Automatic control of intelligent machines in production lines

Abstract

A method for controlling a production process is disclosed, the method including: performing a simulation of the production process for a plurality of machines of a production line for a plurality of configurations of a speed management component, where for each configuration, determining, by the simulation component, a plurality of statuses of the plurality of machines of the production line based on one or more events that change an operational state of the production line and based on speed setpoints of the plurality of machines, calculating, by the digital twin speed management component, based on the determined plurality of statuses and the respective configurations used for the digital twin speed management component, at least one new speed setpoint for at least one of the plurality of machines of the production line, analyzing performance of the production line based on the speed setpoints of the plurality of machines including the at least one new speed setpoint calculated for the at least one of the plurality of machines, and deploying, based on the analysis, one of the plurality of configurations of the speed management component to control the plurality of machines of the production line.

Description

[Technical field]

[0001] The present disclosure relates to a control system for a production line. [Background technology]

[0002] A production line comprises a number of machines that perform a series of sequential operations to manufacture products. The behavior of the machines in the production line may change during the running time of the production line. To optimize the production performance of the production line, the machine settings of the machines in the production line may be changed. The machine settings may be changed manually. However, changing the machine settings of one machine in the production line may affect multiple machines in the production line, for example machines that are located downstream of the machine whose machine settings have been changed. Therefore, manually changing the machine settings of the machines in the production line requires not only personnel but also the availability of experienced personnel to avoid a reduction in the production performance or downtime of the machines in the production line. Summary of the Invention

[0003] According to the present invention, a method of controlling a production process is provided. The method includes performing a simulation of the production process of a plurality of machines of a production line for a plurality of configurations of a speed management component. The simulation of the production process includes determining, for each configuration, a plurality of statuses of the plurality of machines of the production line by the simulation component based on one or more events that change the operating state of the production line and based on the speed setpoints of the plurality of machines, and calculating, by the digital twin speed management component, at least one new speed setpoint for at least one machine of the plurality of machines of the production line based on the determined plurality of statuses and the respective configurations used for the digital twin speed management component. The method further includes analyzing the performance of the production line based on the speed setpoints of the plurality of machines, including the at least one new speed setpoint calculated for the at least one machine of the plurality of machines, and deploying, based on the analysis, one of the plurality of configurations of the speed management component to control the plurality of machines of the production line.

[0004] An improved speed management component may be deployed to control machines in a production line by simulating the production process of multiple machines in the production line based on one or more events that change the operating state of the production line and based on the speed setpoints of the multiple machines. The deployed speed management component may control the machines in the production line taking into account events, e.g., stochastic events, and avoids manual interference with the production line. Thus, a self-tuning control system is provided that can bring multiple benefits. For example, the deployed speed management component frees operators from inefficient manual monitoring and control, providing an intelligent production line with less manual intervention, optimizing the production performance of the production line by increasing the uptime and productivity of the production line, and reducing losses and unplanned downtime due to failures upstream and downstream of the machines that result from not reaching the target rate of individual machines.

[0005] The present invention allows for the rapid and efficient determination and testing of configurations of rate management components to determine a final version of the configuration of the rate management components, which can be used to control machines on the production line. This is achieved by using simulation to perform performance analysis. Furthermore, since the digital twin rate management components do not affect the physical machines on the production line, testing the configuration of the rate management components using simulation provides a safe method for testing the configuration of the rate management components.

[0006] Results of the analysis of the performance of the digital twin speed control components that interact with the simulation components can be compared to the performance of the speed control components that interact directly with the production line machines. This allows for quick identification of improved configurations of the speed control components. Additionally or alternatively, this allows for confirmation or classification of the performance of the speed control components currently interacting with the production line machines. Results of the comparison may be reported to a user.

[0007] According to another aspect of the invention, a system configured to control a production process of a plurality of machines of a production line is provided. The system comprises a simulation component configured to simulate the production process for each of a plurality of configurations of a speed management component by determining a plurality of statuses of the plurality of machines of the production line based on one or more events that change an operational state of the production line and based on speed setpoints of the plurality of machines. The system further comprises a digital twin speed management component configured to simulate the production process for each of the plurality of configurations by calculating at least one new speed setpoint for at least one of the plurality of machines of the production line based on the determined plurality of statuses and the respective configuration used for the digital twin speed management component. The system is configured to analyze a performance of the production line for each of the plurality of configurations based on the speed setpoints of the plurality of machines including the at least one new speed setpoint calculated for the at least one of the plurality of machines, and based on the analysis, deploy one of the plurality of configurations of the speed management component to control the plurality of machines of the production line.

[0008] This allows for the formation of a self-adjusting control system based on the machine's data-driven logic. The speed management component can automatically adjust and modify the machine speed parameters of a production line or basic production unit, EPU, in real time to achieve self-optimized manufacturing performance.

[0009] Self-tuning control systems for continuous production lines, e.g. speed management components, create stability between the individual machines of a continuous production line, ensuring reduced interruptions to flow. By determining the state of the machines in the production line and reacting automatically, an improved self-tuning control system for self-optimized manufacturing performance of a complete continuous production line, e.g. EPU, is provided. This allows the effort of manually monitoring the production line to be significantly reduced. EXAMPLES

[0010] The present invention is defined in the claims. However, below is provided a non-exhaustive list of non-limiting examples. Any one or more of the features of these examples may be combined with any one or more features of the other examples, embodiments, or aspects described herein.

