A production rhythm-based dynamic warehouse intelligent sorting method, system, device and medium

Abstract

The application provides a production rhythm-based dynamic warehouse intelligent sorting method, system, device and medium, which can improve the adaptability of the sorting system to the production rhythm, optimize the inventory turnover efficiency, and reduce the production interruption risk caused by material mismatch or supply delay. The method comprises the following steps: collecting real-time parameters of the production rhythm of each key station on the production line, and obtaining demand information of target materials to be sorted; based on the real-time parameters of the production rhythm and the demand information, predicting the production rhythm prediction parameters and the material demand prediction of each key station in a future period of time, and combining the current inventory state, the equipment availability and the preset optimization parameters to generate a dynamic bin adjustment plan and an optimal sorting task sequence suitable for the production line at present, and scheduling a flexible scheduling mechanism to dynamically adjust the physical position of the corresponding materials, and then calling a sorting mechanism to sort a specified number of target materials from the dynamically adjusted bin to the corresponding target station.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of warehouse management, and in particular to a dynamic warehouse intelligent sorting method, system, device and medium based on production rhythm. BACKGROUND

[0002] The existing warehouse sorting system mainly includes the following three types:

[0003] (1) Fixed path AGV sorting system: This system is usually used in large e-commerce / distribution centers, and uses automated guided vehicles (AGV) to transport and sort according to a predetermined path, to solve the problem of automatic processing of large-scale, standardized orders. However, when facing frequent production plan adjustments or dramatic production rhythm fluctuations, this system has difficulty in path re-planning, and the response is slow, making it difficult to achieve instant and accurate supply of materials.

[0004] (2) Static storage location management system based on WMS preset instructions: This system uses warehouse management system (WMS) to allocate entry and exit instructions for materials, to achieve relatively fixed or only periodic optimization of storage locations, solving the problem of basic tracking and inventory management of materials. However, this system cannot dynamically adjust the inventory layout according to the real-time material consumption rate and demand priority of the production line, which may result in long pickup paths for hot materials or high-quality storage locations occupied by cold materials.

[0005] (3) Sorting system based on modular design: This system uses modular design to obtain a partially flexible modular sorting unit, which can be expanded to a certain extent according to demand, but its scheduling logic is mostly based on historical data or fixed thresholds, and lacks sufficient linkage with the manufacturing execution system (MES), mostly receiving one-way instructions, lacking real-time perception of production rhythm data and predictive scheduling based on it, resulting in a time difference or quantity mismatch between material supply and production demand, which can easily cause waiting or accumulation. That is, the sorting system based on modular design has limited response capability to real-time dynamic changes of the production line.

[0006] In summary, the existing warehouse sorting system has problems of insufficient efficiency and flexibility, weak intelligence and adaptability, and poor system integration and collaboration. These problems make it difficult for the existing warehouse sorting system to adapt to dynamic changes in production rhythm, material flow efficiency is not high, warehouse space is not optimally utilized, and sorting operations do not match the actual production demand with sufficient precision.

[0007] Therefore, there is an urgent need for a comprehensive solution that can effectively combine production line real-time beat data, dynamically adjust warehouse strategy and sorting task priority, and realize accurate, efficient and self-adaptive sorting of materials. SUMMARY

[0008] Embodiments of the present application provide a dynamic warehouse intelligent sorting method and system based on production beat, equipment and medium, to solve the problems existing in the related art, and the technical solutions are as follows:

[0009] In a first aspect, the embodiments of the present application provide a dynamic warehouse intelligent sorting method based on production beat, comprising:

[0010] Collecting production beat real-time parameters of each key station on the production line, and obtaining demand information of target materials to be sorted;

[0011] Based on the production beat real-time parameters and the demand information, calling a production prediction model to predict production beat prediction parameters and material demand prediction of each key station in a future period of time;

[0012] Based on the production beat prediction parameters and the material demand prediction, in combination with the current inventory state, device availability and preset optimization parameters, a dynamic bin adjustment plan and an optimal sorting task sequence suitable for the production line are generated, wherein the dynamic bin adjustment plan contains corresponding materials in the target materials that need to adjust the physical location, and the optimal sorting task sequence contains target stations corresponding to the target materials;

[0013] Based on the dynamic bin adjustment plan, a flexible scheduling mechanism in the production environment is dispatched to dynamically adjust the physical location of the corresponding materials;

[0014] Based on the optimal sorting task sequence, a sorting mechanism in the production environment is called to pick a specified number of target materials from the dynamically adjusted bin and sort them into the corresponding target station.

[0015] In one embodiment, collecting production beat real-time parameters of each key station on the production line and obtaining demand information of target materials to be sorted comprises:

[0016] Calling photoelectric sensors and industrial cameras on each key station to monitor the product passing rate and buffer material inventory level of each key station in real time;

[0017] Calling PLC controllers on each key station to read device status and real-time yield data of each key station;

[0018] Call the information recognition channel at the entrance of the warehouse, automatically scan the unique identification code label on the warehousing material or the unique identification code on the warehousing box, and obtain the detailed information of the warehousing material;

[0019] Call the ERP system to obtain the production plan for a period of time in the future;

[0020] Based on the product passing rate, the buffer zone material inventory level, the equipment state and the real-time yield data, the real-time parameters of the production rhythm are obtained;

[0021] Based on the detailed information of the warehousing material and the production plan, the demand of the target material is determined, and the demand information is obtained.

[0022] In an embodiment, based on the real-time parameters of the production rhythm and the demand information, a production prediction model is called to predict the production rhythm prediction parameters and material demand prediction quantities of each key station in a period of time in the future, including:

[0023] Data cleaning and feature extraction are performed on the real-time parameters of the production rhythm and the demand information to obtain corresponding feature data;

[0024] A specified time series prediction model is used as the production prediction model;

[0025] Based on the corresponding feature data, the specified time series prediction model is called to predict the production rhythm prediction parameters and material demand prediction quantities of each key station in a period of time in the future.

[0026] In an embodiment, based on the production rhythm prediction parameters and the material demand prediction quantities, in combination with the current inventory state, equipment availability and preset optimization parameters, a dynamic bin adjustment plan and an optimal sorting task sequence suitable for the production line at present are generated, including:

[0027] With the preset optimization parameters as the optimization target, a specified multi-objective optimization algorithm is called to generate the dynamic bin adjustment plan and the optimal sorting task sequence based on the production rhythm prediction parameters, the material demand prediction quantities, the current inventory state and the equipment availability.

[0028] In an embodiment, based on the dynamic bin adjustment plan, a flexible scheduling mechanism in the production environment is dispatched to dynamically adjust the physical position of the corresponding material, including:

[0029] The dynamic bin adjustment plan is sent to the WCS, and the WCS is called to determine the current bin, target bin and bin adjustment time of the corresponding material based on the dynamic bin adjustment plan, wherein WCS refers to warehouse control system;

[0030] sending a storage location adjustment instruction to the WCS, instructing the WCS to schedule the flexible scheduling mechanism to transfer the corresponding material from the current storage location to the target storage location within the storage location adjustment time, so as to complete the dynamic adjustment of the geographical position of the corresponding material.

[0031] In an embodiment, the optimal sorting task sequence includes an optimal sorting operation sequence and an optimal path planning of the target material handling; based on the optimal sorting task sequence, a sorting mechanism in a production environment is called to sort out a specified number of the target materials from the dynamically adjusted storage location and then sort them into the corresponding target workstations, including:

[0032] sending sorting instructions to the sorting mechanism in sequence according to the optimal sorting operation sequence of the optimal sorting task sequence, to instruct the sorting mechanism to pick up a specified number of the target materials from the dynamically adjusted storage location according to the received sorting instructions, and then sort the picked-up target materials in sequence to the specified positions of the corresponding target workstations;

[0033] calling the sorting mechanism to distribute the sorted target materials from the specified positions to the corresponding target workstations according to the optimal path planning of the optimal sorting task sequence.

[0034] In an embodiment, the method further includes:

[0035] monitoring the actual flow condition of the target materials and the actual production rhythm variation of the production line in real time to obtain real-time monitoring results;

[0036] when a production abnormality is determined to occur based on the real-time monitoring results, triggering an alarm, and returning to execute the production prediction model to predict the production rhythm prediction parameters and the material demand prediction quantity of each key workstation in a future period of time based on the production rhythm real-time parameters and the demand information.

