Production and transportation combined scheduling method, electronic device, storage medium and program product

Abstract

This invention provides a production handling joint scheduling method, electronic device, storage medium, and program product, relating to the field of workshop scheduling technology. The method includes: obtaining basic parameters for joint scheduling; constructing a production handling collaborative optimization model based on a master-slave game theory-based two-layer collaborative optimization, the production handling collaborative optimization model including an upper-layer model and a lower-layer model; using Gaussian processes to quantitatively model the instantaneous speed during overhead crane handling, and introducing chance-constrained programming theory to transform the random instantaneous speed constraint into a deterministic linear constraint satisfying a pre-set confidence level; based on the basic parameters for joint scheduling, using a reinforcement learning-guided global-local neighborhood search and tabu search algorithm, performing a nested collaborative solution on the production handling collaborative optimization model to obtain the target production handling joint scheduling scheme. This invention reduces equipment waiting time while also considering overhead crane capacity consumption, effectively improving the overall collaborative operation efficiency of the workshop.

Description

Technical Field

[0001] This invention relates to the field of workshop scheduling technology, specifically to a production handling joint scheduling method, electronic equipment, storage medium, and program product. Background Technology

[0002] In the manufacturing of glass panels for microelectronic displays, the processing steps are complex and the process is highly continuous, placing extremely stringent requirements on the timeliness of material handling in production equipment. Overhead Shuttles (OHS), as the core equipment for high-altitude, three-dimensional material handling, directly determine the overall output and quality of the display production line through their handling efficiency.

[0003] However, existing production and handling optimization methods have the following limitations in practical applications: (1) Traditional scheduling studies usually treat production scheduling on equipment and overhead crane handling assignment as independent sub-problems. However, in actual display production lines, the dynamic handling demand caused by production and the actual carrying capacity of the overhead crane system (idle waiting, congestion, etc.) are strongly coupled in both directions. Separate optimization can easily lead to theoretical solutions not being able to be executed on time when the actual overhead crane carrying capacity is limited, resulting in panel handling delays.

[0004] (2) Multiple overhead cranes move freely on a shared closed-loop track. Due to obstacle avoidance and safe distance restrictions, the actual speed of the cranes fluctuates randomly. Existing deterministic scheduling models usually set the crane speed as a constant, which cannot accurately simulate this dynamic logistics interference, resulting in poor scheduling robustness.

[0005] (3) The joint scheduling of production and transportation in display manufacturing is a typical master-slave game problem. Traditional single-layer optimization algorithms cannot handle the interaction of two-layer models, where upper-layer decision-making is issued and lower-layer state feedback is provided. In addition, when faced with a two-layer, multi-level decision space, conventional heuristic algorithms are prone to getting stuck in local optima due to fixed search operators and lack of adaptive coordination mechanisms, and cannot achieve efficient collaborative optimization of upper and lower layer solutions.

[0006] Therefore, how to construct a game theory model that can characterize the closed feedback relationship between production and transportation, quantify the random speed disturbances in the transportation process to improve the anti-interference ability of the scheme, and design an efficient solution algorithm that can adaptively find the best solution are the problems that urgently need to be solved in the field of intelligent scheduling in microelectronics manufacturing workshops. Summary of the Invention

[0007] This invention provides a production handling joint scheduling method, electronic device, storage medium, and program product.

[0008] According to one aspect of the present invention, a production handling joint scheduling method is provided, comprising: acquiring basic parameters for joint scheduling; constructing a production handling collaborative optimization model based on a master-slave game-theoretic two-layer collaborative optimization, wherein the production handling collaborative optimization model includes an upper-layer model and a lower-layer model, the upper-layer model optimizing workpiece sequencing decisions and equipment allocation decisions with the objective of minimizing the maximum completion time, wherein the workpiece sequencing decisions are used to determine the execution order of the corresponding processes for each workpiece, and the equipment allocation decisions are used to determine the equipment used to execute each process; the lower-layer model optimizing crane allocation decisions with the objective of minimizing the total idle travel time of each crane, wherein the crane allocation decisions are used to determine the movement of workpieces after the completion of a process to the next process. The target location of the overhead crane; the instantaneous speed during the crane's transport process is quantitatively modeled using a Gaussian process, and chance-constrained programming theory is introduced to transform the random instantaneous speed constraint into a deterministic linear constraint that meets a pre-set confidence level; based on the joint scheduling basic parameters, the production transport collaborative optimization model is solved in a nested manner using a global-local neighborhood search and tabu search algorithm guided by reinforcement learning, resulting in the target production transport joint scheduling scheme; in the solution process, during the global search phase, the selection probability of the neighborhood search operator is adaptively adjusted based on a reinforcement learning mechanism; during the local search phase, a tabu search strategy is executed on the critical path based on a priority weight mechanism.