[0011] Example 1: A method for controlling a production process comprising: performing a simulation of the production process for a plurality of machines of a production line for a plurality of configurations of a speed management component, where for each configuration, determining by the simulation component a plurality of statuses of the plurality of machines of the production line based on one or more events that change an operational state of the production line and based on speed setpoints of the plurality of machines, calculating by the digital twin speed management component at least one new speed setpoint for at least one of the plurality of machines of the production line based on the determined plurality of statuses and the respective configurations used for the digital twin speed management component, and analyzing a performance of the production line based on the speed setpoints of the plurality of machines including the at least one new speed setpoint calculated for the at least one of the plurality of machines. The method further includes deploying one of the plurality of configurations of the speed management component based on the analysis to control the plurality of machines of the production line. Example 2: The method of embodiment 1, wherein a maximum buffer capacity of at least one machine in the production line is a configuration parameter of a plurality of configurations of the speed management component, and a simulation of the production process of the plurality of machines in the production line is performed for a plurality of maximum buffer capacities of the at least one machine in the production line. Example 3: The method of claim 1 or 2, further comprising determining a dynamic optimal buffer capacity of at least one machine in the production line of the production process by optimizing a maximum buffer capacity of the at least one machine with respect to at least one of: 1) a level of microstops or speed mismatch of the machines in the production line, 2) a speed of the machines in the production line, and 3) an operation phase based on the simulation, wherein the dynamic optimal buffer capacity varies over time based on at least one of: 1) an actual level of microstops or speed mismatch of the machines in the production line, 2) a speed of the machines in the production line, and 3) an operation phase. Example 4: 4. The method of embodiment 3, wherein the dynamic optimum buffer capacity of the at least one machine does not exceed the maximum buffer capacity of the at least one machine. Example 5: The method according to any one of embodiments 1 to 4, wherein the simulation comprises a Markov Chain Monte Carlo, MCMC, simulation. Example 6: The method of example 3, wherein machine learning is used to determine the dynamic optimum buffer capacity of at least one machine. Example 7: 4. The method according to claim 2 or 3, wherein the maximum buffer capacity of the at least one machine is the maximum buffer capacity of all machines in the production line. Example 8: The method according to any one of embodiments 1 to 7, wherein at least one new speed set point for at least one machine among a plurality of machines in the production line is obtained by the simulation component from the digital twin speed control component, and the simulation component and the digital twin speed control component form a feedback loop. Example 9: The method according to any one of the preceding claims, wherein the simulation of the production process is performed in real time. Example 10: The method according to any one of the preceding claims, wherein the digital twin speed control component is directly connected to the simulation component. Example 11: The method according to one of the preceding embodiments, wherein at least one speed setpoint of the speed setpoints of a plurality of machines in the production line corresponds to a target speed of the plurality of machines. Example 12: Multiple statuses are - for each machine of a set of machines, a flow model describing the flow of material from one machine or buffer to the next machine or buffer in the production line; - a reliability model for each machine of multiple machines that describes the time the machine is up or down; and - a quality model representing a quantity rejected by each machine of the plurality of machines due to quality issues. Example 13: The method according to any one of Examples 1 to 12, wherein the flow model is a deterministic model. Example 14: The method according to any one of embodiments 1 to 13, wherein the reliability model is a statistical model. Example 15: The method according to any one of the preceding claims, wherein the quality model is a statistical model. Example 16: There are several statuses, - the speed of one of the machines; - the buffer level of one of the machines, - Efficiency of one machine among several machines; -Material list or process order, PO, message for one of multiple machines, - parameters of one of the machines, - significant modifications to one of the machines; and - A method according to any one of claims 1 to 15, comprising at least one of the following: - a failure of one of the machines. Example 17: The method of any one of claims 1 to 16, wherein the one or more events include at least one of an operator shutdown of at least one of the plurality of machines, an unplanned shutdown of one of the plurality of machines, or a shortage or excess of product at an infeed or outfeed of one of the plurality of machines. Example 18: The method according to any one of claims 1 to 17, wherein the one or more events are one or more unplanned events that occur with a certain probability. Example 19: Analyzing the performance includes analyzing the performance in terms of at least one of: uptime of the plurality of machines, production volume of the plurality of machines, losses below target rates of the plurality of machines, and unplanned downtime of the plurality of machines. The method according to any one of Examples 1 to 18. Example 20: The method according to any one of Examples 1 to 19, further comprising: obtaining a plurality of live statuses of a plurality of machines in the production line by aggregating machine data from the plurality of machines by an aggregation server connected to the plurality of machines in the production line; calculating, by a speed management component connected to the aggregation server, at least one second speed setting value for at least one of the plurality of machines in the production line based on the plurality of live statuses; and controlling a speed of at least one of the plurality of machines in the production line by setting the at least one second speed setting value of the at least one of the plurality of machines in the production line. Example 21: The method of example 20, further comprising controlling a buffer level of at least one of the plurality of machines by calculating, by a speed management component, a speed setpoint of the at least one of the plurality of machines based on at least one of: 1) an actual level of microstops or speed mismatch of the machines of the production line, 2) a speed of the machines of the production line, and 3) an operating phase. Example 22: One or more events - machine stop from previously estimated random variables, - machine outages from random variables updated by a rate management component connected to an aggregation server that connects to multiple machines; - actual machine outages received from a rate management component that connects to an aggregation server that connects to multiple machines; and -A method according to any one of Examples 20 and 21, comprising one of the following: a machine stop obtained from a file. Example 23: The method of any one of Examples 20 to 22, further comprising comparing the performance of the digital twin rate control component with the performance of the rate control component and reporting the results of the comparison. Example 24: A method according to any one of Examples 20 to 23, wherein controlling the speed of at least one of a plurality of machines in the production line includes adjusting the speed of the one of the plurality of machines based on at least one of i) a status of a downstream machine located downstream of the machine in the production line, and ii) a status of an upstream machine located upstream of the machine in the production line. Example 25: 25. The method according to any one of claims 20 to 24, wherein controlling the speed of at least one of the plurality of machines in the production line includes controlling a speed of each machine of the plurality of machines to synchronize the speeds of the plurality of machines. Example 26: The method according to any one of claims 20 to 25, wherein controlling the speed of at least one of the multiple machines in the production line includes synchronously reducing the speeds of the multiple machines when one machine has a failure. Example 27: A method as described in any one of Examples 20 to 26, wherein controlling the speed of at least one of a plurality of machines in the production line includes synchronously increasing the speeds of the plurality of machines upon start-up of the plurality of machines. Example 28: The method according to any one of Examples 20 to 27, wherein controlling the speed of at least one of the multiple machines in the production line includes synchronizing the speeds of the multiple machines when one machine deviates from a target speed during the production process. Example 29: The method according to any one of Examples 20 to 28, further comprising controlling, by the speed management component, a buffer level of at least one machine in the production line according to a dynamic optimal buffer capacity of at least one machine in the production process based on at least one of: 1) the level of microstops or speed mismatch of the machines in the production line, 2) the speed of the machines in the production line, and 3) the operation phase. Example 30: A system configured to control a production process of a plurality of machines of a production line, the system comprising: a simulation component configured to simulate the production process for each of a plurality of configurations of a speed management component by determining a plurality of statuses of the plurality of machines of the production line based on one or more events that change an operating state of the production line and based on speed setpoints of the plurality of machines, and a digital twin speed management component configured to simulate the production process for each of the plurality of configurations by calculating at least one new speed setpoint for at least one of the plurality of machines of the production line based on the determined plurality of statuses and a respective configuration used for the digital twin speed management component. The system is configured to analyze a performance of the production line for each of the plurality of configurations based on the speed setpoints of the plurality of machines including the at least one new speed setpoint calculated for the at least one of the plurality of machines, and based on the analysis, deploy one of the plurality of configurations of the speed management component to control the plurality of machines of the production line. Example 31: The system of Example 30, wherein a maximum buffer capacity of at least one machine in the production line is a configuration parameter of multiple configurations of the speed control component, and a simulation of the production process of the multiple machines in the production line is performed for multiple configurations of the speed control component and for multiple maximum buffer capacities of the at least one machine in the production line. Example 32: The system is configured to determine a dynamic optimal buffer capacity of at least one machine in a production line of a production process by optimizing, based on a simulation, a maximum buffer capacity of the at least one machine with respect to at least one of: 1) a level of microstops or speed mismatch of the machines in the production line, 2) a speed of the machines in the production line, and 3) an operating phase, wherein the dynamic optimal buffer capacity changes over time based on at least one of: 1) an actual level of microstops or speed mismatch of the machines in the production line, 2) a speed of the machines in the production line, and 3) an operating phase. A system described in one of Examples 30 and 31. Example 33: The system of Example 32, wherein the dynamic optimum buffer capacity does not exceed the maximum buffer capacity. Example 34: The system of any one of Examples 30 to 33, wherein the dynamic optimal buffer capacity is determined by using Markov Chain Monte Carlo, MCMC, simulation. Example 35: The system of one of Examples 32 and 33, wherein artificial intelligence is used to determine the dynamic optimal buffer capacity. Example 36: The system of any one of Examples 30 to 35, further comprising: an aggregation server configured to obtain a plurality of live statuses of a plurality of machines in the production line by aggregating machine data from the plurality of machines; and a speed management component configured to receive at least one live status of the plurality of live statuses from the aggregation server, calculate at least one second speed setting value of at least one machine in the plurality of machines in the production line based on the at least one live status, and control a speed of at least one machine in the plurality of machines in the production line by setting the at least one second speed setting value for the at least one machine in the plurality of machines in the production line. Example 37: The event - machine stop from previously estimated random variables, - machine outages from random variables updated by a rate management component connected to an aggregation server that connects to multiple machines; - the actual machine stop received from the speed management component, and -The system described in Example 36, comprising at least one of the following: -Machine stop obtained from a file. Example 38: The system of any one of Examples 30 to 37, wherein each speed setpoint of the speed setpoints is received from a digital twin speed management component that connects directly to the simulation component. Example 39: The system according to any one of embodiments 30 to 38, wherein the speed setpoints correspond to target speeds for a plurality of machines. Example 40: The simulation components are: - A flow model that describes the flow of material from one machine or buffer to the next machine or buffer in the production line; - A reliability model that describes the time the machine is up or down, and -A quality model representing the quantity rejected by each machine due to quality issues. Example 41: The system of Example 40, wherein the flow model is a deterministic model. Example 42: The system of one of embodiments 40 and 41, wherein the reliability model is a statistical model. Example 43: The system according to any one of embodiments 40 to 42, wherein the quality model is a statistical model. Example 44: There are several statuses, - the speed of one of the machines; - the buffer level of one of the machines, - Efficiency of one machine among several machines; -Material list for one of the machines, - parameters of one of the machines, - significant modifications to one of the machines; and -A system described in any one of Examples 30 to 443, including at least one of the following: -A failure of one of the multiple machines. Example 45: 37. The system of example embodiment 36, wherein the speed management component is configured to synchronize speeds of multiple machines. Example 46: The system of any one of Examples 36 and 45, wherein the speed management component is configured to synchronously reduce the speeds of multiple machines if one machine fails. Example 47: The system of any one of Examples 36, 45, and 46, wherein the speed management component is configured to synchronously increase the speeds of the multiple machines upon start-up of the multiple machines. Example 48: The system of any one of Examples 36, 45-47, wherein the speed management component is configured to synchronize speeds of multiple machines when one machine deviates from a target speed during production. Example 49: The system of any one of Examples 36, 45 to 48, wherein the speed management component is configured to adjust the speed of one of the multiple machines based on at least one of a status of a downstream machine located downstream of the machine in the production line and a status of an upstream machine located upstream of the machine in the production line. Example 50: The system of any one of Examples 30 to 49, wherein the simulation component is configured to simulate the production process in real time. Example 51: The system of any one of Examples 30 to 50, wherein the events include at least one of an operator shutdown, an unplanned shutdown, and a shortage or excess of product at the infeed or outfeed of the machine. Example 52: The system of any one of embodiments 30 to 51, wherein the digital twin speed management component and the simulation component are deployed within a container. Example 53: The system of any one of Examples 30 to 52, wherein the system comprises a plurality of machines in a production line, the plurality of machines comprising at least one of a crimper, a buffer, a combiner, a cutting and turning unit, a packer line, a wrapper, and a bundler. Example 54: The system of any one of Examples 36, 45 to 48, wherein the digital twin rate management component, the rate management component, and the simulation component run as modules within a containerized environment on the edge device. Example 55: The system of any one of embodiments 30 to 54, wherein the production line comprises a basic production unit, an EPU. Example 56: The system described in any one of Examples 30 to 55, which interfaces with and communicates with multiple machines using the Open Platform Communication, OPC, Unified Architecture, UA, and Tobacco Machine Communication, TMC standards.

