Production plan and workshop scheduling integrated optimization method for hydraulic cylinder manufacturing enterprise

Abstract

The invention discloses a production plan and workshop scheduling integrated optimization method of a hydraulic cylinder manufacturing enterprise. The method comprises the following steps of performing a series of assumptions on a production plan and workshop scheduling integrated model of the hydraulic cylinder manufacturing enterprise; establishing a multi-objective optimization function taking minimization of machining time as a main objective and minimization of machining cost as a secondary objective, and setting constraint conditions of the multi-objective optimization function; an improved particle swarm algorithm is adopted to solve the multi-objective optimization function, the improved particle swarm algorithm adopts double-layer coding to represent the positions of particles, the first layer of the double-layer coding is the machining sequence of each workpiece, and the second layer of the double-layer coding is machining equipment of each machining sequence; and the improved particle swarm optimization algorithm performs iterative optimization on the positions of the particles, and outputs a scheme which meets the completion time and has the minimum processing cost as an optimal production plan and workshop scheduling scheme.

Description

Technical Field

[0001] The invention relates to the technical field of workshop scheduling, and in particular to a production plan and workshop scheduling integrated optimization method for a hydraulic cylinder manufacturing enterprise. Background Art

[0002] With the diversification of market demand, hydraulic cylinder manufacturers are facing the transformation of production mode. The traditional standard mass production method no longer meets the needs of the current market, and enterprises must turn to a customized production model oriented to orders. This model makes hydraulic cylinder products present the characteristics of single-piece small batches and diversity, which increases the complexity of production management. In this context, the production process of hydraulic cylinders has become more complicated, which puts higher requirements on the information construction and intelligent production planning decisions of enterprises.

[0003] At present, many small and medium-sized hydraulic cylinder companies are still at a low level in terms of information construction, and the formulation of production plans mainly depends on the experience of dispatchers and the production information recorded by the system. These companies have not yet realized intelligent production planning decisions, resulting in limited production efficiency. In order to meet this challenge, some companies have begun to try to integrate production planning with workshop scheduling, in order to formulate a more scientific and efficient production plan by comprehensively considering factors such as orders, production, outsourcing, storage, transportation and market. This integrated approach aims to improve the processing efficiency of enterprises and provide new ideas for the formulation of production plans.

[0004] Although the method of integrating production planning and workshop scheduling has advantages in theory, there are still some problems in practical application. First, existing research often does not consider comprehensively when establishing integrated models, has a single optimization goal, and fails to fully cover all key factors in the production process. Secondly, the search ability of the algorithm used is limited, and there is a lack of effective diversity maintenance mechanism in the search process. It is easy to fall into the local optimal solution and cannot find the global optimal or near-optimal solution. In addition, the direct use of existing algorithms may not be able to dynamically adjust the search strategy according to changes in the actual production process, resulting in the algorithm's poor adaptability when facing a complex and changing production environment. This makes it difficult for existing technologies to provide completely feasible and efficient solutions when solving the production planning and scheduling problems faced by hydraulic cylinder manufacturers. Therefore, how to improve the existing technology to make it more in line with actual production needs is a topic that hydraulic cylinder manufacturers need to explore further. Summary of the invention

[0005] The present invention proposes an integrated optimization method for production planning and workshop scheduling of a hydraulic cylinder manufacturing enterprise, which solves the problems of single optimization target and poor adaptability of existing integrated optimization methods.

[0006] In order to solve the above technical problems, the present invention provides a method for integrating production planning and workshop scheduling of a hydraulic cylinder manufacturing enterprise, comprising the following steps:

[0007] A multi-objective optimization function is established with minimizing processing time as the main objective and minimizing processing cost as the secondary objective, and constraints of the multi-objective optimization function are set; an improved particle swarm algorithm is used to solve the multi-objective optimization function, and the improved particle swarm algorithm uses a double-layer coding to represent the position of particles, the first layer of the double-layer coding is the processing sequence of each workpiece, and the second layer is the processing equipment for each processing sequence; the improved particle swarm algorithm iteratively optimizes the position of particles, and outputs a solution that meets the completion time and minimizes the processing cost as the optimal production plan and workshop scheduling solution.