[0037] In a second aspect, the embodiments of the present application further provide a dynamic warehouse intelligent sorting system based on production rhythm, including:

[0038] a production rhythm perception module, configured to collect production rhythm real-time parameters of each key workstation on a production line, and obtain demand information of target materials to be sorted;

[0039] a data fusion and decision processing module configured to predict production cycle time prediction parameters and material demand prediction quantities of each of the key workstations in a future period of time based on the production cycle time real-time parameters and the demand information, and generate a dynamic bin adjustment plan and an optimal sorting task sequence suitable for the production line based on the production cycle time prediction parameters, the material demand prediction quantities, current inventory status, device availability, and preset optimization parameters, wherein the dynamic bin adjustment plan includes corresponding materials of the target materials that need to adjust physical positions, and the optimal sorting task sequence includes target workstations corresponding to the target materials;

[0040] a dynamic adaptive warehousing module configured to dynamically adjust physical positions of the corresponding materials based on the dynamic bin adjustment plan by scheduling a flexible scheduling mechanism in a production environment;

[0041] an intelligent sorting execution module configured to call a sorting mechanism in the production environment to pick a specified number of the target materials from the dynamically adjusted bin and sort them into the corresponding target workstations based on the optimal sorting task sequence.

[0042] In an embodiment, when the production cycle perception module is used to collect production cycle time real-time parameters of each key workstation on the production line and obtain demand information of target materials to be sorted, it is specifically configured to:

[0043] call photoelectric sensors and industrial cameras on each of the key workstations to monitor product passing rates and buffer material inventory levels of each of the key workstations in real time;

[0044] call PLC controllers on each of the key workstations to read device states and real-time yield data of each of the key workstations;

[0045] call an information recognition channel at a warehouse entrance to automatically scan unique identification code labels on incoming materials or unique identification codes on incoming material boxes to obtain detailed information of the incoming materials;

[0046] call an ERP system to obtain a production plan for a future period of time;

[0047] obtain the production cycle time real-time parameters based on the product passing rates, the buffer material inventory levels, the device states, and the real-time yield data;

[0048] determine the demand of the target materials based on the detailed information of the incoming materials and the production plan to obtain the demand information.

[0049] In an embodiment, the data fusion and decision processing module is specifically configured to:

[0050] perform data cleaning and feature extraction on the production rhythm real-time parameters and the demand information to obtain corresponding feature data;

[0051] use a specified time series prediction model as the production prediction model;

[0052] invoke the specified time series prediction model to predict production rhythm prediction parameters and material demand prediction amounts of each of the key workstations in a future period of time based on the corresponding feature data.

[0053] In an embodiment, the data fusion and decision processing module is specifically configured to:

[0054] invoke a specified multi-objective optimization algorithm to generate the dynamic bin adjustment plan and the optimal sorting task sequence based on the production rhythm prediction parameters, the material demand prediction amounts, the current inventory state, and the equipment availability, with the preset optimization parameters as the optimization objective.

[0055] In an embodiment, the dynamic adaptive warehousing module is specifically configured to:

[0056] send the dynamic bin adjustment plan to a WCS, and invoke the WCS to determine a current bin, a target bin, and a bin adjustment time of the corresponding material based on the dynamic bin adjustment plan, wherein WCS refers to a warehouse control system;

[0057] send a bin adjustment instruction to the WCS, and instruct the WCS to schedule the flexible scheduling mechanism to transfer the corresponding material from the current bin to the target bin within the bin adjustment time, so as to complete the dynamic adjustment of the geographical position of the corresponding material.

[0058] In an implementation, the optimal sorting task sequence comprises an optimal sorting operation sequence and an optimal path planning of the target material handling; when the intelligent sorting execution module is used to call the sorting mechanism to pick up a specified number of the target materials from the dynamically adjusted storage locations and sort them to the corresponding target workstations based on the optimal sorting task sequence, the intelligent sorting execution module is specifically configured to:

[0059] send sorting instructions to the sorting mechanism according to the optimal sorting operation sequence of the optimal sorting task sequence, to instruct the sorting mechanism to pick up a specified number of the target materials from the dynamically adjusted storage locations according to the received sorting instructions, and then sort the picked-up target materials to the specified positions of the corresponding target workstations in sequence;

[0060] call the sorting mechanism to distribute the sorted target materials from the specified positions to the corresponding target workstations according to the optimal path planning of the optimal sorting task sequence.

[0061] In an implementation, the system further comprises a central coordination and monitoring module, which is configured to:

[0062] monitor the actual flow of the target materials and the actual production rhythm changes of the production line in real time to obtain real-time monitoring results;

[0063] when a production abnormality is determined based on the real-time monitoring results, trigger an alarm, and return to calling a production prediction model to predict the production rhythm prediction parameters and the material demand prediction of each key workstation in a future period of time based on the production rhythm real-time parameters and the demand information by the data fusion and decision processing module.

[0064] In a third aspect, the embodiments of the present application further provide an electronic device, which comprises a memory and a processor, the memory stores instructions, the instructions are loaded and executed by the processor to implement the method in any of the implementations of the above aspects, wherein the memory and the processor communicate with each other through an internal connection path.

[0065] In a fourth aspect, the embodiments of the present application further provide a computer readable storage medium, which stores a computer program, when the computer program runs on a computer, the method in any of the implementations of the above aspects is implemented.

[0066] Compared with the prior art, the present application at least has the following beneficial effects by real-time sensing and predicting the production rhythm, dynamically adjusting the warehouse layout to adapt to the material demand changes, and accurately matching and coordinating the sorting task and the production demand:

[0067] (I) The self-adaptability to the production rhythm is improved: the production rhythm changes caused by factors such as quick response to production plan changes and equipment efficiency fluctuations can be responded to, the material supply strategy can be dynamically adjusted, the waiting or accumulation caused by the mismatch between material supply and production demand is significantly reduced, and the overall flow and flexibility of the production line are improved. According to preliminary simulation analysis, the average material waiting time can be reduced by 15%-30%.

[0068] (II) The utilization rate of storage space and operation efficiency are optimized: through dynamic bin management and intelligent formation of hot areas based on real-time data driving, the access path of high-frequency materials is effectively shortened, and the storage space is more reasonably allocated, which is expected to improve the comprehensive storage space utilization rate by 10%-20% and reduce the average unit material handling distance.

[0069] (III) The sorting accuracy and material turnover rate are improved: the intelligent sorting execution module combines accurate path planning and advanced identification technology to greatly reduce the probability of wrong sorting and missing sorting. At the same time, the close linkage with the production rhythm accelerates the overall turnover speed of materials and reduces the work-in-process inventory.

[0070] (IV) The deep integration of warehouse sorting and production process is realized: the present application no longer regards warehouse sorting as an isolated link, but as an organic part of the intelligent manufacturing system, realizes real-time collaboration with production plan and execution layer through data driving, which helps to improve the overall production and operation efficiency, reduce operation cost, and provides strong support for on-demand production and lean manufacturing.

[0071] In summary, the present application can improve the adaptability of the sorting system to the production rhythm, optimize the inventory turnover efficiency, and reduce the risk of production interruption caused by material mismatch or supply delay, so as to provide a comprehensive solution that can effectively combine real-time rhythm data of the production line, dynamically adjust warehouse strategy and sorting task priority, and realize accurate, efficient and adaptive sorting of materials.

[0072] The above summary is merely intended to illustrate the present application and is not intended to limit in any way. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features of the present application will be readily apparent from the drawings and detailed description below. BRIEF DESCRIPTION OF DRAWINGS

[0073] In the drawings, like reference numerals refer to same or similar functionalities throughout the several views. The drawings are not necessarily to scale. It is to be understood that the drawings only depict several embodiments in accordance with the disclosure and should not be considered to limit the scope of the disclosure.

[0074] Figure 1A flowchart of an example of a production rhythm-based dynamic warehouse intelligent sorting method according to an embodiment of the present application is shown in FIG. 1. As shown in FIG. 1, the method can include the following steps:

[0075] Figure 2 A flowchart of another example of a production rhythm-based dynamic warehouse intelligent sorting method according to an embodiment of the present application is shown in FIG. 2. As shown in FIG. 2, the method can include the following steps:

[0076] Figure 3 A block diagram of a production rhythm-based dynamic warehouse intelligent sorting system according to an embodiment of the present application is shown in FIG. 3. As shown in FIG. 3, the system can include the following components:

[0077] Figure 4 A block diagram of an electronic device according to an embodiment of the present application is shown in FIG. 4. As shown in FIG. 4, the device can include the following components: DETAILED DESCRIPTION

[0078] In the following, only certain exemplary embodiments are simply described. As those skilled in the art can recognize, the described embodiments can be modified in various different ways without departing from the spirit or scope of the present application. Therefore, the drawings and the description are considered to be exemplary in nature rather than limiting.

[0079] Figure 1 A flowchart of a production rhythm-based dynamic warehouse intelligent sorting method according to an embodiment of the present application is shown in FIG. 1. As shown in FIG. 1, the method can include the following steps: Figure 1

[0080] S110, collect real-time parameters of production rhythms of each key station on a production line, and obtain detailed information of a target material to be sorted.