[0012] The embodiments will now be further described with reference to the figures. [Brief description of the drawings]

[0013] [Figure 1] FIG. 1 shows a schematic example of a production line machine. [Diagram 2] FIG. 2 shows an example diagram of the state machine of the machine. [Diagram 3] FIG. 3 shows an exemplary configuration of a system for controlling the production process of multiple machines of a production line. [Figure 4] FIG. 4 shows a process flow diagram of a method for controlling a production process. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] 1 shows a schematic example of a production line 100, also referred to as an EPU in the following description. The production line may consist of two or more machines that perform a series of successive operations to produce an output, e.g., an end product or final product. Each of the machines is capable of at least one of receiving material at the respective machine infeed and performing an operation, and discharging an output that can be received by the next machine in the production line 100.

[0015] The machines of a production line and the states of these machines can be modeled by state machine diagrams. The state machine diagrams of the individual machines of a production line can be used to simulate the production line. For example, a simulation model of an EPU can be built based on the state machines of the EPU's machines. State machine diagrams can model the behavior of a single object, e.g., machine speed, which specifies the different speed states that these machines can go through during production based on events.

[0016] The layout of production line 100 is shown in FIG. 1. The production line comprises machines 110, 120, 132, 134, 142, 144, 152, 154, 162 and 164. Each machine may comprise one or more machines. Machines 110-164 in the production line may comprise Tobacco Machine Communication (TMC) enabled machines. Arrows in FIG. 1 indicate the infeed and outfeed of each machine in production line 100. During normal production, the travel time of objects or goods from the outfeed of one machine to the infeed of the next machine may not be taken into account. These objects or goods that are moved between machines may represent the passive capacity of the production line that is always present during production.

[0017] The machine 110 may be a crimper. The crimper may be a combination of two sub-machines: a crimper unit CU and a machine that produces tobacco rods, which is the final product of the crimper. Any stochastic stoppage due to a CU failure may also be observed in the machine that produces tobacco rods (because the upstream machine is stopped). The crimper and the machine that produces tobacco rods can only run at the same speed. The crimper and the machine that produces tobacco rods may be modeled as a state machine, where the state may change based on events, e.g., stochastic stoppage data. These stochastic stoppage data may include stoppages due to a CU failure or a loader failure of the machine 110. The final product from the machine 110 may be tobacco rods that are fed into the machine 120. The machine 120 may be a buffer.

