier in turn. If the supplier Raw materials Maximum Supply Greater than or equal to factory Raw materials Procurement needs , then select supplier and get Otherwise, continue to select the next supplier to satisfy the factory Raw materials The quantity of raw materials transported to the factory by all suppliers under the procurement demand is taken as the i-th production individual The procurement plan of the system; where the dimension of the procurement plan is Ns;

[0039] Step 2.3.2, Factory The goal is to minimize the sum of the CO2 emissions generated by the delivery of a unit product to the warehouse center and the CO2 emissions generated by the storage of a unit product in the warehouse center. After sorting all warehouse centers in ascending order, each warehouse center after sorting is judged in turn. If the warehouse center The upper limit of storage capacity Greater than or equal to the quantity of product f produced by factory k , then select Warehouse Center and get Otherwise, continue to select the next warehouse center, so as to obtain all warehouse centers that meet the quantity of factory k to produce product f. The storage quantity and as the i-th production individual The production allocation plan of ; wherein the dimension of the production allocation plan is Nm;

[0040] Step 2.3.3, with the goal of minimizing the CO2 emissions generated by the warehouse center transporting unit products to retailers, sort all warehouse centers in ascending order, and then judge each warehouse center in turn. If the warehouse center Meet the Retailers About Products Demand , then select Warehouse Center and get Otherwise, continue to the next warehouse center to satisfy the retailer About Products All warehouse centers transport products under demand To retailers The number of and as the i-th production individual Storage and transportation plan; where the dimension of the storage and transportation plan is Nd;

[0041] Step 2.3.4: According to the i-th production individual The procurement plan, production allocation plan, storage and transportation plan are calculated using formula (1) Total CO2 emissions .

[0042] Further, the step 2.4 includes:

[0043] Step 2.4.1. From Three randomly selected production individuals , using formula (8) to obtain the i-th variant production individual :

[0044] (8)

[0045] Step 2.4.2: Update using formula (9) The Bit encoding :

[0046] (9)

[0047] Step 2.4.3: Use formula (10) to get the i-th offspring individual ;

[0048] (10)

[0049] In formula (9), express The Bit encoding; express The Bit encoding; Represents a random number between 0 and 1;

[0050] Step 2.4.4: Use formula (11) to get the i-th individual in the t+1 generation population :

[0051] (11).

[0052] Further, the step 3.3 includes:

[0053] Step 3.3.0, initialize b=1;

[0054] Step 3.3.1, judgment Characterizing the procurement program Exceeding the supplier Raw materials Maximum Supply , if it exceeds, then use formula (12) to Make adjustments and get suppliers Transport of raw materials To the factory Adjusted quantity , otherwise, let = ;

[0055] Step 3.3.2: After assigning b+1 to b, if b>Ns, execute step 3.3.3; otherwise, return to step 3.3.1 and execute sequentially;

[0056] (12)

[0057] Step 3.3.3, let b = 1;

[0058] Step 3.3.4, judgment Characterizing the procurement program Does it satisfy formula (5)? ​​If it does, let , and execute step 3.3.6, otherwise, execute step 3.3.5;

[0059] Step 3.3.5: With the goal of minimizing the sum of the CO2 emissions generated when suppliers obtain unit raw materials and the CO2 emissions generated when suppliers transport unit raw materials to the factory, sort all suppliers in ascending order and determine whether each supplier satisfies formula (13). If so, select the corresponding supplier. , and calculate the supplier Transport of raw materials To the factory The adjusted amount + , so that ; Otherwise, continue to determine the next supplier;

[0060] (13)

[0061] Step 3.3.6, after assigning b+1 to b, if b>Ns, execute step 3.3.7, otherwise execute step 3.3.4;

[0062] Step 3.3.7, judgement Characterizing the production allocation scheme Whether the production capacity of factory k for product f is exceeded , if it exceeds, then use formula (14) to Make adjustments and get the factory Transport Products To the warehouse center Adjusted quantity , otherwise, let = ;

[0063] (14)