[0008] Preferably, the improved particle swarm algorithm iteratively optimizes the positions of particles and outputs a solution that meets the completion time and minimizes the processing cost as the optimal production plan and workshop scheduling solution, including the following steps:

[0009] Step S1: Initialize the particle swarm and encode each particle using a double-layer encoding structure;

[0010] Step S2: Calculate the fitness of each particle, update the individual optimal value of each particle and the group optimal value of the particle swarm;

[0011] Step S3: Perform crossover mutation on particles and update the speed and position of particles;

[0012] Step S4: Determine whether the current number of iterations reaches the set maximum number of iterations. If not, dynamically update the inertia weight and learning factor of the particle swarm algorithm according to the current number of iterations, randomly distribute the particles that exceed the boundary of the search space, and return to step S2. Otherwise, execute step S5.

[0013] Step S5: Get the workshop scheduling plan, allocate the production plan according to the workshop scheduling plan, and determine whether the completion time of the current production plan meets the set constraints. If so, output the current plan as the optimal production plan and workshop scheduling plan. If not, return to step S1 to generate a new workshop scheduling plan.

[0014] Preferably, performing crossover mutation on particles in step S3 comprises the following steps:

[0015] Step S31: Randomly select two particles as parents to participate in the crossover operation;

[0016] Step S32: randomly select several genes in the first parent generation, and copy the selected genes to the same positions in the offspring generation;

[0017] Step S33: Copy the genes that are missing from the offspring in the second parent generation to the offspring in order of position.

[0018] Preferably, the expression for updating the inertia weight in step S4 is:

[0019]

[0020] Where ω is the inertia weight; ω min is the minimum value of inertia weight; ω max is the maximum value of the inertia weight; t is the current iteration number; T max is the maximum number of iterations; τ is the exponential coefficient.

[0021] Preferably, the expression for updating the dynamic learning factor in step S4 is:

[0022]

[0023] In the formula, are the individual learning factor and social learning factor at the tth iteration respectively; are the initial and final values ​​of the individual learning factor, respectively; are the initial value and the final value of the social learning factor respectively; t is the current iteration number; T max is the maximum number of iterations.

[0024] Preferably, the expression for randomly distributing particles beyond the boundary of the search space in step S4 is:

[0025]

[0026] In the formula, u b and l b are the upper and lower boundaries of the search area; λ is An arbitrary random number between An arbitrary random number between ; is the value of particle i at the tth iteration.

[0027] Preferably, the expression of the secondary objective is:

[0028]

[0029] Where I is the total number of parts that need to be processed; n is the number of parts that need to be processed by factory P1 as specified in the production plan, and its initial value is I; c ikr For process O ik On the rth machine M r Processing cost on ikr is a decision variable, if process O ik By machine M r1 if processing is done, 0 otherwise; i,k For workpiece J i The kth process, {O i,k |k=1,2,…,K}, K is workpiece J i The total number of processes; M is the processing machine set of factory P1, M = {M r |r=1,2,…,R}, R is the total number of processing machines in factory P1; w1 is the unit price of processing a single part in factory P2; w2 is the unit price required to transport a single part from factory P2 to factory P1.

[0030] Preferably, the expression for allocating the number of processed parts in factory P1 and factory P2 is:

[0031]

[0032] In the above formula, q1 is the number of workpieces that need to be processed by factory P1; q2 is the number of workpieces that need to be processed by factory P2; I is the total number of workpieces that need to be processed; I k is the number of workpieces currently being processed; T is the maximum processing time; f2 is the objective function for minimizing the processing time.

[0033] Preferably, the main objective is expressed as:

[0034] f2=max{min{max{Et ikr}},EP2};

[0035] In the formula, Et ikr For process O ik On the rth machine M r EP2 is the time taken for factory P2 to process all parts and transport them to factory P1.

[0036] Preferably, the constraint condition is expressed as:

[0037] EP2<min{max{Et ikr}};

[0038] E ikr =max{Et i(k-1)r , Et ikr－hl}+t ikr σ;

[0039] St ikr =Et i(k-1)r ;

[0040] St ikr ≥St i(k-1)r ;

[0041] St ikr -Stikr-hl ≥t hlr ;

[0042] min{max{Et ikr}}≤T;

[0043] In the above formula, EP2 is the time it takes for factory P2 to process all parts and transport them to factory P1; Et ikr For process O ik On the rth machine M r Completion time on Et ikr－hl For process O ik On the rth machine M r The previous process O hl Completion time; t ikr For process O ik The processing time; σ is the dynamic adjustment parameter; St ikr For process O ik On the rth machine M r Processing time on St ikr-hl For process O ik On the rth machine M r The previous process O hl The start time of hlr For the immediately preceding process O hl processing time; T is the set maximum processing time.