[0018] In one example, the state machine of the machine 110 may include a first state, e.g., a crimper down state, that corresponds to the machine state when the machine is stopped. The machine 110 may enter this state due to the following conditions: The first condition is that the machine 110 experiences a probabilistic stop (i.e., machine failure). The failure may be due to upstream and downstream machine failures. The duration and occurrence timing of these stoppages may be simulated using a probability generator. The second condition is that the machine 120 stops due to being full. The second state of the machine 110, e.g., a crimper full speed state, may correspond to the machine 110 running at production speed. The third state of the machine 110, e.g., a crimper slow speed state, may correspond to a reduced production speed value. The fourth state of the machine 110, e.g., a modified state, may be entered based on the quality of the machine 110 output.

[0019] The infeed to the machine 120, e.g., a buffer, may be the output from the machine 110, e.g., tobacco rods. The buffer may be set to a maximum buffer capacity. The maximum buffer capacity may correspond to the maximum amount of tobacco rods that the buffer can achieve. This value may be predefined or set by a user.

[0020] The production line 100 may be composed of three parts. A first part may be composed of machines 110 and 120, a second part may be composed of machines 132, 142, 152 and 162, and a third part may be composed of machines 134, 144, 154 and 164. The second and third parts may represent first and second branches of the production line and may share the output from the first part. The machines 132, 142, 152 and 162 may be of the same type as the machines 134, 144, 154 and 164.

[0021] Machines 132 and 132 may be combiners. The input to the combiner may be the output from machine 120, such as tobacco rods.

[0022] The machines of the production line may be represented by a state machine with two or more states. For example, machine 132 may be represented by a state machine with four states, each of which corresponds to a different speed at which machine 132 may run. Each state of the machine's state machine may be entered based on an event associated with the machine, such as a probabilistic stop of the machine, or an event associated with another machine of the production line, such as an event of an upstream or downstream machine. Machines 142 and 144 may be buffers, machines 152 and 154 may be cutting and turning units, and machines 162 and 164 may be wrapper and bundler units. The production process may be a process for manufacturing tobacco products.

[0023] According to one embodiment, a production line may include a first production line including machine 110, e.g., a crimper, machine 132, e.g., a combiner, machine 152, e.g., a cutting and turning unit, and machine 162, e.g., a wrapper and bundler unit, and a second production line including machine 110, e.g., a crimper, machine 134, e.g., a combiner, machine 154, e.g., a cutting and turning unit, and machine 164, e.g., a wrapper and bundler unit. The production line may perform the production process without a buffer between at least one of machines 110 and 132, 110 and 134, 132 and 152, and 134 and 154.

[0024] According to one embodiment, the production line may consist of only one production line including machine 110, e.g., a crimper, machine 132, e.g., a combiner, machine 152, e.g., a cutting and turning unit, and machine 162, e.g., a wrapper and bundler unit. This production line can perform the production process without a buffer between machines 110 and 132. Additionally or alternatively, this production line can perform the production process without a buffer between machines 132 and 152.

[0025] According to one embodiment, the production line may include machine 110, e.g., a crimper, machine 132, e.g., a combiner, machine 142, e.g., a buffer, machine 152, e.g., a cutting and turning unit, and machine 162, e.g., a wrapper and bundler unit. The production line can perform the production process without a buffer between machines 110 and 132.

[0026] According to one embodiment, a production line may include a first production line including machine 110, e.g., a crimper, machine 132, e.g., a combiner, machine 142, e.g., a buffer, machine 152, e.g., a cutting and turning unit, and machine 162, e.g., a wrapper and bundler unit, and a second production line including machine 110, e.g., a crimper, machine 134, e.g., a combiner, machine 144, e.g., a buffer, machine 154, e.g., a cutting and turning unit, and machine 164, e.g., a wrapper and bundler unit. This production line may perform the production process without a buffer between at least one of machines 110 and 132 and 110 and 134.

[0027] 2 illustrates an example diagram of a state machine 200 for a machine in a production line, e.g., machine 152. The state machine diagram 200 may include a first state 210 that corresponds to a state when machine 152 is stopped. Machine 152 may enter this state from any of states 220 and 230. For example, machine 152 may enter this state based on a first event, e.g., a machine failure of machine 152, or a failure of an upstream machine, e.g., machine 142, or a downstream machine, e.g., machine 162. When simulating the behavior of machine 152, the first event may be simulated by a probability generator. The probability generator may simulate the duration and timing of occurrence of the first event. Arrows 202 and 206 in the state machine represent the occurrence of the first event. A second event may occur due to a lack of output from an upstream machine, e.g., machine 142. Arrow 204 in the state machine 200 represents the occurrence of the second event. State 220 may correspond to a state in which machine 152 is running at a target speed for machine 152. Arrow 212 may represent a start-up or restart of machine 152. The production rate of machine 152 may be synchronized with the infeed or speed of another machine, such as machine 162. For example, if machine 162 (a machine downstream of machine 152) is running at a slow speed, the speed of machine 152 may be slowed down by entering state 230 (see arrow 222). Similarly, the speed of machine 152 may be increased from the slow speed state 230 to state 220 (see arrow 224).

[0028] FIG. 3 illustrates an exemplary configuration of a system 300 for controlling a production process of a plurality of machines 110-164 of a production line. The system 300 may be comprised of the machines 110-164 of the production line, and an edge device 310. All the machines 110-164 of the production line may be coupled to the edge device 310. The system 300 comprises a simulation component 312 configured to simulate a production process for each of the plurality of configurations of the speed management component 304 by determining a plurality of statuses of the plurality of machines 110-164 of the production line based on one or more events that change the operating state of the production line and based on the speed setpoints of the plurality of machines 110-164. The system 300 further comprises a digital twin speed management component 314 configured to simulate a production process for each of the plurality of configurations by calculating at least one new speed setpoint of at least one of the plurality of machines 110-164 of the production line based on the determined plurality of statuses and the respective configuration used for the digital twin speed management component 314. The performance of the production line for each of the multiple configurations may be analyzed based on the speed setpoints of the multiple machines 110-164, including the calculated at least one new speed setpoint for at least one of the multiple machines. Based on the analysis, a configuration of the multiple configurations of the speed management component 304 is deployed to control the multiple machines 110-164 of the production line. The digital twin speed management component 314 is directly coupled to the simulation component 312. Each of the speed setpoints for the speed setpoints used by the simulation component 312 may be received from the digital twin speed management component 314.

[0029] The system 300 may further comprise an aggregation server 302 and a speed management component 304. The aggregation server 302 is configured to obtain a plurality of live statuses of the plurality of machines 110-164 of the production line by aggregating machine data from the plurality of machines. This may be done via a bus 320. The speed management component 304 may be configured to receive at least one live status of the plurality of live statuses from the aggregation server 302, calculate at least one second speed setpoint of the at least one machine of the plurality of machines 110-164 of the production line based on the at least one live status, and control a speed of the at least one machine of the plurality of machines 110-164 of the production line by setting the at least one second speed setpoint of the at least one machine of the plurality of machines 110-164 of the production line.

[0030] The digital twin rate management component 314, the rate management component 304, and the simulation component 312 may run as modules within a containerized environment on the edge device 310. They may interface and intercommunicate with multiple machines 110-164 using Open Platform Communication, OPC, Unified Architecture, UA, and Tobacco Machine Communication, TMC standards.