[0009] According to at least one embodiment of the production handling joint scheduling method of the present invention, the basic parameters of the joint scheduling include: the equipment layout topology of the microelectronics manufacturing workshop, the equipment spacing, the standard operating speed of the overhead crane, the handling time of the overhead crane between each piece of equipment, the number of processes contained in the workpiece, the standard processing time of each process, and the equipment that can be used for each process.

[0010] The production handling joint scheduling method according to at least one embodiment of the present invention is represented by the following upper-level model: ,

[0011] The constraints of the upper-level model include:

[0012]

[0013]

[0014]

[0015]

[0016]

[0017] in, This represents the objective function of the upper-level model; Indicates workpiece i Completion time; Indicates equipment k For the process O i(pi) The end time of the processing task; O i(pi) Indicates workpiece i The last process; pi represents the maximum process index of workpiece i; i This represents the workpiece index; n represents the total number of workpieces; min represents the minimum value; max represents the maximum value; Indicate process O ij Is it in the device? k The markings on the upper processing area; k and k’ All represent device indexes; m represents the total number of devices; Indicates equipment k For the process O ij The end time of the processing task; Indicates equipment k For the process O pq The start time of the processing task; Represents the maximum constant parameter; Indicate process O ij With process O pq In the equipment k The order of processing is indicated; Indicates equipment k For the process O ij The start time of the processing task; Indicates overhead crane v For the process O ij The end time of the transportation mission; Indicates equipment k For the process O ij The end time of the processing task; Indicates equipment k For the process O ij The processing time for the processing task; Indicates that the overhead crane is from the equipment k Move to device k’ Time; Indicates equipment k With equipment k’ The distance between them; This indicates the normal operating speed of the overhead crane under undisturbed conditions. Indicates equipment k’ For the process O i(j-1) The start time of the processing task; Indicate process O i(j-1) Is it in the device? k’ The markings on the upper processing area; O i(j-1) Indicates workpiece i The j -1 process; O ij Indicates workpiece i The j One process; v Indicates the overhead crane index; j Indicates the process index.

[0018] The production handling joint scheduling method according to at least one embodiment of the present invention is represented by the following lower-level model:

[0019] The constraints on the objective function of the lower-level model include:

[0020]

[0021]

[0022]

[0023]

[0024]

[0025]

[0026]

[0027] in, This represents the objective function of the lower-level model; Represents the mathematical expectation operator; Indicate process O ij With process O pq By overhead crane v The order of transportation; Indicates overhead crane v For the process O pqThe end time of the empty transport phase; Indicates overhead crane v For the process O pq The start time of the empty transport phase; h represents the total number of overhead cranes; Indicate process O ij Is it by overhead crane? v Transportation markings; p Indicates the workpiece index; q Indicates the process index; O pq Indicates workpiece p The q One process; Indicates overhead crane v For the process O ij The start time of the empty transport phase; Indicates overhead crane v For the process O pq The end time of the load transportation phase; Indicates overhead crane v For the process O pq The end time of the empty transport phase; Indicates overhead crane v For the process O pq The start time of the empty transport phase; Indicate process O ij Location of the unloading point; Indicate process O ij Location of the loading point; This represents the distance calculation operator; v t Indicates that the overhead crane is t Instantaneous velocity at a given moment; Represents a Gaussian process; ~ indicates that it follows a distribution; This represents the mean of instantaneous velocity fluctuations during a Gaussian process; The variance of instantaneous velocity; l A length scale representing the time correlation of transport volume; and Both represent time. Indicates overhead crane v For the process O ij The start time of the load transportation phase; Indicates overhead crane v For the process O ijThe end time of the empty transport phase; Indicates overhead crane v’ For the process O i(j-1) The end time of the transportation mission.

[0028] The production handling joint scheduling method according to at least one embodiment of the present invention transforms the random instantaneous speed constraint into a deterministic linear constraint that satisfies a preset confidence level, including: Will Convert to ,in, v eff Indicates effective speed. This represents the ηth quantile of the standard normal distribution, where η represents the confidence level; and uses... v eff Replacement Formula and In v t .

[0029] According to at least one embodiment of the present invention, a production and transportation joint scheduling method performs nested collaborative solution of the production and transportation collaborative optimization model, comprising: generating an initial population using a hybrid initialization strategy and passing it to the lower-level model; the lower-level model receiving parameters passed from the upper-level model, solving the objective function of the lower-level model, and feeding back the solved crane allocation decision-related parameters to the upper-level model; the upper-level model receiving the parameters fed back from the lower-level model, calculating the objective function of the upper-level model, and then performing a selection operation; the upper-level model performing a global search phase, adaptively adjusting the probability combination of neighborhood search operators based on a reinforcement learning mechanism; the upper-level model performing a local search phase, executing a tabu search strategy on the critical path based on a priority weight mechanism to generate a new solution; the upper-level model updating the intelligent cells according to the generated new solution and passing it to the lower-level model; determining whether the maximum number of iterations has been reached, if not, continuing iteration, if reached, and outputting the optimal production and transportation joint scheduling scheme as the target production and transportation joint scheduling scheme.

