A job shop intelligent scheduling system based on digital twin technology

Abstract

The present invention discloses a job shop intelligent scheduling system based on digital twin technology, including a job shop digital twin model and a job shop workshop interacting with the job shop digital twin model, the job shop digital twin model generates a scheduling strategy and issues it to the job shop workshop; the job shop workshop accepts the scheduling strategy in real time and completes the production process according to the scheduling strategy, and the job shop digital twin model monitors the production process of the job shop workshop in real time. If a disturbance factor is monitored during the production process, the original scheduling strategy is checked based on discrete event simulation technology to see whether it meets the production scheduling standard. If not, the original scheduling strategy is updated. The present invention utilizes digital twin technology to not only reproduce the workshop production process in real time, improve the monitoring capability of its production process, but also improve the ability of the scheduling system to cope with disturbances during operation.

Description

Technical Field

[0001] The present invention belongs to the fields of intelligent manufacturing technology and industrial engineering technology, and specifically relates to a job shop intelligent scheduling system based on digital twin technology. Background Art

[0002] In the market environment of personalized product demand, higher requirements are put forward for the flexibility, robustness, and processing efficiency of the job shop scheduling system. The construction of an intelligent scheduling system has become a new method and new model to meet the above requirements. With the development and application of automation equipment and information technology, digital and networked job shops have been widely implemented, providing an important foundation for building an intelligent scheduling system and improving the intelligence level of job shops.

[0003] In the actual production process of the job shop, disturbances occur frequently and in various forms. External disturbances include order changes, random product release, and delivery date changes, while internal disturbances include machine failures, machine degradation, and job rework. These disturbances pose a major challenge to the adaptability of production scheduling. If they are not discovered in time and handled at a reasonable time, they are likely to lead to reduced production efficiency and serious waste of resources.

[0004] In the face of the above disturbance problems, traditional job shop scheduling systems focus more on the performance of the scheduling methods used under one or more scheduling standards, and are highly targeted. However, under one or more scheduling standards, due to the limitations of a single scheduling method, it has weak adaptability to many heterogeneous disturbance factors and job diversity; and in the dynamic scheduling process, there is a lack of real-time monitoring of the impact of disturbance factors on the scheduling strategy, and the scheduling strategy cannot be updated in time when it deteriorates. Summary of the invention

[0005] In order to solve the problems existing in the traditional job shop scheduling system, the present invention provides a job shop intelligent scheduling system based on digital twin technology, with the purpose of real-time monitoring the occurrence of production disturbances and the moment when they affect the scheduling strategy under different scheduling standards, combining priority scheduling rules to form a new scheduling method and then updating the scheduling strategy, effectively responding to the disturbance factors that arise, and improving the robustness and production efficiency of the production system to a certain extent.

[0006] The technical solutions adopted by the present invention to solve the above-mentioned technical problems are as follows:

[0007] A job shop intelligent scheduling system based on digital twin technology, including a job shop digital twin model and a job shop workshop interacting with the job shop digital twin model;

[0008] The job shop digital twin model collects production information of the job shop workshop in real time during the production process, generates a scheduling strategy according to the production tasks issued to the job shop workshop and the collected production information, and issues the scheduling strategy to the job shop workshop;

[0009] The job shop accepts the scheduling strategy in real time and completes the production process according to the scheduling strategy, and the job shop digital twin model monitors the production process of the job shop in real time. If disturbance factors are detected during the production process, the discrete event simulation technology is used to check whether the original scheduling strategy meets the production scheduling standards. If not, the original scheduling strategy is updated.