[0081] In an implementation, step S110 can be performed at a preset frequency (e.g., a frequency of seconds or a frequency of minutes, such as 1 minute). The key stations on the production line can include, but are not limited to, a stamping station, a welding station, a painting station, an entry station of an assembly line, and an exit station of the assembly line.

[0082] In an implementation, the implementation process of step S110 can include the following steps:

[0083] S111, call photoelectric sensors and industrial cameras at each key station to monitor real-time product passing rates and buffer material inventory levels of each key station.

[0084] In a specific implementation, photoelectric sensors and industrial cameras can be deployed at each key station in advance to monitor real-time product passing rates and buffer material inventory levels of each key station. That is, when step S111 is performed, real-time product passing rates and buffer material inventory levels of each key station can be monitored by calling photoelectric sensors and industrial cameras at each key station.

[0085] ​The production rate of the product can be understood as core data reflecting the production rhythm in terms of the number of products. The buffer inventory level can be understood as core data reflecting the production rhythm in terms of the running state of the equipment.

[0086] In S112, the PLC controller of each key station is called to read the equipment state and real-time yield data of each key station.

[0087] In specific implementation, the PLC controller of each key station can be connected through industrial Ethernet. In S112, the PLC controller of each key station is called to read the equipment state and real-time yield data of each key station. The real-time yield data can include real-time production rate.

[0088] In S113, the information recognition channel at the warehouse entrance is called to automatically scan the unique identification code on the incoming material or the unique identification code on the incoming material box to obtain the detailed information of the incoming material.

[0089] In specific implementation, the information recognition channel can be set in advance at the warehouse entrance as an automatic material information recognition channel.

[0090] As an example, the unique identification code can be a QR code (i.e. a two-dimensional code) or an RFID tag. The type of information recognition channel can be determined according to the unique identification code on the material and the material box.

[0091] For example, if the unique identification code is an RFID tag, the information recognition channel can be an RFID read-write channel. The RFID read-write channel can be composed of an RFID reader array.

[0092] In specific implementation, the detailed information of the incoming material can include but is not limited to material ID, batch, material type, quantity, incoming time, supplier information and initial storage location.

[0093] It should be understood that by performing S111-S113, real-time online product quantity and offline rhythm data of each key station can be collected.

[0094] In S114, the ERP system is called to obtain the production plan for a future period of time.

[0095] In specific implementation, the ERP system refers to an Enterprise Resource Planning (ERP) system. The production plan for a future period of time (e.g. 8 hours in the future) can be made through the ERP system.

[0096] As an example, the production plan can include vehicle model, configuration, planned output, and target workstations, priority (which can be determined by the urgency of demand), expected delivery time window, etc. corresponding to each material.

[0097] In the above steps S111-S114, the corresponding data can be transmitted in real time through the MQTT protocol. For example, in step S111, the product passing rate and buffer material inventory level of each key workstation are transmitted in real time through the MQTT protocol, in step S112, the device status and real-time production data of each key workstation are transmitted in real time through the MQTT protocol, in step S113, the incoming material details are transmitted in real time through the MQTT protocol, and in step S114, the production plan for a future period of time is transmitted in real time through the MQTT protocol.

[0098] S115, based on the product passing rate, the buffer material inventory level, the device status and the real-time production data, obtaining the production rhythm real-time parameter.

[0099] In specific implementation, the device comprehensive efficiency (Overall Equipment Effectiveness, OEE) can be calculated based on the product passing rate, the real-time production data and the production time (i.e. the length of time corresponding to the future period of time in the above step S114), and then the OEE, the real-time production rate and the buffer material inventory level are taken as the production rhythm real-time parameter. It can be understood that the production rhythm real-time parameter includes the OEE, the real-time production rate and the buffer material inventory level.

[0100] S116, based on the incoming material details and the production plan, determining the demand of the target material to obtain the demand information.

[0101] In specific implementation, based on the incoming material details and the production plan, the demand of the target material can be determined according to the related demand that needs to be processed by the target material, and the demand information is obtained.

[0102] As an example, the demand information can include but is not limited to: material ID, demand quantity (which can be determined by the quantity in the incoming material details or the planned output in the production plan), target workstation, expected delivery time window and priority.

[0103] The target workstation can be a key workstation or a workstation other than the key workstations on the production line, and the embodiments of the present application do not limit it.

[0104] In specific implementation, step S116 and step S115 can be executed synchronously or in a corresponding order, and the embodiments of the present application do not limit it.

[0105] In the embodiment of the present application, by performing step S110, the production rhythm and material demand on the production line can be perceived in real time.

[0106] S120, based on the production rhythm real-time parameter and the demand information, calling a production prediction model to predict the production rhythm prediction parameter and the material demand prediction of each key station in a future period of time.

[0107] In an embodiment, the implementation process of step S120 can include the following steps:

[0108] S121, data cleaning and feature extraction are performed on the production rhythm real-time parameter and the demand information to obtain corresponding feature data.

[0109] In specific implementation, the way of data cleaning and feature extraction on the production rhythm real-time parameter and the demand information is the same as or similar to the existing data cleaning and feature extraction in machine learning, which will not be repeated here.

[0110] S122, a specified time series prediction model is used as the production prediction model.

[0111] In specific implementation, multiple time series prediction models can be built in, which can all be used as production prediction models. Subsequently, according to actual demand, a corresponding time series prediction model can be selected as the production prediction model.

[0112] As an example, the multiple time series prediction models can include a time series model based on LSTM (Long Short Term Memory Network) and a difference autoregressive moving average model (Autoregressive Integrated Moving Average Model, ARIMA). The ARIMA model is also called sum autoregressive moving average model.

[0113] In this example, the specified time series prediction model can be a time series model based on LSTM or an ARIMA model.

[0114] S123, based on the corresponding feature data, the specified time series prediction model is called to predict the production rhythm prediction parameter and the material demand prediction of each key station in a future period of time.

[0115] In specific implementation, the specified time series prediction model can use the existing one, that is, in the embodiment of the present application, only the input parameters of the specified time series prediction model are modified to obtain the corresponding output parameters, and the internal processing logic is not modified, so the prediction processing process of the specified time series prediction model will not be described in detail.

[0116] In a specific implementation, the future period of time in step S123 can be set according to actual needs, for example, it can be 1-4 hours in the future.

[0117] In the embodiments of the present application, by performing steps S121-S123, the time series analysis and the machine learning prediction model (such as LSTM) can be used to make short-term prediction of the production rhythm, so as to predict the production rhythm trend and material consumption rate in a future period of time.

[0118] That is, in the embodiments of the present application, by performing step S120, the production rhythm data and the ERP plan data of each key workstation on the production line are integrated to predict the production rhythm trend and material consumption rate in a future period of time, such as predicting the average production rhythm T predicted and the material demand M predicted (item, workstation, qty) of each key workstation in the next hour, wherein item represents material, workstation represents workstation number, and qty represents quantity.

[0119] S130, based on the production rhythm prediction parameter, the material demand prediction, combined with the current inventory state, the equipment availability and the preset optimization parameter, a dynamic bin adjustment plan and an optimal sorting task sequence suitable for the production line are generated.

[0120] In an embodiment, the preset optimization parameter can include, but is not limited to, at least two of the following parameters: minimizing total material waiting time, maximizing warehouse space turnover rate, minimizing total material handling distance, maximizing on-time delivery rate, minimizing inventory holding cost, minimizing average order delivery delay, and maximizing equipment utilization.

[0121] In an embodiment, the dynamic bin adjustment plan can contain the corresponding material of the target material that needs to adjust the physical location. The optimal sorting task sequence contains the target workstation corresponding to the target material.

[0122] In an embodiment, the dynamic bin adjustment plan (i.e., which materials should be moved to which bin) and the optimal sorting task sequence can be generated based on the production rhythm prediction parameter, the material demand prediction, the current inventory state, and the equipment availability by calling a specified multi-objective optimization algorithm with the preset optimization parameter as the optimization target.

[0123] As an example, the specified multi-objective optimization algorithm can be, but is not limited to, Genetic Algorithms (GA) or Simulated Annealing (SA). For example, in practical applications, the specified multi-objective optimization algorithm can also be a particle swarm algorithm or a reinforcement learning algorithm, etc.

[0124] As an example, when the specified multi-objective optimization algorithm is a simulated annealing algorithm, and the preset optimization parameters include minimizing the average order delivery delay and maximizing the equipment utilization, the simulated annealing algorithm can be used to solve the optimization target of minimizing the average order delivery delay and maximizing the equipment utilization.

[0125] The algorithm input includes: production rhythm prediction parameter T predicted, material demand prediction M predicted, current real-time inventory S current (item, location), robot / shuttle status R status.

[0126] The algorithm output: future one-hour bin adjustment plan L_adjust (item, from_loc, to_loc, priority) and sorting task sequence P_task (item, qty, from_loc, to_workstation, sequence_no, due_time).

