Joint optimization method for flexible job shop scheduling and maintenance decision considering operating conditions

By constructing a time-varying equipment reliability model and a hybrid discrete particle swarm search method, the production scheduling and maintenance decisions of the flexible manufacturing system are optimized, solving the problem of random uncertainty in equipment failure, achieving efficient equipment maintenance and stable production operation, and reducing total costs.

CN118941046BActive Publication Date: 2026-11-17TENGZHOU SHANDONG DAHAN INTELLIGENT TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411111504.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-14
Publication Date
2026-11-17
Estimated Expiration
2044-08-14

AI Technical Summary

Technical Problem

In existing flexible manufacturing systems, the multi-stage nonlinear failure characteristics exhibited by equipment under different product processing conditions exacerbate the random uncertainty of equipment failures. This makes it difficult for the joint optimization of production scheduling and maintenance decisions to respond to equipment maintenance needs under varying operating conditions, and it also fails to effectively balance system production efficiency with equipment maintenance needs.

Method used

A reliability model for equipment under time-varying operating conditions is constructed. Combining hybrid discrete particle swarm optimization and variable neighborhood search methods, a two-stage joint optimization method for production scheduling and maintenance decision-making is designed. By optimizing equipment maintenance planning and production scheduling, the optimal maintenance time interval and actual maintenance time are determined, thereby reducing downtime losses caused by equipment failures and achieving fault-free operation and stable production of equipment.

Benefits of technology

It significantly reduced downtime losses caused by equipment failures, alleviated customer dissatisfaction with delayed delivery, balanced the utilization rate among multiple devices, optimized the total cost of production and maintenance, and improved the production efficiency and equipment availability of flexible workshops.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941046B_ABST
    Figure CN118941046B_ABST
Patent Text Reader

Abstract

A joint optimization method for flexible job shop scheduling and maintenance decision-making in the field of intelligent manufacturing, considering operating conditions, is proposed. In the first stage, a production scheduling model is constructed based on the workshop equipment's production capacity, product process routes, and order task requirements. An initial production plan is generated based on a hybrid discrete particle swarm optimization and variable neighborhood search method. In the second stage, based on the maintenance decision model, the equipment maintenance requirements under this production plan are analyzed, and a maintenance plan is formulated. When production tasks and maintenance activities on the same equipment conflict, the production plan is dynamically optimized based on the variable neighborhood search method, and the equipment status is updated in real time. Simultaneously, an economically optimal production plan and maintenance schedule are formulated. This invention achieves a dynamic balance between system production efficiency and equipment maintenance requirements while ensuring fault-free equipment operation and production process continuity. It significantly reduces maintenance costs caused by equipment degradation, reduces customer dissatisfaction with delivery delays, and promotes the balanced utilization of multiple machines within the workshop.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to a technology in the field of intelligent manufacturing, specifically a joint optimization method for flexible workshop scheduling and maintenance decisions that takes into account working conditions. Background Technology

[0002] Flexible manufacturing systems are widely used in industries such as automotive, aerospace, and semiconductors. Flexible job shop scheduling enables the mixed production of various product types by rationally allocating system resources. However, equipment exhibits multi-stage nonlinear failure characteristics under different product processing conditions, exacerbating the random uncertainty of equipment failures and thus affecting product delivery time. Therefore, joint optimization of production scheduling and maintenance decisions is crucial for balancing system production efficiency and equipment maintenance needs. Current research on flexible job shop scheduling and maintenance decisions mainly focuses on optimizing production scheduling results using predetermined maintenance plans, which has the following shortcomings: it does not consider the actual operating performance of equipment and cannot respond to the uncertainty of equipment maintenance needs under varying operating conditions; furthermore, it ignores the coupling effect between production tasks and maintenance activities, and static joint optimization schemes cannot achieve global economic optimality. Summary of the Invention

[0003] This invention addresses the aforementioned shortcomings of existing technologies by proposing a joint optimization method for flexible job shop scheduling and maintenance decisions that considers operating conditions. A equipment reliability model under time-varying operating conditions is constructed, considering maintenance cost rate and equipment availability indicators to achieve equipment maintenance planning. A two-stage joint optimization method for production scheduling and maintenance decisions is designed. This invention can significantly reduce downtime losses caused by equipment failures, alleviate customer dissatisfaction with delayed deliveries, and balance the utilization rate among multiple pieces of equipment. By formulating cost-effective production and maintenance plans, it ensures fault-free equipment operation and uninterrupted, stable production.

[0004] This invention is achieved through the following technical solution:

[0005] like Figure 1 As shown, this invention relates to a joint optimization method for flexible job shop scheduling and maintenance decision-making considering operating conditions, characterized by comprising the following steps:

[0006] Step S1: Construct a production scheduling model for the flexible operation workshop based on the workshop equipment production capacity, product process route, and order task requirements. Solve the production scheduling model by combining hybrid discrete particle swarm optimization and variable neighborhood search methods to determine the initial production plan.

[0007] Step S2: Determine the processing tasks and their timing arrangements for each piece of equipment based on the initial production plan. Analyze the time-varying operating conditions formed by the equipment under different processing tasks and operating states. Establish an equipment maintenance decision model that considers the evolution of equipment failure rate under time-varying operating conditions. Solve the equipment maintenance decision model to determine the optimal maintenance time interval and the corresponding optimal equipment maintenance time point for each maintenance cycle of the equipment.

[0008] Step S3: Determine whether the maintenance period corresponding to the optimal equipment maintenance time point conflicts with the processing period of the process on the same equipment. If there is no conflict, determine the optimal equipment maintenance time point as the actual equipment maintenance time point. If there is a conflict, calculate the cost surplus of preventive maintenance activities and the cost surplus of corrective maintenance activities corresponding to the advance maintenance plan and the delayed maintenance plan respectively, and determine the actual equipment maintenance time point based on the cost surplus comparison of the two candidate plans.

[0009] Step S4: Based on the actual maintenance time of the equipment, establish a joint optimization model for production scheduling and maintenance decision-making. With the goal of minimizing the total delay penalty cost, equipment load balancing penalty cost and total maintenance cost, the variable neighborhood search method is used to optimize the initial production plan. Based on the optimized production plan, the equipment operating conditions, equipment failure rate and maintenance plan are updated synchronously to obtain the joint optimization results of the production plan and maintenance plan.