[0030] According to another aspect of the present invention, an electronic device is provided, comprising: a memory storing execution instructions; and a processor executing the execution instructions stored in the memory, such that the processor performs a production handling joint scheduling method according to any embodiment of the present invention.

[0031] According to another aspect of the present invention, a readable storage medium is provided, wherein the readable storage medium stores execution instructions, which, when executed by a processor, are used to implement the production handling joint scheduling method of any embodiment of the present invention.

[0032] According to another aspect of the present invention, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the production and material handling joint scheduling method of any embodiment of the present invention.

[0033] The present invention has the following beneficial effects: (1) Construct a master-slave game two-layer optimization framework to realize the nested interaction between the upper-level processing scheduling and the lower-level handling and dispatching schemes. While reducing equipment waiting time, it also takes into account the overhead crane capacity consumption, effectively improving the overall collaborative operation efficiency of the workshop.

[0034] (2) By using Gaussian processes and chance-constrained programming, random transportation constraints are transformed into deterministic linear constraints, avoiding the high computational complexity of directly solving Gaussian processes. This ensures that the scheduling scheme still has a high on-time rate even when the crane speed fluctuates dynamically, thus reducing the risk of logistics delays.

[0035] (3) The nested algorithm designed integrates the advantages of reinforcement learning and tabu search. By adaptively evolving the global operator probability through reinforcement learning, the blindness of manual parameter tuning is avoided; by combining the priority weighting of critical path processes to perform local tabu search, the optimization bottleneck is accurately located, which improves the convergence speed and solution quality of the algorithm in a two-layer complex search space. Attached Figure Description

[0036] The accompanying drawings illustrate exemplary embodiments of the invention and, together with the description thereof, serve to explain the principles of the invention. These drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification.

[0037] Figure 1 This is a schematic diagram of a workshop layout according to one embodiment of the present invention.

[0038] Figure 2 This is a flowchart illustrating a production handling joint scheduling method according to an embodiment of the present invention.

[0039] Figure 3 This is a schematic diagram of the model solving process according to one embodiment of the present invention.

[0040] Figure 4 This is a schematic diagram of the equipment that can be selected for each process of a workpiece according to an embodiment of the present invention.

[0041] Figure 5 This is a schematic diagram showing the processing time of each process of a workpiece on each piece of equipment according to an embodiment of the present invention.

[0042] Figure 6 This is a schematic diagram of the overhead crane transport time according to one embodiment of the present invention.

[0043] Figure 7 This is a Gantt chart of a scheduling scheme according to an embodiment of the present invention.

[0044] Figure 8 This is a schematic structural block diagram of an electronic device employing a processor-based hardware implementation according to an embodiment of the present invention. Detailed Implementation

[0045] The present invention will now be described in further detail with reference to the accompanying drawings and examples. It should be understood that the specific examples described herein are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be noted that, for ease of description, only the parts relevant to the present invention are shown in the accompanying drawings.

[0046] It should be noted that, unless otherwise specified, the embodiments and features described in this invention can be combined with each other. The technical solution of this invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0047] Figure 1 A schematic diagram of a workshop layout according to an embodiment of the present invention is shown. This embodiment uses a microelectronic display screen (OLED panel) manufacturing workshop as an application scenario. The workshop adopts a high-altitude three-dimensional layout, with multiple processing equipment distributed in the ground processing area, responsible for the process processing of glass panel workpieces; storage platforms (Stockers) and handling robotic arms are provided next to the equipment for temporary storage of workpieces; multiple overhead cranes (OHS) run on the aerial tracks above the equipment, responsible for the automatic aerial transport of workpieces between the loading area and the processing equipment, as well as between different processing equipment.

[0048] Based on the above three-dimensional layout of the workshop, the following assumptions are proposed to facilitate the construction of subsequent models.

[0049] (1) The processing sequence of each process for the same workpiece is fixed, and the processing sequence of different workpieces does not affect each other; (2) The processing equipment can only be accessed by a single overhead crane; (3) Once the overhead crane starts working, the transportation task will not be interrupted; (4) The overhead crane can reach any position on the track and is a two-way transport; (5) The overhead crane transports the workpiece from the loading area to the equipment in the first processing stage and moves it between the equipment in subsequent stages.

[0050] (6) When the next process of the same workpiece is processed on a single machine, transportation is not required; (7) The equipment and crane were initially idle; (8) Each machine can only process one workpiece, and each workpiece can only pass through one machine; (9) Each processing equipment has a storage platform next to it to store the workpieces to be processed; (10) The time for the robotic arm in the storage platform to pick up the workpiece is included in the time during which the workpiece is assigned to a designated overhead crane for transport; (11) The workpieces are all in the loading zone at the initial moment.

[0051] Figure 2 A schematic diagram of the overall process of a production handling joint scheduling method according to an embodiment of the present invention is shown. Figure 1 The method shown includes steps S1 to S4. This method can be executed by electronic devices such as mobile phones and tablets.

