Production prediction method for flexible manufacturing system

Abstract

The invention discloses a production prediction method for a flexible manufacturing system. The production prediction method comprises the steps of 1, setting algorithm preconditions and scheduling strategy requirements, 2, constructing a core element model, and 3, executing production whole-process simulation prediction. The method effectively solves the problem of production prediction of a flexible manufacturing system caused by uncertainty of varieties, processes and cutters and influence of a scheduling strategy, can simulate the whole production processing process, outputs multi-dimensional results such as work order completion time, a blank and cutter preparation list, a production Gantt chart, a production trend chart and an equipment starting rate, and improves the production efficiency. And reliable data support is provided for a user to adjust a production plan and configure resources.

Description

Technical Field

[0001] This invention relates to the field of flexible manufacturing technology, specifically to a production prediction method for flexible manufacturing systems. Background Technology

[0002] As the manufacturing industry shifts towards multi-variety, small-batch production, Flexible Manufacturing Systems (FMS) are widely used due to their flexible production capabilities. However, compared to traditional automated production lines (which have a single product type and a linear production process, and can be easily predicted using the formula "work order completion time ≈ work order start time + output × average processing cycle time"), FMS suffers from uncertainties in product type, process, and tooling, and is also affected by scheduling strategies, leading to a significant increase in the difficulty of production forecasting.

[0003] Existing technologies largely focus on FMS scheduling strategies, such as genetic algorithms, particle swarm optimization, and simulated annealing particle swarm optimization, but rarely explore production forecasting in depth. Production forecasting, as a crucial step before production starts, provides users with core data such as work order completion times, tool lists, and material preparation lists, directly impacting production plan adjustments and resource allocation efficiency.

[0004] Therefore, we propose a production forecasting method for flexible manufacturing systems. Summary of the Invention

[0005] The purpose of this invention is to provide a production prediction method for flexible manufacturing systems, which solves the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a production prediction method for a flexible manufacturing system, comprising the following steps:

[0007] Step 1: Setting Algorithm Prerequisites and Scheduling Strategy Requirements: The algorithm prerequisites include complete material set, complete tool set, zero tool change time, raw material input, replacement of tools with new tools, automatic material replenishment mechanism, and single handling equipment; The scheduling strategy requirements include outputting the time taken for each robot scheduling cycle and the change in pallet position after each cycle, outputting the time taken from machine tool startup to processing completion after machine tool startup, and supporting at most one machine tool or loading station loading per scheduling cycle, outputting the processing time when loading the machine tool, and outputting the operation time and product status after operation when loading the loading station;

[0008] Step 2: Construct the core element model: Define the attributes of work orders, production plans, robots, machine tools, loading stations, and pallet storage. Work orders include product type, start time, production quantity, and priority coefficient. Robots include maximum movement speed, acceleration, and current position. Machine tools include remaining processing time, loading / unloading positions, loading time, and unloading time. Loading stations include remaining operation time and loading / unloading positions. Pallet storage includes loading / unloading positions and whether pallets are present.

[0009] Step 3: Perform full-process production simulation and prediction: Collect data from all equipment on the production line and process data from all parts of the work order, initialize the system time to 0, and repeatedly execute the scheduling algorithm to generate robot action instructions, update the system time, correct the remaining time of the equipment, manage tool life and record milestone events until all work orders are completed and output the prediction results.

[0010] As a preferred embodiment of the present invention, the automatic material replenishment mechanism in step one is as follows: if the number of products required by the work order is greater than the number of blanks in the line, after the finished product leaves the line, the algorithm automatically generates blanks to be put into the line according to the machine tool capacity and the work order priority; the single handling equipment is a robot or a rail transport vehicle.

[0011] In a preferred embodiment of the present invention, the scheduling strategy in step one requires that the product status output when the loading station loads materials includes no material and material. If the product is material, the type of blank must be specified.

[0012] In a preferred embodiment of the present invention, the start time of the work order in step two satisfies the following: if the start time is earlier than the predicted start time, then the start time of the work order is equal to the predicted start time; if the start time is later than the predicted start time, then the work order will not be produced before the start time; the production quantity of the work order satisfies the following: when the quantity of finished products produced is greater than or equal to the production quantity required by the work order, the work order will stop production; the priority coefficient of the work order is used to increase the probability of the product corresponding to the work order being sent to the machine tool for processing by multiplying the current score by the priority coefficient during scheduling calculation.