[0031] Each machine 110-164 may be equipped with a server, such as a TMC OPC UA server, TMC-S, running as a module on the respective machine in a containerized environment. In this way, the machine parameters can be accessed, read and written. This allows the development and continuous improvement of the control mechanisms without the involvement of the original equipment manufacturer OEM. Furthermore, an aggregation server 302, e.g. a TMC OPC UA aggregation server, TMC-AS, running as a module on an edge device in the containerized environment. TMC-AS acts as a client of all TMC-Ses of machines 110-164 and can aggregate the servers of the machines controlled by the speed management.

[0032] In this description, the simulation component 312, also referred to as a real-time probabilistic simulator, TMC-AS DT, may be a first microservice running on an existing edge device in a containerized environment. The digital twin rate management component 314, as well as the rate management component 304, also referred to as a rate management controller, SM, may be a second microservice running on an existing edge device in a containerized environment.

[0033] A real-time stochastic simulator, TMC-AS DT, can be coded as a replica of a group of machines 110-164 and validated against actual production time series. TMC-AS DT may consist of a deterministic state machine of the product flow, while other relevant production events are statistically modeled and inferred from the machine data.

[0034] The TMC-AS DT may be considered as an aggregation server 302, e.g., a TMC aggregation server, where the following TMC type instances and their connections to the underlying systems are replaced with dynamic simulations in which the quantities change over time based on the speed of the machine modules. The simulations include: 1) the materials loaded into the machines are consumed (deterministically) according to the machine speed and status, as well as the material list information; 2) the production quantities are calculated (deterministically) as a function of the set speed of the machine and the material list information, and the speed increases and decreases are simulated with the machine state changes; and 3) the amount of material in the buffer is updated (deterministically) as a result of the combined inflow and outflow of materials. The machine status may be simulated stochastically over time, i.e., by emulating the random behavior of unplanned machine stops. The state machines of the production line machines may be driven by stochastic events.

[0035] A complete data set for one production shift period from a group of real production machines may be used for the simulation. Furthermore, machined outages may be classified into outage classes. For each outage class and for every outage, a frequency may be estimated as a probability function of the mean time between failures, MTBF, and mean time to repair, MTTR. Events are drawn from such distributions to change the machine status in the simulation, which leads to the deterministic part of the model. For example, if an operator outage occurs after an average of 48 minutes of uninterrupted machine production, and the machine returns to normal production status within 2 minutes, the stop and restart events are drawn from suitable random variables that match the measured average values. Overall, i.e. over a sufficiently long operating period, TMC-AS DT is expected to behave as a group of machines of a real production shift, i.e. to achieve very similar production levels and similar outage times. The TMC-AS DT microservice can be deployed as a container. A containerized image of TMC-AS DT can be deployed to the target edge device 310.

[0036] During the development of a speed management controller, SM, e.g., the speed management component 304, configurations of the speed management component 304 in the form of a digital twin speed management component 314 can be tested in terms of performance against the TMC-AS DT without affecting the operation of the physical machines 110-164. Simulations for different configurations of the speed management component 304 in the form of a digital twin speed management component 314 can be repeated so that a satisfactory performance level is achieved during the simulation. The final accepted version may be deployed to the machines 132-164 of the operational physical group and the timings adjusted. The speed management component 304 can be considered as a deployed version of the digital twin speed management component 314.

[0037] A speed management controller, SM, microservice may be developed to synchronize the speed of the TMC-AS machines or machine modules while checking the status of the buffers (if any) and the efficiency of the machines. The machine modules may be equipped with machine buffers within the machine. More specifically, the SM may be a modular microservice that adjusts the upstream and downstream speeds depending on the buffer levels and the efficiency of the machine, and issues commands for automatic start of the machine. The strength of the feedback and feedforward loops may be adjustable and may be set by the user.

[0038] The TMC-AS DT and SM microservices may interact with the TMC OPC UA aggregation server, TMC-AS, to read / write information to / from the TMC OPC UA server, TMC-S, of the machines 132-164. The TMC-AS DT and SM microservices may need to expose the system 300 real-time values ​​and event information to the historian.

[0039] The final implementation version of the real-time probabilistic simulator, TMC-AS DT, is used to define what the EPU performance would have been without implementing rate management. This definition is needed to compare the EPU performance improvement with the rate management solution. The real-time probabilistic simulator, TMC-AS DT, sends and receives the following to and from the EPU TMC OPC UA Aggregation Server (implemented under the scope of the EPU Digital Foundation): Receiving: 1. EPU machine value via OPC UA (client / server) communication, 2. Process Order (PO) message (material list) via OPC UA (client / server) communication. send: 1.Time series and log information via OPC UA (client / server) communication.

[0040] The final implementation version of the speed management controller (Speed ​​Management Component 304) is used to control the EPU machine via the EPU TMC OPC UA Aggregation Server (implemented under the scope of the EPU Digital Foundation). The final implementation version of the SM controller also incorporates part of the TMC-AS DT (312 and 314) logic. This is the predictive analysis logic required for the SM controller to predict and control the EPU machine independently. The speed management controller (Speed ​​Management Component 304) sends and receives the following to and from the EPU TMC OPC UA Aggregation Server (implemented under the scope of the EPU Digital Foundation): Receiving: 1. EPU machine value via OPC UA (client / server) communication, and 2. Process Order (PO) message (material list) via OPC UA (client / server) communication. send: 1. Time series and log information via OPC UA (client / server) communication, and 2. EPU machine value via OPC UA (client / server) communication.

[0041] A process order, a PO message, can be sent to the machine before production starts. It contains a description of which brand, which market the quantity is being produced for, how much quantity needs to be produced with this PO, etc. This is an additional check because for certain formats the maximum speed may be different. For example, for a format that needs to have an insert in the pack, the maximum speed of the machine needs to be reduced. Thus the PO message provides a cross check.

[0042] The speed management controller logic (Speed ​​Management Component 304) can be developed by using a real-time stochastic simulator, TMC-AS DT. This controller logic is developed targeting the following aspects and functional improvements of the EPU:

[0043] The SM logic includes a combination of deterministic self-tuning control loops that monitor the performance of the EPU machines during production. These monitoring may be based on machine status information that may be received via machine integration. Based on these observations, the SM logic varies the EPU machines to obtain optimal performance from the EPU under existing conditions during production. In addition, SM logic may be developed to check the status of buffers (if present) to balance the speed of the TMC-AS machine modules with the efficiency of the machines. Based on the SM logic produced, or with or without SM logic, a comparative analysis of the EPU for various configurations of the speed management components can be performed. Based on the results of this analysis, the SM logic can be used or further improved. The analysis may include the determination of performance indicators such as increased machine uptime, increased production, losses not reaching the target rate, amount of unplanned downtime due to upstream and downstream failures, etc.