[0127] That is, in this example, the current real-time inventory S current (item, location) is used to represent the current inventory state, and the robot / shuttle status R status is used to represent the equipment availability. The future one-hour bin adjustment plan L_adjust (item, from_loc, to_loc, priority) is a dynamic bin adjustment plan, and the sorting task sequence P_task (item, qty, from_loc, to_workstation, sequence_no, due_time) is an optimal sorting task sequence. Wherein, item represents material, from_loc represents source bin (i.e. current bin), to_loc represents target bin, priority represents task priority, qyt represents quantity, to_workstation represents sorting position, sequence_no represents sorting task sequence number, and due_time represents expected time.

[0128] As another example, when the preset optimization parameter includes minimizing total material waiting time and maximizing warehouse space turnover rate, and the specified multi-objective optimization algorithm is a genetic algorithm, the optimization target can be to minimize total material waiting time and maximize warehouse space turnover rate, and the production cycle prediction parameter, the material demand prediction quantity, the actual storage location of each material in the current warehouse (which can be obtained by interacting with the warehouse control system (WCS)), the ABC classification of the material, the availability and movement cost of the flexible scheduling mechanism in the production environment are comprehensively considered, and a dynamic storage location adjustment plan and an optimal sorting task sequence are generated every certain period of time (for example, 15 minutes).

[0129] That is, in this example, the actual storage location of each material in the current warehouse is used to represent the current inventory status, and the availability and movement cost of the flexible scheduling mechanism in the production environment are used to represent the equipment availability.

[0130] In the embodiments of the present application, the current and predicted production demand, inventory status, and equipment availability are comprehensively evaluated by performing step S130, so as to dynamically generate a dynamic storage location adjustment plan and an optimal sorting task sequence.

[0131] S140, based on the dynamic storage location adjustment plan, scheduling the flexible scheduling mechanism in the production environment to dynamically adjust the physical location of the corresponding material.

[0132] In an embodiment, the flexible scheduling mechanism can include but is not limited to a plurality of shuttle vehicles and a plurality of elevators that can shuttle in four directions. The specific number of shuttle vehicles and elevators can be set according to actual needs, and the embodiments of the present application do not limit this.

[0133] In an embodiment, the implementation process of step S140 can include the following steps:

[0134] S141, sending the dynamic storage location adjustment plan to the WCS, and calling the WCS to determine the current storage location, target storage location, and storage location adjustment time of the corresponding material based on the dynamic storage location adjustment plan.

[0135] In specific implementation, the priority of the corresponding material can be determined according to the demand information in step S110, and then the dynamic storage location adjustment plan corresponding to the corresponding material with high priority is sent to the WCS first. After receiving the dynamic storage location adjustment plan, the WCS determines the current storage location, target storage location, and storage location adjustment time of the corresponding material based on the dynamic storage location adjustment plan.

[0136] As an example, after receiving the dynamic storage location adjustment plan, the WCS identifies that the A-class material X (such as a high-value or key component) corresponding to the dynamic storage location adjustment plan is currently located in the cold zone C-03-05, but the production tact prediction parameter of the target work station W01 corresponding to the A-class material X will be greatly increased. It can be determined that the current storage location of the A-class material X is the cold zone C-03-05, the target storage location is a fast cache area (such as the hot cache area H-01-01), and the storage location adjustment time is 30 seconds (which can be determined according to the increase of the production tact prediction parameter of the target key work station W01).

[0137] S142, send a storage location adjustment instruction to the WCS, instructing the WCS to schedule a flexible scheduling mechanism to move the corresponding material from the current storage location to the target storage location within the storage location adjustment time, so as to complete the dynamic adjustment of the geographical position of the corresponding material.

[0138] In a specific implementation, the WCS can feed back the current storage location, target storage location, and storage location adjustment time of the corresponding material determined according to the dynamic storage location adjustment plan. After receiving the feedback information of the WCS, a storage location adjustment instruction can be sent to the WCS.

[0139] As an example, after receiving the storage location adjustment instruction, the WCS schedules a flexible scheduling mechanism to move the A-class material X from the cold zone C-03-05 to the hot cache area H-01-01 within 30 seconds. The flexible scheduling mechanism can be the flexible scheduling mechanism closest to the cold zone C-03-05 in the production environment. For example, a shuttle vehicle and an elevator closest to the cold zone C-03-05 can be scheduled to work cooperatively to move the A-class material X from the cold zone C-03-05 to the hot cache area H-01-01 within 30 seconds, so as to quickly complete the automatic movement of the A-class material X.

[0140] In the embodiments of the present application, by performing step S140, the material with high frequency demand can be migrated from the backup storage area to the fast picking area or the hot cache area (both the fast picking area and the hot cache area belong to the fast cache area) close to the corresponding production work station, and vice versa, the material with low demand frequency in the short term can be moved to the slow cache area (such as the slow picking area or the cold zone), so as to realize the dynamic caching of the material, the pre-allocation of the hot material to the high-frequency access area, the centralized storage of the cold material, and the like, thereby forming a warehouse layout dynamically matched with the production tact, and optimizing the utilization of the storage space and improving the material picking efficiency. That is, step S140 is used to realize the flexible storage and dynamic allocation of the material.

[0141] S150, based on the optimal picking task sequence, calling a picking mechanism in the production environment to pick a specified number of target materials from the dynamically adjusted storage location and then pick them to the corresponding target work station.

[0142] In an embodiment, the sorting mechanism can include, but is not limited to, a high-speed sorting device (such as a cross-belt sorter, a carousel sorter, a robotic sorting arm) or a robotic system (including several robots, which include AGVs (intelligent transport robots)). It should be understood that the sorting mechanism is equipped with a 3D vision guidance system and an intelligent path guidance system (both systems are available), to achieve material identification and precise positioning.

[0143] In an embodiment, the optimal sorting task sequence can include, but is not limited to, an optimal sorting operation sequence and a best path planning of target material transport.

[0144] In an embodiment, the implementation process of step S150 can include the following steps:

[0145] S151, according to the optimal sorting operation sequence of the optimal sorting task sequence, send sorting instructions to the sorting mechanism in sequence, to instruct the sorting mechanism to pick up a specified number of target materials from the dynamically adjusted storage locations according to the received sorting instructions, and then sort the picked target materials to the specified positions of the corresponding target workstations in sequence.

[0146] In specific implementation, the specified position can be a sorting chute or an AGV corresponding to the target workstation, so that the target materials can be conveniently distributed.

[0147] S152, according to the optimal path planning of the optimal sorting task sequence, call the sorting mechanism to distribute the sorted target materials from the specified position to the corresponding target workstation.

[0148] In specific implementation, according to the optimal path planning of the optimal sorting task sequence, the sorting mechanism can be called to distribute the sorted target materials from the specified area to the sorting buffer of the target key workstation, such as a specified buffer, a specific JIS (Just-In-Sequence) sorting compartment, a line-side warehouse or an out-of-warehouse assembly point (such as a direct out-of-warehouse port).

[0149] As an example, taking the sorting instruction of picking 3 pieces of material A (target material) from the H-01-01 storage location and delivering to the sequencing buffer channel of the assembly line Z03 station (target station) as an example, in this case, the WCS has dispatched a flexible scheduling mechanism (such as a four-way shuttle vehicle) to move the material box containing material A to the H-01-01 storage location. After receiving the sorting instruction, the sorting mechanism calls the robot sorting arm R01 to pick 3 pieces of material A from the H-01-01 storage location and sort them into the sorting chute or AGV corresponding to the assembly line Z03 station. At this time, the vision secondary identification can be used to confirm whether the 3 pieces of material picked are 3 pieces of material A. After confirming that the 3 pieces of material are indeed 3 pieces of material A, the sorting mechanism (such as AGV) is called to deliver the 3 pieces of material A to the sequencing buffer channel of the assembly line Z03 station according to the optimal path planning of the optimal sorting task sequence.

[0150] In step S152, when it is confirmed that the target material sorting is completed, the sorting completion time can be recorded, and the material inventory status can be updated.

[0151] In the embodiments of the present application, by performing step S150, the target material can be sorted, sequenced (if JIS supply is required), and delivered to the sequencing buffer channel matched with the target key station in sequence and accurately according to the production rhythm requirements of each key station.

[0152] In an applicable scenario provided by the embodiments of the present application, as shown in Figure 1 and Figure 2 The dynamic warehouse intelligent sorting method based on production rhythm provided by the embodiments of the present application can further include the following steps:

[0153] S160, real-time monitoring of the actual flow of the target material and the actual production rhythm change of the production line to obtain a real-time monitoring result.