[0013] In a preferred embodiment of the present invention, the machine tool in step two is a single-station machining scenario, i.e., it has one spindle and only one worktable for machining operations; the remaining machining time of the machine tool satisfies the following: when the remaining machining time is 0, the machine tool is in an idle state and can accept loading and unloading; when the remaining machining time is greater than 0, the machine tool is in a machining state and cannot perform loading and unloading; the loading station is a single-station scenario, i.e., only one worktable is used for manual loading and unloading; the remaining operation time of the loading station satisfies the following: when the remaining operation time is 0, the loading station is in an idle state and the robot can provide service; when the remaining operation time is greater than 0, the loading station is in a manual operation state and cannot perform loading and unloading; the presence or absence of pallets in the pallet storage satisfies the following: when there are pallets, the workpieces at the corresponding station can only be taken away; when there are no pallets, the corresponding station can place workpieces.

[0014] In a preferred embodiment of the present invention, the modification of the remaining time of the equipment in step three is specifically as follows: in the i-th round of scheduling, the remaining time of the j-th equipment in the previous round is obtained, and the remaining time of the j-th equipment in the current round is calculated in combination with the robot time in the current round; when the remaining time of the equipment in the previous round is greater than 0 and the remaining time in the current round is 0, the time when the equipment completes processing or operation is recorded.

[0015] In a preferred embodiment of the present invention, the management of tool life in step three is specifically as follows: in the i-th round of scheduling, the remaining life of the n-th tool of the j-th machine tool in the previous round is obtained, and the life consumption of the tool in this processing is subtracted to obtain the remaining life of the tool in the current round; when the remaining life of the tool is less than or equal to the tool warning life, a tool warning is triggered; when the remaining life of the tool is 0, a tool change is triggered, and the remaining life of the tool is reset to the maximum life of the tool.

[0016] In a preferred embodiment of the present invention, the prediction results in step three include work order completion time, material preparation plan, material preparation list, tool preparation plan, tool preparation list, production Gantt chart, equipment utilization rate, and production trend chart; the prediction efficiency of the algorithm reaches 400 processes / second.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0018] This invention effectively solves the production forecasting problem in flexible manufacturing systems caused by the uncertainty of product variety, process, and tooling, as well as the influence of scheduling strategies. It can simulate the entire production and processing process and output multi-dimensional results such as work order completion time, blank and tooling preparation list, production Gantt chart, production trend chart, and equipment utilization rate, providing reliable data support for users to adjust production plans and allocate resources.

[0019] This method is highly adaptable, requiring no specific quantity of machine tools, loading stations, or material storage facilities, and can flexibly adapt to FMS of different scales. Its predictive efficiency reaches 400 processes per second, meeting real-time on-site needs. By accurately predicting the timing and quantity of material and tool preparation, it can avoid resource shortages or excessive reserves, reducing production downtime and waste. Combined with equipment utilization rate prediction, it can also optimize equipment configuration, reduce idle costs, and significantly improve the efficiency and scientific nature of FMS production management. Attached Figure Description

[0020] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0021] Figure 1 This is an operational diagram of the production prediction method for a flexible manufacturing system according to the present invention. Detailed Implementation

[0022] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0023] like Figure 1 As shown, a production forecasting method for a flexible manufacturing system is implemented as follows:

[0024] Prerequisite settings and scheduling strategy confirmation:

[0025] Material availability: Ensure sufficient quantity of blanks required for production, without considering the time spent manually searching for blanks off-line;

[0026] Tooling stock is complete: Tool reserves meet production needs, without considering the time spent manually searching for tools off-line;

[0027] Tool change time is 0: The machine tool is equipped with a sister tool, and the tool is changed from the side door of the tool magazine, avoiding the interruption of the machining process due to waiting for tool change;

[0028] Raw material input: All workpieces entering the line from the loading station are unprocessed raw materials, excluding rework parts or semi-finished products;

[0029] Replace the tool with a new tool: The tool replaced manually is in brand new condition, and the current lifespan of the tool after replacement is equal to its maximum lifespan.

[0030] Automatic material replenishment mechanism: If the quantity of products required by the work order is greater than the quantity of blanks in the line, after the finished products leave the line, the algorithm automatically generates the blanks that need to be put into the line based on the machine tool production capacity and the work order priority.

[0031] Single handling equipment: Only one handling equipment (such as a robot or rail transport vehicle) is configured to handle the loading and unloading of workpieces within the production line.

[0032] Scheduling strategy function matching

[0033] After each robot scheduling cycle is completed, the time taken for this cycle and the changes in the pallet position are output. After the machine tool is started, the time required from machine tool start-up to processing completion is output. Each scheduling cycle supports a maximum of one machine tool or loading station loading: if loading is done on the machine tool, the corresponding processing time must be output; if loading is done on the loading station, the operation time and the product status of the loading station after the operation (no material / material available, if material available, the type of blank must be specified).

