Multi-objective optimization method for multi-factory production scheduling under uncertain shipping time

Abstract

The present invention provides a multi-objective optimization method, system, storage medium and electronic device for multi-factory production scheduling under uncertain sea shipping time, and relates to the technical field of multi-factory production scheduling. Under the premise of determining the sea shipping time, the embodiment of the present invention constructs a production scheduling model for a multi-factory production network based on the multi-factory production scheduling resources and a single batch of overseas orders; in view of the uncertainty of large ship sea shipping time, a 1-norm sphere uncertainty set is designed to construct a multi-objective robust optimization model for multi-factory production scheduling, and optimize the total delivery time deviation and total delivery cost multi-objective of a single batch of overseas orders. The model takes into account the pre-established sea shipping plan and the long and uncertain sea shipping time; in addition, the model is converted into an equivalent model, and a two-stage hybrid heuristic algorithm is designed to solve it, and a reasonable and feasible production scheduling suggestion is given for a single batch of overseas orders.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-factory production scheduling, and in particular to a multi-objective optimization method, system, storage medium and electronic equipment for multi-factory production scheduling under uncertain shipping time. Background Art

[0002] With the rapid development of global trade and emerging markets, more and more manufacturers are choosing to shift from traditional centralized production to distributed, multi-factory production networks. Distributed, multi-factory production, with its geographically dispersed nature, offers manufacturers the potential to reduce costs, improve efficiency, and achieve energy conservation and emission reductions.

[0003] Compared to production scheduling within a single factory, the multi-site nature of multiple factories not only leads to differences in raw material, labor, and warehousing costs, but also presents significant challenges for long-distance logistics. When placing an order, overseas customers typically consider their own sales and inventory levels, specifying a desired delivery date and expecting manufacturers to provide efficient and timely product delivery. If the order delivery time is significantly ahead of the overseas customer's delivery time, this can result in expensive storage costs at international ports. If the order delivery time is significantly later than the overseas customer's delivery time, this can lead to significant delays and losses for the customer.

[0004] When scheduling manufacturing tasks driven by overseas orders, ocean freight is often overlooked as a significant factor. First, the impact of shipping schedules. Expected shipping schedules and long shipping lead times significantly constrain manufacturers' processing time, impacting production allocation, scheduling, and storage. Second, ocean freight uncertainty is a widespread issue, severely impacting the reliability of the global supply chain. When actual shipping dates deviate from published schedules, shippers, manufacturers, and their customers all face significant losses due to delays. Summary of the Invention

[0005] (1) Technical problems solved

[0006] In response to the shortcomings of the existing technology, the present invention provides a multi-objective optimization method, system, storage medium and electronic device for multi-factory production scheduling under uncertain sea shipping time, which solves the technical problem of ignoring the impact of sea shipping time uncertainty on multi-factory production scheduling.

[0007] (2) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] A multi-objective optimization method for multi-factory production scheduling under uncertain ocean shipping time, including:

[0010] S1. Obtain multi-factory production scheduling resources and single-batch overseas orders;

[0011] S2. On the premise of determining the shipping time, construct a production scheduling model for the multi-factory production network based on the multi-factory production scheduling resources and the single batch of overseas orders;

[0012] S3. Based on the production scheduling model, the uncertain shipping time is represented by the 1-norm sphere uncertainty set, and a multi-objective robust optimization model for production scheduling is constructed;

[0013] S4. Convert the multi-objective robust optimization model into an equivalent model, use a heuristic algorithm to solve the equivalent model, and obtain a multi-factory production scheduling robust optimization result.

[0014] Preferably, the multi-objective robust optimization model in S3 includes:

[0015] (1) The first objective function of the robust objective is to minimize the 1-norm of the difference between the actual delivery date of all overseas orders in a single batch of overseas orders and the expected delivery date of overseas customers when the shipping time deviation is the worst:

[0016]

[0017]

[0018]

[0019] (2) The second objective function is to minimize the total delivery cost of all overseas orders in a single batch of overseas orders:

[0020]

[0021]

[0022]

[0023]

[0024]

[0025] Where J represents the set of overseas orders, J = {1, 2, ..., n}; I represents the set of factories, I = {1, 2, ..., m}; G represents the set of domestic ports, G = {1, 2, ..., p}; S represents the set of large ships, S = {1, 2, ..., q};

[0026] Assuming the uncertainty parameter actual shipping time δ s In a ball, is the center of the sphere, ρ is the radius of the sphere, Δ s is the error value; let Matrix δ=[δ1,δ2,…,δ s ] is the matrix of actual shipping time for each ship, matrix is the matrix of the historical average shipping time of each ship, the matrix Δ=[Δ1,Δ2,…,Δ s ] is the deviation matrix between the actual shipping time of each ship and its historical average shipping time;

[0027] There are constraints on the actual shipping time for all large ships:

[0028] |Δ s |≤ρ s

[0029] δ is an s-dimensional real vector; then the 1-norm sphere uncertainty set of the actual shipping time of the large ship is expressed as

[0030]

[0031] a s Indicates the date of departure of the large vessel by sea;

[0032] y ijs is a decision variable. If the manufacturing task of overseas order j is assigned to factory i and the shipping task of overseas order j is assigned to ship s, it takes 1; otherwise, it takes 0. js , if the shipping task of overseas order j is assigned to large ship s, take 1; otherwise take 0; d j The date when the customer requested the goods for overseas order j;

[0033] q j represents the number of products in overseas order j; They represent the processing cost, warehousing cost, domestic land transportation cost, and international shipping cost of unit product j of overseas order respectively;

[0034] represents the processing cost per unit of product of overseas order j in factory i;

[0035] x ijj′ is a decision variable. If the manufacturing tasks of overseas orders j and j′ are both assigned to factory i, and overseas order j is completed and then followed by overseas order j′, it takes 1; otherwise, it takes 0.

[0036] represents the daily storage cost of unit product j for overseas orders; represents the storage time of overseas order j;

[0037] represents the domestic transportation cost per unit of product between factory i and port g;

[0038] If a large ship s departs from port g, Tsg =1; otherwise T sg =0;

[0039] It represents the transportation cost of each unit product of large ship;

[0040] The multi-objective robust optimization model in S3 also includes:

[0041] Constraints:

[0042]

[0043]

[0044]

[0045]

[0046]

[0047]

[0048]

[0049]

[0050]

[0051]

[0052]

[0053]

[0054]

[0055]

[0056] Among them, s j 、s j′ denote the actual delivery dates of overseas orders j and j′ respectively; M represents a very large constant;

[0057] tr ig represents the land transportation time from factory i to port g;

[0058] pro ij represents the processing time of overseas order j in factory i.