[0154] In an embodiment, the core functions of the warehouse management system (WMS) and the WCS can be built in, and the two-way data interaction with the manufacturing execution system (MES) can be performed through a unified API interface to realize the issuance of instructions and the collection of states. In this case, the system state visualization interface (such as a Web-based digital twin board) can be provided to real-time display the material flow, equipment state, inventory heat map, production rhythm matching degree, and other key performance indicators (KPIs) of the entire warehouse sorting area, and through monitoring these data, the actual flow of the target material and the actual production rhythm change of the production line can be realized.

[0155] The production rhythm matching degree can be represented by a deviation between the production rhythm real-time parameter and the production rhythm prediction parameter. The smaller the deviation, the higher the production rhythm matching degree, and the larger the deviation, the lower the production rhythm matching degree.

[0156] S170, when it is determined that the production is abnormal based on the real-time monitoring result, triggering an alarm, and returning to step S120.

[0157] As a first example, when it is determined that the production rhythm deviates from the prediction significantly based on the real-time monitoring result (which may be caused by sudden equipment failure or order emergency change), it can be determined that the production is abnormal. For example, when the real-time monitoring result contains a monitoring result that the deviation between the production rhythm real-time parameter T_actual and the production rhythm prediction parameter T_predicted (|T_actual-T_predicted| / T_predicted) is greater than a preset threshold (for example, 15%) for more than a preset time (for example, 5 minutes), it is determined that the production rhythm deviates from the prediction significantly, and it is determined that the production is abnormal. The preset threshold and the preset time can be set according to actual needs, and the embodiments of the present application are not limited thereto.

[0158] As a second example, when it is determined that an exception occurs in the sorting execution process based on the real-time monitoring result, it can be determined that the production is abnormal. For example, when the real-time monitoring result contains monitoring results such as material identification error, path blockage, etc., it is determined that the production is abnormal.

[0159] As a third example, when it is determined that a key device (such as a certain shuttle vehicle) is stopped due to failure for more than a specified time (such as 3 minutes) based on the real-time monitoring result, it can be determined that the production is abnormal. The specified time can be set according to actual needs, and the embodiments of the present application are not limited thereto.

[0160] In an embodiment, in combination with the above three examples and possible application scenarios on the production line, it can be concluded that when the real-time monitoring result contains the following monitoring situations, it is determined that the production is abnormal:

[0161] (1) The production rhythm deviates from the prediction significantly: the actual value of the production rhythm deviates from the predicted value by more than a preset threshold.

[0162] (2) Abnormal material state: such as material shortage, quality problem or position error.

[0163] (3) Device state change: such as AGV failure, charging demand or passage blockage.

[0164] (4) Execution exception: sorting task execution failure or serious delay.

[0165] (5) Periodic optimization: reaching a specified optimization time length (in this case, routine optimization is performed according to a preset time interval).

[0166] The above five monitoring conditions can be used as a closed-loop adaptive adjustment triggering condition to trigger an alarm.

[0167] In an embodiment, when it is determined based on the real-time monitoring result that a production abnormality occurs, an alarm is triggered, abnormal information and current system snapshot data are fed back, and then step S120 is returned to execute. In this way, an emergency re-planning (i.e., returning to execute step S120) can be forcibly triggered, so that the new system state and production constraints can be adapted.

[0168] In another embodiment, when it is determined based on the real-time monitoring result that a production abnormality occurs, an alarm is triggered, and the abnormality is also processed, and then step S120 is returned to execute.

[0169] For example, in S150, if the robot sorting arm R01 fails to grab the material Y from the bin location C-03-05 due to an abnormal material posture, the vision system will attempt to adjust the grabbing posture several times (e.g., 2 times). If it still fails, it will report abnormal information (e.g., material ID, bin location, abnormal type) for feedback. At this time, the real-time monitoring result can include the following monitoring condition: the material Y (material ID) in the bin location C-03-05 fails to be grabbed due to an abnormal material posture (abnormal type).

[0170] Based on the monitoring condition, the following abnormality processing can be performed:

[0171] 1) Suspend other tasks involving this material;

[0172] 2) instruct another standby robot or manual intervention for processing;

[0173] 3) if the material has a substitute and the production is urgent, consider using the substitute when returning to execute S120 to make a decision.

[0174] In the embodiments of the present application, by executing steps S160 and S170, the decision optimization process can be dynamically adjusted or re-executed when a production abnormality occurs, realizing closed-loop adaptive adjustment of warehouse sorting, and ensuring that the production environment can be continuously adapted to changes.

[0175] For ease of understanding, the closed-loop adaptive adjustment process in the dynamic warehouse intelligent sorting method based on production rhythm provided in the embodiments of the present application will be further described below in combination with an automobile chassis assembly production line.

[0176] Assume that in a certain automobile chassis assembly production line, it is originally planned to produce A type chassis at a rate of 200 pieces per hour, and the material sorting strategy optimization has been carried out based on this production rhythm. But in the production process, due to temporary quality problems in the upstream process, the actual production rate is reduced to 120 pieces per hour, and the production task of B type chassis (80 pieces per hour) is inserted urgently. In this case, the closed-loop adaptive adjustment corresponding to the automobile chassis assembly production line can include the following steps:

[0177] Step 1: Change detection and anomaly identification

[0178] In step 1, real-time monitoring data can be as shown in Table 1 below.

[0179] Table 1

[0180]

[0181] At this time, the photoelectric sensor installed at the key station of the production line can monitor the significant change of the actual production rate in real time, and the MES system pushes the work order change information. By comparing the actual production rhythm with the predicted production rhythm, it is found that the deviation exceeds the preset threshold of 15%, and the closed-loop adjustment mechanism is automatically triggered.

[0182] Step 2: Snapshot saving and state evaluation

[0183] In step 2, the related content of the system state snapshot can be as shown in Table 2 below.

[0184] Table 2

[0185]

[0186] When the closed-loop adjustment mechanism is automatically triggered, a complete snapshot of the current running state can be captured immediately, including all sorting tasks being executed and waiting, AGV position and state, inventory situation, etc. At the same time, the interruption cost of the task being executed and the influence of continuing execution are evaluated.

[0187] Step 3: Re-execute step S120 decision optimization process

[0188] The input parameters of the above step S120 decision re-execution can be as shown in Table 3 below.

[0189] Table 3

[0190]

[0191] The execution process of the above step S120 decision re-execution can include real-time recalculation of time series prediction model and re-optimization of dynamic sorting strategy engine.

[0192] The related process of real-time recalculation of the time series prediction model is as follows:

[0193] Input the changed production data;

[0194] Update the beat bias matrix B = 0.5·F_workorder' + 0.3·F_equipment' + 0.2·F_production';

[0195] Recalculate the material demand time series distribution of the next 2 hours.

[0196] The related process of re-optimization of the dynamic sorting strategy engine is as follows:

[0197] Recalculate the priority score of all materials to be sorted;

[0198] A-type material priority is reduced: Score_A = 0.6·60 + 0.3·40 + 0.1·70 = 56.0;

[0199] B-type material priority is increased: Score_B = 0.6·85 + 0.3·50 + 0.1·30 = 70.5;

[0200] Generate a new sorting task queue and execution plan.

[0201] That is, in step 3, the above S120 decision optimization process can be re-run according to the newly detected production beat and work order changes. The time series prediction model receives the updated data, recalculates the material demand prediction of B-type chassis, and takes it into comprehensive consideration. At the same time, the dynamic sorting strategy engine re-evaluates the priority of all tasks to be executed, and the B-type material obtains a higher priority due to the urgent insertion order.

[0202] Step 4: Dynamic adjustment of storage location

[0203] In step 4, the related content of the storage location dynamic adjustment instruction can be as shown in Table 4.

[0204] Table 4

[0205]

[0206] In step 4, the calculation results of the hot area reallocation can be as shown in Table 5.

[0207] Table 5

[0208]

[0209] That is, in step 4, based on the new material demand prediction and priority score, a dynamic adjustment instruction of the storage location can be generated. The B-type chassis special parts originally placed in the remote storage location are urgently scheduled to the high-frequency area close to the production line, and part of the A-type parts are moved from the high-frequency area to the medium-frequency area, realizing dynamic optimization allocation of storage location resources.

[0210] Step 5: Task queue rearrangement and resource reallocation

[0211] In step 5, the task queue rearrangement result can be shown in Table 6 as follows.

[0212] Table 6

[0213]

[0214] In step 5, the resource reallocation result can be shown in Table 7 as follows.

[0215] Table 7

[0216]

[0217] That is, in step 5, according to the new priority score, all to-be-executed and newly added sorting tasks can be reordered. For the tasks that have already started to be executed, unless the interruption cost is particularly low, they are allowed to continue to be completed. At the same time, the AGV resources are reallocated, part of the AGVs are scheduled to the new high-priority B-type chassis material sorting task, and the originally standby AGVs are activated to increase the processing capacity.