[0010] The beneficial effects of the job shop intelligent scheduling system based on digital twin technology proposed in the present invention are:

[0011] By applying digital twin technology to the job shop, the present invention not only reproduces the production process of the job shop in real time and improves the monitoring capability of the production process, but also integrates the process of obtaining the scheduling method and generating the scheduling strategy, thereby improving the ability of the scheduling system to cope with disturbances during operation; multiple priority scheduling rules are selected from the constructed priority scheduling rule library, and the scheduling method formed by the flexible combination of the priority scheduling rules overcomes the limitation of the insufficient adaptability of a single scheduling method under one or more scheduling standards; a new scheduling method is used each time the scheduling strategy needs to be updated under the influence of disturbance factors, thereby avoiding the adverse effects of the original scheduling strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 This is an overall flow chart of a job shop intelligent scheduling system based on digital twin technology of the present invention in one of the embodiments;

[0013] Figure 2 This is the architecture diagram of the job shop digital twin model;

[0014] Figure 3 Generate, execute, and update flow charts for scheduling strategies;

[0015] Figure 4 This is the production operation flow chart of the job shop. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 are within the scope of protection of the present invention.

[0017] In one embodiment, if Figure 1 As shown, the present invention provides a job shop intelligent scheduling system based on data twin technology, which includes a job shop digital twin model and a job shop workshop interacting with the job shop.

[0018] The job shop digital twin model collects production information in the job shop workshop production process in real time, where the production information includes production process information, information on production factors in the workshop, and operating parameters of production factors in the workshop. The production process information includes the process flow of the operation, the handling flow of the operation, the extraction flow of raw materials, and the storage flow of finished products. The information on production factors in the workshop includes processing machine information, handling tool information, stacker information, clamping tool information, and information on processing machine operations. The operating parameters of production factors in the workshop include processing machine operating parameters, handling tool transportation parameters, stacker operating parameters, and clamping tool operating parameters. Then, a scheduling strategy is generated based on the production tasks issued to the job shop workshop and the collected production information, and the scheduling strategy is issued to the job shop workshop.

[0019] The job shop accepts the scheduling strategy in real time and completes the production process according to the scheduling strategy. The job shop digital twin model monitors the production process of the job shop in real time. If disturbance factors are detected during the production process, the discrete event simulation technology is used to check whether the original scheduling strategy meets the production scheduling standards. If not, the original scheduling strategy is updated.

[0020] When building a job shop digital twin model, the data generated by the workshop production process, workshop element parameters, etc. are collected in real time and processed and analyzed to achieve real-time and high-fidelity reproduction of the job shop production process and workshop elements in the virtual space. In order to meet the high efficiency of the production process operation, the production scheduling method and production scheduling strategy are embedded in the constructed virtual model, and finally a job shop digital twin model with three layers including method layer, strategy layer and execution layer is constructed, such as Figure 2 As shown in the figure, the three layers feed back and complement each other. The powerful virtual-real interaction, comprehensive measurement, and comprehensive analysis and prediction capabilities of digital twin technology can overcome the limitations of traditional job shop scheduling systems to a certain extent.

[0021] The method layer is responsible for generating an overall scheduling strategy based on production tasks and resource constraints in the job shop. During the production process, when the method layer receives information from the strategy layer that the scheduling strategy needs to be updated, the method layer generates a new and more suitable scheduling method based on the current production tasks and workshop resources and updates the original scheduling strategy with the scheduling strategy generated by the new scheduling method.

[0022] After the production task is issued, the available production resources in the workshop are counted, and based on the discrete event simulation technology, a priority scheduling rule PDR is selected from the priority scheduling rule library PDR_set based on the process. m p Determine the machine that processes each job and the order in which the jobs are processed before each machine. All selected priority scheduling rules are combined into a scheduling method PDR_set m , this scheduling method is the scheduling method of the job shop in the current production scenario. Each process is assigned a priority scheduling rule PDR m p Generate a sub-scheduling strategy, and all sub-scheduling strategies are combined into an overall scheduling strategy. Each sub-scheduling strategy is responsible for scheduling the production process of each process. Because one of the core purposes of scheduling is to select machines to process jobs in the production system and to sort the jobs before the machines, and the priority scheduling rules have the advantages of extensiveness, diversity and simplicity, the scheduling method formed by their flexible combination can overcome the limitations of insufficient adaptability of a single scheduling method.