[0010] The equipment maintenance decision model defines equipment maintenance parameters, decision variables, and intermediate variables; based on the parameters, an equipment reliability model is established to calculate the equipment failure rate, and the maintenance cost rate and equipment availability are determined according to the equipment failure rate; with the optimization objectives of minimizing the maintenance cost rate and maximizing equipment availability, the equipment maintenance decision model is solved to determine the optimal maintenance time interval and the corresponding optimal equipment maintenance time for each maintenance cycle; when the optimal equipment maintenance time conflicts with the process processing time, the actual equipment maintenance time is determined by comparing the cost surplus corresponding to the early maintenance plan and the delayed maintenance plan, specifically including:

[0011] Step ①: Define the parameters as follows: B = {r1, r2, …, r K} represents the set of equipment maintenance cycles, with the preventive cycle index of equipment k being r; the duration of a single preventive maintenance activity for equipment k is represented as . The corresponding cost of preventative maintenance activities is When equipment k fails, the duration of corrective maintenance activities is expressed as follows: The corresponding corrective maintenance activities cost is The number of failures that occur before the z-th processing position of equipment k is denoted as u. kz W is a very large positive integer;

[0012] Step 2: Define the decision variable as: yrkz ∈{0,1},y rkz =1 indicates that equipment k is in the r-th maintenance cycle, and preventative maintenance activities are performed after the z-th processing position; otherwise, y rkz =0; >0 indicates the actual preventive maintenance time of equipment k in the r-th maintenance cycle;

[0013] Step ③: Define the intermediate variable as: λ rk (t|X) represents the failure rate function of device k in the r-th maintenance cycle; This represents the optimal maintenance time interval for device k in the r-th maintenance cycle. This represents the actual maintenance time interval of device k in the r-th maintenance cycle. This represents the optimal maintenance time for device k in the r-th maintenance cycle. This represents the optimal end time for maintenance operations on device k in the r-th maintenance cycle. This indicates the actual end time of maintenance work for device k in the r-th maintenance cycle;

[0014] Step 4: Establish a device reliability model under time-varying operating conditions, λ rk (t|X) =αλ k (t)f(βX), where X = [X1, X2, …, X f , …, X F [Indicates the impact of the processing on equipment M] k A set of factors contributing to failure rate, β = [β1, β2,…, β F ] represents the coefficient of the corresponding factor in X;

[0015] Step 5, Equipment M k The baseline failure rate function is represented by the Weibull distribution λ. k (t)= , where θ k and η k Let these represent the shape parameter and scale parameter in the k-Weibull distribution of the device, respectively;

[0016] Step 6: Construct an equipment maintenance planning model V with the objectives of minimizing maintenance cost rate and maximizing equipment availability. rk = By finding the derivative equation of the model and making it equal to 0, the optimal maintenance interval for each cycle of the equipment can be obtained. The maintenance cost rate is defined as follows: Equipment availability is defined as = By making d respectively / dT rk=0 and d / dT rk =0, the minimum maintenance cost rate value can be obtained. and maximum device availability value ;

[0017] Step 7: Determine the time period during which equipment k performs maintenance activities in the r-th cycle based on the optimal maintenance time interval. , When this period is during the standby period of device k, the actual start time of the maintenance activity performed by device k in the r-th cycle is... = Conversely, when [ , When the process conflicts with the normal processing of the operation at the z-th position of equipment k, calculate the cost surplus of preventive maintenance activities and the cost surplus of corrective maintenance activities corresponding to the advance maintenance plan and the delayed maintenance plan, respectively.

[0018] Step 8: When the preventive maintenance activity of equipment k in the r-th cycle is scheduled to be implemented before the start of processing at the z-th position, the start point of the maintenance activity is denoted as t. a When satisfied At that time, t a Expressed as a formula When satisfied At that time, t a Expressed as a formula , where x ijkz For binary decision variables, when process O ij When x is assigned to the z-th position of device k for processing, ijkz =1, otherwise x ijkz =0, and They are respectively process O ij Start time and completion time This represents the start time of processing at the z-th position of device k. The completion time of the (z-1)th position of device k is represented by ; the cost surplus of corrective maintenance activities is represented by . Expressed as a formula , equivalent to The cost surplus of preventive maintenance activities is expressed as Expressed as a formula ;

[0019] Step 9: When the preventive maintenance work of equipment k in the r-th cycle is scheduled to be carried out after the completion of the z-th position, the start point of the maintenance activity is denoted as t. d Expressed as a formula The cost surplus of corrective maintenance activities is expressed as Expressed as a formula , equivalent to The cost surplus of preventive maintenance activities is expressed as Expressed as a formula ;

[0020] Step 10, when Decision variable y rk(z-1) =1, =t a Conversely, y rkz =1, =t d .

[0021] The production scheduling model refers to an integer programming model established with the objective of minimizing order delay penalty costs and inter-equipment load balancing penalty costs, in order to obtain the optimal initial production scheduling scheme, specifically including:

[0022] Step 1: Define the parameters as follows: M = {M1, M2, …, M K Let {} be the set of workshop equipment, index k; P = {P1, P2, …, P} I} represents a set of products, with index i; O i = {O i1 O i2 , …, } is product P i The process set, indexed as j; M ij For process O i Optional equipment set, process O ij Processing time PT ijk Depending on the selected device M K Z = {z1, z2, …, z} K} represents the set of processing positions for the equipment, with index z; D i For product P i Delivery period, For product P i Penalty per unit time for late delivery; C b Penalties for load balancing between devices;

[0023] Step 2: Define the decision variable as: x ijkz x ∈{0,1} ijkz =1 indicates process O ij Process at the z-th position of device k, otherwise = 0; >0 indicates process O ij Start processing time; >0 indicates process O ij Completion time;

[0024] Step 3: Establish the model. The optimization objective is to minimize the total cost. The total cost consists of the total delay penalty cost and the equipment load balancing penalty cost. The total delay penalty cost refers to the sum of the delay delivery penalties for all products in the order task. It is indicated that TTC i The penalty for product i's delay is calculated as follows: if the product's delivery time is later than the required delivery date, a corresponding penalty cost is incurred, which is proportional to the delay time; otherwise, the delay penalty is zero. The equipment load balancing penalty cost is proportional to the load balancing level of all equipment in the system, denoted by TBC. The smaller the variance of the load of all equipment, the smaller this cost item. The optimization objective and the total cost of each item are expressed as follows: ,in , , ;

[0025] Step 4: Establish constraints to ensure that each process of the product can only be assigned to one position on one machine for processing, represented as: ;

[0026] Step 5: Establish constraints to ensure that the processing of product steps strictly follows the technological process. That is, the start time of a certain step must not be earlier than the completion time of the previous step. This is represented as... ;

[0027] Step 6: Establish constraints to ensure that once a process begins, it cannot be interrupted before completion. This is represented as... ;

[0028] Step 7: Establish constraints to ensure equipment M k Each processing position can only be used to process one product in a process, denoted as... ;

[0029] Step 8: Establish constraints to ensure that the processing position of each piece of equipment is sequentially assigned to each process of the product, represented as follows: ;

[0030] Step 9: Establish constraints to ensure that there is no overlap between operations assigned to the same equipment. That is, the start time of the current processing position on each piece of equipment must not be earlier than the completion time of the previous processing position. This is represented as... ;

[0031] Step 10: Considering the above parameters, decision variables, optimization objectives, and constraints, the flexible job shop scheduling model is established as follows: ,in , ,in The first constraint of the model is The second constraint is The third constraint is The fourth constraint is The fifth constraint is The sixth constraint is .