[0218] Step 6: Abnormal processing and manual cooperation

[0219] In step 6, if it is found in the re-planning process that part of the key parts required by the B-type chassis are in short supply, which may cause production interruption, an automatic material shortage exception handling process can be triggered, and a manual cooperation task is generated. The AR auxiliary system immediately provides task guidance for the warehouse operator to help him quickly find and pick the short supply materials in the standby warehouse.

[0220] Exemplarily, in step 6, the abnormal processing strategy can be shown in Table 8 as follows.

[0221] Table 8

[0222]

[0223] Step 7: Execution monitoring and effect evaluation

[0224] In step 7, the execution monitoring data can be shown in Table 9 as follows.

[0225] Table 9

[0226]

[0227] That is, in step 7, the execution effect of the closed-loop adaptive adjustment can be continuously monitored, and key performance indicators can be collected in real time. Data shows that through timely adjustment, the challenges brought by production rhythm changes have been successfully coped with, high material on-time delivery rate has been maintained, and AGV empty rate has been effectively reduced. At the same time, new problems such as AGV power shortage can be continuously monitored, and corresponding processing suggestions can be generated.

[0228] Step 8: Parameter self-learning update (optional)

[0229] In step 8, the decision parameters can be updated through reinforcement learning mechanism based on the actual effect of closed-loop adaptive adjustment. The reward value is calculated according to the factors such as successfully coping with production changes and maintaining high on-time delivery rate, and the priority score weight and abnormality detection threshold are adjusted accordingly. At the same time, the model performance indicators are recorded and evaluated to provide data support for subsequent algorithm optimization.

[0230] For example, in step 8, the parameters updated by reinforcement learning can be as shown in Table 10 below.

[0231] Table 10

[0232]

[0233] In summary, through dynamic adjustment or re-execution of the decision optimization process, closed-loop adaptive adjustment of warehouse sorting can be achieved to ensure that it can continuously adapt to changes in the production environment.

[0234] As can be seen from the above description, compared with the prior art, the dynamic warehouse intelligent sorting method based on production rhythm provided by the embodiments of the present application can achieve the following beneficial effects:

[0235] (I) Improve the adaptive ability to production rhythm: quickly respond to production rhythm changes caused by production plan changes, equipment efficiency fluctuations and other factors, dynamically adjust material supply strategy, significantly reduce waiting or accumulation caused by mismatch between material supply and production demand, and improve the overall smoothness and flexibility of the production line. According to preliminary simulation analysis, the average material waiting time can be reduced by 15%-30%.

[0236] (II) Optimize warehouse space utilization and operation efficiency: through dynamic bin location management and intelligent formation of hot areas based on real-time data driving, the access path of high-frequency materials is effectively shortened, and the warehouse space is more reasonably allocated, which is expected to improve the comprehensive warehouse space utilization rate by 10%-20% and reduce the average unit material handling distance.

[0237] (Three) improve the sorting accuracy and material turnover rate: intelligent sorting execution module combined with accurate path planning and advanced recognition technology, greatly reduces the probability of wrong sorting and missing sorting. At the same time, the close linkage with the production rhythm accelerates the overall turnover speed of the material, reduces the work-in-process inventory backlog.

[0238] (Four) realize the deep integration of warehouse sorting and production process: this application no longer regards warehouse sorting as an isolated link, but as an organic part of the intelligent manufacturing system, realizes real-time collaboration with production planning and execution through data driving, which helps to improve the overall production and operation efficiency, reduce operation cost, and provides strong support for on-demand production and lean manufacturing.

[0239] In summary, the dynamic warehouse intelligent sorting method based on production rhythm provided by the embodiment of the application can improve the adaptability of the sorting system to the production rhythm, optimize the inventory turnover efficiency, and reduce the risk of production interruption caused by material mismatch or supply delay, so as to provide a comprehensive solution that can effectively combine real-time rhythm data of the production line, dynamically adjust warehouse strategy and sorting task priority, and realize accurate, efficient and adaptive sorting of materials.

[0240] Figure 3 The structure block diagram of the dynamic warehouse intelligent sorting system based on production rhythm according to an embodiment of the application is shown. As shown in the figure, the system can include: Figure 3

[0241] The production rhythm sensing module 10 is used to collect the real-time parameters of the production rhythm of each key station on the production line, and obtain the demand information of the target material to be sorted;

[0242] The data fusion and decision processing module 20 is used to predict the production rhythm prediction parameters and material demand prediction of each key station in a future period of time based on the real-time parameters of the production rhythm, the demand information, and the production prediction model; based on the production rhythm prediction parameters and the material demand prediction, combined with the current inventory state, the device availability and the preset optimization parameters, a dynamic bin adjustment plan suitable for the production line and an optimal sorting task sequence are generated, wherein the dynamic bin adjustment plan contains the corresponding material in the target material that needs to adjust the physical location, and the optimal sorting task sequence contains the target station corresponding to the target material.

[0243] The dynamic adaptive warehouse module 30 is used to dynamically adjust the physical location of the corresponding material based on the dynamic bin adjustment plan, and schedule the flexible scheduling mechanism in the production environment.

[0244] The intelligent sorting execution module 40 is used to call the sorting mechanism in the production environment to pick a specified number of target materials from the dynamically adjusted bin based on the optimal sorting task sequence, and then sort them into the corresponding target station.​

[0245] In an embodiment, the production rhythm perception module 10, when used for collecting production rhythm real-time parameters of each key station on the production line and obtaining demand information of target materials to be sorted, specifically is used for: calling photoelectric sensors and industrial cameras at each key station to monitor the product passing rate and the buffer material inventory level of each key station in real time; calling PLC controllers at each key station to read the device status and real-time yield data of each key station; calling an information recognition channel at the warehouse entrance to automatically scan the unique identification code label on the incoming material or the unique identification code on the incoming material box to obtain detailed information of the incoming material; calling an ERP system to obtain a production plan for a future period of time; obtaining production rhythm real-time parameters based on the product passing rate, the buffer material inventory level, the device status, and the real-time yield data; and determining the demand of the target material based on the detailed information of the incoming material and the production plan to obtain the demand information.

[0246] As an example, as shown in Figure 3 The production rhythm perception module 10 can be used as a perception layer of a dynamic warehouse intelligent sorting system based on production rhythm, which can include a production line real-time data acquisition interface 11 and a material information automatic identification unit 12. The production line real-time data acquisition interface 11 is used to communicate with photoelectric sensors, industrial cameras, and PLCs at each key station to monitor the product passing rate and the buffer material inventory level of each key station in real time, read the device status and real-time yield data of each key station, and obtain a production plan for a future period of time. The material information automatic identification unit 12 is used to communicate with an information recognition channel at the warehouse entrance to automatically scan the unique identification code label on the incoming material or the unique identification code on the incoming material box to obtain detailed information of the incoming material.

[0247] That is, the production rhythm perception module 10 can integrate, for example, an RFID reader array, a machine vision sensor, a PLC data interface, and a communication interface with a MES / ERP system.

[0248] In an embodiment, the data fusion and decision processing module 20, when used for calling a production prediction model to predict production rhythm prediction parameters and material demand prediction quantities of each key station in a future period of time based on production rhythm real-time parameters and demand information, specifically is used for: performing data cleaning and feature extraction on the production rhythm real-time parameters and the demand information to obtain corresponding feature data; using a specified time series prediction model as the production prediction model; and calling the specified time series prediction model to predict the production rhythm prediction parameters and the material demand prediction quantities of each key station in the future period of time based on the corresponding feature data.

[0249] In one implementation, when the data fusion and decision processing module 20 generates a dynamic storage location adjustment plan and an optimal sorting task sequence suitable for the current production line based on production cycle prediction parameters, material demand prediction quantities, current inventory status, equipment availability, and preset optimization parameters, it specifically uses the preset optimization parameters as the optimization objective to call a specified multi-objective optimization algorithm to generate a dynamic storage location adjustment plan and an optimal sorting task sequence based on production cycle prediction parameters, material demand prediction quantities, current inventory status, and equipment availability.

[0250] As an example, such as Figure 3 As shown, the data fusion and decision processing module 20 can serve as the decision layer of a dynamic intelligent warehousing and sorting system based on production cycle time. It can include a production cycle time prediction unit 21 and a dynamic scheduling optimization algorithm engine 22. Specifically, the production cycle time prediction unit 21 uses a production prediction model to predict the production cycle time parameters and material demand forecasts for each key workstation over a future period, based on real-time production cycle time parameters and demand information. The dynamic scheduling optimization algorithm engine 22 uses the production cycle time prediction parameters and material demand forecasts, combined with current inventory status, equipment availability, and preset optimization parameters, to generate a dynamic warehouse location adjustment plan and optimal sorting task sequence suitable for the current production line.