[0023] The strategy layer is responsible for transmitting the scheduling strategy generated by the method layer to the execution layer and supervising the production process of the execution layer. During the supervision process, according to the external disturbance factors from the outside world and / or the abnormal phenomena fed back by the execution layer (i.e., the internal disturbance factors of the jobshop), the strategy layer self-checks the adaptability of the original scheduling strategy performance, that is, based on the discrete event simulation technology, it checks whether the original scheduling strategy meets the production scheduling standards. If not, that is, the scheduling strategy is no longer suitable for the current production environment, then the information that the scheduling strategy needs to be updated is fed back to the method layer.

[0024] The execution layer is directly connected to the real job shop, responsible for receiving the scheduling strategy issued by the strategy layer, and applying the scheduling strategy to the production process of the job shop. At the same time, the execution layer collects data that can represent the production process and status of the elements of the real production workshop, such as the operating parameters of the machine, the actual processing time of the machine, the use of raw materials, the completion of the finished product, the geometric information of the production elements of the workshop, the operating parameters of the production elements of the workshop, etc., so that the execution layer can map the real production workshop in real time and with high fidelity, analyze the production process status of the production workshop, and feedback the abnormal status information (disturbance factors) to the strategy layer.

[0025] The scheduling strategy generation process from the method layer to the strategy layer and the scheduling application process from the strategy layer to the execution layer are a process of task flow transmission.

[0026] The feedback from the execution layer and external disturbances to the strategy layer and the feedback from the strategy layer to the method layer is a process of information flow transmission.

[0027] The transmission process of task flow and information flow enables the interaction between the various layers of the job shop digital twin model, which is conducive to timely detection and response to disturbance factors in the production process.

[0028] The reasonable selection and use of production scheduling methods is the guarantee of the efficiency and order of the entire production process, and is responsible for the generation of scheduling strategies. The method layer in the job shop digital twin model is based on a heuristic algorithm that is widely present in quantity and variety—priority scheduling rule PDR. First, the priority scheduling rule library PDR_set is constructed. The priority scheduling rule PDR contained in it is selected according to the parameter data para1 collected in the production process and the parameter data para2 required for solving the scheduling standard. Para1 and para2 are combined into parameter para, where the parameters of the production process are such as the processing time of the machine processing the job, the transportation time of the job between the processes, etc., and the parameters required for solving the scheduling standard are such as the delivery date of the job under the maximum delay time scheduling standard, and the weight coefficient of the job under the total weighted completion time scheduling standard.

[0029] Next, discrete event simulation technology is used to select P priority scheduling rules PDR with superior performance from the priority scheduling rule library PDR_set in units of processes to form a scheduling method PDR_set. m , where P is the number of processes in the job shop, and the priority scheduling rule PDR assigned to the pth process is the pth sub-scheduling method PDR m p , PDR_set, PDR_set m and PDR m p The relationship is shown in formula (1). m p The process of generating a sub-scheduling strategy on the pth process includes setting the selection strategy of the machine to be processed by the job and the priority strategy of the jobs waiting to be processed in front of the machine. The process of setting the selection strategy of the machine to be processed by the job is shown in formula (2), where para ijp represents the parameter value of job i in process p machine j, f d (para ijp ) indicates using the dth PDR to solve the parameter value para ijpThe priority value obtained, J represents the number of machines included in the pth process, and the maximum priority value obtained finally corresponds to j, which is the machine selected by the job to be processed. Similarly, the process of setting the priority strategy for jobs waiting to be processed in front of the machine is shown in Figure (3), I represents the number of jobs waiting to be processed in front of the machine, and the maximum priority value obtained finally corresponds to i, which is the job that is processed by the machine first. PDR m p The multiple sub-scheduling strategies obtained are combined to form an overall scheduling strategy.

[0030]

[0031]

[0032]

[0033] The execution of the scheduling strategy is to monitor the actual production process, collect the production operation and workshop element parameter data in real time, and is responsible for analyzing the data collected from the actual production workshop, timely discovering and uploading abnormal phenomena (internal disturbances), such as equipment failures, equipment degradation, etc. Among them, because the scheduling method (priority scheduling rule combination) for generating the scheduling strategy is centered around the job and processing machine, the failures, degradation and other disturbances of auxiliary equipment such as transportation tools and stackers indirectly affect the performance of the scheduling strategy. In order to make the above-mentioned disturbance factors directly express the impact on the current scheduling strategy, and then promote the self-examination of the scheduling strategy performance, the time other than the processing machine processing operation is collectively referred to as non-value-added time, that is, the preparation time before the job is processed, such as transportation time and handling time, which realizes the transformation of the disturbance factors that indirectly affect the scheduling strategy into time parameters that directly affect the performance of the scheduling strategy.