[0064] Step 3.3.8: After assigning b+1 to b, if b>Nm+Ns, execute step 3.3.9; otherwise, return to step 3.3.7 and execute sequentially;

[0065] Step 3.3.9, let b = Ns + 1;

[0066] Step 3.3.10, judgement Characterizing the production allocation scheme Does it satisfy formula (6)? If it does, let ; and execute step 3.3.13, otherwise, execute step 3.3.11;

[0067] Step 3.3.11: Use formula (15) to get the factory Transport Products To the warehouse center Adjusted quantity , so that ;

[0068] (15)

[0069] Step 3.3.12: Assign b+1 to b. If b>Ns+Nm, execute step 3.3.13; otherwise, return to step 3.3.10 and execute sequentially.

[0070] Step 3.3.13, judgement Characterization of storage and transport scenarios Whether the storage capacity of the warehouse center for the product has been exceeded , if it exceeds, then use formula (16) to Make adjustments to get the warehouse center Transport Products To retailers Adjusted quantity ,otherwise, = ;

[0071] (16)

[0072] Step 3.3.14: Assign b+1 to b. If b>Nm+Ns+Nd, execute step 3.3.15. Otherwise, return to execute step 3.3.13 in sequence.

[0073] Step 3.3.15, let b = Ns + Nm + 1;

[0074] Step 3.3.16, judgment Characterization of storage and transport solutions Does it satisfy formula (7)? If it does, let , and execute step 3.3.18, otherwise, execute step 3.3.17;

[0075] Step 3.3.17: Use formula (17) to get the transport product of warehouse center j To retailers Adjusted quantity , so that ;

[0076] (17)

[0077] Step 3.3.18, let b+1 be assigned to b, if b Ns+Nm+Nd, then execute step 3.3.16; otherwise, it means the It generation population is obtained The i-th individual .

[0078] An electronic device of the present invention includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the production and transportation collaborative optimization method based on a two-stage differential evolution four-layer supply chain network, and the processor is configured to execute the program stored in the memory.

[0079] The present invention provides a computer-readable storage medium, and the computer-readable storage medium stores a computer program, which is characterized in that when the computer program is run by a processor, the steps of the production and transportation collaborative optimization method based on a two-stage differential evolution four-layer supply chain network are executed.

[0080] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0081] 1. The present invention takes the minimum total CO2 emissions in the supply chain network as the goal, considers the flow and capacity constraints in the network, and establishes a multi-raw material, multi-product, four-layer supply chain production and transportation collaborative optimization model. The model designs a two-stage decision optimization for decision variables. The first stage optimizes the optimal production plan set, and the second stage optimizes the product and raw material transportation plan in the network. Using the differential evolution algorithm for global search and local repair, through multiple iterations, the approximate optimal production and transportation plan is finally obtained, thereby reducing the total CO2 emissions, achieving effective control of carbon emissions in the entire chain and efficient allocation of resources, improving social satisfaction, and providing efficient and feasible important decision support for the green design and optimization of the supply chain.

[0082] 2. The present invention optimizes factory production in the first stage, and obtains production individuals in the population based on the greedy strategy, which ensures the quality of the production individuals in the first stage and also ensures its diversity. In the second stage, a repair strategy is implemented for the transportation individuals that do not meet the constraints, and the network product transportation plan is dynamically adjusted through local search, which improves the quality of the transportation individuals in the population. It helps the supply chain make reasonable decisions and obtain the best production and transportation plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 This is a four-layer supply chain network structure diagram of the present invention;

[0084] Figure 2This is a schematic diagram of the method for collaborative optimization of production and transportation in the supply chain of the present invention. DETAILED DESCRIPTION