[0059] Preferably, the equivalent model in S4 includes:

[0060] Goal 1:

[0061]

[0062] Goal 2:

[0063]

[0064] Constraints:

[0065]

[0066]

[0067]

[0068]

[0069]

[0070]

[0071]

[0072]

[0073]

[0074]

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081] Among them, b j The value representing the shipping date of the large vessel assigned to overseas order j minus the date the customer requested the goods.

[0082] Preferably, a two-stage hybrid heuristic algorithm is used to solve the equivalent model, including:

[0083] Phase 1:

[0084] Without considering the selection of the processing plant and the start date of processing, the equivalent model is simplified and relaxed, a heuristic algorithm is designed to search and obtain multiple shipping task solutions, and a population containing multiple chromosomes with assigned large ship numbers is obtained by encoding;

[0085] Phase 2:

[0086] The last generation of the above population is sorted and screened as the partial initial population containing multiple chromosomes assigned to large ship numbers in this stage, and a partial initial population containing multiple chromosomes assigned to factory numbers and a partial initial population containing multiple chromosomes assigned to processing start dates are generated. A multi-objective optimization heuristic algorithm is designed to obtain the Pareto optimal solution, and the solution is decoded as the final robust optimization result for multi-factory production scheduling.

[0087] Preferably, the first stage specifically includes:

[0088] S411. In the equivalent model, separate the decision variable y related to the allocation of large ships. js The relevant objectives and constraints are simplified into the following sub-model:

[0089] Goal 1:

[0090]

[0091] Goal 2:

[0092]

[0093] Constraints:

[0094]

[0095]

[0096] S412, solving the above sub-model, randomly generating an initial population of the first stage including multiple n columns of chromosomes, calculating the fitness of each individual in the population, and setting Gen1 = 0;

[0097] Z i =aV 1 +bV 2

[0098]

[0099] Among them, a and b are constants, which are the weights of target one and target two respectively; f i s1 represents the fitness value of the first stage corresponding to the i-th chromosome; Nind1 represents the population size in the first stage;

[0100] S413, determine whether the termination condition is met. If so, sort the current population in non-ascending order according to fitness and output it as the last generation population Chrom1; otherwise, go to S414;

[0101] S414, duplicate the elite individual into Nind1 / 2 to form a stallion;

[0102] S415, roulette wheel is used to select Nind1 / 2 individuals from the parent population after removing the elite individuals;

[0103] S416, merge the stallions with the selected population;

[0104] S417, performing a two-point crossover operation on the merged population;

[0105] S418. Perform a breeder mutation operation on the population after the crossover operation to generate a new generation of population, and calculate the fitness of each individual in the population; set Gen1 = Gen1 + 1, and return to S413.

[0106] Preferably, the second stage specifically includes:

[0107] S421. Generate an initial population of chromosome units in the second stage, which includes a plurality of chromosomes each consisting of three n columns:

[0108] The first Nind2 individuals of the last generation population Chrom1 in the first stage are used as part of the initial population for assigning large ship numbers to chromosome 3 in the second stage NSGAII algorithm; where Nind2 represents the population size in the second stage, and Nind2 << Nind1;

[0109] A priority selection mechanism for processing factories using processing and transportation costs as judgment indicators is introduced to generate a portion of the initial population of chromosome 1 assigned factory numbers. This priority selection mechanism means setting a certain selection probability to generate the initial population of chromosomes assigned base numbers. The principle of setting the selection probability is: the factory with the lower the sum of production and domestic land transportation costs, the higher the selection probability is set.

[0110] Under the ideal assumption that the storage time is 0, the start date of processing is obtained by working backwards from the overseas order delivery deadline;

[0111]

[0112] S422, performing constraint check on the chromosomes in the initial population generated in S421; if it fails, making corrections according to the preset constraint processing rules;

[0113] S423. Calculate the fitness of each individual in the population, setting Gen2 = 0; wherein, the fitness function of the second stage is composed of the non-dominated sorting level and the crowding degree. The higher the non-dominated level, the lower the crowding degree, and the higher the fitness of the chromosome individual;

[0114] S424: Determine whether the preset termination condition is met. If so, output the Pareto optimal solution in the current population and decode it as the final multi-factory production scheduling robust optimization result; otherwise, proceed to S425.

[0115] S425, retain the parent population and perform a single-point crossover operation;

[0116] S426, executing the breeder mutation operation;

[0117] S427. Perform constraint check on the chromosomes in the current population; if it fails, perform correction according to the preset constraint processing rules to generate a progeny population containing Nind2 individuals;

[0118] S428. Merge the offspring and the parent generation, calculate the fitness of each individual in the combined population, select the first Nind2 individuals with higher fitness; set Gen2 = Gen2 + 1, and go to S424.

[0119] Preferably, for constraint one: the departure time of the large ship must be later than the arrival time of the order at the domestic port; the preset constraint processing rules specifically include: reducing the proportion of chromosomes that violate constraint one in the initial population through the generation rule of the initial population; and correcting the chromosomes that violate constraint one to chromosomes that comply with constraint one by directly correcting the chromosomes;

[0120] Preferably, for constraint 2: two order tasks assigned to the same processing factory cannot be processed at the same time; the preset constraint processing rules specifically refer to:

[0121] The first step is to determine the order set to be judged. If the set is empty, end; otherwise, go to the second step;

[0122] Step 2: Determine whether the processing time of the last overseas order conflicts with the processing time of the second-to-last overseas order; if so, skip to step 3; if not, skip to step 4;

[0123] In step 3, if there is no conflict, the last order is deleted from the order set and the process returns to step 2. If there is a conflict, the processing time of the last overseas order is moved forward to avoid conflict while keeping the domestic land transportation time unchanged. For the time difference, the order is placed in a warehouse near the processing factory for storage pending domestic transportation. The last order is deleted from the order set and the process returns to step 2.

[0124] Step 4: Jump to the next processing plant or the next chromosome and repeat the first step.

[0125] A multi-objective optimization system for multi-factory production scheduling under uncertain ocean shipping time, including:

[0126] The acquisition module is used to obtain multi-factory production scheduling resources and single-batch overseas orders;

[0127] A construction module is used to construct a production scheduling model for a multi-factory production network based on the multi-factory production scheduling resources and a single batch of overseas orders, under the premise of determining the sea freight time;

[0128] An optimization module is used to construct a multi-objective robust optimization model for production scheduling based on the production scheduling model and represent uncertain shipping time by using a 1-norm sphere uncertainty set;

[0129] The solution module is used to convert the multi-objective robust optimization model into an equivalent model, use a heuristic algorithm to solve the equivalent model, and obtain a multi-factory production scheduling robust optimization result.