[0044] The benefits of the present invention include at least:

[0045] 1. Construct an integrated optimization model for production planning and workshop scheduling, with minimizing processing time as the primary goal and minimizing processing cost as the secondary goal, while taking into account the constraints of the planning stage and scheduling processing;

[0046] 2. By adopting double-layer coding in the particle swarm algorithm, the processing sequence and equipment allocation of the workpiece are taken into consideration at the same time, which improves the representation ability and optimization space of the algorithm; by introducing the operator crossover method of the genetic algorithm and re-randomly distributing the particles beyond the boundary, the diversity and optimization ability of the population are enhanced; the learning factor and inertia weight in the particle swarm algorithm are dynamically adjusted according to the number of iterations, which improves the adaptability and convergence speed of the algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a flow chart of a method according to an embodiment of the present invention;

[0048] Figure 2 Schematic diagram of two integration strategies and core conflicts in an embodiment of the present invention;

[0049] Figure 3 is a schematic diagram of an integrated model in an embodiment of the present invention;

[0050] Figure 4 A schematic diagram of double-layer coding according to an embodiment of the present invention;

[0051] Figure 5 A schematic diagram of a PBX cross connection in an embodiment of the present invention;

[0052] Figure 6 is a flow chart of an improved particle swarm algorithm in an embodiment of the present invention;

[0053] Figure 7 A schematic diagram of a convergence curve of preliminary scheduling according to an embodiment of the present invention;

[0054] Figure 8 A Gantt chart for preliminary scheduling of an embodiment of the present invention;

[0055] Fig. 9 A Gantt chart for the second scheduling of an embodiment of the present invention;

[0056] Fig.10 This is a Gantt chart of the third scheduling of an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the protection scope of the present invention.

[0058] like Figure 1 As shown, an embodiment of the present invention provides a method for integrating production planning and workshop scheduling of a hydraulic cylinder manufacturing enterprise, comprising the following steps:

[0059] Step S1: Analyze the processing characteristics of hydraulic cylinder manufacturing enterprises, such as small batches of single pieces, multiple varieties, and high degree of product customization. Use the integrated modeling method as the optimization solution for integrating its production plan with workshop scheduling. Adopt an integrated strategy of micro-planning to refine macro-planning, and establish a collaborative optimization theoretical model.

[0060] Specifically, the integrated modeling method has significant research value for the integrated optimization of production planning and workshop scheduling in hydraulic cylinder manufacturing enterprises. This method integrates the constraints of the planning layer and the scheduling layer to build a unified large model, which effectively solves key problems such as capacity balance, resource allocation, machine utilization and processing efficiency.

[0061] In actual production, scheduling usually aims to complete product production tasks and focuses on short-term production activities, while planning focuses on the long-term development of the enterprise and considers medium- and long-term tasks. Figure 2 As shown, due to the different time ranges considered by the two, the traditional processing method easily leads to ignoring the benefit goals at the planning level during the scheduling process, or ignoring the pursuit of minimum processing time at the scheduling level when making plans. In order to overcome this problem, an embodiment of the present invention proposes an integrated strategy that combines micro planning with macro planning.

[0062] Under this strategy, micro-planning is responsible for refining the production tasks of each cycle in the macro-plan and arranging machines for processing, so as to achieve accurate scheduling of specific processes without affecting inventory and other resources. This not only enables medium- and long-term planning to obtain positive feedback from micro-planning, thereby making more detailed and favorable decisions, but also ensures that the current processing tasks can be successfully completed based on the most advanced scheduling algorithms, resulting in better solutions. Through this integrated approach, production planning and workshop scheduling are coordinated, improving the company's production efficiency and market competitiveness.

[0063] For enterprises, on-time delivery of orders is the foundation of their long-term development. Adding micro-plans to the integrated model in the form of decision variables can solve the problem of difficulty in mutual conversion and coordination when integrating production plans and scheduling. The original tasks of the macro-cycle model are still defined as the tasks of the integrated model. While keeping the scheduling tasks unchanged, the ideas of processors and decision makers are kept consistent to obtain the final solution. The structure of the integrated model of production planning and workshop scheduling is as follows: Figure 3 shown.