[0085] In this embodiment, a production and transportation collaborative optimization method based on a four-layer supply chain network with two-stage differential evolution is to combine the supplier set Raw materials collection provided Transport to the factory for assembly Produce and obtain product collection After that, it will be transported to the warehouse center for collection. Store and finally transport to retailers for collection In the four-layer supply chain network, according to the above process, the supply chain network structure diagram is as follows: Figure 1 As shown, where Any supplier, Any raw material, Represents any factory, Any product, represents any warehouse center, Any retailer; will produce the product Raw materials required The consumption is recorded as ,supplier Raw materials The maximum supply is recorded as ,supplier Get unit raw materials The CO2 emissions generated when ,factory Production products The upper limit of production capacity is ,factory Storing surplus units of raw materials The amount of CO2 emissions generated is recorded as ,factory Production unit products The amount of CO2 emissions generated is recorded as , warehouse center About Products The upper limit of storage capacity is recorded as , warehouse center Storage Unit Products The amount of CO2 emissions generated is recorded as , Retailer About Products The demand for ;supplier Transport unit raw materials To the factory The resulting CO2 emissions are recorded as ,factory Unit Product To the warehouse center The amount of CO2 emissions generated is recorded as , warehouse center Transport unit product To retailers The amount of CO2 emissions generated is recorded as Specifically, Figure 2 As shown, the production and transportation collaborative optimization method is carried out in the following steps:

[0086] Step 1: Build a production and transportation collaborative optimization model:

[0087] According to the coupling relationship between production decision variables and transportation decision variables in the production and transportation process of the four-layer supply chain network, the production and transportation decision variables are divided into the decision variables of the factory output in the first stage and the transportation plan decision variables of the raw materials or final products in the second stage. The decision variables of the factory output include: the number of product f produced by factory k ; The decision variables for the transportation plan of raw materials or final products include: supplier Transport of raw materials To the factory Number of ,factory Transport Products To the warehouse center Number of , warehouse center Transport Products To retailers Number of ;

[0088] Taking the minimum total CO2 emissions in the supply chain network as the goal, the objective function of the production and transportation collaborative optimization model is constructed using formula (1): , specifically including CO2 emissions from raw material suppliers, CO2 emissions from factory final product production, CO2 emissions from network transportation, and CO2 emissions from storage of raw materials and final products:

[0089] (1)

[0090] The constraints of the production and transportation collaborative optimization model are constructed using equations (2) to (5):

[0091] (2)

[0092] (3)

[0093] (4)

[0094] (5)

[0095] (6)

[0096] (7)

[0097] Formula (2) indicates that the quantity of raw materials provided by any supplier cannot exceed its supply capacity ;

[0098] Formula (3) indicates that the quantity of final products produced by any factory cannot exceed its production capacity ;

[0099] Formula (4) indicates that the number of final products stored in any warehouse center cannot exceed its storage capacity ;

[0100] Formula (5) means that any factory receives the supplier set The amount of raw materials provided needs to be greater than the amount of raw materials required to produce the final product;

[0101] Formula (6) shows that the quantity of final products produced by any factory and the quantity of products transferred to the warehouse center are The number of is equal;

[0102] Formula (7) indicates that any retailer from the warehouse center set The final product transferred needs to be less than the retailer's demand for the final product .

[0103] Step 2: Based on the differential evolution algorithm, optimize the decision variables of the factory output in the first stage to obtain the optimal production plan set:

[0104] Step 2.1, initialization parameters, including: population size is , the current number of iterations is , the maximum number of iterations is ; The mutation operator is and crossover operator ; Initialize t=1;

[0105] Step 2.2: The tth generation population The code:

[0106] Step 2.2.1: Take the production plan consisting of the quantity of each product produced by each factory as a production individual, so that the t-generation population consists of N individuals. ,in, represents the tth generation population The i-th production individual in , and the encoding method of each production individual is integer encoding, the encoding length dim of each production individual is the product of the number of factories and the number of product types, and each bit of the encoding in each production individual represents the quantity of a certain product produced by a certain factory;

[0107] Step 2.2.2, let the upper bound of any a-th bit code be , the lower bound is ; a∈[1,dim]; and ∈ , represents the upper bound set, ∈ , Represents the lower bound set.