[0251] In one implementation, the dynamic adaptive warehousing module 30, when used to dynamically adjust the physical location of corresponding materials based on a dynamic storage location adjustment plan and by scheduling a flexible scheduling mechanism in the production environment, specifically performs the following: sending a dynamic storage location adjustment plan to the WCS; calling the WCS to determine the current storage location, target storage location, and storage location adjustment time of the corresponding materials based on the dynamic storage location adjustment plan, wherein the WCS refers to the warehouse control system; and sending a storage location adjustment instruction to the WCS, instructing the WCS to schedule a flexible scheduling mechanism to move the corresponding materials from the current storage location to the target storage location within the storage location adjustment time, so as to complete the dynamic adjustment of the geographical location of the corresponding materials.

[0252] In one implementation, the optimal sorting task sequence includes an optimal sorting operation sequence and an optimal path planning for the handling of target materials. When the intelligent sorting execution module 40, based on the optimal sorting task sequence, calls upon the sorting mechanism in the production environment to pick a specified quantity of target materials from dynamically adjusted storage locations and then sorts them to the corresponding target workstations, it specifically performs the following: It sends sorting instructions to the sorting mechanism sequentially according to the optimal sorting operation sequence of the optimal sorting task sequence, instructing the sorting mechanism to retrieve a specified quantity of target materials from the dynamically adjusted storage locations according to the received sorting instructions, and then sequentially sorts the retrieved target materials to the specified positions of the corresponding target workstations; It calls upon the sorting mechanism to deliver the sorted target materials from the specified positions to the corresponding target workstations according to the optimal path planning of the optimal sorting task sequence.

[0253] As an example, as shown in Figure 3 The dynamic adaptive warehouse module 30 and the intelligent sorting execution module 40 can serve as the execution layer of the production rhythm-based dynamic warehouse intelligent sorting system. The dynamic adaptive warehouse module 30 can include a reconfigurable modular storage unit 31 and a flexible scheduling execution mechanism 32. The intelligent sorting execution module 40 can include a high-speed automatic sorting device 41 and a path guidance and material tracking system 42.

[0254] The reconfigurable modular storage unit 31 can include, but is not limited to, a modular shelf, an adjustable height storage location, and a four-way shuttle plate shelf system. The flexible scheduling mechanism 32 is responsible for accurately and efficiently completing material sorting according to instructions and can include multiple shuttle vehicles and multiple elevators, and can also include collaborative mobile robots AMR. The high-speed automatic sorting device 41 can include cross-belt sorters, gyro wheel sorters, robotic sorting arms, etc., and the path guidance and material tracking system 42 can include a robotic system including several robots (including AGVs).

[0255] In an embodiment, the production rhythm-based dynamic warehouse intelligent sorting system further includes a central coordination and monitoring module 50, which is configured to: monitor the actual flow of target materials and the actual production rhythm changes of the production line in real time to obtain real-time monitoring results; when a production abnormality is detected based on the real-time monitoring results, trigger an alarm, and return the production rhythm prediction parameters and the material demand prediction amount of each key station in a future period of time predicted by the data fusion and decision processing module 20 based on production rhythm real-time parameters and demand information, and calling a production prediction model.

[0256] It should be understood that in the production rhythm-based dynamic warehouse intelligent sorting system, the data of the production rhythm perception module 10 is the input of the data fusion and decision processing module 20; the instructions of the data fusion and decision processing module 20 drive the dynamic adaptive warehouse module 30 and the intelligent sorting execution module 40; the execution status of the dynamic adaptive warehouse module 30 and the intelligent sorting execution module 40 is fed back to the central coordination and monitoring module 50, and can be further fed back to the data fusion and decision processing module 20 to form a closed-loop control.

[0257] As an example, the central coordination and monitoring module 50 can be developed based on a micro-service architecture to contain the core functions of WMS and WCS, and interact with the MES through a unified API interface for bidirectional data exchange, for responsible for accurately and efficiently completing material sorting according to instructions, for responsible for coordinating the synchronous and efficient collaborative work of the above-mentioned production rhythm perception module 10, data fusion and decision processing module 20, dynamic adaptive warehousing module 30 and intelligent sorting execution module 40, and through real-time monitoring of system running state, detecting abnormal conditions, and dynamically adjusting or re-executing decision optimization process through data fusion and decision processing module 20 according to feedback data, to ensure that the system can continuously adapt to changes in the production environment. This closed-loop adaptive adjustment mechanism can make the entire production rhythm-based dynamic warehousing intelligent sorting system form a complete PDCA cycle (Plan-Do-Check-Act), realizing intelligent self-optimization.

[0258] As shown in the example, Figure 3 The central coordination and monitoring module 50 can serve as the monitoring layer of the production rhythm-based dynamic warehousing intelligent sorting system, which can include a system state visualization interface 51 and a collaborative control and bus 52. That is, the system state visualization interface 51 is used to display the material flow, equipment state, inventory heat map, production rhythm matching degree and other key performance indicators of the entire warehousing sorting area in real time, and the collaborative control and bus 52 is used for bidirectional data interaction of other modules. The central coordination and monitoring module 50 can also provide necessary human-computer interaction interfaces through the system state visualization interface 51 and the collaborative control and bus 52 to support manual intervention and system parameter adjustment.

[0259] As an example, assuming that the production rhythm of the paint coating station (P01) of a certain model of the production line is expected to increase by 20% in the next 1 hour due to order adjustment, resulting in a significant increase in the demand for a specific color paint (material X), then each module in the production rhythm-based dynamic warehousing intelligent sorting system can execute the following processing logic:

[0260] 1. The production rhythm perception module 10 obtains the plan adjustment information through the MES interface, and detects that the work-in-process passing rate of the P01 station has an upward trend through its own sensors.

[0261] 2. The production rhythm prediction unit 21 of the data fusion and decision processing module 20 confirms this rhythm change trend, and the dynamic scheduling optimization algorithm engine 22 judges that the current storage of material X in the backup area is insufficient to meet the predicted demand, and its current bin picking path is long.

[0262] 3. Data fusion and decision processing module 20 generates instructions: a) Dynamically migrate the most recent batch of material X (partial or all) in the inventory from the more distant D-10-05 storage location to the fast buffer area F-02 near the exit of P01 workstation; b) Increase the priority of material X sorting tasks for P01 workstation within the next hour.

[0263] 4. The central coordination and monitoring module 50 schedules the shuttle of the dynamic adaptive warehousing module 30 to perform the material X transfer operation. At the same time, the intelligent sorting execution module 40 will prioritize the sorting request of material X sent to workstation P01 to ensure that there is sufficient material reserve in the line-side buffer next to workstation P01 before the actual peak of production demand arrives.

[0264] The functions of each module in the dynamic intelligent warehousing sorting system based on production cycle time in this application embodiment can be found in the corresponding description in the above method, and will not be repeated here.

[0265] Figure 4 A structural block diagram of an electronic device according to an embodiment of this application is shown. Figure 4 As shown, the electronic device includes a memory 310 and a processor 320. The memory 310 stores instructions, which are loaded and executed by the processor 320 to implement the dynamic warehouse intelligent sorting method based on production cycle time in the above embodiment. The number of memories 310 and processors 320 can be one or more.

[0266] The electronic device also includes:

[0267] The communication interface 330 is used to communicate with external devices and perform data exchange and transmission.

[0268] If the memory 310, processor 320, and communication interface 330 are implemented independently, they can be interconnected via a bus to communicate with each other. This bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0269] Optionally, if the memory 310, the processor 320 and the communication interface 330 are integrated on a chip, the memory 310, the processor 320 and the communication interface 330 can complete the communication among each other through an internal interface.

[0270] The embodiment of the present application provides a computer readable storage medium, and the computer readable storage medium stores a computer program. When the computer program runs on a computer, the method provided in the embodiment of the present application is implemented.

[0271] The embodiment of the present application further provides a chip, which comprises a processor, and the processor is used for calling and running instructions stored in a memory, so that a communication device installed with the chip executes the method provided in the embodiment of the present application.

[0272] The embodiment of the present application further provides a chip, which comprises an input interface, an output interface, a processor and a memory, and the input interface, the output interface, the processor and the memory are connected through internal connection paths. The processor is used for executing code in the memory. When the code is executed, the processor is used for executing the method provided in the embodiment of the present application.

[0273] It should be understood that the processor described above can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. It should be noted that the processor can be a processor supporting an advanced RISC machine (ARM) architecture.