[0032] The hybrid discrete particle swarm optimization and variable neighborhood search method includes the following steps:

[0033] Step a: Initialize parameters, including particle swarm size, maximum number of iterations, crossover probability, and mutation probability;

[0034] Step b: A particle or feasible solution is represented by a two-layer encoding method, where the two layers of encoding are the process sequence and the equipment sequence, denoted as OS and MS respectively; the elements in OS represent the index of the processed product, and the number of times the index appears represents the number of processes for the corresponding product; the elements in the MS sequence represent the processing equipment selected for each process in OS;

[0035] Step c: Optimize the randomly generated initial particles using the discrete particle swarm optimization method, updating the particles with the objective function as the fitness value. This process consists of three parts: The first part randomly selects either the OS sequence or MS sequence of the particles for mutation. For the OS sequence, an element at one random position is inserted into another position, changing the order of the elements in the OS sequence. For the MS sequence, multiple random positions are selected to change the processing equipment selection for the corresponding process, generating new particles represented as follows: ;

[0036] Step d, Part Two: Randomly select an optimal particle for each individual during the crossover operation learning iteration process. Information; the first crossover operation is: randomly dividing the order task P into two non-empty subsets Pi. 1 and P 2 From the current particle Take out the contents contained in P in sequence 1 Products in the set are copied to new particles, preserving the element information of the process sequence and equipment sequence; in the individual optimal particle Remove the part contained in P 1 The product information in the file will be copied to the new particles in the original order. The second crossover operation is as follows: randomly select the first S positions of the particle encoding sequence, and from the current particle... The products from the first S positions are sequentially extracted and copied to new particles, preserving the element information of the process sequence and equipment sequence; in the individual optimal particle... Remove the particles contained in the current particle. The product information in the first S positions is used to copy the remaining products to the new particles in the original order. ;

[0037] Step e, Part 3: Randomly select a crossover operation to learn the globally optimal particle during the iteration process. Information; the first crossover operation is: randomly dividing the order task P into two non-empty subsets Pi. 1 and P 2 From the current particle Take out the contents contained in P in sequence 1 Products in the set are copied to new particles, preserving the element information of the process sequence and equipment sequence; in the globally optimal particle... Remove the part contained in P 1 The product information in the file will be copied to the new particles in the original order. The second crossover operation is as follows: randomly select the first S positions of the particle encoding sequence, and from the current particle... The product from the first S positions is sequentially extracted and copied to a new particle, preserving the element information of the process sequence and equipment sequence; in the globally optimal particle... Remove the particles contained in the current particle. The product information in the first S positions is used to copy the remaining products to the new particles in the original order. ;

[0038] Step f: When the iteration termination condition is met, select the top 10% of particles by fitness value as input for the variable neighborhood search method;

[0039] Step g: In the variable neighborhood search method, a deep local search is performed on the current particle through mutation and crossover operations. For mutation operation, an element in the MS sequence is randomly selected and the processing equipment selection of the corresponding process is changed; for crossover operation, a position in the OS sequence is randomly selected and the element at that position is inserted into other positions.

[0040] Step h: When the iteration termination condition is met, output the particle with the optimal fitness value.

[0041] The joint optimization model for production scheduling and maintenance decisions described in step S4 includes parameter definition, objective function establishment, and constraint construction, and is solved based on the joint optimization model, specifically including:

[0042] Step 1: Define M as the set of workshop equipment, k as the equipment index; P as the set of products, i as the product index; O i Let B be the set of processes for product i, and j be the process index; let B be the set of equipment maintenance cycles, and r be the maintenance cycle index; let Z be the set of equipment processing positions, and z be the processing position index; x ijkz For binary decision variables, when process Oij When x is assigned to the z-th position of device k for processing, ijkz =1, otherwise x ijkz =0; y rkz As a binary decision variable, when the preventive maintenance activity of equipment k in the r-th maintenance cycle is scheduled to be implemented after the z-th processing position, y rkz =1, otherwise y rkz =0; and These represent process O respectively. ij The start and end times of processing; and These represent the optimal start time for preventive maintenance and the actual start time for preventive maintenance of equipment k in the r-th maintenance cycle, respectively. and Let λ represent the optimal maintenance time interval and the actual maintenance time interval for device k in the r-th maintenance cycle, respectively; rk (t|X) represents the failure rate function of device k in the r-th maintenance cycle; and These represent the duration of a preventive maintenance activity and the duration of a corrective maintenance activity for equipment k, respectively. and Let u represent the cost of one preventative maintenance activity and the cost of one corrective maintenance activity for equipment k, respectively; kz Represents the number of failures that occur in equipment k before the z-th processing position; W is a very large positive integer; total delay penalty cost. and equipment load balancing penalty costs The expression is the same as in the first-stage production scheduling model; the total maintenance cost is expressed as... TMC k Let K be the maintenance cost of device k; the optimization objective and the total maintenance cost are expressed as: ,in ;

[0043] Step 2: Establish constraints to ensure that each device M in each maintenance cycle k Only one maintenance activity can be performed at most, as indicated by: ;

[0044] Step 3: Establish constraints to ensure that the maintenance cycle of each piece of equipment is scheduled in sequence, represented as... ;

[0045] Step 4: Establish constraints to ensure that the start time of the actual equipment maintenance activity is later than the completion time of the current processing position's operation, denoted as... ;

[0046] Step 5: Establish constraints to ensure that the start time of the next operation at the equipment location is later than the end time of the maintenance activity at the current location, i.e., for It should meet the start time of the next process position of the equipment. Greater than or equal to the end time of the maintenance activity at the current location ;

[0047] Step 6: Combining the above parameters, decision variables, optimization objective, constraints, and the constraints in the first-stage production scheduling model, the second-stage joint optimization model for production scheduling and maintenance decisions is established as follows: The overall objective is expressed as... ,in ,in The first constraint is The second constraint is The third constraint is The fourth constraint is .in , .