[0108] Step 2.3: Use greedy selection to select the i-th production individual Perform a preliminary evaluation and obtain the i-th production individual Minimum CO2 emissions , thus obtaining the minimum CO2 emission set of the tth generation ;

[0109] Step 2.3.1, the goal is to minimize the sum of the CO2 emissions generated by the supplier when obtaining the unit of raw materials and the CO2 emissions generated by the supplier when transporting the unit of raw materials to the factory. This choice is for the factory to purchase and transport the CO2 emissions generated by the small, after all suppliers are sorted in ascending order, each supplier is judged in turn, if the supplier Raw materials Maximum Supply Greater than or equal to factory Raw materials Procurement needs , then select supplier and get Otherwise, continue to select the next supplier to satisfy the factory Raw materials The quantity of raw materials transported to the factory by all suppliers under the procurement demand is taken as the i-th production individual procurement plan; where the dimension of the procurement plan is Ns.

[0110] Step 2.3.2, Factory The goal is to minimize the sum of the CO2 emissions generated by the delivery of a unit product to the warehouse center and the CO2 emissions generated by the storage of a unit product in the warehouse center. After sorting all warehouse centers in ascending order, each warehouse center after sorting is judged in turn. If the warehouse center The upper limit of storage capacity Greater than or equal to the quantity of product f produced by factory k , then select Warehouse Center and get Otherwise, continue to select the next warehouse center, so as to obtain all warehouse centers that meet the quantity of factory k to produce product f. The storage quantity and as the i-th production individual The production allocation plan of ; wherein the dimension of the production allocation plan is Nm;

[0111] Step 2.3.3, with the goal of minimizing the CO2 emissions generated by the warehouse center transporting unit products to retailers, sort all warehouse centers in ascending order, and then judge each warehouse center in turn. If the warehouse center Meet the Retailers About Products Demand , then select Warehouse Center and get Otherwise, continue to the next warehouse center to satisfy the retailer About Products All warehouse centers transport products under demand To retailers The number of and as the i-th production individual Storage and transportation plan; where the dimension of the storage and transportation plan is Nd;

[0112] Step 2.3.4: According to the i-th production individual The procurement plan, production allocation plan, storage and transportation plan are calculated using formula (1) Total CO2 emissions .

[0113] Step 2.4: According to ,right After mutation and crossover operations, the t-th generation offspring population is obtained , and then use greedy selection to Calculate total CO2 emissions , and and Compare each production individual at the corresponding position in the t+1 generation, and select the production individual with the smallest total emission to join the t+1 generation population. :

[0114] Step 2.4.1. From Three randomly selected production individuals , using formula (8) to obtain the i-th variant production individual :

[0115] (8)

[0116] Step 2.4.2: Update using formula (9) The Bit encoding , this method is simple and guarantees that the generated solution is always within the legal range:

[0117] (9)

[0118] Step 2.4.3: Use formula (10) to get the i-th offspring individual ;

[0119] (10)

[0120] In formula (9), express The Bit encoding; express The Bit encoding; Represents a random number between 0 and 1.

[0121] Step 2.4.4 Use formula (11) to get the i-th individual in the t+1 generation population :

[0122] (11)

[0123] Step 2.5, after assigning t+1 to t, if t> , then the final Generation population This is the optimal production plan set; otherwise, return to step 2.4 and execute sequentially.

[0124] Step 3: Based on the differential evolution algorithm, optimize the decision variables of the transportation plan for raw materials or final products in the first and second stages to obtain the optimal transportation plan set:

[0125] Step 3.1, initialization parameters, including: population size is , the current number of iterations is The maximum number of iterations is , initialize the number of iterations ; Here, the population size of the second stage is consistent with that of the first stage, and the transport individuals of the second stage population correspond to the production individuals of the first stage population in the same position, and the two constitute the optimized production and transportation plan.

[0126] Step 3.2: Itth generation population The code:

[0127] Step 3.2.1. Generation population The i-th individual The procurement plan, production allocation plan, storage and transshipment plan form a transportation plan and serve as the i-th transportation individual , thus obtaining the It generation population , and the encoding method of each transport individual is integer encoding, The length of the code for each transported individual; express No. bit encoding; this setting utilizes the final population of the first stage and optimize the transportation plan based on this information.

[0128] Step 3.2.2, let any b-th bit be encoded The upper bound of , the lower bound is ; b∈[1,D]; and ∈ , represents the upper bound set, ∈ , Represents the lower bound set.