[0130] A storage medium stores a computer program for multi-objective optimization of multi-factory production scheduling under uncertain shipping time, wherein the computer program enables a computer to execute the multi-objective optimization method for multi-factory production scheduling under uncertain shipping time as described above.

[0131] An electronic device, comprising:

[0132] one or more processors;

[0133] Memory; and

[0134] One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including a multi-objective optimization method for executing multi-factory production scheduling under uncertain shipping time as described above.

[0135] (3) Beneficial effects

[0136] The present invention provides a multi-objective optimization method, system, storage medium, and electronic device for multi-factory production scheduling under uncertain ocean shipping time. Compared with existing technologies, this method has the following advantages:

[0137] This paper designs a 1-norm sphere uncertainty set for large-vessel shipping times and constructs a multi-objective robust optimization model for multi-factory production scheduling. This model optimizes the total delivery time deviation and total delivery cost for a single batch of overseas orders. This model considers pre-defined shipping plans and long, uncertain shipping times. Furthermore, it transforms the model into an equivalent model and solves it using a two-stage hybrid heuristic algorithm, providing reasonable and feasible production scheduling recommendations for a single batch of overseas orders. BRIEF DESCRIPTION OF THE DRAWINGS

[0138] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0139] Figure 1 A schematic diagram of a multi-objective optimization method for multi-factory production scheduling under uncertain ocean shipping time provided by an embodiment of the present invention;

[0140] Figure 2 A schematic diagram of a delivery process for a single batch of overseas orders provided by an embodiment of the present invention;

[0141] Figure 3 A schematic diagram of a two-stage hybrid heuristic algorithm provided in an embodiment of the present invention;

[0142] Figure 4 A schematic diagram of the coding of one n-column chromosome in the first stage provided by an embodiment of the present invention;

[0143] Figure 5 A schematic diagram of encoding a chromosome unit consisting of three n-column chromosomes in the second stage provided by an embodiment of the present invention;

[0144] Figure 6 A schematic diagram of the process flow of the first-stage heuristic algorithm provided by an embodiment of the present invention;

[0145] Figure 7 A schematic diagram of the flow of the second-stage heuristic algorithm provided by an embodiment of the present invention;

[0146] Figure 8 A schematic diagram of the generation rules of the initial population in the second-stage heuristic algorithm provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0147] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0148] The embodiments of the present application solve the technical problem of ignoring the influence of uncertainty factors of sea shipping time on multi-factory production scheduling by providing a multi-objective optimization method, system, storage medium and electronic device for multi-factory production scheduling under uncertainty of sea shipping time.

[0149] The technical solution in the embodiments of the present application is to solve the above technical problems, and the overall idea is as follows:

[0150] like Figure 1 As shown, this embodiment of the present invention designs a 1-norm sphere uncertainty set for large vessel shipping times and constructs a multi-objective robust optimization model for multi-factory production scheduling. This model optimizes the total delivery deviation and total delivery cost for a single batch of overseas orders. This model considers pre-defined shipping plans and long, uncertain shipping times. Furthermore, it transforms the model into an equivalent model and designs a two-stage hybrid heuristic algorithm to solve it, providing reasonable and feasible production scheduling recommendations for a single batch of overseas orders.

[0151] In order to better understand the above technical solution, the above technical solution will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0152] Example:

[0153] An embodiment of the present invention provides a multi-objective optimization method for multi-factory production scheduling under uncertain ocean shipping time, comprising:

[0154] S1. Obtain multi-factory production scheduling resources and single-batch overseas orders;

[0155] S2. On the premise of determining the shipping time, construct a production scheduling model for the multi-factory production network based on the multi-factory production scheduling resources and the single batch of overseas orders;

[0156] S3. Based on the production scheduling model, the uncertain shipping time is represented by the 1-norm sphere uncertainty set, and a multi-objective robust optimization model for production scheduling is constructed;

[0157] S4. Convert the multi-objective robust optimization model into an equivalent model, use a heuristic algorithm to solve the equivalent model, and obtain a multi-factory production scheduling robust optimization result.

[0158] In an embodiment of the present invention, the multi-objective robust optimization model takes into account the pre-established shipping plan and the long and uncertain shipping time; in addition, the model is converted into an equivalent model, and a two-stage hybrid heuristic algorithm is designed to solve it, providing reasonable and feasible production scheduling suggestions for a single batch of overseas orders.

[0159] The following is a detailed introduction to the various steps of the above technical solution:

[0160] First, it is necessary to explain and supplement the description of the multi-factory production scheduling problem involved in the embodiment of the present invention, as follows:

[0161] Taking single-batch overseas order groups as the optimization target, each overseas order must go through four key stages: domestic production, domestic warehousing, domestic logistics, and international logistics, before ultimately being delivered to the overseas customer. A single-batch overseas order group is a unit of orders placed by multiple overseas users within the same or similar time period. Each overseas order contains only one required product model and corresponds to only one overseas user. Each customer specifies a desired delivery time to the manufacturer. The domestic production phase of each overseas order is called a manufacturing task, and each manufacturing task can only be assigned to one factory for processing. Each factory has an independent assembly line capable of processing any manufacturing task for an overseas order with the same quality. Each manufacturing task for an overseas order can be assigned to any factory. A factory can only process one manufacturing task for an overseas order at a time, and preemption or pause in processing is not allowed. Each factory is equipped with a warehouse for storing finished products. Each overseas order is shipped by a single vessel, each with a designated domestic departure port, shipping time, and shipping time. The delivery time for the order is the arrival time of the product at the international destination port. Based on the above prerequisites, the delivery process of a single batch of overseas orders is as follows: Figure 2 shown.

[0162] A single batch of overseas orders from a manufacturing company consists of n orders, represented by the set J = {1, 2, …, n}, which can be assigned to m distributed factories for processing, with the factory set represented by I = {1, 2, …, m}. From the moment an order is placed, a single batch of orders must complete four core steps: manufacturing, storage, land transportation, and sea transportation, ultimately reaching the overseas customer for delivery.

[0163] In step S1, obtain multi-factory production scheduling resources and single-batch overseas orders;

[0164] The multi-factory production scheduling resources and single-batch overseas orders specifically involve the following symbology:

[0165] gather:

[0166] gather describe J Overseas order set, J = {1, 2, ..., n} I Factory set, I = {1, 2, ..., m} G Domestic port set, G = {1, 2, ..., p} S Large ships gather, S={1,2,…,q}

[0167] Decision variables

[0168]

[0169] Related parameters

[0170]

[0171]

[0172] In step S2, on the premise of determining the sea shipping time, a production scheduling model for a multi-factory production network is constructed according to the multi-factory production scheduling resources and a single batch of overseas orders.