[0048] Technical effect

[0049] This invention addresses the shortcomings of traditional flexible job shop scheduling and maintenance decision-making research, which fails to adequately consider equipment failure behaviors dependent on production and the strong coupling between production tasks and maintenance activities. It proposes a joint optimization method for flexible job shop scheduling and maintenance decision-making that considers operating conditions. Specifically, the project constructs a reliability model for equipment considering time-varying operating conditions, proposes an equipment maintenance planning model based on maintenance cost rate and availability indicators to solve for the optimal maintenance interval for each piece of equipment, and proposes a cost-saving maintenance decision model to address the non-parallelism of production tasks and maintenance activities for the same equipment. For the first time, it establishes an integer programming model with the optimization objective of minimizing total cost, including total delay penalty cost, inter-equipment load balancing penalty cost, and total maintenance cost, systematically considering the processing capacity of flexible workshop equipment, order task requirements, product process routes, and production-maintenance coupling constraints. It innovatively constructs a two-stage joint optimization framework for flexible workshop scheduling and maintenance decisions, and obtains highly economical and operable production and maintenance solutions through a hybrid discrete particle swarm optimization and variable neighborhood search method. This significantly reduces the total production and maintenance cost of flexible workshops, including total delay penalty cost, inter-equipment load balancing penalty cost, and total maintenance cost, compared to traditional optimization methods, providing valuable guidance for manufacturing enterprises in formulating and optimizing production and maintenance solutions for flexible workshops in real time. Attached Figure Description

[0050] Figure 1 This is a flowchart of the present invention;

[0051] Figure 2 This is a Gantt chart of the initial production scheme obtained in the first stage of this invention based on a hybrid discrete particle swarm optimization and variable neighborhood search method.

[0052] Figure 3 This is a diagram showing the equipment failure rate evolution under the initial production scheme in this embodiment of the invention, regardless of whether operating conditions are considered.

[0053] Figure 4 This is a comparison chart of the optimal equipment and actual maintenance activity execution time periods in an embodiment of the present invention;

[0054] Figure 5 This is a Gantt chart of the flexible workshop production and maintenance scheme optimized based on the variable neighborhood search method in this embodiment of the invention. Detailed Implementation

[0055] This embodiment relates to a joint optimization method for flexible job shop scheduling and maintenance decision-making that considers operating conditions, including the following steps:

[0056] Step 1: Based on the requirements of flexible workshop production scheduling, design decision variables and optimization objectives: This invention optimizes the total cost, including total delay penalty cost and equipment load balancing penalty cost, by rationally planning the production scheme of flexible workshop order tasks. Therefore, the decision variables and optimization objectives are designed as follows:

[0057] (1-1) The decision variables include decision variables related to the order production plan, x ijkz Let x be a binary decision variable determining whether process j of product i is processed on equipment k. ijkz =1 indicates yes, otherwise x ijkz =0; Decision on the start time of process j for product i; Decision-making process j for product i.

[0058] (1-2) The optimization objective is to minimize the total delay penalty cost. The total cost is the sum of the load balancing penalty cost (TBC) and the equipment load balancing penalty cost (TTC). i The late delivery penalty for product i is calculated as follows: if the product's delivery time is later than the required delivery date, a corresponding penalty cost is incurred, which is proportional to the late delivery time; otherwise, the late delivery penalty is zero. The Total Cost Balance (TBC) is proportional to the load balancing level of all equipment in the workshop; the smaller the variance of the equipment load, the smaller this cost item. Therefore, the problem optimization objective and the cost design for each part are as follows: ,in , ,in Among them, M, P, O iLet k represent the set of equipment in the workshop, the set of products in the order task, and the set of processes for product i, respectively. k is the equipment number, i is the product number in the order task, j is the process number for the product, and M is the process number for the product. ij Indicates process O ij Optional device set, PT ijk Indicates process O ij In device M K The time required for processing; Z represents the set of processing positions of the workshop equipment, D i Product P i Delivery period, For product P i Penalty per unit time for late delivery, C b The penalty for load balancing between devices is indicated by the parameters mentioned above, which are all obtained based on the actual production conditions in the workshop.

[0059] Step 2: Based on the actual production situation, construct the constraints of the model as follows:

[0060] (2-1) Establish constraints to ensure that each process of the product can only be assigned to one position on one machine for processing, expressed as: ;

[0061] (2-2) Establish constraints to ensure that the processing of product steps strictly follows the technological process, that is, the start time of a certain step must not be earlier than the completion time of the previous step of the product, expressed as: ;

[0062] (2-3) Establish constraints to ensure that a process cannot be interrupted once it starts and before it is completed, represented as: ;

[0063] (2-4) Establish constraints to ensure that equipment M k Each processing position can only be used to process one product in a process, denoted as... ;

[0064] (2-5) Establish constraints to ensure that the processing position of each piece of equipment is sequentially assigned to each process step of the product, represented as follows: ;

[0065] (2-6) Establish constraints to ensure that there is no overlap between operations assigned to the same equipment, that is, the start time of the current processing position of each equipment must not be earlier than the completion time of the previous processing position, expressed as: ;

[0066] Step 3: Solve the integer programming model with the optimization objective of minimizing total cost using a hybrid discrete particle swarm optimization and variable neighborhood search method, and generate the initial production plan before the workshop starts processing, specifically including:

[0067] (3-1) Set the particle swarm size, maximum number of iterations, crossover probability, and mutation probability parameters;

[0068] (3-2) A two-layer coding method is used to represent a particle or feasible solution, specifically including a process sequence and a device sequence, denoted as OS and MS respectively; the elements in OS represent the index of the processed product, and the number of times the index appears represents the number of processes for the corresponding product; the elements in the MS sequence represent the processing device selected for each process in OS;

[0069] (3-3) Update the particles using the objective function as the fitness value. Randomly select either the OS sequence or MS sequence of the particles for mutation. For the OS sequence, randomly select an element at one position and insert it into other positions, changing the order of the elements in the OS sequence. For the MS sequence, randomly select multiple positions to change the processing equipment selection for the corresponding process, generating new particles represented as follows: ;

[0070] (3-4) Randomly select the optimal particle in the crossover operation learning iteration process. The information. The first crossover operation is: randomly divide the order task P into two non-empty subsets Pi. 1 and P 2 From the current particle Take out the contents contained in P in sequence 1 Products in the set are copied to new particles, preserving the element information of the process sequence and equipment sequence; in the individual optimal particle Remove the part contained in P 1 The product information in the file will be copied to the new particles in the original order. The second type of crossover operation involves randomly selecting the first S positions of the particle encoding sequence, and then... The products from the first S positions are sequentially extracted and copied to new particles, preserving the element information of the process sequence and equipment sequence; in the individual optimal particle... Remove the particles contained in the current particle. The product information in the first S positions is used to copy the remaining products to the new particles in the original order. ;

[0071] (3-5) Randomly select a cross-operation learning iteration process to find the globally optimal particle. The information. The first crossover operation is: randomly divide the order task P into two non-empty subsets Pi. 1 and P 2 From the current particle Take out the contents contained in P in sequence 1 Products in the set are copied to new particles, preserving the element information of the process sequence and equipment sequence; in the globally optimal particle... Remove the part contained in P1 The product information in the file will be copied to the new particles in the original order. The second type of crossover operation involves randomly selecting the first S positions of the particle encoding sequence, and then... The product from the first S positions is sequentially extracted and copied to a new particle, preserving the element information of the process sequence and equipment sequence; in the globally optimal particle... Remove the particles contained in the current particle. The product information in the first S positions is used to copy the remaining products to the new particles in the original order. .