[0129] Step 3.3: Use the repair strategy to Make corrections and get the corrected Transport units ; Thus, the revised It generation population is obtained ,in, express The Bit encoding; initialize b=1;

[0130] Step 3.3.1, judgment Characterizing the procurement program Exceeding the supplier Raw materials Maximum Supply , if it exceeds, then use formula (12) to Make adjustments and get suppliers Transport of raw materials To the factory Adjusted quantity , otherwise, let = ;

[0131] Step 3.3.2: After assigning b+1 to b, if b>Ns, execute step 3.3.3; otherwise, return to step 3.3.1 and execute sequentially;

[0132] (12)

[0133] Step 3.3.3, let b = 1;

[0134] Step 3.3.4, judgment Characterizing the procurement program Does it satisfy formula (5)? ​​If it does, let , and execute step 3.3.6, otherwise, execute step 3.3.5;

[0135] Step 3.3.5: With the goal of minimizing the sum of the CO2 emissions generated when suppliers obtain unit raw materials and the CO2 emissions generated when suppliers transport unit raw materials to the factory, sort all suppliers in ascending order and determine whether each supplier satisfies formula (13). If so, select the corresponding supplier. , and calculate the supplier Transport of raw materials To the factory The adjusted amount + , so that ; Otherwise, continue to determine the next supplier;

[0136] (13)

[0137] Step 3.3.6. After assigning b+1 to b, if b>Ns, execute step 3.3.7, otherwise execute step 3.3.4.

[0138] Step 3.3.7, judgement Characterizing the production allocation scheme Whether the production capacity of factory k for product f is exceeded , if it exceeds, then use formula (14) to Make adjustments and get the factory Transport Products To the warehouse center Adjusted quantity , otherwise, let = ;

[0139] (14)

[0140] Step 3.3.8: After assigning b+1 to b, if b>Nm+Ns, execute step 3.3.9; otherwise, return to step 3.3.7 and execute sequentially.

[0141] Step 3.3.9, let b = Ns + 1;

[0142] Step 3.3.10, judgement Characterizing the production allocation scheme Does it satisfy formula (6)? If it does, let ; and execute step 3.3.13, otherwise, execute step 3.3.11;

[0143] Step 3.3.11: Use formula (15) to get the factory Transport Products To the warehouse center Adjusted quantity , so that ;

[0144] (15)

[0145] Step 3.3.12: Assign b+1 to b. If b>Ns+Nm, execute step 3.3.13; otherwise, return to step 3.3.10 and execute sequentially.

[0146] Step 3.3.13, judgement Characterization of storage and transport scenarios Whether the storage capacity of the warehouse center for the product has been exceeded , if it exceeds, then use formula (16) to Make adjustments to get the warehouse center Transport Products To retailers Adjusted quantity ,otherwise, = ;

[0147] (16)

[0148] Step 3.3.14: Assign b+1 to b. If b>Nm+Ns+Nd, execute step 3.3.15. Otherwise, return to execute step 3.3.13 in sequence.

[0149] Step 3.3.15, let b = Ns + Nm + 1;

[0150] Step 3.3.16, judgment Characterization of storage and transport solutions Does it satisfy formula (7)? If it does, let , and execute step 3.3.18, otherwise, execute step 3.3.17;

[0151] Step 3.3.17: Use formula (17) to get the transport product of warehouse center j To retailers Adjusted quantity , so that ;

[0152] (17)

[0153] Step 3.3.18, let b+1 be assigned to b, if b Ns+Nm+Nd, then execute step 3.3.16; otherwise, it indicates that the It generation population is obtained. The i-th individual .

[0154] Step 3.4: Use differential evolution algorithm to After crossover and mutation operations, the It-th generation offspring population is obtained. ;

[0155] Step 3.5: Use the repair strategy to Each individual in is corrected to obtain the corrected offspring population ;

[0156] Step 3.6: Use formula (1) to calculate Total CO2 emissions Z( )and Total CO2 emissions Z( ), where the second-stage population includes the transportation plan-related decisions, which correspond to the production plan in the first stage one by one, and and Compare each transport individual at the corresponding position in the , and select the transport individual with the smallest total emission to join the It+1 generation transport population ;

[0157] Step 3.5: Assign to ,like , then the final Generation population This is the optimal transportation solution set. Otherwise, return to step 3.4 and execute sequentially.