[0173] In this step, we focus on the sea transportation link and propose a multi-objective deterministic model for production scheduling in a multi-factory production network. This model coordinates the scheduling of single-batch overseas order groups while optimizing the overall on-time delivery and total delivery cost of overseas orders.

[0174] The deterministic model includes:

[0175] Goal 1: Minimize the total difference between the actual delivery date of all overseas orders in a single batch of overseas orders and the expected delivery date of overseas customers. In other words, maximize the total on-time delivery of all overseas orders, minimize the overseas warehousing costs and delay costs of overseas orders, and meet customer needs:

[0176]

[0177]

[0178]

[0179]

[0180]

[0181]

[0182]

[0183] Wherein, formula (2) indicates that the actual delivery date of each overseas order is the start processing date s of the order j , processing days Order warehouse storage days Land transportation days Shipping days Formula (3) Overseas order processing days It is related to the product type of overseas order j and is proportional to the processing capacity and product quantity of the assigned factory i. Formula (4) indicates that the warehouse storage time for overseas order j is the selected ship's shipping date minus the end processing date of overseas order j. The ideal situation is that the warehouse storage time for overseas order j is 0, meaning that once the manufacturing task of overseas order j is completed at the base, it is shipped to the domestic port to catch the ship departing that day. Formula (6) indicates that the land transportation days for overseas order j depend on the factory assigned to overseas order j and the selected ship's shipping port.

[0184] The second goal is to minimize the total delivery cost of all overseas orders in a single batch of overseas orders. The delivery cost of each overseas order is composed of production cost, warehouse storage cost, domestic land transportation cost and international shipping cost.

[0185]

[0186]

[0187]

[0188]

[0189]

[0190] Wherein, formula (9) represents the unit product processing cost of overseas order j: It is related to the factory i assigned to the order and the type of product itself. Formula (10) shows that the unit product storage cost of overseas order j is the product of the unit product storage cost per unit time of overseas order j and the storage time. Formula (11) shows the domestic land transportation cost per unit product of overseas order j The international shipping cost per unit of product for overseas order j is related to the factory assigned to it and the selected shipping port of the large vessel. Formula (12) represents the international shipping cost per unit of product for overseas order j, which is related to the selected large vessel. The shipping cost of the large vessel is set to increase as the shipping time within a fixed distance decreases.

[0191] Constraints:

[0192]

[0193]

[0194]

[0195]

[0196]

[0197]

[0198]

[0199]

[0200]

[0201]

[0202]

[0203]

[0204]

[0205]

[0206] Among them, constraint (13) indicates that any order task has no order relationship with itself; constraint (14) indicates that each order task has a unique order task in front of it, and is only assigned to a unique factory; constraint (15) indicates that each order task has a unique order task behind it, and is only assigned to a unique factory; constraint (16) indicates that each factory has only one first order task or is not assigned an order task; constraint (17) indicates that each factory has only one last order task or is not assigned an order task; constraint (18) indicates that for any factory, any order task cannot be both in front of and behind another order task at the same time; constraint (19) indicates that for any order task , there must be the same number of pre-order tasks and post-order tasks; constraint (20) indicates the value constraint of the decision variable when the factory has not arranged any task; constraint (21) indicates that the start processing time of each order task is greater than 0; constraint (22) indicates that any two order tasks of the same factory cannot be processed at the same time; constraint (23) indicates that for any task, only one factory can be selected for processing, and only one large ship can be selected for overseas transportation; constraint (24) indicates a one-to-one correspondence between two decision variables; constraint (25) indicates that the departure time selected for each order task must be greater than the arrival time of the order task at the domestic port; constraint (26) indicates the value range of the decision variable.

[0207] In step S3, based on the production scheduling model, the uncertain shipping time is represented by the 1-norm sphere uncertainty set, and a multi-objective robust optimization model for production scheduling is constructed.

[0208] Normally, during the actual delivery of an order, overseas shipping time may be affected by many factors such as sea weather, wind direction, and waves. Compared to pre-planned shipping schedules, there are often deviations. Ocean shipping schedules are extremely unpredictable and difficult to predict. Fluctuations in overseas shipping times can cause the original manufacturing task scheduling plan to deviate from the optimal one. Therefore, robust optimization methods are used to solve production scheduling problems under uncertain shipping times. Robust optimization considers the best solution under the worst-case scenario, representing a conservative perspective. The resulting optimization solution is not necessarily optimal, but when the uncertain parameters are perturbed, the resulting solution remains feasible.

[0209] Aiming at the uncertain parameter of large vessel shipping time, the embodiment of the present invention designs a 1-norm sphere uncertainty set and establishes a multi-objective robust optimization model for the joint scheduling of production tasks and logistics tasks.

[0210] The 1-norm sphere uncertainty set: Assuming the uncertainty parameter actual shipping time δ s In a ball, is the center of the sphere, ρ is the radius of the sphere (the maximum standard deviation of the shipping time deviation of all large ships), Δ s is the error value (the deviation between the large ship s and the historical average shipping time); let Matrix δ=[δ1,δ2,…,δ s ] is the matrix of actual shipping time for each ship, matrix is the matrix of the historical average shipping time of each ship, the matrix Δ=[Δ1,Δ2,…,Δ s ] is the deviation matrix between the actual shipping time of each ship and its historical average shipping time;

[0211] There are constraints on the actual shipping time for all large ships:

[0212] |Δ s |≤ρ s (27)

[0213] δ is an s-dimensional real vector, i.e., the maximum singular value. The value of parameter ρ is given in advance, reflecting that the standard deviation of all large ship shipping time deviations does not exceed ρ. When , this problem is a deterministic problem, indicating that the shipping time of all large ships takes the standard value; when When , it means that the shipping time of large ships deviates from the standard value to a certain extent. According to the risk preference of the decision maker, adjust ρ s Therefore, the uncertainty set of the 1-norm sphere of the actual shipping time of the large ship is expressed as:

[0214]

[0215] The multi-objective robust optimization model includes:

[0216] The first goal is to minimize the sum of squares of the difference between the actual delivery date of all overseas orders in a single batch of overseas orders and the expected delivery date of overseas customers when considering the worst case of shipping time deviation. In the case of uncertainty, the parameter Adjusted to δ s , and the uncertain parameter shipping time only exists in formula (7) of the objective function 1 of the original deterministic model.