[0052] S1: Obtain the basic parameters for joint scheduling.

[0053] The joint scheduling basic parameters are the data (data involved in the model constraints) needed for solving the subsequent production handling collaborative optimization model. As one possible implementation, the joint scheduling basic parameters may include: the equipment layout topology of the microelectronics manufacturing workshop, equipment spacing, standard overhead crane operating speed, overhead crane handling time between equipment, the number of processes contained in the workpiece, the standard processing time of each process, and the equipment that can be used for each process.

[0054] S2: Construct a production handling collaborative optimization model based on master-slave game theory and two-layer collaborative optimization.

[0055] The production handling collaborative optimization model consists of an upper-level model and a lower-level model. The upper-level model aims to minimize the maximum completion time by optimizing workpiece sequencing and equipment allocation decisions. Specifically, the workpiece sequencing decision determines the execution order of the corresponding processes for each workpiece, and the equipment allocation decision determines the equipment used to perform each process. The lower-level model aims to minimize the total idle travel time of each overhead crane by optimizing the crane allocation decision. The crane allocation decision determines the crane used to move completed workpieces to the next target location.

[0056] The upper-level model needs to comprehensively consider the processing options on different equipment and the processing connections between processes, and construct constraints between completion time and each process, transportation, and waiting time. Based on this, the following upper-level model was constructed: , (1) The constraints of the upper-level model include: (2) (3) (4) (5) (6) (7) in, This represents the objective function of the upper-level model; Indicates workpiece i Completion time; Indicates equipment k For the process O i(pi) The end time of the processing task; O i(pi) Indicates workpiece i The last process; pi represents the maximum process index of workpiece i; i This represents the workpiece index; n represents the total number of workpieces; min represents the minimum value; max represents the maximum value; Indicate process O ij Is it in the device? k The markings on the upper processing area; k and k’ All represent device indexes; m represents the total number of devices; Indicates equipment k For the process O ij The end time of the processing task; Indicates equipment k For the process O pq The start time of the processing task; Represents the maximum constant parameter; Indicate process O ij With process O pq In the equipment k The order of processing is indicated; Indicates equipment k For the process O ij The start time of the processing task; Indicates overhead crane v For the process O ij The end time of the transportation mission; Indicates equipment k For the process O ij The end time of the processing task; Indicates equipment k For the process O ij The processing time for the processing task; Indicates that the overhead crane is from the equipment k Move to devicek’ Time; Indicates equipment k With equipment k’ The distance between them; This indicates the normal operating speed of the overhead crane under undisturbed conditions. Indicates equipment k’ For the process O i(j-1) The start time of the processing task; Indicate process O i(j-1) Is it in the device? k’ The markings on the upper processing area; O i(j-1) Indicates workpiece i The j -1 process; O ij Indicates workpiece i The j One process; v j represents the overhead crane index; j represents the process index.

[0057] Among the above constraints, formula (2) is used to limit each process to only one machine for processing; formula (3) corresponds to the sequence constraint of the processes on the same machine; formula (4) is used to limit the start time of the process to be later than the end time of its load stroke; formula (5) is used to limit the workpiece to not be interrupted once processing begins; formula (6) is used to limit the distance between machines; formula (7) corresponds to the process route constraint of the workpiece (including transportation time).

[0058] The lower-level model aims for balanced and efficient overhead crane handling tasks, requiring the determination of which overhead crane will perform the handling task for each process. Building upon the workpiece sequencing and equipment allocation decisions provided by the upper-level model, the allocation and execution order of overhead crane handling tasks for each workpiece are further refined. Considering that the overhead crane handling process typically exhibits stable, small-amplitude fluctuations, and that the instantaneous velocity during overhead crane handling follows a Gaussian distribution, the following lower-level model is constructed: (8) The constraints of the lower-level model include: (9) (10) (11) (12) (13) (14) (15) (16) in, This represents the objective function of the lower-level model; Represents the mathematical expectation operator; Indicate process O ij With process O pq By overhead crane v The order of transportation; Indicates overhead crane v For the process O pq The end time of the empty transport phase; Indicates overhead crane v For the process O pq The start time of the empty transport phase; h represents the total number of overhead cranes; Indicate process O ij Is it by overhead crane? v Transportation markings; p Indicates the workpiece index; q Indicates the process index; O pq Indicates workpiece p The q One process; Indicates overhead crane v For the process O ij The start time of the empty transport phase; Indicates overhead crane v For the process O pq The end time of the load transportation phase; Indicates overhead crane v For the process O pq The end time of the empty transport phase; Indicates overhead crane v For the process O pq The start time of the empty transport phase; Indicate process O ij Location of the unloading point; Indicate process O ij Location of the loading point; This represents the distance calculation operator; v t Indicates that the overhead crane is t Instantaneous velocity at a given moment; Represents a Gaussian process; ~ indicates that it follows a distribution; This represents the mean of the instantaneous transport velocity fluctuations during a Gaussian process; The variance representing the transport speed; l A length scale representing the time correlation of transport volume; and Both represent time. Indicates overhead crane v For the process O ij The start time of the load transportation phase; Indicates overhead crane v For the process O ij The end time of the empty transport phase; Indicates overhead crane v’ For the process O i(j-1) The end time of the transportation mission.