[0158] In this embodiment, an electronic device includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the above method, and the processor is configured to execute the program stored in the memory.

[0159] In this embodiment, a computer-readable storage medium stores a computer program on the computer-readable storage medium, and the computer program executes the steps of the above method when executed by a processor.

Claims

1. A production and transportation collaborative optimization method based on a four-layer supply chain network with two-stage differential evolution, characterized in that: It is to bring together suppliers Raw materials collection provided Transport to the factory for assembly Produce and obtain product collection After that, it will be transported to the warehouse center for collection. Store and finally transport to retailers for collection In the four-layer supply chain network, let Any supplier, Any raw material, Represents any factory, Any product, represents any warehouse center, Any retailer that will produce the product Raw materials needed The consumption is recorded as ,supplier Raw materials The maximum supply is recorded as ,supplier Get unit raw materials The CO2 emissions generated when ,factory Production products The upper limit of production capacity is ,factory Storing surplus units of raw materials The amount of CO2 emissions generated is recorded as ,factory Production unit products The amount of CO2 emissions generated is recorded as , warehouse center About Products The upper limit of storage capacity is recorded as , warehouse center Storage Unit Products The amount of CO2 emissions generated is recorded as , Retailer About Products The demand for ;supplier Transport unit raw materials To the factory The resulting CO2 emissions are recorded as ,factory Unit Product To the warehouse center The amount of CO2 emissions generated is recorded as , warehouse center Shipping unit product To retailers The amount of CO2 emissions generated is recorded as The production and transportation collaborative optimization method is carried out in the following steps: Step 1: Build a production and transportation collaborative optimization model: According to the coupling relationship between production decision variables and transportation decision variables in the production and transportation process of the four-layer supply chain network, the production and transportation decision variables are divided into the first-stage factory output decision variables and the second-stage raw material or final product transportation plan decision variables. The factory output decision variables include: the number of product f produced by factory k ; The decision variables for the transportation plan of the raw materials or final products include: supplier Transport of raw materials To the factory Number of ,factory Transport Products To the warehouse center Number of , warehouse center Transport Products To retailers Number of ; Taking the minimum total CO2 emissions in the supply chain network as the goal, the objective function of the production and transportation collaborative optimization model is constructed using formula (1): : (1) The constraints of the production and transportation collaborative optimization model are constructed using equations (2) to (5): (2) (3) (4) (5) (6) (7) Step 2: Based on the differential evolution algorithm, optimize the decision variables of the factory output in the first stage to obtain the optimal production plan set: Step 2.1, initialization parameters, including: population size is , the current number of iterations is , the maximum number of iterations is ; The mutation operator is and crossover operator ; Initialize t=1; Step 2.2: The tth generation population The code: Step 2.2.1: Take the production plan consisting of the quantity of each product produced by each factory as a production individual, so that the t-generation population consists of N individuals. ,in, represents the tth generation population The i-th production individual in , and the encoding method of each production individual is integer encoding, the encoding length dim of each production individual is the product of the number of factories and the number of product types, and each bit of the encoding in each production individual represents the quantity of a certain product produced by a certain factory; Step 2.2.2, let the upper bound of any a-th bit code be , the lower bound is ; a∈[1,dim]; and ∈ , represents the upper bound set, ∈ , represents the lower bound set; Step 2.3: Use greedy selection to select the i-th production individual Perform a preliminary evaluation and obtain the i-th production individual Minimum CO2 emissions , thus obtaining the minimum CO2 emission set of the tth generation ; Step 2.4: According to ,right After mutation and crossover operations, the t-th generation offspring population is obtained , and then use greedy selection to Calculate total CO2 emissions , and and Compare each production individual at the corresponding position in the t+1 generation, and select the production individual with the smallest total emission to join the t+1 generation population. : Step 2.5, after assigning t+1 to t, if t> , then the final Generation population This is the optimal production plan set; otherwise, return to step 2.4 and execute sequentially; Step 3: Based on the differential evolution algorithm, optimize the decision variables of the transportation plan for raw materials or final products in the first and second stages to obtain the optimal transportation plan set: Step 3.1, initialization parameters, including: population size is , the current number of iterations is The maximum number of iterations is , initialize the number of iterations ; Step 3.2: Itth generation population The code: Step 3.2.