[0044] EPU speed synchronization is a function provided by the speed management controller (304). The speed management component 304 can ensure that all machines in a production line will decelerate in a synchronized manner in the event of a single machine failure. Similarly, the speed management component 304 can ensure that all machines will start up in a synchronized manner (i.e., accelerate) to a target speed once a machine failure is resolved. Auto restart and standby features may exist on the machines to create a synchronized deceleration and startup behavior.

[0045] Additionally, the speed management component 304 can ensure that the speeds of all machines are synchronized during run time if one machine deviates from the target speed during production. For example, if a significant change event causes the packer to slow down, the remaining EPU machines will synchronize their speeds to keep the entire EPU line in balance.

[0046] The EPU machine's production speed acceleration and deceleration capabilities may not be identical, therefore the speed management component 304 can take into account the machine's acceleration and deceleration capabilities to synchronize the machine speed to avoid additional process stoppages during machine deceleration, start-up, and speed deceleration.

[0047] The buffer storage capacity of various amounts of the production line may be analyzed based on the performance of the SM (speed management component 314) against the capacity of the EPU machine.

[0048] Under the existing capacity of the EPU machine, a certain amount of buffer storage capacity may be required, because the EPU machine always accelerates and decelerates, and the acceleration and deceleration values ​​of various EPU machines are not the same, so without this certain amount of buffer storage capacity, the machine will stop due to lack or excess of production at the infeed or outfeed.

[0049] The maximum capacity of the buffer of the production line may be controlled by the system. The maximum capacity may be a parameter of the system. The SM controller (speed management component 304) can be configured to optimize the production line using the maximum buffer usage value based on the technical limitations of the EPU machine.

[0050] For example, the buffer (machine 142) has a maximum capacity of 25% and its active capacity is 10%. If in this situation the CTU (machine 152) fails, the combiner (machine 132) will run (using the speed management component 304) until the buffer capacity reaches 25% (while the CTU is still faulty). The combiner will then be put into standby by the downstream machine 142.

[0051] An optimization analysis of buffer minimization utilization can be performed with respect to its impact on the performance of an EPU machine with a speed management controller (digital twin speed management component 314). Thus, the impact of buffer reduction can be evaluated by using a simulation of the machine using key performance indicators (KPIs). This evaluation may be provided as a report. Based on the results of this evaluation, a value for the maximum capacity of the buffer storage may be determined. The maximum capacity of the production line may be a configuration parameter of the controller.

[0052] The machines 132-164 may be controlled by a rate management component 304 and may be connected to a bus 320. Furthermore, materials and / or products may flow from one machine to the next machine in a production line to produce an end product.

[0053] In FIG. 3, the simulation component 312 and the digital twin speed management component 314 may be considered as digital systems since they do not act on the physical machines 110-164. The aggregation server 302 and the speed management component 304 may be considered as physical systems since they act on the physical machines 110-164. The simulation component 312 may replicate the interface of the aggregation server 302. The aggregation server 302 aggregates live data of the machines 110-164. The speed management component 304 receives live data of the machines 110-164 or status of the machines 110-164, where the status is based on the live data of the machines 110-164. The aggregation server 302 receives data, e.g., speed setpoints of the machines or KPIs, from the speed management component 304. The digital twin speed management component 314 receives simulation data of the machines 110-164 or simulation status of the machines 110-164 from the simulation component 312. Alternatively, the digital twin speed management component 314 receives live data of the machines 110-164 or live status of the machines 110-164. The simulation component 312 receives data from the digital twin speed management component 314, for example, speed setpoints of the machines or KPIs.

[0054] The digital twin speed management component 314 and the speed management component 304 obtain information (simulated or live) from the machines 110-164. Based on the obtained information, the digital twin speed management component 314 and the speed management component 304 estimate 1) the status of the flow model (where the material is) and 2) the internal statistical parameters of the machine reliability model (defining MTBF / MTTR) and the quality model (defining rejects).

[0055] Digital twin speed management component 314 and speed management component 304 support 1) a single link-up mode, and 2) a double link-up mode. In single link-up mode, either the second portion comprising machines 132, 142, 152, and 162 or the third portion comprising machines 134, 144, 154, and 164 is operational. In double link-up mode, both the second and third portions are operational. The mode can be selected by a user or by the control system.

[0056] In double link-up mode, the bottleneck of the production line may be the crimper. The small amount of product lost at the bottleneck cannot be recovered anywhere else in the production line. The flow coming from the buffer after the crimper may be split into two flows (a second part and a third part) in proportion to the line efficiency. The line efficiency may be continuously updated based on downtime data obtained from the machine. If the efficiency of one part of the line decreases over time, the flow rate may be reduced proportionately.

[0057] In single link-up mode, the bottleneck may be the packer. In both single link-up and double link-up modes, the digital twin speed management component 314 and the speed management component 304 calculate the speed setpoint of the machine based on the upstream and downstream status of the machine. This is called machine-to-machine, M2M, logic.

[0058] The M2M logic works as follows: 1) If the upstream machine is stopped, the machine speed is reduced to avoid running out of material. The digital twin speed management component 314 or speed management component 304 considers the upstream storage (either buffer or product between machines) and the next predicted failure. The next predicted failure is based on a machine reliability model, which can be built based on actual machine data and can be continuously updated as machine stop data is collected. 2) If the upstream machine is running, the digital twin speed management component 314 or speed management component 304 calculates an optimal buffer level that avoids the machine shutting off or running out of material until the next predicted failure event on either the machine or the upstream machine. To achieve the optimal buffer level, the digital twin speed management component 314 or speed management component 304 calculates the machine speed setpoint taking into account the upstream machine speed.

[0059] The digital twin speed management component 314 or the speed management component 304 may apply M2M logic starting with the last machine on the production line and can work backwards to the infeed machines. Because blockages are less likely, the line path can be run in the reverse direction so that a downstream machine failure does not change the speed of the upstream machine.

[0060] The simulation component 312 may be a containerized application that simulates the behavior of the EPU, i.e., it simulates the machine status including up and down times, speeds, and buffer levels to allow for the consumption of input materials and the production of output materials. The simulation component 312 can be considered as a digital replica of the actual EPU, which is why it is called a Digital Twin (DT).

[0061] The DT can operate in a variety of modes. 1. Fixed probability mode: The DT simulates the process drawing machine stopping from pre-estimated random variables. Each machine speed setpoint is received from the digital twin speed management component 314 that is directly connected to the DT, otherwise the target speed may be used. 2. Update Probability Mode: The DT simulates the process drawing machine stopping from random variables updated by the speed management component 304 that connects to the aggregation server. Each machine speed setpoint is received from the digital twin speed management component 314 that connects directly to the DT, otherwise the target speed may be used. 3. Live mode: The DT simulates the process with actual machine stops received from the speed management component 304 that connects to the aggregation server 302. Each machine speed setpoint is received from the digital twin speed management component 314 that connects directly to the DT, otherwise the target speed is used. 4. From file mode: The DT simulates the process with machine stops taken from a file. Each machine speed setpoint is received from the Digital Twin speed management component 314 which connects directly to the DT, otherwise the target speed is used.