[0059] Among the above constraints, formula (9) is used to limit that each handling task must be assigned to one crane; formula (10) is used to limit that the task chain of the crane is continuous; formula (11) is used to limit that a single crane must unload the previous task before it can set off to pick up the next task; formula (12) is used to limit the calculation method of the crane's no-load travel time; formula (13) is used to limit that after the order of tasks is determined, the no-load arrival time of the subsequent tasks must meet the constraints of the laws of physical motion; formula (14) is used to limit that the instantaneous speed of the crane during the handling process follows a Gaussian process distribution; formula (15) is used to limit the connection constraints between crane handling tasks; formula (16) is used to limit that the start time of a certain process of a workpiece must be later than the end time of the previous process.

[0060] S3: The instantaneous velocity during the overhead crane transport process is quantitatively modeled using Gaussian processes, and chance-constrained programming theory is introduced to transform the random instantaneous velocity constraint into a deterministic linear constraint that satisfies a pre-set confidence level.

[0061] Formula (14) in the constraints of the lower-level model above is used to quantitatively model the instantaneous velocity during the crane handling process using a Gaussian process. Since instantaneous velocity is introduced in formulas (12) and (13)... v t It belongs to the category containing random distribution ( v tNonlinear programming models containing the reciprocal of a random distribution are extremely difficult to solve directly, and the stability of the solution cannot be guaranteed. To address this technical problem, this embodiment introduces chance-constrained programming theory, transforming the random instantaneous velocity constraint into a deterministic linear constraint that satisfies a pre-set confidence level, thereby making the model solvable while ensuring scheduling robustness.

[0062] The derivation process for transforming the random instantaneous velocity constraint into a deterministic linear constraint that satisfies a pre-set confidence level is as follows: Based on a preset confidence level η (e.g., 95%), the probability that the overhead crane transport will be completed before the workpiece processing begins is not less than that confidence level η.

[0063] (17) in, Indicates equipment k For the process O ij The start time of the processing task; Indicates overhead crane v For the process O ij The start time of the transportation mission; Indicates the transportation distance; It represents probability.

[0064] make Let represent the time window allocated to this transportation task. The constraint is equivalent to requiring the crane's instantaneous speed to support the completion of the transportation within Δt. Therefore: (18) According to the characteristics of a Gaussian process, the instantaneous velocity satisfies... Standardize it as Substituting into the above formula, we get: (19) Where Z represents a random variable that follows a standard normal distribution.

[0065] According to the properties of the standard normal distribution, if the probability is greater than η, then the critical value must be less than or equal to the 1-η quantile. Using symmetry, we can obtain: (20) in, It represents the ηth quantile of the standard normal distribution.

[0066] By rearranging the above inequalities, we can obtain the effective speed. v eff The calculation formula is as follows: (twenty one) The above derivation proves that, given a confidence level η, using the corrected constant (effective velocity) v eff Deterministic scheduling is equivalent to satisfying the probabilistic constraints of a stochastic process.

[0067] In one example, based on historical enterprise data, the normal operating speed of the overhead crane under undisturbed conditions is μ = R = 120 m / min. The system scheduling is assumed to have an extremely low tolerance for transportation delays, with a confidence level of η = 95%, corresponding to the standard normal distribution quantile Z. 0.95 ≈1.645. Referring to the operational fluctuation characteristics of overhead cranes in actual workshop environments, the speed standard deviation σ is set to 2 m / min (i.e., the speed mainly fluctuates within the range of 120±4 m / min). The calculated effective speed is: (twenty two) Based on the above transformation, the random component in the original production and handling collaborative optimization model will be derandomized, that is... Convert to and use v eff Replacement Formula and In v t This not only eliminates the computational complexity introduced by the Gaussian process, but also ensures that the generated scheduling scheme will not experience delays due to speed fluctuations with a 95% probability, demonstrating good engineering robustness.

[0068] S4: Based on the joint scheduling basic parameters, the production and transportation collaborative optimization model is solved by a nested approach using a global-local neighborhood search and tabu search algorithm guided by reinforcement learning, so as to obtain the target production and transportation joint scheduling scheme.

[0069] Traditional single-layer optimization algorithms are structurally incapable of handling the multi-level decision relationships of the two-layer game in this embodiment. Furthermore, conventional heuristic algorithms, when faced with a complex two-layer decision space, suffer from fixed search operators and a lack of adaptive coordination mechanisms, making them prone to getting trapped in local optima. To address this technical problem, this embodiment proposes a novel reinforcement learning-guided global-local neighborhood search and tabu search algorithm (RLGLNSA).