1. Generation population The i-th individual The procurement plan, production allocation plan, storage and transshipment plan form a transportation plan and serve as the i-th transportation individual , thus obtaining the It generation population , and the encoding method of each transport individual is integer encoding, The length of the code for each transported individual; express No. Bit encoding; Step 3.2.2, let any b-th bit be encoded The upper bound of , the lower bound is ; b∈[1,D]; and ∈ , represents the upper bound set, ∈ , represents the lower bound set; Step 3.3: Use the repair strategy to Make corrections and get the corrected Transport units ; Thus, the revised It generation population is obtained ,in, express The Bit encoding; Step 3.4: Use differential evolution algorithm to After crossover and mutation operations, the It-th generation offspring population is obtained. ; Step 3.5: Use the repair strategy to Each individual in is corrected to obtain the corrected offspring population ; Step 3.6: Use formula (1) to calculate Total CO2 emissions Z( )and Total CO2 emissions Z( ), and and Compare each transport individual at the corresponding position in the , and select the transport individual with the smallest total emission to join the It+1 generation transport population ; Step 3.5: Assign to ,like , then the final Generation population This is the optimal transportation solution set. Otherwise, return to step 3.4 and execute sequentially.

2. According to claim 1, a production and transportation collaborative optimization method based on a four-layer supply chain network with two-stage differential evolution is characterized in that: The step 2.3 includes: Step 2.3.1, with the goal of minimizing the sum of the CO2 emissions generated when suppliers obtain unit raw materials and the CO2 emissions generated when suppliers transport unit raw materials to the factory, sort all suppliers in ascending order, and then judge each supplier in turn. If the supplier Raw materials Maximum Supply Greater than or equal to factory Raw materials Procurement needs , then select supplier and get Otherwise, continue to select the next supplier to satisfy the factory Raw materials The quantity of raw materials transported to the factory by all suppliers under the procurement demand is taken as the i-th production individual The procurement plan of the system; where the dimension of the procurement plan is Ns; Step 2.3.2, Factory The goal is to minimize the sum of the CO2 emissions generated by the delivery of a unit product to the warehouse center and the CO2 emissions generated by the storage of a unit product in the warehouse center. After sorting all warehouse centers in ascending order, each warehouse center after sorting is judged in turn. If the warehouse center The upper limit of storage capacity Greater than or equal to the quantity of product f produced by factory k , then select Warehouse Center and get Otherwise, continue to select the next warehouse center, so as to obtain all warehouse centers that meet the quantity of factory k to produce product f. The storage quantity and as the i-th production individual The production allocation plan of ; wherein the dimension of the production allocation plan is Nm; Step 2.3.3, with the goal of minimizing the CO2 emissions generated by the warehouse center transporting unit products to retailers, sort all warehouse centers in ascending order, and then judge each warehouse center in turn. If the warehouse center Meet the Retailers About Products Demand , then select Warehouse Center and get Otherwise, continue to the next warehouse center to satisfy the retailer About Products All warehouse centers transport products under demand To retailers The number of and as the i-th production individual Storage and transportation plan; where the dimension of the storage and transportation plan is Nd; Step 2.3.4: According to the i-th production individual The procurement plan, production allocation plan, storage and transportation plan are calculated using formula (1) Total CO2 emissions .

3. According to claim 2, a production and transportation collaborative optimization method based on a four-layer supply chain network with two-stage differential evolution is characterized in that: The step 2.4 comprises: Step 2.4.