[0072] (3-6) When the iteration termination condition is reached, select the top 10% of particles with fitness values ​​as input for the variable neighborhood search method;

[0073] (3-7) In the variable neighborhood search method, the current particle is searched locally through mutation and crossover operations. For mutation operation, an element in the MS sequence is randomly selected and the corresponding processing equipment selection is changed. For crossover operation, a position in the OS sequence is randomly selected and the element at that position is inserted into other positions.

[0074] (3-8) When the iteration termination condition is met, output the particle with the optimal fitness value as the initial production scheduling scheme.

[0075] Step 4: Based on the requirements of the joint scheduling problem of flexible workshop scheduling and equipment maintenance decision-making, design decision variables and optimization objectives: This invention optimizes the total cost, including total delay penalty cost, equipment load balancing penalty cost, and total maintenance cost, by considering time-varying operating conditions and the strong coupling effect between production tasks and maintenance activities. It also formulates maintenance plans and optimizes the initial production scheme of the first stage. Therefore, the decision variables and optimization objectives are designed as follows:

[0076] (4-1) Set the decision variable: y rkz Let y be a binary decision variable for deciding whether to perform preventative maintenance on equipment k after the z-th processing position in the r-th maintenance cycle. rkz =1 indicates yes, otherwise y rkz =0; The actual preventive maintenance time for decision-making equipment k in the r-th maintenance cycle;

[0077] (4-2) Obtain the baseline failure rate distribution function λ of each piece of equipment in the workshop. k (t)= Select the equipment M that affects the processing process k A set of factors for failure rate X = [X1, X2, …, X f , …, X FFurthermore, experiments were designed to determine the extent to which these factors affect the equipment failure rate, β = [β1, β2, …, β]. F ];

[0078] (4-3) Based on the equipment failure rate λ under time-varying operating conditions rk (t|X) =αλ k f(βX) represents the maintenance cost rate. and equipment availability = Indicators for building equipment maintenance planning models V rk = By finding the derivative equation of the model and making it equal to 0, the optimal maintenance interval for each cycle of the equipment can be obtained. ; where the preventive cycle index for device k is r, This indicates the duration of a single preventative maintenance activity for equipment k. This indicates the unit cost of preventative maintenance activities. Indicates the duration of corrective maintenance activities, This indicates the unit cost of corrective maintenance activities;

[0079] (4-4) Determine the time period during which equipment k performs maintenance activities in the r-th cycle based on the optimal maintenance time interval. , Determine whether the time period is within the standby period of device k. If so, the actual start time of the maintenance activity performed by device k in the r-th cycle is... = Otherwise, further construct cost surpluses for preventive maintenance activities and corrective maintenance activities, and optimize the actual maintenance timing of equipment based on the cost surplus optimization method;

[0080] (4-5) Assuming that the preventive maintenance activity of equipment k in the r-th cycle is scheduled to be implemented before the start of processing at the z-th position, construct the start time t of the maintenance activity. a When satisfied At that time, t a Expressed as a formula When satisfied At that time, t a Expressed as a formula The cost surplus of corrective maintenance activities is expressed as Expressed as a formula , equivalent to The cost surplus of preventive maintenance activities is expressed as Expressed as a formula ;in, This represents the optimal maintenance time interval for device k in the r-th maintenance cycle. This represents the actual maintenance time interval of device k in the r-th maintenance cycle; x ijkz For binary decision variables, when process O ij When x is assigned to the z-th position of device k for processing, ijkz =1, otherwise x ijkz =0, and They are respectively process O ij Start time and completion time This represents the start time of processing at the z-th position of device k. This represents the completion time of the (z-1)th position of device k;

[0081] (4-6) Assuming that the preventive maintenance operation of equipment k in the r-th cycle is scheduled to be carried out after the completion of the z-th position, construct the start time t of the maintenance activity. d Expressed as a formula The cost surplus of corrective maintenance activities is expressed as Expressed as a formula , equivalent to The cost surplus of preventive maintenance activities is expressed as Expressed as a formula ;

[0082] (4-7) Based on the cost-saving optimization method and the economic benefits of advancing or delaying equipment maintenance, the actual maintenance activity timing for each piece of equipment is determined. Decision variable y rk(z-1) =1, =t a Otherwise, y rkz =1, =t d .

[0083] (4-8) With the goal of minimizing the total delay penalty cost, equipment load balancing penalty cost, and total maintenance cost, the initial production plan is optimized based on the aforementioned variable neighborhood search method, and a workshop equipment maintenance plan is simultaneously formulated. The total delay penalty cost... and equipment load balancing penalty costs The expression is the same as in the first-stage production scheduling model; the total maintenance cost is expressed as... TMC k Let K be the maintenance cost of device k; the optimization objective and total maintenance cost for the second stage are expressed as: ,in, Where M is the set of workshop equipment, k is the equipment index; P is the set of products, i is the product index; O iLet B be the set of processes for product i, and j be the process index; let B be the set of equipment maintenance cycles, and r be the maintenance cycle index; let Z be the set of equipment processing positions, and z be the processing position index; x ijkz For binary decision variables, when process O ij When x is assigned to the z-th position of device k for processing, ijkz =1, otherwise x ijkz =0; y rkz As a binary decision variable, when the preventive maintenance activity of equipment k in the r-th maintenance cycle is scheduled to be implemented after the z-th processing position, y rkz =1, otherwise y rkz =0; and These represent process O respectively. ij The start and end times of processing; and These represent the optimal start time for preventive maintenance and the actual start time for preventive maintenance of equipment k in the r-th maintenance cycle, respectively. and Let λ represent the optimal maintenance time interval and the actual maintenance time interval for device k in the r-th maintenance cycle, respectively; rk (t|X) represents the failure rate function of device k in the r-th maintenance cycle; and These represent the duration of a preventive maintenance activity and the duration of a corrective maintenance activity for equipment k, respectively. and Let u represent the cost of one preventative maintenance activity and the cost of one corrective maintenance activity for equipment k, respectively; kz This represents the number of failures that occur in device k before the z-th processing position; W is a very large positive integer.