[0217] The degree of delivery deviation of all products in target 1 is described by the matrix 1 norm. That is, the first objective function of the robust target is to minimize the 1 norm of the difference between the actual delivery date of all overseas orders in a single batch of overseas order groups and the expected delivery date of overseas customers when the shipping time deviation is the worst:

[0218]

[0219]

[0220]

[0221] The second objective is to minimize the total delivery cost of all overseas orders in a single batch of overseas orders. The delivery cost of each overseas order is composed of production cost, warehouse storage cost, domestic land transportation cost, and international shipping cost. That is, the second objective function with the goal of minimizing the total delivery cost of all overseas orders in a single batch of overseas orders is:

[0222]

[0223]

[0224]

[0225]

[0226]

[0227] Constraints:

[0228]

[0229]

[0230]

[0231]

[0232]

[0233]

[0234]

[0235]

[0236]

[0237]

[0238]

[0239]

[0240]

[0241]

[0242] In step S4, the multi-objective robust optimization model is converted into an equivalent model, and a heuristic algorithm is used to solve the equivalent model to obtain a multi-factory production scheduling robust optimization result.

[0243] Convert the multi-objective robust optimization model into an equivalent model:

[0244] Goal 1

[0245]

[0246] Goal 2 is

[0247]

[0248] Constraints:

[0249]

[0250]

[0251]

[0252]

[0253]

[0254]

[0255]

[0256]

[0257]

[0258]

[0259]

[0260]

[0261]

[0262]

[0263]

[0264]

[0265] Among them, b j The value representing the shipping date of the large vessel assigned to overseas order j minus the date the customer requested the goods.

[0266] For the objective function (29), where:

[0267]

[0268] when And Δ s =±ρ s ,for Established, the maximum value of the above formula can be reached.

[0269] The proof is as follows:

[0270] To make the maximum value reachable, that is:

[0271]

[0272] Must meet:

[0273]

[0274]

[0275] The first equality condition is:

[0276] Same number;

[0277] The second condition for equality to hold is:

[0278] Δ s =±ρ s (72)

[0279] so:

[0280]

[0281] Through the above analysis, it is proved that the original robust optimization multi-objective model is equivalent to the equivalent model after transformation. Figure 3 The two-stage hybrid heuristic algorithm shown solves the equivalent model, including:

[0282] Phase 1:

[0283] Without considering the processing plant selection and the start date of processing, simplify the equivalent model, design a heuristic algorithm to search and obtain multiple shipping task solutions, and encode to obtain a population containing multiple chromosomes with assigned large ship numbers ( Figure 4 As an example, a chromosome with n columns in the first stage is given. The position number of each column of the chromosome represents the overseas order number, and the value of each column represents the number of the large ship selected for the overseas order).

[0284] Phase 2:

[0285] The last generation of the above population is sorted and screened as a partial initial population containing multiple chromosomes assigned to large ship numbers, and a partial initial population containing multiple chromosomes assigned to factory numbers and a partial initial population containing multiple chromosomes assigned to processing dates are generated ( Figure 5 As an example, a chromosome unit consisting of 3 n-column chromosomes is given in the second stage. In a chromosome unit, the position number of each row represents the number of the overseas order, the first chromosome represents the allocation of processing plants for each overseas order, the second chromosome represents the start processing date of the order task of the overseas order in the processing plant, and the third chromosome represents the allocation of sea shipping vessels for each overseas order. A heuristic algorithm is designed to obtain the Pareto optimal solution and decode it as the final robust optimization result of multi-factory production scheduling.

[0286] The goal of the first stage is to solve and allocate the decision variables y of the large ship ijs Related solutions, search for a group of multi-order shipping task solutions with high on-time delivery rate and good shipping economic benefits; such as Figure 6 As shown, specifically including:

[0287] S411, in the equivalent model, the large ship decision variable y of the original model is ijs Remove the information containing the base and adjust the decision variable to y js , and it is known Then we can separate the decision variable y js The relevant objectives and constraints are simplified into the following sub-model:

[0288] Goal 1:

[0289]

[0290] Goal 2:

[0291]

[0292] Constraints:

[0293]

[0294]

[0295] S412, solving the above sub-model, randomly generating an initial population of the first stage including multiple n columns of chromosomes, calculating the fitness of each individual in the population, and setting Gen1 = 0;

[0296] Z i =aV 1 +bV 2 (78)

[0297]

[0298] Among them, a and b are constants, which are the weights of target one and target two respectively; f i s1 represents the fitness value of the first stage corresponding to the i-th chromosome; Nind1 represents the population size in the first stage;

[0299] S413, determine whether the termination condition is met. If so, sort the current population in non-ascending order according to fitness and output it as the last generation population Chrom1; otherwise, go to S414;

[0300] S414, duplicate the elite individual into Nind1 / 2 to form a stallion;

[0301] S415, roulette wheel is used to select Nind1 / 2 individuals from the parent population after removing the elite individuals;

[0302] S416, merge the stallions with the selected population;

[0303] S417, performing a two-point crossover operation on the merged population;

[0304] S418. Perform a breeder mutation operation on the population after the crossover operation to generate a new generation of population, and calculate the fitness of each individual in the population; set Gen1 = Gen1 + 1, and return to S413.

[0305] S418. Execute the breeder mutation operation to generate a new generation of population and calculate the fitness of each individual in the population; set Gen1 = Gen1 + 1, and return to S413.

[0306] This step adopts the breeder GA mutation strategy, using the compression rate (MutShrink) and gradient partitioning (Gradient) to control the mutation distance. The greater the compression rate and the fewer the gradient partitioning, the greater the mutation amplitude and the stronger the global search capability.

[0307] In the second stage, a heuristic algorithm is designed to solve the equivalent model of the multi-objective robust optimization model of multi-factory production scheduling under uncertain shipping time; Figure 7 As shown, specifically including:

[0308] S421, such as Figure 8 As shown, the initial population of chromosome units consisting of 3 n-column chromosomes is generated in the second stage:

[0309] The first step is to use the first Nind2 individuals of the last generation population Chrom1 in the first stage as part of the initial population for assigning large ship numbers to chromosome 3 in the second stage NSGAII algorithm; where Nind2 represents the population size in the second stage, and it should be noted that Nind1 is much larger than Nind2.

[0310] The second step is to introduce a priority selection mechanism for processing factories with processing and transportation costs as the judgment indicators to generate part of the initial population of chromosome 1 with factory numbers assigned to it. This priority selection mechanism means setting a certain selection probability to generate the initial population of chromosomes with base numbers assigned to chromosome 1. The principle of setting the selection probability is: the factory with the lower the sum of production and domestic land transportation costs, the higher the selection probability is set.