1. From Three randomly selected production individuals , using formula (8) to obtain the i-th variant production individual : (8) Step 2.4.2: Update using formula (9) The Bit encoding : (9) Step 2.4.3: Use formula (10) to get the i-th offspring individual ; (10) In formula (9), express The Bit encoding; express The Bit encoding; Represents a random number between 0 and 1; Step 2.4.4 Use formula (11) to get the i-th individual in the t+1 generation population : (11)。 4. According to claim 3, a production and transportation collaborative optimization method based on a four-layer supply chain network with two-stage differential evolution is characterized in that: The step 3.3 comprises: Step 3.3.0, initialize b=1; Step 3.3.1, judgment Characterizing the procurement program Exceeding the supplier Raw materials Maximum Supply , if it exceeds, then use formula (12) to Make adjustments and get suppliers Transport of raw materials To the factory Adjusted quantity , otherwise, let = ; Step 3.3.2: After assigning b+1 to b, if b>Ns, execute step 3.3.3; otherwise, return to step 3.3.1 and execute sequentially; (12) Step 3.3.3, let b = 1; Step 3.3.4, judgment Characterizing the procurement program Does it satisfy formula (5)? ​​If it does, let , and execute step 3.3.6, otherwise, execute step 3.3.5; Step 3.3.5: With the goal of minimizing the sum of the CO2 emissions generated when suppliers obtain unit raw materials and the CO2 emissions generated when suppliers transport unit raw materials to the factory, sort all suppliers in ascending order and determine whether each supplier satisfies formula (13). If so, select the corresponding supplier. , and calculate the supplier Transport of raw materials To the factory The adjusted amount + , so that ; Otherwise, continue to determine the next supplier; (13) Step 3.3.6, after assigning b+1 to b, if b>Ns, execute step 3.3.7, otherwise execute step 3.3.4; Step 3.3.7, judgment Characterizing the production allocation scheme Whether the production capacity of factory k for product f is exceeded , if it exceeds, then use formula (14) to Make adjustments and get the factory Transport Products To the warehouse center Adjusted quantity , otherwise, let = ; (14) Step 3.3.8: After assigning b+1 to b, if b>Nm+Ns, execute step 3.3.9; otherwise, return to step 3.3.7 and execute sequentially; Step 3.3.9, let b = Ns + 1; Step 3.3.10, judgement Characterizing the production allocation scheme Does it satisfy formula (6)? If it does, let ; and execute step 3.3.13, otherwise, execute step 3.3.11; Step 3.3.11: Use formula (15) to get the factory Transport Products To the warehouse center Adjusted quantity , so that ; (15) Step 3.3.12: Assign b+1 to b. If b>Ns+Nm, execute step 3.3.13; otherwise, return to step 3.3.10 and execute sequentially. Step 3.3.13, judgement Characterization of storage and transport scenarios Whether the storage capacity of the warehouse center for the product has been exceeded , if it exceeds, then use formula (16) to Make adjustments to get the warehouse center Transport Products To retailers Adjusted quantity ,otherwise, = ; (16) Step 3.3.14: Assign b+1 to b. If b>Nm+Ns+Nd, execute step 3.3.

15. Otherwise, return to execute step 3.3.13 in sequence. Step 3.3.15, let b = Ns + Nm + 1; Step 3.3.16, judgment Characterization of storage and transport solutions Does it satisfy formula (7)? If it does, let , and execute step 3.3.18, otherwise, execute step 3.3.17; Step 3.3.17: Use formula (17) to get the transport product of warehouse center j To retailers Adjusted quantity , so that ; (17) Step 3.3.18, let b+1 be assigned to b, if b Ns+Nm+Nd, then execute step 3.3.16; otherwise, it means the It generation population is obtained The i-th individual .

5. An electronic device, comprising a memory and a processor, characterized in that: The memory is used to store a program that supports the processor to execute the production and transportation collaborative optimization method based on a two-stage differential evolution four-layer supply chain network as described in any one of claims 1-4, and the processor is configured to execute the program stored in the memory.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the production and transportation collaborative optimization method based on a two-stage differential evolution four-layer supply chain network described in any one of claims 1 to 4 are executed.

Images