[0070] The model solution process includes an upper-level model solution stage and a lower-level model solution stage. Both stages are optimized using a reinforcement learning-guided global-local neighborhood search and tabu search algorithm. The solution process of the reinforcement learning-guided global-local neighborhood search and tabu search algorithm includes a global search stage and a local search stage. In the global search stage, the selection probability of the neighborhood search operator is adaptively adjusted based on the reinforcement learning mechanism. In the local search stage, a tabu search strategy is executed on the critical path based on a priority weight mechanism.

[0071] During the global search phase, for each smart cell, its workpiece sorting decision can use three operations—insertion, swapping, and path reconnection—to generate a new workpiece sorting decision with a certain probability combination; its equipment allocation decision and crane allocation decision can use a mutation operator with a probability of 0.1 to generate a new equipment allocation decision and crane allocation decision, and select the one with the smallest workpiece completion time as the new smart cell.

[0072] During the local search phase, a tabu search strategy based on priority weights is executed on the smart cells output from the global search phase. This includes: starting from the final process in the joint scheduling Gantt chart, tracing back the occupancy chain of the workpiece on the processing equipment and overhead crane, and constructing a random critical path for the current solution; extracting four feature parameters of the process on the critical path: processing time, transportation time, process start time, and process end time, and calculating the comprehensive priority score of each process according to preset weights (e.g., [0.3, 0.3, 0.2, 0.2]); selecting the top 30% of the key processes with the comprehensive priority scores, keeping their workpiece codes unchanged, and reallocating their feasible equipment and overhead cranes to generate new solutions in the local neighborhood, and storing the process-equipment-overhead crane combination corresponding to the new solution in the tabu list.

[0073] Before the global search phase, the process also includes: generating an initial population using a hybrid initialization strategy: in which, the workpiece sorting decision is generated randomly; in the equipment allocation decision, 70% of the individuals are generated randomly, and 30% of the individuals are generated based on the earliest completion time rule; in the crane allocation decision, 70% of the individuals are generated randomly, and 30% of the individuals are generated based on the shortest transportation time rule.

[0074] like Figure 3 As shown, in one possible implementation, the solution process for the model may include: Step 1: Generate an initial population using a hybrid initialization strategy and pass it to the lower-level model.

[0075] Step 2: The lower-level model receives parameters from the upper-level model and performs lower-level optimization. In this stage, the RLGLNSA algorithm is called to calculate the objective function of the lower-level model, optimize the crane allocation decision, and feed back the optimized crane allocation decision parameters to the upper-level model.

[0076] Step 3: The upper-level model receives the parameters fed back from the lower-level model, calculates the objective function of the upper-level model, and then performs a selection operation (such as selecting high-quality solutions based on non-dominated sorting).

[0077] Step 4: The upper-layer model performs a global search phase: Based on the Q-table, operator probabilities are selected to determine the combination of three operator probabilities: insertion, swapping, and path reconnection. Then, the fitness value is calculated, and the current state and reward are obtained to update the Q-table, achieving adaptive adjustment of operator probabilities.

[0078] Step 5: The upper-level model performs a local search phase: First, it calculates the stochastic critical path of the current solution, and then determines whether the tabu stopping condition has been met. If the tabu stopping condition is met, a new solution is generated directly; if the tabu stopping condition is not met, the optimizable critical processes on the critical path are determined by priority weights, and then the equipment allocation and overhead crane assignment are changed for the determined critical processes to generate a new solution.

[0079] Step 6: The upper-level model updates the smart cells based on the generated new solutions and passes them to the lower-level model.

[0080] Step 7: Determine whether the maximum number of iterations has been reached (set to 200 in this embodiment). If not, return to step 2 to continue iterating. If it has been reached, output the optimal production and handling joint scheduling scheme that takes into account both processing completion time and overhead crane idle travel time.

[0081] In this implementation scheme, the upper and lower layers achieve collaborative optimization through nested calls. The upper layer passes production scheduling decisions to the lower layer model, and the lower layer model optimizes the crane allocation scheme and feeds back the results, forming a closed-loop optimization mechanism.

[0082] This invention can effectively solve the technical problem that existing technologies treat production scheduling and overhead crane handling assignment as independent sub-problems for separate optimization, resulting in execution deviations of theoretical solutions when actual overhead crane capacity is limited, and achieves deep collaboration between processing and logistics.

[0083] The technical effects of this invention will be illustrated below through a specific experiment.

[0084] In the specific experiment, actual production case data from a typical microelectronics company were used: the production line includes 6 to 8 processing stages, equipped with 10 processing machines, 3 overhead cranes for material handling, and a total of 10 workpieces need to be processed. Equipment availability is as follows: Figure 4 As shown, the equipment processing time is as follows: Figure 5 As shown, the processing time for each workpiece at different processing stages and on different equipment is given, and the handling time is as follows: Figure 6 The figure illustrates the time required for the overhead crane to move a workpiece between different positions. Workshop data simulation experiments were conducted using Matlab 2023b on the proposed model and algorithm. The operating environment was Windows 10, i5-9300H CPU@2.40GHz, and 8GB of memory.