[0311] To elaborate, first, for each export order, according to the sum of production and domestic land transportation costs C pro+str , sort the processing factories from small to large, and calculate the priority matrix of the processing factories Nind2 rows and n columns. For example, for an order j, its processing factory priority matrix is In the jth column, the first row is the processing plant with the lowest production and domestic land transportation costs, the second row is the processing plant with the second highest cost, and so on. Therefore, for all export orders, C pro+str Lowest Maquiladora in the Maquiladora Priority Matrix The position numbers in are all the first row.

[0312] The first Nind2 / 2 individuals of the chromosome 1 initial population are all set to the priority matrix of each processing plant. The first row in . After Nind2 / 2 individuals, each export order is set with a certain probability P in Select the processing factory. P is set to [0.5, 0.33, 0.25, 0.2, 0.167, 0.14, 0.125, 0.11, 0.1, 0.09, ...], which means that the probability of selecting The processing factory in the first row corresponding to the order in , the probability of selecting the second row is 0.33, and so on. The smaller the location number of the processing factory, the smaller the probability of being selected.

[0313] Step 3: Under the ideal assumption that the storage time is 0, reverse the overseas order delivery deadline to obtain the start date of processing;

[0314]

[0315] The above three steps will completely generate the initial populations of three decision variables; the three initial populations are linked to each other by the chromosome individual number, that is, the chromosomes in the three populations with the same individual number constitute a chromosome unit.

[0316] S422, performing constraint check on the chromosomes in the initial population generated in S421; if it fails, making corrections according to the preset constraint processing rules;

[0317] (1) Regarding constraint one: the departure time of the large ship must be later than the time when the order arrives at the domestic port; the preset constraint processing rules specifically refer to: reducing the proportion of chromosomes that violate constraint one in the initial population through the generation rules of the initial population (corresponding to step S422); by directly correcting the chromosomes, the chromosomes that violate constraint one are corrected to chromosomes that comply with constraint one (corresponding to the subsequent step S427).

[0318] (2) Regarding constraint 2: two order tasks assigned to the same processing factory cannot be processed at the same time; the preset constraint processing rules specifically refer to:

[0319] The first step is to determine the order set to be judged. If the set is empty, end; otherwise, go to the second step;

[0320] Step 2: Determine whether the processing time of the last overseas order conflicts with the second-to-last overseas order; if so, skip to step 3; if not, skip to step 4;

[0321] In step 3, if there is no conflict, the last order is deleted from the order set and the process returns to step 2. If there is a conflict, the processing time of the last overseas order is moved forward to avoid conflict while keeping the domestic land transportation time unchanged. For the time difference, the order is placed in a warehouse near the processing factory for storage pending domestic transportation. The last order is deleted from the order set and the process returns to step 2.

[0322] Step 4: Jump to the next processing plant or the next chromosome and repeat the first step.

[0323] In an embodiment of the present invention, a CV matrix is ​​set to determine whether constraint one is violated in a population. The CV matrix is ​​a matrix that stores the degree to which individuals in a population violate each constraint, and includes a number of rows corresponding to the number of chromosomes and a number of columns corresponding to the number of constraints.

[0324] S423. Calculate the fitness of each individual in the population, setting Gen2 = 0; wherein, the fitness function of the second stage is composed of the non-dominated sorting level and the crowding degree. The higher the non-dominated level, the lower the crowding degree, and the higher the fitness of the chromosome individual;

[0325] S424: Determine whether the preset termination condition is met. If so, output the Pareto optimal solution in the current population and decode it as the final multi-factory production scheduling robust optimization result; otherwise, proceed to S425.

[0326] S425, retain the parent population and perform a single-point crossover operation;

[0327] In this step, only the chromosome 1 in the chromosome unit is assigned a base number chromosome and the chromosome 3 is assigned a large ship number chromosome, and the single-point crossover operation is performed respectively; in the two chromosome individuals undergoing the crossover operation, a crossover point is randomly set for each to be split, and the genes on the right side of the crossover point are exchanged, thereby obtaining two different chromosomes.

[0328] S426, executing the breeder mutation operation;

[0329] In this step, breeder GA mutation operations are performed only on chromosome 1 assigned with a base number chromosome and chromosome 3 assigned with a large ship number chromosome in the chromosome unit. The compression rate (MutShrink) and gradient division (Gradient) are used to control the mutation distance. The greater the compression rate and the fewer the gradient divisions, the greater the mutation amplitude and the stronger the global search capability.

[0330] The start processing date of chromosome 2 in the chromosome unit is calculated using the formula in step S421 using chromosome 1 and chromosome 3 after crossover and mutation as known conditions without performing crossover and mutation operations.

[0331] S427. Perform constraint check on the chromosomes in the current population; if it fails, perform correction according to the preset constraint processing rules to generate a progeny population containing Nind2 individuals;

[0332] S428. Merge the offspring and the parent generation, calculate the fitness of each individual in the combined population, select the first Nind2 individuals with higher fitness; set Gen2 = Gen2 + 1, and go to S424.

[0333] An embodiment of the present invention provides a multi-objective optimization system for multi-factory production scheduling under uncertain ocean shipping time, comprising:

[0334] The acquisition module is used to obtain multi-factory production scheduling resources and single-batch overseas orders;

[0335] A construction module is used to construct a production scheduling model for a multi-factory production network based on the multi-factory production scheduling resources and a single batch of overseas orders, under the premise of determining the sea freight time;

[0336] An optimization module is used to construct a multi-objective robust optimization model for production scheduling based on the production scheduling model and represent uncertain shipping time by using a 1-norm sphere uncertainty set;

[0337] The solution module is used to convert the multi-objective robust optimization model into an equivalent model, use a heuristic algorithm to solve the equivalent model, and obtain a multi-factory production scheduling robust optimization result.

[0338] An embodiment of the present invention provides a storage medium storing a computer program for multi-objective optimization of multi-factory production scheduling under uncertain sea shipping time, wherein the computer program enables a computer to execute the multi-objective optimization method for multi-factory production scheduling under uncertain sea shipping time as described above.

[0339] An embodiment of the present invention provides an electronic device, including:

[0340] one or more processors;

[0341] Memory; and

[0342] One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including a multi-objective optimization method for executing multi-factory production scheduling under uncertain shipping time as described above.

[0343] It is understandable that the multi-objective optimization system, storage medium, and electronic device for multi-factory production scheduling under uncertain sea shipping time provided in the embodiments of the present invention correspond to the multi-objective optimization method for multi-factory production scheduling under uncertain sea shipping time provided in the embodiments of the present invention. The explanation, examples, and beneficial effects of the relevant contents can refer to the corresponding parts of the multi-objective optimization method for multi-factory production scheduling under uncertain sea shipping time, and will not be repeated here.