[0064] By integrating production planning and workshop scheduling into a unified model, it is possible to consider long-term planning goals, such as resource allocation and capacity balance, and short-term scheduling goals, such as machine utilization and processing efficiency, under one framework. Micro-planning is introduced into the model to refine the macro-plan, ensuring that the production plan at the macro level can be specific to the processing tasks of the machine, and at the same time, the scheduling decisions at the micro level can support the production goals at the macro level.

[0065] Step S2: According to the processing mode in which the main processing factory cannot complete the processing target and allocates the processing to the subsidiary factory during the production process of the hydraulic cylinder enterprise, a series of assumptions are made on the integrated model of production planning and workshop scheduling of the hydraulic cylinder manufacturing enterprise, including:

[0066] a. The company has two factories for processing and manufacturing. Factory P1 is the main processing factory, the preferred factory for parts processing, equipped with an assembly workshop and a finished product warehouse. Factory P2 is an auxiliary processing factory, which only has parts processing functions but no assembly functions. It is far away from the company and is only activated when the production capacity of Factory P1 is insufficient.

[0067] b. The production price of factory P2 is much higher than that of the main factory. Therefore, only when the production capacity of factory P1 reaches its maximum will it enter factory P2 for processing.

[0068] c. The production capacity of factory P2 is infinite by default, and the corresponding processing time plus transportation time is much less than the time consumed by factory P1.

[0069] d. All machines are available at the initial time, that is, time zero, but all workpieces can only be processed at the arrival time.

[0070] e. Each workpiece can only be processed on one machine at a time, and this machine can only process this one workpiece at this time, and each operation cannot be interrupted in the middle.

[0071] f. There is no order constraint between the processes of different workpieces, and different workpieces have the same processing priority.

[0072] g. When the machine processes adjacent steps of the same workpiece, the workpiece movement time can be ignored. The preparation time and movement time are included in the processing time and can be ignored.

[0073] h. The processing cost of the workpiece is only related to the processing machine, has nothing to do with the processing procedure, and does not take labor costs into account.

[0074] i. The total processing time T specified for this batch of parts is fixed, and the workpieces must be delivered on time.

[0075] Step S3: For the problem of integrating production planning and workshop scheduling of hydraulic cylinder manufacturing enterprises, a multi-objective function is established with minimizing processing time as the main objective and minimizing processing cost as the auxiliary objective.

[0076] For the convenience of description, the symbols used in the embodiments of the present invention are as follows:

[0077] J is the set of parts to be processed, J={J i |i=1,2,…,I};

[0078] M is the set of workshop processing machines in factory P1, M={M r |r=1,2,…,R};

[0079] N is the number of parts that can be processed in the workshop of factory P1;

[0080] O ik For workpiece J i The kth process, {O ik |k=1,2,…,K};

[0081] I is the total number of parts that need to be processed;

[0082] n is the number of parts that need to be processed by factory P1 as specified in the production plan, and its initial value is I;

[0083] w1 is the processing unit price of a single part by factory P2;

[0084] w2 is the unit price required to transport a single part from factory P2 to factory P1;

[0085] c ikr For process O ik On the rth machine M r Processing costs on

[0086] E ikr For process O ik On the rth machine M r Completion time on

[0087] E ikr-hl For process O ik On the rth machine M r The previous process O hl completion time;

[0088] St ikr-hl For process O ik On the rth machine M r The previous process O hl The start time of

[0089] St ikr For process O ik On the rth machine M r Processing time on

[0090] x ikr is the decision variable, process O ik By machine M r 1 if processing is done, 0 otherwise;

[0091] EP2 is the time taken by factory P2 to process all parts and transport them to factory P1;

[0092] T is the maximum processing time specified for this batch of parts;

[0093] σ is a dynamic adjustment parameter, which takes a random number between 1 and 1.2.

[0094] The expression of the function with the minimum processing cost as the goal is:

[0095]

[0096] The expression of the function with the minimum processing time as the goal is:

[0097] f2=max{min{max{Etikr}},EP2}.

[0098] Establishing the upper-level goal of minimizing the completion time and the lower-level goal of minimizing the processing cost makes it possible to simultaneously pursue improving production efficiency and reducing production costs during the optimization process.

[0099] Step S4: According to the actual production situation, constrain the minimum processing time objective function, the minimum processing cost objective function and some parameters in the integrated model of production planning and workshop scheduling, including:

[0100] (1) Processing time constraints of factory P2:

[0101] EP2<min{max{Et ikr}};

[0102] The above formula indicates that no matter how much the processing quantity is, the processing time of factory P2 is less than that of factory P1.