[0274] Further, the aforementioned memory can include a read-only memory, and a random access memory, and can further include a non-volatile random access memory. The memory can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memory. The non-volatile memory can include a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically EPROM (EEPROM), or a flash memory. The volatile memory can include a random access memory (RAM), which functions as an external cache. Many forms of RAM are available, by way of example and not limitation. For example, a static RAM (SRAM), a dynamic RAM (DRAM), a synchronous DRAM (SDRAM), a double data rate synchronous DRAM (DDR SDRAM), an enhanced SDRAM (ESDRAM), a synchlink DRAM (SLDRAM), and a direct rambus RAM (DR RAM) can be used.

[0275] In the above-described embodiments, all or a part can be implemented by software, hardware, firmware, or any combination thereof. When implemented as software, it can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions according to the present disclosure are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another computer-readable storage medium.

[0276] In the description of the application, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the application. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in one or more embodiments or examples. In addition, different embodiments or examples described in the specification and characteristics of different embodiments or examples can be combined and combined by those skilled in the art without contradiction.

[0277] In addition, the terms "first", "second" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0278] Any process or method descriptions in flow charts or described elsewhere herein can be understood as representing code modules, segments, or portions of code that include one or more executable instructions for implementing specific logic functions or other processes. And the various embodiments of the application can include additional or fewer steps or processes in comparison to those shown in the figures.

[0279] The logic and / or steps represented in flow charts or otherwise described herein, for example, can be embodied in computer-readable instructions, which can be used to cause one or more processors to perform the actions indicated in the logic flow. The logic and / or steps represented in flow charts or otherwise described herein, for example, can be embodied in computer-readable instructions, which can be used to cause one or more processors to perform the actions indicated in the logic flow.

[0280] It should be understood that various parts of the application can be implemented in hardware, software, firmware or a combination thereof. In the above-described embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. All or part of the steps of the above-described embodiment method can be instructed by the relevant hardware through a program, which can be stored in a computer-readable storage medium, and the program includes one or a combination of the steps of the method embodiment when executed.

[0281] In addition, each of the function units in each embodiment of the present application can be integrated in one processing module, or each unit can be physically present separately, or two or more units can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of a software function module. When the integrated module is realized in the form of a software function module and sold or used as an independent product, it can also be stored in a computer readable storage medium. The storage medium can be a read-only memory, a magnetic disk or an optical disk, etc.

[0282] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of various changes or replacements within the technical scope disclosed in the present application, and these should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A dynamic warehouse intelligent sorting method based on production rhythm, characterized in that, The method comprises the following steps: calling photoelectric sensors and industrial cameras at key stations on a production line to monitor the passing rate of products at each key station and the buffer inventory level in real time; calling PLC controllers at each key station to read the equipment status and real-time production data of each key station; calling an information identification channel at the warehouse entrance to automatically scan the unique identification code label on the incoming material or the unique identification code on the incoming material box to obtain detailed information of the incoming material; calling an ERP system to obtain the production plan for a future period of time; based on the passing rate of the products, the buffer inventory level, the equipment status, and the real-time production data, obtaining real-time parameters of the production rhythm of each key station; based on the detailed information of the incoming material and the production plan, determining the demand for target materials to be sorted to obtain demand information for the target materials; performing data cleaning and feature extraction on the real-time parameters of the production rhythm and the demand information to obtain corresponding feature data; using a specified time series prediction model as a production prediction model; based on the corresponding feature data, calling the specified time series prediction model to predict the production rhythm prediction parameters and material demand prediction values of each key station in a future period of time; taking preset optimization parameters as optimization targets, calling a specified multi-objective optimization algorithm to generate a dynamic bin adjustment plan and an optimal sorting task sequence suitable for the production line based on the production rhythm prediction parameters, the material demand prediction values, the current inventory state, and the equipment availability, wherein the dynamic bin adjustment plan contains corresponding materials among the target materials that need to adjust the physical location, the optimal sorting task sequence contains target stations corresponding to the target materials, and the preset optimization parameters include at least two of the following parameters: minimizing total material waiting time, maximizing warehouse space turnover rate, minimizing total material handling distance, maximizing on-time delivery rate, minimizing inventory holding cost, minimizing average order delivery delay, and maximizing equipment utilization rate; based on the dynamic bin adjustment plan, scheduling a flexible scheduling mechanism in the production environment to dynamically adjust the physical location of the corresponding materials; based on the optimal sorting task sequence, calling a sorting mechanism in the production environment to pick a specified number of target materials from the dynamically adjusted bin and sort them into the corresponding target station.

2. The method of claim 1, wherein, based on the dynamic bin adjustment plan, scheduling a flexible scheduling mechanism in the production environment to dynamically adjust the physical location of the corresponding materials comprises: sending the dynamic bin adjustment plan to a WCS to determine the current bin, target bin, and bin adjustment time of the corresponding materials based on the dynamic bin adjustment plan, wherein WCS refers to a warehouse control system; sending a bin adjustment instruction to the WCS to instruct the WCS to schedule the flexible scheduling mechanism to move the corresponding materials from the current bin to the target bin at the bin adjustment time to complete the dynamic adjustment of the geographical location of the corresponding materials.

3. The method of claim 1, wherein, The optimal sorting task sequence further comprises an optimal sorting operation sequence and an optimal path planning of the target material handling; based on the optimal sorting task sequence, a sorting mechanism in a production environment is called to pick up a specified number of the target materials from the dynamically adjusted storage locations and then sort them to the corresponding target workstations, which comprises: The sorting mechanism is sequentially sent sorting instructions according to the optimal sorting operation sequence of the optimal sorting task sequence, to instruct the sorting mechanism to pick up a specified number of the target materials from the dynamically adjusted storage locations according to the received sorting instructions, and then sequentially sort the picked-up target materials to the specified positions of the corresponding target workstations; According to the optimal path planning of the optimal sorting task sequence, the sorting mechanism is called to distribute the sorted target materials from the specified positions to the corresponding target workstations.

4. The method according to any one of claims 1 to 3, characterized in that, The method further comprises: Real-time monitoring of the actual flow of the target materials and the actual production rhythm changes of the production line to obtain real-time monitoring results; When a production abnormality occurs based on the real-time monitoring results, triggering an alarm and returning to execute: data cleaning and feature extraction on the production rhythm real-time parameters and the demand information to obtain corresponding feature data; using a specified time series prediction model as a production prediction model; based on the corresponding feature data, calling the specified time series prediction model to predict the production rhythm prediction parameters and the material demand prediction of each key workstation in a future period of time.

5. A dynamic warehouse intelligent sorting system based on production rhythm, characterized in that, Comprise: A production rhythm perception module is used to call photoelectric sensors and industrial cameras at each key workstation on a production line to monitor the product passing rate and buffer material inventory level of each key workstation in real time; call the PLC controller on each key workstation to read the device state and real-time yield data of each key workstation; call the information recognition channel at the warehouse entrance to automatically scan the unique identification code label on the incoming material or the unique identification code on the incoming material box to obtain detailed information of the incoming material; call the ERP system to obtain the production plan for a future period of time; Based on the product passing rate, the buffer material inventory level, the device state and the real-time yield data, the production rhythm real-time parameters of each key workstation are obtained; Based on the detailed information of the incoming material and the production plan, the demand for the target material to be sorted is determined to obtain the demand information of the target material; The data fusion and decision processing module is configured to perform data cleaning and feature extraction on the production rhythm real-time parameters and the demand information to obtain corresponding feature data; a specified time series prediction model is used as a production prediction model; based on the corresponding feature data, the specified time series prediction model is called to predict production rhythm prediction parameters and material demand prediction values of each key work station in a future period of time; a preset optimization parameter is used as an optimization target, and a specified multi-objective optimization algorithm is called to generate a dynamic bin adjustment plan and an optimal sorting task sequence suitable for the production line based on the production rhythm prediction parameters, the material demand prediction values, a current inventory state, and device availability, wherein the dynamic bin adjustment plan includes corresponding materials of the target materials that need to adjust physical positions, the optimal sorting task sequence includes target work stations corresponding to the target materials, and the preset optimization parameter includes at least two of the following parameters: minimizing total material waiting time, maximizing warehouse space turnover rate, minimizing total material handling distance, maximizing on-time delivery rate, minimizing inventory holding cost, minimizing average order delivery delay, and maximizing device utilization rate. The dynamic adaptive warehousing module is configured to schedule a flexible scheduling mechanism in a production environment to dynamically adjust physical positions of the corresponding materials based on the dynamic bin adjustment plan. The intelligent sorting execution module is configured to call a sorting mechanism in the production environment to sort out a specified number of the target materials from the dynamically adjusted bin positions based on the optimal sorting task sequence and sort the target materials to corresponding target work stations.

6. An electronic device, comprising: The method comprises the following steps: A memory and a processor, the memory stores instructions, the instructions are loaded and executed by the processor to realize the method of any one of claims 1-4.

7. A computer readable storage medium characterized in that, The computer readable storage medium stores a computer program, when the computer program runs on the computer, the method of any one of claims 1-4 is realized.

Images