[0085] The Gantt chart of the scheduling scheme obtained by the method in this embodiment is as follows: Figure 7 As shown, it visually displays 10 workpieces ( J 1- J 10 ) in 10 processing machines ( M 1- M 10 ) and 3 overhead cranes ( OHS 1- OHS 3) Regarding the scheduling, the workpiece completion time is approximately 164 minutes, and the total idle travel time of each crane is 96 minutes. The job scheduling in the Gantt chart uses multi-color coding to distinguish different jobs; dark gray blocks represent crane load status, and light gray blocks represent crane idle status.

[0086] According to a further embodiment of the present invention, an electronic device is also provided. Figure 8 This diagram illustrates a schematic block diagram of an electronic device employing a processor-based hardware implementation according to an embodiment of the present invention. The hardware structure of the electronic device of the present invention can be implemented using a bus architecture. The bus architecture can include any number of interconnect buses and bridges, depending on the specific application and overall design constraints of the hardware. Bus 1100 connects various circuits including one or more processors 1200, memory 1300, and / or hardware modules. Bus 1100 can also connect various other circuits 1400 such as peripheral devices, voltage regulators, power management circuits, external antennas, etc. Bus 1100 can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Component (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, only one connecting line is used in this diagram, but this does not imply that there is only one bus or one type of bus. Memory 1300 stores a computer program, and when processor 1200 executes the computer program, processor 1200 can perform the methods of the above embodiments of the present invention.

[0087] The present invention also provides a readable storage medium storing a computer program, which, when executed by a processor, is used to implement the methods described above. A "readable storage medium" can be any means capable of containing, storing, communicating, propagating, or transmitting a program for use by or in conjunction with an instruction execution system, apparatus, or device. More specific examples of a readable storage medium include: an electrical connection (electronic device) having one or more wires, a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and portable read-only memory (CDROM), etc.

[0088] This invention also provides a computer program product. The method of this invention can be implemented wholly or partially through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented wholly or partially in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed, the processes or functions of this invention are performed wholly or partially.

[0089] Computer programs or instructions can be stored in a readable storage medium or transferred from one readable storage medium to another. For example, the computer program or instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The readable storage medium can be any available medium capable of access, or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; an optical medium, such as a digital video optical disc; or a semiconductor medium, such as a solid-state drive. The computer-readable storage medium can be a volatile or non-volatile storage medium, or it can include both volatile and non-volatile types of storage media.

[0090] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0091] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0092] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0093] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0094] In the description of this specification, the references to terms such as "one embodiment / mode," "some embodiments / modes," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, or characteristic described in connection with that embodiment / mode or example is included in at least one embodiment / mode or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment / mode or example. Moreover, the specific features, structures, or characteristics described may be combined in any suitable manner in one or more embodiments / modes or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments / modes or examples described in this specification, as well as the features of different embodiments / modes or examples.

[0095] Those skilled in the art should understand that the above embodiments are merely for illustrating the present invention and are not intended to limit the scope of the invention. Those skilled in the art can make other changes or modifications based on the above disclosure, and these changes or modifications still fall within the scope of the present invention.

Claims

1. A method for joint scheduling of production handling, characterized in that, include: Obtain the basic parameters for joint scheduling; A production handling collaborative optimization model based on a master-slave game theory approach is constructed. This model comprises an upper-level model and a lower-level model. The upper-level model aims to minimize the maximum completion time by optimizing workpiece sequencing and equipment allocation decisions. The workpiece sequencing decision determines the execution order of the corresponding processes for each workpiece, while the equipment allocation decision determines the equipment used to execute each process. The lower-level model aims to minimize the total idle travel time of each overhead crane by optimizing the crane allocation decision. The crane allocation decision determines which crane will be used to move the completed workpiece to the next target location. Gaussian processes are used to quantitatively model the instantaneous velocity during overhead crane transport, and chance-constrained programming theory is introduced to transform the random instantaneous velocity constraint into a deterministic linear constraint that satisfies a pre-set confidence level. Based on the joint scheduling parameters, the production and transportation collaborative optimization model is solved in a nested manner using a global-local neighborhood search and tabu search algorithm guided by reinforcement learning, to obtain the target production and transportation joint scheduling scheme. In the solution process, during the global search phase, the selection probability of the neighborhood search operator is adaptively adjusted based on the reinforcement learning mechanism; during the local search phase, a tabu search strategy is executed on the critical path based on the priority weight mechanism.

2. The production handling joint scheduling method as described in claim 1, characterized in that, The basic parameters for joint scheduling include: the equipment layout topology of the microelectronics manufacturing workshop, the equipment spacing, the standard operating speed of the overhead crane, the time for the overhead crane to move between the equipment, the number of processes contained in the workpiece, the standard processing time of each process, and the equipment that can be used for each process.