[0103] (2) The adjacent processes of the i-th workpiece are on the same machine M r Time constraints for processing:

[0104] St ikr ≥St i(k-1)r ;

[0105] St ikr =Et i(k-1)r ;

[0106] E ikr =max{Et i(k-1)r , Et ikr－hl}+t ikr ·σ.

[0107] (3) The difference between the start times of adjacent processes of the i-th workpiece is not less than the processing time of the previous process:

[0108] St ikr -St ikr-hl ≥t hlr .

[0109] (4) Maximum completion time constraint:

[0110] min{max{Et ikr}}≤T.

[0111] In the model, the parameters of production planning and workshop scheduling are constrained to ensure that the scheduling plan is not only optimal in theory, but also meets various restrictions and requirements in actual production.

[0112] Step S5: Using the improved particle swarm algorithm to solve the multi-objective function, outputting a solution that meets the completion time and minimizes the processing cost as the optimal production plan and workshop scheduling solution, including the following steps:

[0113] Step S51: Initialize the particle swarm and encode each particle using a double-layer encoding structure.

[0114] Specifically, the embodiment of the present invention adopts a double-layer structure based on workpiece coding and equipment coding to simultaneously describe the processing sequence information and processing equipment information of the workpiece. The first layer of coding is regarded as the workpiece code, in which the number of times the workpiece appears represents the process of the workpiece; the second layer is the machine code, and each machine code represents the index number of the optional equipment set for the process. Each double-layer coding structure corresponds to a scheduling plan, which represents the processing machine selected for each workpiece and each process.

[0115] like Figure 4 As shown in the figure, if the code of a particle is [1,2,2,1,3,2,3,1,3,2,2,1,3,3,1,2,1,2], then the workpiece code of the particle is [1,2,2,1,3,2,3,1,3], and the equipment code is [2,2,1,3,3,1,2,1,2]. The process code corresponding to the workpiece is {O 11 ,O 21 ,O 22 ,O 12 ,O 31 ,O 23 ,O 32 ,O 13 ,O 33}, according to the optional equipment set corresponding to each process, it can be determined that the equipment numbers corresponding to the equipment code are {2,2,1,3,3,1,2,1,2} in sequence.

[0116] Step S52: Calculate the fitness of each particle, and update the individual optimal value of each particle and the group optimal value of the particle swarm.

[0117] Step S53: Perform crossover mutation on particles and update the speed and position of particles.

[0118] Specifically, the embodiment of the present invention introduces a genetic algorithm into a particle swarm algorithm, performs a position-based PBX crossover on the process code, and improves the particle update method.

[0119] like Figure 5 As shown, several positions are randomly selected in the two parent chromosomes, and the positions may not be continuous. The genes at the selected positions on parent chromosome 1 are copied to the same positions of offspring 1, and then the missing genes in offspring 1 on parent chromosome 2 are filled in according to the position order. Another offspring is obtained in a similar way. The crossover steps of PBX are as follows:

[0120] Step S531: Randomly select two particles as parent chromosomes to participate in the crossover operation.

[0121] Step S532: Randomly select several genes in the first parent generation, and copy the selected genes to the same positions in the offspring generation.

[0122] Step S533: Select the genes with the same position as the first parent generation from the second parent generation, and copy the remaining genes to the offspring in order of position.

[0123] Step S54: Determine whether the current number of iterations reaches the set maximum number of iterations. If not, dynamically update the inertia weight and learning factor of the particle swarm algorithm according to the current number of iterations, randomly distribute the particles that exceed the boundary of the search space, and return to step S52. Otherwise, execute step S55.

[0124] Specifically, an inertia weight power function adaptive adjustment method is proposed and a learning factor update strategy is set. By dynamically adjusting the initial parameters of the particle swarm algorithm, such as the inertia weight and learning factor, the overall performance of the algorithm can be improved. The dynamic adjustment of the initial parameters of the particle swarm algorithm includes:

[0125] (1) Improvement of inertia weight

[0126] In the standard particle swarm algorithm, the inertia weight is introduced, and its initial value is designed to affect the convergence speed and convergence quality of the particle swarm algorithm. In the process of calculation and solution, the traditional particle swarm has the disadvantages of low stability, easy to fall into local optimum and slow convergence. According to multiple simulation experiments, it is found that the inertia weight should not be a fixed value, but should increase with the number of iterations. In the early stage, a larger inertia weight is required to enhance the global optimization ability, and in the later stage, a smaller inertia weight is required, and the local optimization ability is stronger. Therefore, an embodiment of the present invention proposes an inertia weight power function adaptive adjustment method, so that the inertia weight is gradually reduced during the iteration process. Compared with the traditional linear convergence speed, the convergence speed of this method is faster, so that the inertia weight value reaches a smaller value faster, that is, it enters the fine search faster. The update expression of the inertia weight is:

[0127]

[0128] In the formula, ω min is the minimum value of the inertia weight, that is, the final value, and takes ω min =0.4;ω max is the maximum value of the inertia weight, that is, the initial value, and ω max =0.96; t is the current iteration number; T max is the maximum number of iterations; τ is the exponential coefficient, which is 0.5.

[0129] (2) Improvement of learning factors

[0130] The learning factor also affects the performance of the algorithm at all times. As an important parameter in the operation of the algorithm, the learning factor c1 is equivalent to the self-cognition of the particle, which is a reflection of the particle's ability to self-learn and summarize; the learning factor c2 is equivalent to the social cognition of the individual, which is a reflection of the individual's ability to learn from particles with better performance. At the beginning, the particles should have a strong self-cognition ability in order to interconnect with the group, so a larger c1 value and a smaller c2 value can be set. In the later stage, the particles can already express themselves well and need a stronger ability to obtain social information, so a smaller c1 value and a larger c2 value can be set. According to this communication characteristic of the social group specified in the algorithm, the value of the learning factor can be changed during the iteration process to enhance the particle's ability to converge to the global optimal point. As the number of iterations increases, the learning factor change strategy set by the present invention is as follows, so that c1 decreases linearly and c2 increases linearly.

[0131]

[0132] are the initial value and the final value of the individual learning factor, respectively. are the initial value and the final value of the social learning factor respectively.

[0133] The expression for the random distribution of particles beyond the boundary of the search space is:

[0134]

[0135] In the formula, u b and l b are the upper and lower boundaries of the search area; λ is An arbitrary random number between An arbitrary random number between ; is the value of particle i at the tth iteration.

[0136] By dynamically adjusting the inertia weight and learning factor of the particle swarm algorithm and introducing the crossover and mutation operations of the genetic algorithm, the search efficiency and global optimization ability of the algorithm in complex models are improved.

[0137] Step S55: Get the workshop scheduling plan, allocate the production plan according to the workshop scheduling plan, and determine whether the completion time of the current production plan meets the set constraints. If so, output the current plan as the optimal production plan and workshop scheduling plan. If not, return to step S51 to generate a new workshop scheduling plan.

[0138] Specifically, the first run of the particle swarm complete algorithm can only be regarded as a preliminary scheduling, and an optimal processing plan is obtained under the preliminary production plan. The results of the preliminary scheduling are to minimize the maximum completion time and minimize the processing cost. For minimizing the processing cost, a judgment parameter χ is introduced, and the following comparison needs to be made:

[0139]

[0140] Among them, C max is the maximum allowable processing cost.

[0141] If χ is equal to 1, it means that the scheduling plan is reasonable and will continue to be implemented. If χ is equal to 0, it means that the entire contract signing is unreasonable and the procedure will be exited. At the company's operational level, the contract will be returned to the sales department for renegotiation.

[0142] To minimize the maximum completion time, the processing parts can be allocated and the number of parts to be processed in the two factories can be specified:

[0143]

[0144] Among them, q1 is the number of workpieces that factory P1 needs to process; q2 is the number of workpieces that factory P2 needs to process; I k is the processing quantity in the current cycle; I is the total quantity that needs to be processed in the contract signing sequence; δ is the judgment factor. δ equals 1, which means that the processing time requirement is met, and the plan and scheduling results are output. δ equals 0, which means that the production plan is unreasonable, and the processing plan is reallocated, and the scheduling plan of factory P1 is re-formulated.

[0145] After the initial scheduling, the production plan is adjusted according to the results, and then scheduled again until a solution is found that meets both time requirements and optimizes costs, reflecting the dynamic coordination and iterative optimization between planning and scheduling.