[0034] Core Element Model Construction

[0035] Work order model: Each work order corresponds to a product type; if the start time of the work order is earlier than the predicted start time, the predicted start time shall prevail; if it is later than the predicted start time, the work order shall not be produced before the start time; when the quantity of finished products reaches the production quantity required by the work order, the work order stops production; during scheduling calculation, the probability of the product corresponding to the work order entering the machine tool processing is increased by "current score × priority coefficient".

[0036] Production planning model: It consists of multiple work orders, which serve as the task input unit for production forecasting.

[0037] Robot model: Includes three core attributes: maximum movement speed (maximum movement speed on the walking axis), acceleration (the positive and negative acceleration values ​​are equal), and current position (distance from the origin of the walking axis).

[0038] Machine tool model: adopts a single-station machining scenario (1 spindle + 1 worktable); when the remaining machining time is 0, the machine tool is in an idle state and can accept loading and unloading; when the remaining machining time is greater than 0, it is in the machining state; clearly define the fixed loading and unloading points that the robot needs to move to, as well as the time taken for the robot to load (after placing the part, it reaches the safety point) and unload (after picking up the part, it reaches the safety point).

[0039] Loading station model: adopts a single-station scenario (1 workbench for manual loading and unloading); when the remaining operation time is 0, the loading station is idle and can accept robot services; when the remaining operation time is greater than 0, it is in manual operation mode; the fixed points for robot loading and unloading are clearly defined.

[0040] Pallet warehouse model: It contains multiple workstations, and each workstation has a clearly defined fixed point for robot loading and unloading; when there is a pallet, the corresponding workstation can only pick up parts, and when there is no pallet, the corresponding workstation can place parts.

[0041] Production process simulation and prediction execution

[0042] Initial data acquisition and system time initialization

[0043] Collect data from all equipment on the production line, including the robot's speed, acceleration, and initial position; the positions of machine tools, loading stations, and pallet storage; loading and unloading times; and initial states (such as remaining processing time for machine tools and whether there are pallets in the pallet storage). Collect process data for work order parts, including the machine tools and processing time suitable for each process of each product, as well as the tool life consumption value corresponding to each process. Initialize the system time to 0 as the starting time node for prediction.

[0044] Robot motion loop and data update

[0045] Scheduling instruction generation and system time shift: In each round, robot action instructions (such as feeding materials to a designated machine tool / loading station) are generated according to the scheduling strategy. The time ti taken for the robot to complete the action is recorded, the system time is shifted forward by ti, and the pallet position change is updated synchronously.

[0046] Device remaining time update

[0047] Calculation of remaining time for machine tools and loading stations: based on formula T i,j =T i-1,j -t i T i,j T represents the remaining time for scheduling the j-th device (machine tool / loading station) in the i-th round. i-1,j The remaining time for the next round on that device; if T i-1,j >0 and T i,j =0, record the device at P i,jThe processing or operation is completed at all times. After the loading station operation is completed, the corresponding blank is replaced according to the type of fixture and the completion status of the work order.

[0048] Remaining time after loading is updated: When the robot loads the machine tool, the remaining time is updated according to formula T. i,j =t p,k The remaining machining time of the machine tool is updated to the machining time of the k-th process of the p-th product; when loading the loading station, the time is adjusted according to formula T. i,j =t pm,down +t pn,up The remaining operation time of the loading station is updated to be the sum of the unloading time of product pm and the loading time of product pn.

[0049] Tool life management: based on formula l i,j,n =l i-1,j,n -c n Calculate the remaining life (l) of the nth tool on the j-th machine tool in the i-th round of scheduling. i-1,j,n c is the remaining lifespan of the previous cycle. n (This refers to the lifespan consumed during this processing); if l i,j,n ≤l n,alarm Trigger tool warning if l i,j,n If the value is 0, a tool change is triggered, resetting the remaining tool life to the maximum life. n,max And record the tool change time P i,j .

[0050] Milestone event recording: Synchronously record the moments of key milestone events such as machine tool processing completion, robot loading and unloading completion, and tool replacement, as a basis for material preparation and tool preparation time planning.

[0051] Prediction results output

[0052] The robot's motion cycle and data update steps are executed repeatedly until the finished product quantity of all work orders reaches the planned output. The simulation is then stopped and the prediction results are output, including: the completion time of each work order, the material preparation plan (including the time when the blank enters the line), the material preparation list (including the quantity of blanks), the tool preparation plan (including the time when the tool is replaced), the tool preparation list (including the tool specifications and quantity), the production Gantt chart, the equipment utilization rate, and the production trend chart. The algorithm's prediction efficiency reaches 400 processes / second, meeting the needs of on-site use.

[0053] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or basic characteristics. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of equivalents of the claims be included within the present invention.