[0084] (4-9) Establish constraints to ensure that each device M in each maintenance cycle k Only one maintenance activity can be performed at most, as indicated by: ;

[0085] (4-10) Establish constraints to ensure that the maintenance cycle of each piece of equipment is scheduled in sequence, denoted as: ;

[0086] (4-11) Establish constraints to ensure that the start time of the actual equipment maintenance activity is later than the completion time of the operation at the current processing position, denoted as: ;

[0087] (4-12) Establish constraints to ensure that the start time of the next operation at the equipment location is later than the end time of the maintenance activity at the current location, i.e., for It should meet the start time of the next process position of the equipment. Greater than or equal to the end time of the maintenance activity at the current location ;

[0088] The above method was applied to the production and maintenance plan of the flexible operation workshop. The maintenance-related parameters of each piece of equipment are shown in Table 1, and the order task information and the processes that each piece of equipment can process are shown in Table 2.

[0089] Table 1 Maintenance-related parameters for each piece of equipment

[0090] Table 2. Order task information and processing procedures that each piece of equipment can handle.

[0091] First, an initial flexible job shop production scheduling scheme is generated based on a hybrid discrete particle swarm optimization and variable neighborhood search method. The Gantt chart of this scheme is as follows: Figure 2 As shown, the total cost of delay penalties and the cost of load balancing penalties between devices is 1612.45 yuan;

[0092] In the initial production scheduling plan, the equipment failure rate is calculated separately for two scenarios: considering operating conditions and not considering operating conditions. Figure 3 As shown. The failure rate of each piece of equipment increases over time. Considering the operating conditions, the equipment failure rate curve exhibits dynamic and multi-level characteristics, such as... Figure 3 As shown by the red line, the working conditions in the flexible workroom accelerate the probability of equipment failure.

[0093] Based on the maintenance decision model, the timing of maintenance activities for workshop equipment under the initial production scheduling plan is analyzed, and the optimal and actual maintenance activity timings are compared. Figure 4 As shown;

[0094] In this embodiment, devices M8 and M 10 and M 12 The actual maintenance activities will still interrupt the normal processing of subsequent processes on these machines. Therefore, the initial production plan is optimized based on a variable neighborhood search method, and the equipment failure rate and maintenance plan are updated simultaneously. This results in a joint optimization scheme for flexible job shop scheduling and maintenance decisions with a total cost of 8665.865 yuan. (Gantt chart shown). Figure 5 As shown;

[0095] To verify the effectiveness of the proposed joint optimization method for flexible workshop scheduling and maintenance decision-making considering working conditions, other traditional flexible workshop scheduling strategies were applied to this case, and the total costs were compared. Strategy IV is the method proposed in this invention, and other scheduling strategies are as follows:

[0096] Strategy I: Perform static, periodic maintenance on the equipment, setting the maintenance cycle T = 90 minutes. When the equipment has been running for T minutes, insert a maintenance activity directly and mark the maintenance activity's occupancy period as unavailable.

[0097] Strategy II: Schedule maintenance activities based on the optimal maintenance interval of the equipment. When production tasks and maintenance activities conflict, execute the right shift strategy, and the equipment will prioritize completing the current process and postpone the maintenance activity.

[0098] Strategy III: Schedule maintenance activities based on the optimal maintenance interval of the equipment. When production tasks and maintenance activities conflict, implement the left shift strategy, prioritize the maintenance activities on the equipment, and postpone the processing of the current operation.

[0099] The total cost of the production and maintenance schemes derived from the above strategies was calculated, and the results are shown in Table 3. The comparison of total costs shows that the joint optimization method of flexible work shop scheduling and maintenance decision-making that considers working conditions proposed in this invention can achieve significant total cost optimization.

[0100] Table 3. Comparison of total costs for production and maintenance solutions under different strategies

[0101] Compared with existing technologies, the performance improvement of this method lies in the following: It proposes a joint optimization method for flexible job shop scheduling and maintenance decision-making that considers operating conditions. Specifically, it constructs an equipment reliability model considering time-varying operating conditions, proposes an equipment maintenance planning model based on maintenance cost rate and availability indicators to solve for the optimal maintenance interval for each piece of equipment, and proposes a cost-saving maintenance decision-making model to address the non-parallelism of production tasks and maintenance activities for the same equipment. For the first time, it establishes an integer programming model with the optimization objective of minimizing total cost, including total delay penalty cost, inter-equipment load balancing penalty cost, and total maintenance cost, systematically considering the processing capacity of flexible job shop equipment, order task requirements, product process routes, and production-maintenance coupling constraints. It innovatively constructs a two-stage joint optimization framework for flexible job shop scheduling and maintenance decision-making, and obtains highly economical and operable production and maintenance solutions through a hybrid discrete particle swarm optimization and variable neighborhood search method. Compared with traditional methods, the flexible workshop production and maintenance solution obtained by this invention can realize real-time updates of equipment maintenance plans as production plans change, and can simultaneously optimize and adjust production and maintenance plans for the strong coupling relationship between production tasks and maintenance activities, effectively reducing total delay delivery, equipment load balancing penalty costs and maintenance costs.

[0102] The above-described specific implementations can be partially adjusted by those skilled in the art in different ways without departing from the principles and purpose of the present invention. The scope of protection of the present invention is defined by the claims and is not limited to the above-described specific implementations. All implementation schemes within the scope of the claims are bound by the present invention.