[0146] Embodiment 1

[0147] The example data of a hydraulic cylinder manufacturing enterprise is collected. According to the production needs of the enterprise, the latest batch of processing cases are selected and simplified into a 10×8 flexible job shop scheduling problem. The specific simulation data is shown in Table 1. The processing case can be simplified into a flexible job shop scheduling problem with 10 workpieces, 8 equipment, and 35 processes. The experimental environment is on the Windows 10 platform, and its hardware parameters are: CPU 11th Gen Intel(R)Core(TM)i5-1135G7@2.40GHz2.42GHz, programming environment: MATLAB R2019b. The number of particles P N =180, number of iterations F N =500, see the improved particle swarm algorithm for details Figure 6 .

[0148] Table 1 Simulation example data

[0149]

[0150]

[0151] The improved particle swarm algorithm is used to solve the example data. In the preliminary scheduling plan, the optimization process of the algorithm is as follows: Figure 7 The corresponding Gantt chart is shown in Figure 8 As shown. According to the results of the preliminary scheduling, the minimum completion time that can be obtained has exceeded the specified processing time, which means that the task cannot be completed by factory P1 alone, and some parts need to be handed over to factory P2 for processing. The entire program also enters the subsequent production plan allocation function, which will allocate the last part to be processed to factory P2, and then reschedule the remaining parts to obtain a new scheduling plan for factory P1, as shown in Fig. 9 shown.

[0152] If the new production plan and scheduling scheme meets the processing time requirements and its minimum processing cost is less than the maximum allowed processing cost, then the production plan and scheduling scheme are output, which is the optimal solution for the current order.

[0153] The algorithm program will not be exited until the results obtained from multiple scheduling meet the completion time requirements. Fig.10 This is the result of the subsequent scheduling plan, indicating that the production plan should leave 8 workpieces in factory P1 for processing and delegate 2 workpieces to factory P2 for processing in order to complete the processing target. The last scheduling Gantt chart is also the optimal scheduling plan for the main factory processing. Through such repeated scheduling, processing plans with different numbers of parts can be designed, and the one with the shortest processing time under the premise of meeting the completion time requirements can be selected.

[0154] The present invention proposes an integrated optimization method for production planning and workshop scheduling of a hydraulic cylinder manufacturing enterprise. The integrated modeling method is used to successfully build a collaborative optimization model, and an improved intelligent optimization algorithm is used to optimize the production plan and scheduling scheme that meet the actual production needs. The scheduling process can then react to the production plan to make the plan more efficient and high-quality, thereby increasing the enterprise's production efficiency and reducing costs.

[0155] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. Only the preferred embodiments of the present invention are expressed. The description is more specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. As long as there is no contradiction in the combination of these technical features, they should be considered as the scope recorded in this specification.

[0156] It should be noted that, for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these modifications and improvements all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A production planning and workshop scheduling integrated optimization method for a hydraulic cylinder manufacturing enterprise, characterized in that: The following steps are involved: A multi-objective optimization function is established with minimizing processing time as the main goal and minimizing processing cost as the secondary goal, and constraints of the multi-objective optimization function are set; an improved particle swarm algorithm is used to solve the multi-objective optimization function, and the improved particle swarm algorithm uses a double-layer coding to represent the position of particles, the first layer of the double-layer coding is the processing sequence of each workpiece, and the second layer is the processing equipment for each processing sequence; the improved particle swarm algorithm iteratively optimizes the position of particles, and outputs a solution that meets the completion time and minimizes the processing cost as the optimal production plan and workshop scheduling solution.

2. The integrated optimization method for production planning and workshop scheduling of a hydraulic cylinder manufacturing enterprise according to claim 1 is characterized in that: The improved particle swarm algorithm iteratively optimizes the positions of particles and outputs a solution that meets the completion time and minimizes the processing cost as the optimal production plan and workshop scheduling solution, including the following steps: Step S1: Initialize the particle swarm and encode each particle using a double-layer encoding structure; Step S2: Calculate the fitness of each particle, update the individual optimal value of each particle and the group optimal value of the particle swarm; Step S3: Perform crossover mutation on particles and update the speed and position of particles; Step S4: Determine whether the current number of iterations reaches the set maximum number of iterations. If not, dynamically update the inertia weight and learning factor of the particle swarm algorithm according to the current number of iterations, randomly distribute the particles that exceed the boundary of the search space, and return to step S2. Otherwise, execute step S5. Step S5: Get the workshop scheduling plan, allocate the production plan according to the workshop scheduling plan, and determine whether the completion time of the current production plan meets the set constraints. If so, output the current plan as the optimal production plan and workshop scheduling plan. If not, return to step S1 to generate a new workshop scheduling plan.