[0034] The update of the scheduling strategy is the guarantee for the continuous, efficient and orderly progress of the production process, and is responsible for handling the disturbance factors encountered in the production process. Figure 3 , when encountering disturbance factors and feedback to the strategy layer, the scheduling strategy generated by the scheduling method has a certain robustness. Therefore, the strategy layer will use discrete event simulation technology to check whether the original scheduling strategy of the current production status can still meet the production scheduling standards based on the production tasks and workshop resources after the disturbance. If the production scheduling standards are met, the original scheduling strategy will continue to be executed; if the production scheduling standards cannot be met, the information that needs to update the scheduling strategy will be fed back to the method layer. The method layer will re-combine the priority scheduling rules according to the production tasks and production resources at the update time to generate a new scheduling method PDR_set m , and then generate a new scheduling strategy, update the original scheduling strategy with the new scheduling strategy, and then realize the update of the scheduling strategy.

[0035] The scheduling method connects the scheduling strategy into a whole, and the scheduling strategy connects the workshop production factors into a whole. The operation process of the processing workshop is carried out under the guidance of the scheduling strategy. The operation process of the entire scheduling system includes the operation of the jobshop digital twin model and the actual job shop production process and the interaction between the two.

[0036] The operation process of the scheduling system is centered on the processing machine, and the time consumption of the entire production process is divided into two categories: value-added time and non-value-added time. Non-value-added time is the time for the processing machine to process the job, and non-value-added time is the preparation time before the job is processed, including the time for the handling operation of the handling equipment, the finished product storage of the stacker, and the time for the extraction of raw materials.

[0037] The scheduling strategy generated according to the priority scheduling rule is characterized by determining the machines for processing operations and the order of machine processing operations in a certain process. The workshop elements guiding the production process of each sub-scheduling strategy are the processing machines in a process, the handling tools for handling operations, and if the process is the starting process, the stacker for extracting raw materials, and if the process is the final process, the stacker for storing finished products. Among them, the processing machine is the core equipment.

[0038] See also Figure 4 The production operation process of the job shop after receiving the scheduling strategy includes:

[0039] The raw materials are moved from the raw material shelves to the raw material turnover area by stackers, and then transported to the work-in-progress turnover area before the first process of the operation by transportation tools according to the process flow of each operation to wait for processing; the raw materials in the work-in-progress turnover area are placed on the processing machine using a clamping tool, and the processing machine processes them. After processing, they are placed in the semi-finished product turnover area next to the process. The job allocation of the process machine and the processing order of the previous generation of processing jobs of each machine are determined by the sub-scheduling strategy;

[0040] If the process is the last process, the work completed by the processing machine of the process is the finished product, which will be transported to the finished product turnover area by a transport trolley and then transported to the finished product shelf by a finished product stacker;

[0041] If the process is not the last process, the work after processing by the processing machine that can complete the process is a semi-finished product, which is then transported to the work-in-process turnover area of ​​the downstream process by a transport cart to be processed by the processing machine in the downstream process. The processing operation flow of the downstream process is the same as that of the first process.

[0042] By applying digital twin technology to the job shop, the present invention not only reproduces the production process of the job shop in real time and improves the monitoring capability of the production process, but also integrates the process of obtaining the scheduling method and generating the scheduling strategy, thereby improving the ability of the scheduling system to cope with disturbances during operation; multiple priority scheduling rules are selected from the constructed priority scheduling rule library, and the scheduling method formed by the flexible combination of the priority scheduling rules overcomes the limitation of the insufficient adaptability of a single scheduling method under one or more scheduling standards; a new scheduling method is used each time the scheduling strategy needs to be updated under the influence of disturbance factors, thereby avoiding the adverse effects of the original scheduling strategy.