Claims

1. A joint optimization method for flexible job shop scheduling and maintenance decision-making considering operating conditions, characterized in that, Includes the following steps: Step S1: Construct a production scheduling model for the flexible operation workshop based on the workshop equipment production capacity, product process route, and order task requirements. Solve the production scheduling model by combining the hybrid discrete particle swarm optimization and variable neighborhood search methods to determine the initial production plan. Step S2: Determine the processing tasks and their timing arrangements for each piece of equipment based on the initial production plan. Analyze the time-varying operating conditions formed by the equipment under different processing tasks and operating states. Establish an equipment maintenance decision model that considers the evolution of equipment failure rate under time-varying operating conditions. Solve the equipment maintenance decision model to determine the optimal maintenance time interval and the corresponding optimal equipment maintenance time point for each maintenance cycle of the equipment. Step S3: Determine whether the maintenance period corresponding to the optimal equipment maintenance time point conflicts with the processing time of the process on the same equipment. If there is no conflict, determine the optimal equipment maintenance time point as the actual equipment maintenance time point. If there is a conflict, calculate the cost surplus of preventive maintenance activities and the cost surplus of corrective maintenance activities corresponding to the early maintenance plan and the delayed maintenance plan respectively, and determine the actual maintenance time of the equipment based on the cost surplus comparison of the two candidate plans. Step S4: Based on the actual maintenance time of the equipment, establish a joint optimization model for production scheduling and maintenance decision-making. With the goal of minimizing the total delay penalty cost, equipment load balancing penalty cost and total maintenance cost, the variable neighborhood search method is used to optimize the initial production plan. Based on the optimized production plan, the equipment operating conditions, equipment failure rate and maintenance plan are updated synchronously to obtain the joint optimization results of the production plan and maintenance plan. The equipment maintenance decision model defines equipment maintenance parameters, decision variables, and intermediate variables; based on the parameters, an equipment reliability model is established to calculate the equipment failure rate, and the maintenance cost rate and equipment availability are determined according to the equipment failure rate; with the optimization objectives of minimizing the maintenance cost rate and maximizing equipment availability, the equipment maintenance decision model is solved to determine the optimal maintenance time interval and the corresponding optimal equipment maintenance time for each maintenance cycle; when the optimal equipment maintenance time conflicts with the process processing time, the actual equipment maintenance time is determined by comparing the cost surplus corresponding to the early maintenance plan and the delayed maintenance plan, specifically including: Step ①: Define the parameters as follows: B = {r1, r2, …, r K } represents the set of equipment maintenance cycles, with the preventive cycle index of equipment k being r; the duration of a single preventive maintenance activity for equipment k is represented as . The corresponding cost of preventative maintenance activities is When equipment k fails, the duration of corrective maintenance activities is expressed as follows: The corresponding corrective maintenance activities cost is The number of failures that occur before the z-th processing position of equipment k is denoted as u. kz W is a very large positive integer; Step 2: Define the decision variable as: y rkz ∈{0,1},y rkz =1 indicates that equipment k is in the r-th maintenance cycle, and preventative maintenance activities are performed after the z-th processing position; otherwise, y rkz =0; >0 indicates the actual preventive maintenance time of equipment k in the r-th maintenance cycle; Step ③: Define the intermediate variable as: λ rk (t|X) represents the failure rate function of device k in the r-th maintenance cycle; This represents the optimal maintenance time interval for device k in the r-th maintenance cycle. This represents the actual maintenance time interval of device k in the r-th maintenance cycle. This represents the optimal maintenance time for device k in the r-th maintenance cycle. This represents the optimal end time for maintenance operations on device k in the r-th maintenance cycle. This indicates the actual end time of maintenance work for device k in the r-th maintenance cycle; Step 4: Establish a device reliability model under time-varying operating conditions, λ rk (t|X) =αλ k (t)f(βX), where X=[X1, X2, …, X f , …, X F [Indicates the impact of the processing on equipment M] k A set of factors contributing to failure rate, β = [β1, β2,…, β F ] represents the coefficient of the corresponding factor in X; Step 5, Equipment M k The baseline failure rate function is represented by the Weibull distribution λ. k (t)= , where θ k and η k Let these represent the shape parameter and scale parameter in the k-Weibull distribution of the device, respectively; Step 6: Construct an equipment maintenance planning model V with the objectives of minimizing maintenance cost rate and maximizing equipment availability. rk = By finding the derivative equation of the model and making it equal to 0, the optimal maintenance interval for each cycle of the equipment can be obtained. Among them, the maintenance cost rate is defined as Equipment availability is defined as = By making d respectively / dT rk =0 and d / dT rk =0, the minimum maintenance cost rate value can be obtained. and maximum device availability value ; Step 7: Determine the time period during which equipment k performs maintenance activities in the r-th cycle based on the optimal maintenance time interval. , When this period is during the standby period of device k, the actual start time of the maintenance activity performed by device k in the r-th cycle is... = Conversely, when [ , When the process conflicts with the normal processing of the operation at the z-th position of equipment k, calculate the cost surplus of preventive maintenance activities and the cost surplus of corrective maintenance activities corresponding to the advance maintenance plan and the delayed maintenance plan, respectively. Step 8: When the preventive maintenance activity of equipment k in the r-th cycle is scheduled to be performed before the start of processing at the z-th position, the start point of the maintenance activity is denoted as t. a When satisfied At that time, t a Expressed as a formula When satisfied At that time, t a Expressed as a formula , where x ijkz For binary decision variables, when process O ij When x is assigned to the z-th position of device k for processing, ijkz =1, otherwise x ijkz =0, and They are respectively process O ij Start time and completion time This represents the start time of processing at the z-th position of device k. The completion time of the (z-1)th position of device k is represented by ; the cost surplus of corrective maintenance activities is represented as Expressed as a formula , equivalent to The cost surplus of preventive maintenance activities is expressed as Expressed as a formula ; Step 9: When the preventive maintenance work of equipment k in the r-th cycle is scheduled to be carried out after the completion of the z-th position, the start point of the maintenance activity is denoted as t. d Expressed as a formula The cost surplus of corrective maintenance activities is expressed as Expressed as a formula , equivalent to The cost surplus of preventive maintenance activities is expressed as Expressed as a formula ; Step 10, when Decision variable y rk(z-1) =1, =t a Conversely, y rkz =1, =t d .

2. The joint optimization method for flexible job shop scheduling and maintenance decision-making considering operating conditions as described in claim 1, characterized in that, The production scheduling model refers to an integer programming model established with the objective of minimizing order delay penalty costs and inter-equipment load balancing penalty costs, in order to obtain the optimal initial production scheduling scheme, specifically including: Step 1: Define the parameters as follows: M = {M1, M2, …, M K Let {} be the set of workshop equipment, index k; P = {P1, P2, ..., P} I } represents a set of products, with index i; O i = {O i1 O i2 , …, } is product P i The process set, indexed as j; M ij For process O i Optional equipment set, process O ij Processing time PT ijk Depending on the selected device M K Z = {z1, z2, …, z} K } represents the set of processing positions for the equipment, with index z; D i For product P i Delivery period, For product P i Penalty per unit time for late delivery; C b Penalties for load balancing between devices; Step 2: Define the decision variable as: x ijkz x ∈{0,1} ijkz =1 indicates process O ij Process at the z-th position of device k, otherwise = 0; >0 indicates process O ij Start processing time; >0 indicates process O ij Completion time; Step 3: Establish the model. The optimization objective is to minimize the total cost. The total cost consists of the total delay penalty cost and the equipment load balancing penalty cost. The total delay penalty cost refers to the sum of the delay delivery penalties for all products in the order task. It is indicated that TTC i The penalty for product i's delay is calculated as follows: if the product's delivery time is later than the required delivery date, a corresponding penalty cost is incurred, which is proportional to the delay time; otherwise, the delay penalty is zero. The equipment load balancing penalty cost is proportional to the load balancing level of all equipment in the system, denoted by TBC. The smaller the variance of the load of all equipment, the smaller this cost item. The optimization objective and the total cost of each item are expressed as follows: ,in , , ; Step 4: Establish constraints to ensure that each process of the product can only be assigned to one position on one machine for processing, represented as: ; Step 5: Establish constraints to ensure that the processing of product steps strictly follows the technological process. That is, the start time of a certain step must not be earlier than the completion time of the previous step. This is represented as... ; Step 6: Establish constraints to ensure that once a process begins, it cannot be interrupted before completion. This is represented as... ; Step 7: Establish constraints to ensure device M k Each processing position can only be used to process one product in a process, denoted as... ; Step 8: Establish constraints to ensure that the processing position of each piece of equipment is sequentially assigned to each process of the product, represented as follows: ; Step 9: Establish constraints to ensure that there is no overlap between operations assigned to the same equipment. That is, the start time of the current processing position on each piece of equipment must not be earlier than the completion time of the previous processing position. This is represented as... ; Step 10: Considering the above parameters, decision variables, optimization objectives, and constraints, the flexible job shop scheduling model is established as follows: ,in , ,in The first constraint of the model is The second constraint is The third constraint is The fourth constraint is The fifth constraint is The sixth constraint is .