[0344] In summary, compared with the existing technology, the present invention has the following beneficial effects:

[0345] This embodiment of the present invention designs a 1-norm sphere uncertainty set for large-vessel shipping times and constructs a multi-objective robust optimization model for multi-factory production scheduling. This model optimizes the total delivery deviation and total delivery cost for a single batch of overseas orders. This model considers pre-defined shipping plans and long, uncertain shipping times. Furthermore, it transforms the model into an equivalent model and solves it using a two-stage hybrid heuristic algorithm, providing reasonable and feasible production scheduling recommendations for a single batch of overseas orders.

[0346] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0347] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A multi-objective optimization method for multi-factory production scheduling under uncertain shipping time, characterized by: include: S1. Obtain multi-factory production scheduling resources and single-batch overseas orders; S2. On the premise of determining the shipping time, construct a production scheduling model for the multi-factory production network based on the multi-factory production scheduling resources and the single batch of overseas orders; S3. Based on the production scheduling model, a multi-objective robust optimization model for production scheduling is constructed by representing the uncertain shipping time using a 1-norm sphere uncertainty set; S4. Converting the multi-objective robust optimization model into an equivalent model, solving the equivalent model using a heuristic algorithm, and obtaining a multi-factory production scheduling robust optimization result; The multi-objective robust optimization model in S3 includes: (1) The first objective function of the robust objective is to minimize the 1-norm of the difference between the actual delivery date of all overseas orders in a single batch of overseas orders and the expected delivery date of overseas customers when the shipping time deviation is the worst: (2) The second objective function is to minimize the total delivery cost of all overseas orders in a single batch of overseas orders: Where J represents the set of overseas orders, J = {1, 2, ..., n}; I represents the set of factories, i = {1, 2, ..., m}; G represents the set of domestic ports, G = {1, 2, ..., p}; S represents the set of large ships, S = {1, 2, ..., q}; Assuming the uncertainty parameter actual shipping time δ s In a ball, is the center of the sphere, ρ is the radius of the sphere, Δ s is the error value; let Matrix δ=[δ1,δ2,…,δ s ] is the matrix of actual shipping time for each ship, matrix is the matrix of the historical average shipping time of each ship, the matrix Δ=[Δ1,Δ2,…,Δ s ] is the deviation matrix between the actual shipping time of each ship and its historical average shipping time; There are constraints on the actual shipping time for all large ships: |D s |≤ρ s δ is an s-dimensional real vector; then the 1-norm sphere uncertainty set of the actual shipping time of the large ship is expressed as a s Indicates the date of departure of the large vessel by sea; y ijs is a decision variable. If the manufacturing task of overseas order j is assigned to factory i and the shipping task of overseas order j is assigned to ship s, it takes 1; otherwise, it takes 0. js , if the shipping task of overseas order j is assigned to large ship s, take 1; otherwise take 0; d j The date when the customer requested the goods for overseas order j; q j represents the number of products in overseas order j; They represent the processing cost, warehousing cost, domestic land transportation cost, and international shipping cost of unit product j of overseas order respectively; represents the processing cost per unit of product of overseas order j in factory i; x ijj′ is a decision variable. If the manufacturing tasks of overseas orders j and j′ are both assigned to factory i, and overseas order j is completed and then followed by overseas order j′, it takes 1; otherwise, it takes 0. represents the daily storage cost of unit product j for overseas orders; represents the storage time of overseas order j; represents the domestic transportation cost per unit of product between factory i and port g; If a large ship s departs from port g, T sg =1; otherwise T sg =0; It represents the transportation cost per unit product of large ship.

2. The multi-objective optimization method for multi-factory production scheduling under uncertain ocean shipping time according to claim 1, characterized in that: The multi-objective robust optimization model in S3 also includes: Constraints: Among them, s j 、s j′ denote the actual delivery dates of overseas orders j and j′ respectively; M represents a very large constant; tr ig represents the land transportation time from factory i to port g; pro ij represents the processing time of overseas order j in factory i.

3. The multi-objective optimization method for multi-factory production scheduling under uncertain ocean shipping time according to claim 2, characterized in that: The equivalent models in S4 include: Goal 1: Goal 2: Constraints: Among them, b j It represents the shipping date of the large vessel assigned to overseas order j minus the date when the customer requested the goods.

4. The multi-objective optimization method for multi-factory production scheduling under uncertain ocean shipping time according to claim 3, characterized in that: A two-stage hybrid heuristic algorithm is used to solve the equivalent model, including: Phase 1: Without considering the selection of the processing plant and the start date of processing, the equivalent model is simplified and relaxed, a heuristic algorithm is designed to search and obtain multiple shipping task solutions, and a population containing multiple chromosomes with assigned large ship numbers is obtained by encoding; Phase 2: The last generation of the above population is sorted and screened as the partial initial population containing multiple chromosomes assigned to large ship numbers in this stage, and a partial initial population containing multiple chromosomes assigned to factory numbers and a partial initial population containing multiple chromosomes assigned to processing start dates are generated. A multi-objective optimization heuristic algorithm is designed to obtain the Pareto optimal solution, and the solution is decoded as the final robust optimization result for multi-factory production scheduling.

5. The multi-objective optimization method for multi-factory production scheduling under uncertain ocean shipping time according to claim 4, characterized in that: The first phase specifically includes: S411. In the equivalent model, separate the decision variable y related to the allocation of large ships. js The relevant objectives and constraints are simplified into the following sub-model: Goal 1: Goal 2: Constraints: S412, solving the above sub-model, randomly generating an initial population of the first stage including multiple n columns of chromosomes, calculating the fitness of each individual in the population, and setting Gen1 = 0; Z i =aV 1 +bV 2 Among them, a and b are constants, which are the weights of target one and target two respectively; f i s1 represents the fitness value of the first stage corresponding to the i-th chromosome; Nind1 represents the population size in the first stage; S413, determine whether the termination condition is met. If so, sort the current population in non-ascending order according to fitness and output it as the last generation population Chrom1; otherwise, go to S414; S414, duplicate the elite individual into Nind1 / 2 to form a stallion; S415, roulette wheel is used to select Nind1 / 2 individuals from the parent population after removing the elite individuals; S416, merge the stallions with the selected population; S417, performing a two-point crossover operation on the merged population; S418. Perform a breeder mutation operation on the population after the crossover operation to generate a new generation of population, and calculate the fitness of each individual in the population; set Gen1 = Gen1 + 1, and return to S413.