[0043] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0044] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these 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 job shop intelligent scheduling system based on digital twin technology, characterized in that: Includes the job shop digital twin model and the job shop workshop that interacts with it; The job shop digital twin model collects production information of the job shop workshop in real time during the production process, generates a scheduling strategy according to the production tasks issued to the job shop workshop and the collected production information, and issues the scheduling strategy to the job shop workshop; The job shop accepts the scheduling strategy in real time and completes the production process according to the scheduling strategy, and the job shop digital twin model monitors the production process of the job shop in real time. If a disturbance factor is detected during the production process, the discrete event simulation technology is used to check whether the original scheduling strategy meets the production scheduling standard. If not, the original scheduling strategy is updated; The job shop digital twin model includes a method layer, a strategy layer and an execution layer; The method layer is responsible for generating an overall scheduling strategy based on the production tasks and resource constraints in the job shop. During the production process, when the method layer receives information from the strategy layer that a scheduling strategy needs to be updated, the method layer generates a new scheduling method based on the current production tasks and workshop resources and updates the original scheduling strategy with the scheduling strategy generated by the new scheduling method. The process of generating an overall scheduling strategy according to the production task and the resource constraints in the job shop comprises the following steps: Construct a priority dispatch rule base PDR_set, which contains priority dispatch rules PDR that are selected based on parameter data para1 collected during the production process and parameter data para2 required to solve the dispatch standard; Using discrete event simulation technology, P priority scheduling rules PDR are selected from the priority scheduling rule base PDR_set based on the process. m p Combined into scheduling method PDR_set m , where P is the number of processes in the job shop; Using the scheduling method PDR_set m Generate an overall scheduling strategy, which includes multiple sub-scheduling strategies; The strategy layer is responsible for transmitting the scheduling strategy generated by the method layer to the execution layer and supervising the production process of the execution layer. In the supervision process, according to external disturbance factors from the outside world and / or internal disturbance factors of the job shop fed back by the execution layer, the discrete event simulation technology is used to check whether the original scheduling strategy meets the production scheduling standard. If not, the information that the scheduling strategy needs to be updated is fed back to the method layer. The execution layer is responsible for receiving the scheduling strategy issued by the strategy layer, applying the scheduling strategy to the production process of the job shop, and feeding back the internal disturbance factors of the job shop to the strategy layer.

2. According to claim 1, a job shop intelligent scheduling system based on digital twin technology is characterized in that: The production operation process of the job shop after receiving the scheduling strategy includes: The raw materials are moved from the raw material shelves to the raw material turnover area by stackers, and then transported to the work-in-progress turnover area before the first process of the operation by transportation tools according to the process flow of each operation to wait for processing; the raw materials in the work-in-progress turnover area are placed on the processing machine using a clamping tool, and the processing machine processes them. After processing, they are placed in the semi-finished product turnover area next to the process. The job allocation of the process machine and the processing order of the previous generation of processing jobs of each machine are determined by the sub-scheduling strategy; If the process is the last process, the work completed by the processing machine of the process is the finished product, which will be transported to the finished product turnover area by a transport trolley and then transported to the finished product shelf by a finished product stacker; If the process is not the last process, the work after processing by the processing machine that can complete the process is a semi-finished product, which is then transported to the work-in-process turnover area of ​​the downstream process by a transport cart to be processed by the processing machine in the downstream process. The processing operation flow of the downstream process is the same as that of the first process.

3. According to claim 1, a job shop intelligent scheduling system based on digital twin technology is characterized in that: The production information includes production process information, information on production factors in the workshop and operating parameters of production factors in the workshop. The production process information includes the process flow of operations, the handling flow of operations, the extraction flow of raw materials and the storage flow of finished products. The information on production factors in the workshop includes processing machine information, handling tool information, stacker information, clamping tool information and information on processing operations by processing machines. The operating parameters of production factors in the workshop include processing machine operating parameters, handling tool transportation parameters, stacker operating parameters and clamping tool operating parameters.

Images