3. The integrated optimization method for production planning and workshop scheduling of a hydraulic cylinder manufacturing enterprise according to claim 2 is characterized in that: The crossover mutation of particles in step S3 includes the following steps: Step S31: Randomly select two particles as parents to participate in the crossover operation; Step S32: randomly select several genes in the first parent generation, and copy the selected genes to the same positions in the offspring generation; Step S33: Copy the genes that are missing from the offspring in the second parent generation to the offspring in order of position.

4. The integrated optimization method for production planning and workshop scheduling of a hydraulic cylinder manufacturing enterprise according to claim 2 is characterized in that: The expression for updating the inertia weight in step S4 is: Where ω is the inertia weight; ω min is the minimum value of inertia weight; ω max is the maximum value of the inertia weight; t is the current iteration number; T max is the maximum number of iterations; τ is the exponential coefficient.

5. The integrated optimization method for production planning and workshop scheduling of a hydraulic cylinder manufacturing enterprise according to claim 2 is characterized in that: The expression for updating the dynamic learning factor in step S4 is: In the formula, are the individual learning factor and social learning factor at the tth iteration respectively; are the initial and final values ​​of the individual learning factor, respectively; are the initial value and the final value of the social learning factor respectively; t is the current iteration number; T max is the maximum number of iterations.

6. The integrated optimization method for production planning and workshop scheduling of a hydraulic cylinder manufacturing enterprise according to claim 2 is characterized in that: The expression for randomly distributing particles beyond the boundary of the search space in step S4 is: In the formula, u b and l b are the upper and lower boundaries of the search area; λ is An arbitrary random number between An arbitrary random number between ; is the value of particle i at the tth iteration.

7. The integrated optimization method for production planning and workshop scheduling of a hydraulic cylinder manufacturing enterprise according to claim 1 is characterized in that: The expression of the secondary objective is: In the formula, I is the total number of parts that need to be processed; n is the number of parts that need to be processed by factory P1 as specified in the production plan, and its initial value is I; c ikr For process O ik On the rth machine M r Processing cost on ikr is a decision variable, if process O ik By machine M r 1 if processing is done, 0 otherwise; i,k For workpiece J i The kth process, {O i,k |k=1,2,…,K}, K is workpiece J i The total number of processes; M is the processing machine set of factory P1, M = {M r |r=1,2,...,R}, R is the total number of processing machines in factory P1; w1 is the unit price of processing a single part in factory P2; w2 is the unit price required to transport a single part from factory P2 to factory P1.

8. The integrated optimization method for production planning and workshop scheduling of a hydraulic cylinder manufacturing enterprise according to claim 7 is characterized in that: The expression for allocating the number of processed parts in factory P1 and factory P2 is: In the above formula, q1 is the number of workpieces that need to be processed by factory P1; q2 is the number of workpieces that need to be processed by factory P2; I is the total number of workpieces that need to be processed; I k is the number of workpieces currently being processed; T is the maximum processing time; f2 is the objective function for minimizing the processing time.

9. The integrated optimization method for production planning and workshop scheduling of a hydraulic cylinder manufacturing enterprise according to claim 1 is characterized by: The main objective is expressed as: f2=max{min{max{And ikr }},EP2}; In the formula, Et ikr For process O ik On the rth machine M r EP2 is the time taken for factory P2 to process all parts and transport them to factory P1.

10. The integrated optimization method for production planning and workshop scheduling of a hydraulic cylinder manufacturing enterprise according to claim 1, characterized in that: The constraint condition is expressed as: EP2<min{max{And ikr }}; And ikr =max{And i(k-1)r ,And ikr－hl }+t ikr ·σ; St ikr =And i(k-1)r ; St ikr ≥St i(k-1)r ; St ikr -St ikr-hl ≥t hlr ; min{max{And ikr }}≤T; In the above formula, EP2 is the time it takes for factory P2 to process all parts and transport them to factory P1; Et ikr For process O ik On the rth machine M r Completion time on Et ikr－hl For process O ik On the rth machine M r The previous process O hl Completion time; t ikr For process O ik The processing time; σ is the dynamic adjustment parameter; St ikr For process O ik On the rth machine M r Processing time on St ikr-hl For process O ik On the rth machine M r The previous process O hl The start time of hlr For the immediately preceding process O hl processing time; T is the set maximum processing time.

Images