[0062] The live mode can be used to run what-if scenarios online, i.e., while the EPU is in production, using different speed management components 304 or no speed management components 304 at all. The what-if scenarios are useful for comparing and contrasting the performance of different control strategies, i.e., different speed control algorithms implemented by the speed management components 304.

[0063] The fixed probability mode can be used to generate offline simulations for debugging and evaluating the expected performance of a rate management component 304 configuration.

[0064] DT implements a production system that combines machines, software components that perform product processing functions, and buffers, software components that store the output. Such production systems are not simple. DT may implement a Monte Carlo simulation approach.

[0065] For each machine, the DT (simulation component 312) may implement 1) a flow model, i.e., a deterministic model describing the flow of material from the previous buffer or machine to the next buffer or machine, 2) a reliability model, i.e., a statistical model describing the up / down times of the machine, and 3) a quality model, i.e., a statistical model describing the quantities rejected by each machine due to quality issues.

[0066] The flow model can also implement backflow, i.e., interruption of flow when the machine is out of material (no input material from an empty upstream buffer) or blocked (no output material to a full downstream buffer). Out of material and blocked events are also called external stoppages and are the primary focus of reduction in the speed management component 304.

[0067] Since the DT exposes the same OPC UA interface that the aggregation server exposes, the containerized rate management component 304 or the digital twin rate management component 314 can connect to an aggregation server that connects to either a real EPU or a DT without any modification. The advantage of exposing the same interface is that the management component can be tuned digitally offline and if the performance is satisfactory, it can be used online without touching it, which makes online and offline performance not fully comparable.

[0068] 4 illustrates a process flow diagram of a method 400 for controlling a production process. The method 400 includes performing 410 a simulation of the production process of a plurality of machines of a production line for a plurality of configurations of a speed management component. For each configuration, the simulation may include determining (412) a plurality of statuses of the plurality of machines of the production line by the simulation component based on one or more events that change the operational state of the production line and based on the speed setpoints of the plurality of machines, calculating (414) at least one new speed setpoint for at least one of the plurality of machines of the production line by the digital twin speed management component based on the determined plurality of statuses and the respective configurations used for the digital twin speed management component, and analyzing (416) the performance of the production line based on the speed setpoints of the plurality of machines, including the at least one new speed setpoint calculated for the at least one of the plurality of machines. In an embodiment, step 416 may not be part of the simulation.

[0069] In step 420, a configuration of the multiple configurations of speed management components is developed for controlling the multiple machines in the production line based on the analysis.

[0070] In step 430, a dynamic optimal buffer capacity of at least one machine in the production line for the production process is determined by optimizing, based on the simulation, a maximum buffer capacity of at least one machine in the production line with respect to at least one of: 1) a level of micro-stops or speed mismatch of the machines in the production line, 2) a speed of the machines in the production line, and 3) an operating phase, where the dynamic optimal buffer capacity changes over time based on at least one of: 1) an actual level of micro-stops or speed mismatch of the machines in the production line, 2) a speed of the machines in the production line, and 3) an operating phase.

[0071] Further, the method 400 may include a method for controlling a speed of at least one machine in the production line. The method for controlling a speed of at least one machine in the production line includes steps 440-460. In step 440, a plurality of live statuses of a plurality of machines in the production line are obtained by aggregating machine data or live data from the plurality of machines. An aggregation server, e.g., aggregation server 302, coupled to the plurality of machines in the production line may perform step 440. In step 450, at least one second speed setpoint is calculated for at least one machine of the plurality of machines in the production line based on the plurality of live statuses. A speed management component, e.g., speed management component 304, coupled to the aggregation server may perform step 450. In step 460, a speed of at least one machine of the plurality of machines in the production line is controlled by setting at least one second speed setpoint for at least one machine of the plurality of machines in the production line. The speed management component may perform step 460.

[0072] At step 470, the method 400 includes comparing the performance of the digital twin rate management components to the performance of the rate management components, and at step 480, the method includes reporting the results of the comparison. The results of the comparison may affect the deployment of step 420. For example, a configuration of the multiple configurations of the rate management components is deployed to control multiple machines of the production line based on the results of the analysis and comparison. A configuration of the multiple configurations of the rate management components may be deployed only if the configuration of the rate management components improves at least one of the efficiency and performance of the production line.

[0073] Thus, one configuration of the multiple configurations of the rate management component can be tested by using this configuration for the digital twin rate management component that interacts with the simulation component. This does not affect the machines in the production line. If the configuration meets the requirements, this configuration can be used for the aggregation server and the rate management component that interacts with the machines in the production line. For example, the configuration may be deployed to an edge device or a production line to control the machines in the production line.

[0074] The simulation component may be validated. Using an offline version of the validated simulation component, several different versions or configurations of the rate management controller may be created and tested with the simulation component by using historical production data from the production line. During development, there may be no impact on the physical machine. The final approved version or configuration based on the improvement of the rate management controller may be finalized or deployed to the gateway (edge ​​device in the containerized environment) by the SRE, site reliability engineer. The terms "rate management controller" and "rate management component" may be used interchangeably herein.

[0075] During development of a rate governing controller version or configuration, a request from a user may be to see the impact of a reduced buffer fill level being used as the maximum capacity during production. This analysis may be performed as various rate governing controller versions or configurations are created before a final version or configuration is selected.

[0076] One other function of the speed management controller is that whatever the reduced capacity of the buffer as a maximum capacity, the speed management controller observes (while running on the edge device) the previous operating efficiency of the machines and the predictability of the machines and based on that adjusts the current filling level of the buffer (which must be within the newly defined maximum filling level of the buffer) to minimize dependencies between the machines.

[0077] For purposes of this specification and the appended claims, unless otherwise indicated, all numbers expressing amounts, quantities, percentages, and the like are understood to be modified in all instances by the term "about." Also, all ranges include the maximum and minimum points disclosed, and include any intermediate ranges therein, which may or may not be specifically recited herein. Thus, in this context, the number A is understood as A±5%. Within this context, the number A may be considered to include a numerical value that is within the typical standard error for the measurement of the property that the number A modifies. The number A may, in some cases, as used in the appended claims, deviate by the percentages recited above, provided that the amount by which A deviates does not materially affect the basic and novel properties of the claimed invention. Also, all ranges include the maximum and minimum points disclosed, and include any intermediate ranges therein, which may or may not be specifically recited herein.

[0078] The specific embodiments and examples described above are illustrative of the invention but do not limit it, it being understood that other embodiments of the invention may be made and that the specific embodiments and examples described herein are not exhaustive.

[0079] As used herein, the terms "upstream" and "forward," as well as "downstream" and "rearward," are used to describe the relative location of a machine or module in a production line with respect to the direction in which product flows or moves through the production line.

[0080] The objective of the present invention is to create a self-tuning control system based on the machine's data-driven logic, which can automatically adjust and modify the machine speed parameters of the elementary production unit (EPU) in real time to achieve self-optimized manufacturing performance.