3. The production handling joint scheduling method as described in claim 1, characterized in that, The upper-level model is represented as: ， The constraints of the upper-level model include: in, This represents the objective function of the upper-level model; Indicates workpiece i Completion time; Indicates device k For the process O i(pi) The end time of the processing task; O i(pi) Indicates workpiece i The last process; pi represents the maximum process index of workpiece i; i This represents the workpiece index; n represents the total number of workpieces; min represents the minimum value; max represents the maximum value; Indicate process O ij Is it in the device? k The markings on the upper processing area; k and k’ All represent device indexes; m represents the total number of devices; Indicates device k For the process O ij The end time of the processing task; Indicates equipment k For the process O pq The start time of the processing task; Represents the maximum constant parameter; Indicate process O ij With process O pq In the equipment k The order of processing is indicated; Indicates equipment k For the process O ij The start time of the processing task; Indicates overhead crane v For the process O ij The end time of the transportation mission; Indicates device k For the process O ij The end time of the processing task; Indicates device k For the process O ij The processing time for the processing task; Indicates that the overhead crane is from the equipment k Move to device k’ Time; Indicates equipment k With equipment k’ The distance between them; This indicates the normal operating speed of the overhead crane under undisturbed conditions. Indicates equipment k’ For the process O i(j-1) The start time of the processing task; Indicate process O i(j-1) Is it in the device? k’ The markings on the upper processing area; O i(j-1) Indicates workpiece i The j -1 process; O ij Indicates workpiece i The j One process; v Indicates the overhead crane index; j Indicates the process index.

4. The production handling joint scheduling method as described in claim 3, characterized in that, The lower-level model is represented as: The constraints on the objective function of the lower-level model include: in, This represents the objective function of the lower-level model; Represents the mathematical expectation operator; Indicate process O ij With process O pq By overhead crane v The order of transportation; Indicates overhead crane v For the process O pq The end time of the empty transport phase; Indicates overhead crane v For the process O pq The start time of the empty transport phase; h represents the total number of overhead cranes; Indicate process O ij Is it by overhead crane? v Transportation markings; p Indicates the workpiece index; q Indicates the process index; O pq Indicates workpiece p The q One process; Indicates overhead crane v For the process O ij The start time of the empty transport phase; Indicates overhead crane v For the process O pq The end time of the load transportation phase; Indicates overhead crane v For the process O pq The end time of the empty transport phase; Indicates overhead crane v For the process O pq The start time of the empty transport phase; Indicate process O ij Location of the unloading point; Indicate process O ij Location of the loading point; This represents the distance calculation operator; v t Indicates that the overhead crane is t Instantaneous velocity at a given moment; Represents a Gaussian process; ~ indicates that it follows a distribution; This represents the mean of instantaneous velocity fluctuations during a Gaussian process; The variance of instantaneous velocity; l A length scale representing the time correlation of transport volume; and Both represent time. Indicates overhead crane v For the process O ij The start time of the load transportation phase; Indicates overhead crane v For the process O ij The end time of the empty transport phase; Indicates overhead crane v’ For the process O i(j-1) The end time of the transportation mission.

5. The production handling joint scheduling method as described in claim 4, characterized in that, Transforming the random instantaneous velocity constraint into a deterministic linear constraint that satisfies a pre-set confidence level includes: Will Convert to ,in, v eff Indicates effective speed. This represents the ηth quantile of the standard normal distribution, where η represents the confidence level; and uses... v eff Replacement Formula and In v t .

6. The production handling joint scheduling method as described in claim 1, characterized in that, The production handling collaborative optimization model is solved using a nested, multi-level collaborative approach, including: An initial population is generated using a hybrid initialization strategy and then passed to the lower-level model. After receiving the parameters passed from the upper-level model, the lower-level model solves the objective function of the lower-level model and feeds back the solved parameters related to the crane allocation decision to the upper-level model. The upper-level model receives parameters from the lower-level model, calculates the objective function of the upper-level model, and then performs a selection operation; The upper-level model performs the global search phase, adaptively adjusting the probability combination of neighborhood search operators based on a reinforcement learning mechanism; The upper-level model performs a local search phase, using a priority weighting mechanism to execute a tabu search strategy on the critical path to generate a new solution; The upper-level model updates the smart cells based on the generated new solutions and passes them to the lower-level model; Determine if the maximum number of iterations has been reached. If not, continue iterating. If the maximum number of iterations has been reached, output the optimal production and transportation joint scheduling scheme as the target production and transportation joint scheduling scheme.

7. An electronic device, characterized in that, include: The memory stores execution instructions; as well as A processor that executes the execution instructions stored in the memory, causing the processor to perform the production handling joint scheduling method according to any one of claims 1 to 6.

8. A readable storage medium, characterized in that, The readable storage medium stores execution instructions, which, when executed by a processor, are used to implement the production handling joint scheduling method according to any one of claims 1 to 6.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the production handling joint scheduling method as described in any one of claims 1 to 6.

Images