[0054] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A production forecasting method for a flexible manufacturing system, characterized in that: The methods and steps include the following: Step 1: Setting Algorithm Prerequisites and Scheduling Strategy Requirements: The algorithm prerequisites include complete material set, complete tool set, zero tool change time, raw material input, replacement of tools with new tools, automatic material replenishment mechanism, and single handling equipment; The scheduling strategy requirements include outputting the time taken for each robot scheduling cycle and the change in pallet position after each cycle, outputting the time taken from machine tool startup to processing completion after machine tool startup, and supporting at most one machine tool or loading station loading per scheduling cycle, outputting the processing time when loading the machine tool, and outputting the operation time and product status after operation when loading the loading station; Step 2: Construct the core element model: Define the attributes of work orders, production plans, robots, machine tools, loading stations, and pallet storage. Work orders include product type, start time, production quantity, and priority coefficient. Robots include maximum movement speed, acceleration, and current position. Machine tools include remaining processing time, loading / unloading positions, loading time, and unloading time. Loading stations include remaining operation time and loading / unloading positions. Pallet storage includes loading / unloading positions and whether pallets are present. Step 3: Perform full-process production simulation and prediction: Collect data from all equipment on the production line and process data from all parts of the work order, initialize the system time to 0, and repeatedly execute the scheduling algorithm to generate robot action instructions, update the system time, correct the remaining time of the equipment, manage tool life and record milestone events until all work orders are completed and output the prediction results.

2. The production forecasting method for a flexible manufacturing system according to claim 1, characterized in that: The automatic material replenishment mechanism described in step one is as follows: if the number of products required by the work order is greater than the number of blanks in the line, after the finished product leaves the line, the algorithm automatically generates blanks to be added to the line based on the machine tool capacity and the work order priority; the single handling equipment is a robot or a rail transport vehicle.

3. The production forecasting method for a flexible manufacturing system according to claim 1, characterized in that: In the scheduling strategy described in step one, the product status output when the loading station loads materials includes "no material" and "material available". If it is "material available", the type of blank must be specified.

4. The production forecasting method for a flexible manufacturing system according to claim 1, characterized in that: The start time of the work order in step two satisfies the following: if the start time is earlier than the predicted start time, the start time of the work order is equal to the predicted start time; if the start time is later than the predicted start time, the work order will not be produced before the start time. The production quantity of the work order satisfies the following: when the quantity of finished products produced is greater than or equal to the production quantity required by the work order, the work order will stop production. The priority coefficient of the work order is used to increase the probability of the product corresponding to the work order being sent to the machine tool for processing by multiplying the current score by the priority coefficient during scheduling calculation.

5. The production forecasting method for a flexible manufacturing system according to claim 1, characterized in that: The machine tool described in step two is a single-station machining scenario, meaning it has one spindle and only one worktable for machining operations. The remaining machining time of the machine tool satisfies the following conditions: when the remaining machining time is 0, the machine tool is idle and can accept loading and unloading; when the remaining machining time is greater than 0, the machine tool is in machining mode and cannot perform loading and unloading. The loading station is a single-station scenario, meaning it has only one worktable for manual loading and unloading. The remaining operation time of the loading station satisfies the following conditions: when the remaining operation time is 0, the loading station is idle and can be serviced by a robot; when the remaining operation time is greater than 0, the loading station is in manual operation mode and cannot perform loading and unloading. The presence or absence of pallets in the pallet storage satisfies the following conditions: when there are pallets, the workpieces at the corresponding station can only be taken away; when there are no pallets, the corresponding station can place workpieces.

6. The production forecasting method for a flexible manufacturing system according to claim 1, characterized in that: The specific steps for correcting the remaining time of the equipment in step three are as follows: In the i-th round of scheduling, obtain the remaining time of the j-th equipment in the previous round, and calculate the remaining time of the j-th equipment in this round by combining it with the robot's time in this round; when the remaining time of the equipment in the previous round is greater than 0 and the remaining time in this round is 0, record the time when the equipment completes processing or operation.

7. The production forecasting method for a flexible manufacturing system according to claim 1, characterized in that: The tool life management described in step three is as follows: In the i-th round of scheduling, obtain the remaining life of the n-th tool of the j-th machine tool in the previous round, subtract the life consumption of the tool in this machining, and obtain the remaining life of the tool in the current round; when the remaining life of the tool is less than or equal to the tool warning life, trigger a tool warning; when the remaining life of the tool is 0, trigger a tool change and reset the remaining life of the tool to the maximum life of the tool.

8. The production forecasting method for a flexible manufacturing system according to claim 1, characterized in that: The prediction results in step three include work order completion time, material preparation plan, material preparation list, tool preparation plan, tool preparation list, production Gantt chart, equipment utilization rate, and production trend chart; the prediction efficiency of the algorithm reaches 400 processes / second.

Images