[0081] During production on the shop floor, the process of manually adjusting machine parameters according to the machine's performance requires manual monitoring. Furthermore, the process of manual adjustment depends on the experience and ability of the machine personnel. This can lead to poor production performance and personnel spending time on inefficient activities. However, a self-adjusting control system for a continuous production line, e.g. a speed management system, creates stability between the individual machines in a continuous production line and ensures reduced interruptions to the flow. By determining the state of the machines in the production line and reacting automatically, an improved self-adjusting control system for self-optimized manufacturing performance of a complete continuous production line, e.g. an EPU, is provided. This allows the work of manually monitoring the production line to be significantly reduced.

Claims

1. A method for controlling the production process, This involves simulating the production process of multiple machines on a production line for multiple configurations of speed control components, with each configuration being: Based on one or more events that change the operating state of the production line, and based on the speed setting values ​​of multiple machines, the simulation components determine multiple statuses of the multiple machines on the production line. Based on the determined multiple statuses and the respective configurations used in the digital twin speed management component, the digital twin speed management component calculates at least one new speed setpoint for at least one of the multiple machines in the production line. The process includes: analyzing the performance of the production line based on the speed setting values ​​of the plurality of machines, including the calculated at least one new speed setting value for at least one of the plurality of machines; In order to control the plurality of machines in the production line, based on the analysis, one configuration of the plurality of configurations of the speed control component is deployed, The method described above is By using an aggregation server connected to the multiple machines of the production line, machine data is collected from the multiple machines to obtain multiple live statuses of the multiple machines of the production line, Based on the aforementioned multiple live statuses, the speed management component connected to the aggregation server calculates at least one speed setting value for at least one of the multiple machines in the production line, The method further includes controlling the speed of at least one of the machines in the production line by setting at least one speed setting value for at least one of the machines in the production line, and The aforementioned one or more events - Actual machine stoppages received from speed management components connected to the aggregation server connected to the multiple machines, and A method comprising: one of the following: machine stop, which is updated by a random variable connected to the aggregation server connected to the plurality of machines.

2. The method according to claim 1, wherein the maximum buffer capacity of at least one machine in the production line is a configuration parameter of the plurality of configurations of the speed control component, and the simulation of the production process of the plurality of machines in the production line is performed for the plurality of maximum buffer capacities of at least one machine in the production line.

3. The method according to claim 1, further comprising determining the dynamically optimal buffer capacity of at least one machine in the production line of the production process by optimizing the maximum buffer capacity of at least one machine with respect to at least one of 1) the level of microstops or speed mismatches of the machines in the production line, 2) the speed of the machines in the production line, and 3) the operating phase, wherein the dynamically optimal buffer capacity changes over time based on at least one of 1) the actual level of microstops or speed mismatches of the machines in the production line, 2) the speed of the machines in the production line, and 3) the operating phase.

4. The aforementioned multiple statuses, - For each of the multiple machines, a flow model representing the flow of material from one machine or buffer in the production line to the next machine or the next buffer, - For each of the aforementioned multiple machines, a reliability model representing the time the machine is operating or stopped, and The method according to claim 1, determined based on at least one of quality models representing the quantity that failed to pass by each machine of the plurality of machines due to quality problems.

5. The aforementioned multiple statuses, - The speed of one of the aforementioned machines, - The buffer level of one of the aforementioned machines, - The efficiency of one of the aforementioned machines, - One of the material list, process order, PO, and message for one of the aforementioned machines, - Parameters of one of the aforementioned machines, - A significant change to one of several machines, and The method according to claim 1, comprising at least one of the following: failure of one of the plurality of machines.

6. The method according to claim 1, wherein the one or more events include at least one of the following: one or more unplanned events occurring with a certain probability; a shutdown by an operator of at least one of the plurality of machines; an unplanned shutdown of one of the plurality of machines; a shortage or surplus of products in the infeed or outfeed of one of the plurality of machines; a machine shutdown from a previously estimated random variable; and a machine shutdown obtained from a file.

7. The performance of the aforementioned digital twin speed management component is compared with the performance of the aforementioned speed management component, The method according to one of claims 1 to 6, further comprising reporting the results of the comparison.

8. A system configured to control the production processes of multiple machines on a production line, wherein the system A simulation component, A simulation component is configured to simulate the production process for each of the multiple configurations of the speed management component by determining multiple statuses of the multiple machines of the production line based on one or more events that change the operating state of the production line and based on the speed setting values ​​of the multiple machines, A digital twin speed management component, A digital twin speed management component is configured to simulate the production process of each of the multiple configurations by calculating at least one new speed setting value for at least one of the multiple machines on the production line, based on the multiple statuses determined and the twin speed management component, An aggregation server configured to acquire multiple live statuses of the multiple machines on the production line by aggregating machine data from the multiple machines, The aforementioned speed control component, The aggregation server receives at least one live status from the plurality of live statuses, Based on at least one of several live statuses, calculate at least one speed setting value for at least one of the machines in the production line. The speed management component is configured to control the speed of at least one of the machines in the production line by setting at least one speed setting value for at least one of the machines in the production line, The aforementioned one or more events Actual machine stoppages received from speed management components connected to an aggregation server that connects to multiple machines, and This includes one of the machine stops from a random variable updated by a speed management component connected to an aggregation server connected to multiple machines, and The aforementioned system Based on the speed setting values ​​of the plurality of machines, including the calculated at least one new speed setting value of at least one of the plurality of machines, the performance of the production line of each of the plurality of configurations is analyzed. A system configured to deploy one of the multiple configurations of the speed control component based on the analysis, in order to control the multiple machines of the production line.

9. The system according to claim 8, wherein the maximum buffer capacity of at least one machine in the production line is a configuration parameter of the plurality of configurations of the speed control component, and the simulation of the production process of the plurality of machines in the production line is performed for the plurality of configurations of the speed control component and for the plurality of maximum buffer capacities of at least one machine in the production line.

10. The system according to claim 8, wherein the system is configured to determine the dynamically optimal buffer capacity of at least one machine on the production line for the production process by optimizing the maximum buffer capacity of the at least one machine with respect to at least one of 1) the level of microstops or speed mismatches of the machines on the production line, 2) the speed of the machines on the production line, and 3) the operating phase, wherein the dynamically optimal buffer capacity changes over time based on at least one of 1) the actual level of microstops or speed mismatches of the machines on the production line, 2) the speed of the machines on the production line, and 3) the operating phase.

11. The system comprises the plurality of machines of the production line, The plurality of machines comprises at least one of a crimping machine, buffer, combiner, cutting and rotating unit, packer line, wrapper, and bundler, The system according to claim 8, which interacts with and communicates with the plurality of machines using Open Platform Communication, OPC, Unified Architecture, UA, and Tobacco Machine Communication, TMC standards.

12. The system according to one of claims 8 to 11, wherein the digital twin speed management component, the speed management component, and the simulation component are executed as modules in a containerized environment on an edge device.

Images