3. The joint optimization method for flexible job shop scheduling and maintenance decision-making considering operating conditions as described in claim 1, characterized in that, The hybrid discrete particle swarm optimization and variable neighborhood search method includes the following steps: Step a: Initialize parameters, including particle swarm size, maximum number of iterations, crossover probability, and mutation probability; Step b: A particle or feasible solution is represented by a two-layer encoding method, where the two layers of encoding are the process sequence and the equipment sequence, denoted as OS and MS respectively; the elements in OS represent the index of the processed product, and the number of times the index appears represents the number of processes for the corresponding product; the elements in the MS sequence represent the processing equipment selected for each process in OS; Step c: Optimize the randomly generated initial particles using the discrete particle swarm optimization method, updating the particles with the objective function as the fitness value. This process consists of three parts: The first part randomly selects either the OS sequence or MS sequence of the particles for mutation. For the OS sequence, an element at one random position is inserted into another position, changing the order of the elements in the OS sequence. For the MS sequence, multiple random positions are selected to change the processing equipment selection for the corresponding process, generating new particles represented as follows: ; Step d, Part Two: Randomly select an optimal particle for each individual during the crossover operation learning iteration process. Information; the first crossover operation is: randomly dividing the order task P into two non-empty subsets Pi. 1 and P 2 From the current particle Take out the contents contained in P in sequence 1 Products in the set are copied to new particles, preserving the element information of the process sequence and equipment sequence; in the individual optimal particle Remove the part contained in P 1 The product information in the file will be copied to the new particles in the original order. The second crossover operation is as follows: randomly select the first S positions of the particle encoding sequence, and from the current particle... The products from the first S positions are sequentially extracted and copied to new particles, preserving the element information of the process sequence and equipment sequence; in the individual optimal particle... Remove the particles contained in the current particle. The product information in the first S positions is used to copy the remaining products to the new particles in the original order. ; Step e, Part 3: Randomly select a crossover operation to learn the globally optimal particle during the iteration process. Information; the first crossover operation is: randomly dividing the order task P into two non-empty subsets Pi. 1 and P 2 From the current particle Take out the contents contained in P in sequence 1 Products in the set are copied to new particles, preserving the element information of the process sequence and equipment sequence; in the globally optimal particle... Remove the part contained in P 1 The product information in the file will be copied to the new particles in the original order. The second crossover operation is as follows: randomly select the first S positions of the particle encoding sequence, and from the current particle... The product from the first S positions is sequentially extracted and copied to a new particle, preserving the element information of the process sequence and equipment sequence; in the globally optimal particle... Remove the particles contained in the current particle. The product information in the first S positions is used to copy the remaining products to the new particles in the original order. ; Step f: When the iteration termination condition is met, select the top 10% of particles by fitness value as input for the variable neighborhood search method; Step g: In the variable neighborhood search method, a deep local search is performed on the current particle through mutation and crossover operations. For mutation operation, an element in the MS sequence is randomly selected and the processing equipment selection of the corresponding process is changed; for crossover operation, a position in the OS sequence is randomly selected and the element at that position is inserted into other positions. Step h: When the iteration termination condition is met, output the particle with the optimal fitness value.

4. The joint optimization method for flexible job shop scheduling and maintenance decision-making considering operating conditions as described in claim 1, characterized in that, The joint optimization model for production scheduling and maintenance decisions described in step S4 includes parameter definition, objective function establishment, and constraint construction, and is solved based on the joint optimization model, specifically including: Step 1: Define M as the set of workshop equipment, k as the equipment index; P as the set of products, i as the product index; O i Let B be the set of processes for product i, and j be the process index; let B be the set of equipment maintenance cycles, and r be the maintenance cycle index; let Z be the set of equipment processing positions, and z be the processing position index; x ijkz For binary decision variables, when process O ij When x is assigned to the z-th position of device k for processing, ijkz =1, otherwise x ijkz =0; y rkz As a binary decision variable, when the preventive maintenance activity of equipment k in the r-th maintenance cycle is scheduled to be implemented after the z-th processing position, y rkz =1, otherwise y rkz =0; and Representing process O ij The start and end times of the processing; and These represent the optimal start time for preventive maintenance and the actual start time for preventive maintenance of equipment k in the r-th maintenance cycle, respectively. and Let λ represent the optimal maintenance time interval and the actual maintenance time interval for device k in the r-th maintenance cycle, respectively; rk (t|X) represents the failure rate function of device k in the r-th maintenance cycle; and These represent the duration of a preventive maintenance activity and the duration of a corrective maintenance activity for equipment k, respectively. and Let u represent the cost of one preventative maintenance activity and the cost of one corrective maintenance activity for equipment k, respectively; kz Represents the number of failures that occur in equipment k before the z-th processing position; W is a very large positive integer; total delay penalty cost. and equipment load balancing penalty costs The expression is the same as in the first-stage production scheduling model; the total maintenance cost is expressed as... TMC k Let K be the maintenance cost of device k; the optimization objective and the total maintenance cost are expressed as: ,in ; Step 2: Establish constraints to ensure that each device M in each maintenance cycle k Only one maintenance activity can be performed at most, as indicated by: ; Step 3: Establish constraints to ensure that the maintenance cycle of each piece of equipment is scheduled in sequence, represented as... ; Step 4: Establish constraints to ensure that the start time of the actual equipment maintenance activity is later than the completion time of the current processing position's operation, denoted as... ; Step 5: Establish constraints to ensure that the start time of the next operation at the equipment location is later than the end time of the maintenance activity at the current location, i.e., for It should meet the start time of the next process position of the equipment. Greater than or equal to the end time of the maintenance activity at the current location ; Step 6: Combining the above parameters, decision variables, optimization objective, constraints, and the constraints in the first-stage production scheduling model, the second-stage joint optimization model for production scheduling and maintenance decisions is established as follows: The overall objective is expressed as... ,in ,in The first constraint is The second constraint is The third constraint is The fourth constraint is ,in , .

Citation Information

Patent Citations

  • Flexible flow shop dynamic scheduling and maintenance optimization method considering random interference

    CN118011978A

  • Multi-target flexible workshop dynamic scheduling method and device under double-resource constraint and electronic equipment

    CN118333313A