6. The multi-objective optimization method for multi-factory production scheduling under uncertain ocean shipping time according to claim 4 or 5, characterized in that: The second phase specifically includes: S421. Generate an initial population of chromosome units in the second stage, which includes a plurality of chromosomes each consisting of three n columns: The first Nind2 individuals of the last generation population Chrom1 in the first stage are used as part of the initial population for assigning large ship numbers to chromosome 3 in the second stage NSGAII algorithm; where Nind2 represents the population size in the second stage, and Nind2 << Nind1; A priority selection mechanism for processing factories using processing and transportation costs as judgment indicators is introduced to generate a portion of the initial population of chromosome 1 assigned factory numbers. This priority selection mechanism means setting a certain selection probability to generate the initial population of chromosomes assigned base numbers. The principle of setting the selection probability is: the factory with the lower the sum of production and domestic land transportation costs, the higher the selection probability is set. Under the ideal assumption that the storage time is 0, the start date of processing is obtained by working backwards from the overseas order delivery deadline; S422, performing constraint check on the chromosomes in the initial population generated in S421; if it fails, making corrections according to the preset constraint processing rules; S423. Calculate the fitness of each individual in the population, setting Gen2 = 0; wherein, the fitness function of the second stage is composed of the non-dominated sorting level and the crowding degree. The higher the non-dominated level, the lower the crowding degree, and the higher the fitness of the chromosome individual; S424: Determine whether the preset termination condition is met. If so, output the Pareto optimal solution in the current population and decode it as the final multi-factory production scheduling robust optimization result; otherwise, proceed to S425. S425, retain the parent population and perform a single-point crossover operation; S426, executing the breeder mutation operation; S427. Perform constraint check on the chromosomes in the current population; if it fails, perform correction according to the preset constraint processing rules to generate a progeny population containing Nind2 individuals; S428. Merge the offspring and the parent generation, calculate the fitness of each individual in the combined population, select the first Nind2 individuals with higher fitness; set Gen2 = Gen2 + 1, and go to S424.

7. The multi-objective optimization method for multi-factory production scheduling under uncertain ocean shipping time according to claim 6, characterized in that: (1) Regarding constraint 1: the departure time of the large ship must be later than the time when the order arrives at the domestic port; the preset constraint processing rules specifically include: reducing the proportion of chromosomes that violate constraint 1 in the initial population through the generation rule of the initial population; and correcting the chromosomes that violate constraint 1 to chromosomes that comply with constraint 1 by directly correcting the chromosomes; And / or (2) for constraint 2: two order tasks assigned to the same processing factory cannot be processed at the same time; the preset constraint processing rules specifically refer to: The first step is to determine the order set to be judged. If the set is empty, end; otherwise, go to the second step; The second step is to determine whether the processing time of the last overseas order conflicts with the second-to-last overseas order; If yes, skip to step 3; if no, skip to step 4; In step 3, if there is no conflict, the last order is deleted from the order set and the process returns to step 2. If there is a conflict, the processing time of the last overseas order is moved forward to avoid conflict while keeping the domestic land transportation time unchanged. For the time difference, the order is placed in a warehouse near the processing factory for storage pending domestic transportation. The last order is deleted from the order set and the process returns to step 2. Step 4: Jump to the next processing plant or the next chromosome and repeat the first step.

8. A multi-objective optimization system for multi-factory production scheduling under uncertain shipping time, characterized by: include: The acquisition module is used to obtain multi-factory production scheduling resources and single-batch overseas orders; A construction module is used to construct a production scheduling model for a multi-factory production network based on the multi-factory production scheduling resources and a single batch of overseas orders, under the premise of determining the sea freight time; An optimization module is used to construct a multi-objective robust optimization model for production scheduling based on the production scheduling model and represent uncertain shipping time by using a 1-norm sphere uncertainty set; A solution module, configured to transform the multi-objective robust optimization model into an equivalent model, solve the equivalent model using a heuristic algorithm, and obtain a multi-factory production scheduling robust optimization result; Wherein, the multi-objective robust optimization model includes: (1) The first objective function of the robust objective is to minimize the 1-norm of the difference between the actual delivery date of all overseas orders in a single batch of overseas orders and the expected delivery date of overseas customers when the shipping time deviation is the worst: (2) The second objective function is to minimize the total delivery cost of all overseas orders in a single batch of overseas orders: Where J represents the set of overseas orders, J = {1, 2, ..., n}; I represents the set of factories, I = {1, 2, ..., m}; G represents the set of domestic ports, G = {1, 2, ..., p}; S represents the set of large ships, S = {1, 2, ..., q}; Assuming the uncertainty parameter actual shipping time δ s In a ball, is the center of the sphere, ρ is the radius of the sphere, Δ s is the error value; let Matrix δ=[δ1,δ2,…,δ s ] is the matrix of actual shipping time for each ship, matrix is the matrix of the historical average shipping time of each ship, the matrix Δ=[Δ1,Δ2,…,Δ s ] is the deviation matrix between the actual shipping time of each ship and its historical average shipping time; There are constraints on the actual shipping time for all large ships: |D s |≤ρ s δ is an s-dimensional real vector; then the 1-norm sphere uncertainty set of the actual shipping time of the large ship is expressed as a s Indicates the date of departure of the large vessel by sea; y ijs is a decision variable. If the manufacturing task of overseas order j is assigned to factory i and the shipping task of overseas order j is assigned to ship s, it takes 1; otherwise, it takes 0. js , if the shipping task of overseas order j is assigned to large ship s, take 1; otherwise take 0; d j The date when the customer requested the goods for overseas order j; q j represents the number of products in overseas order j; They represent the processing cost, warehousing cost, domestic land transportation cost, and international shipping cost of unit product j of overseas order respectively; represents the processing cost per unit of product of overseas order j in factory i; x ijj′ is a decision variable. If the manufacturing tasks of overseas orders j and j′ are both assigned to factory i, and overseas order j is completed and then followed by overseas order j′, it takes 1; otherwise, it takes 0. represents the daily storage cost of unit product j for overseas orders; represents the storage time of overseas order j; represents the domestic transportation cost per unit of product between factory i and port g; If a large ship s departs from port g, T sg =1; otherwise T sg =0; It represents the transportation cost per unit product of large ship.

9. A storage medium, characterized in that: It stores a computer program for multi-objective optimization of multi-factory production scheduling under uncertain sea shipping time, wherein the computer program enables the computer to execute the multi-objective optimization method for multi-factory production scheduling under uncertain sea shipping time as described in any one of claims 1 to 7.

10. An electronic device, characterized in that: include: one or more processors; Memory; as well as One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including a method for executing the multi-objective optimization method for multi-factory production scheduling under uncertain shipping time according to any one of claims 1 to 7.

Images