Intelligent factory integrated operation management method based on multi-system integration

Abstract

The invention particularly relates to an intelligent factory integrated operation management method based on multi-system integration, and relates to the technical field of intelligent factory operation management. Performing intelligent planning and robustness decision making; performing production execution and monitoring; and carrying out equipment full-life-cycle management. According to the method, the limitation of a traditional static plan is thoroughly broken through through a collaborative mechanism of multi-dimensional risk index linkage, dynamic time buffering and a plan chain; on one hand, the DTB performs real-time calculation based on a supply chain risk index, an equipment health index and a process stability index, approaches reference buffer to avoid resource waste when the risk is low, automatically expands to absorb disturbance when the risk is high, and ensures dynamic balance of plan efficiency and disturbance resistance; and on the other hand, the pre-generated plan chain is rapidly triggered when the disturbance exceeds the limit, global rearrangement is not needed, and continuous production can be guaranteed only through local adjustment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent factory operation management, and in particular to an integrated operation management method for an intelligent factory based on multi-system integration. BACKGROUND

[0002] Under the wave of intelligent manufacturing, the construction of intelligent factories is accelerating, but operation management still faces many challenges:

[0003] On the one hand, the problem of multi-system data island is prominent, production sites, supply chains, quality control, personnel scheduling, and other systems are independent of each other, and device sensor data, order information, and material status are difficult to connect, resulting in scattered data collection, difficult governance, and management layers difficult to obtain comprehensive and accurate operation data, lacking strong support when making decisions, and plans and execution links often disjointed due to information gaps;

[0004] On the other hand, the production plan is weak in disturbance resistance, traditional plan generation relies on experience or simple algorithms, and does not fully consider dynamic risks such as device health fluctuations, supply chain delays, and process parameter drifts. In the face of sudden equipment failure, material delivery delay, and other situations, there is a lack of a rapid response mechanism, and local disturbances can easily spread to global production stagnation, order delivery rate and production efficiency are severely affected, and it is difficult to adapt to market demand for flexible and efficient production.

[0005] Therefore, a need exists for an integrated operation management method for an intelligent factory based on multi-system integration to address the aforementioned problems. SUMMARY

[0006] The purpose of the present application is to solve the above-mentioned problems, and an integrated operation management method for an intelligent factory based on multi-system integration is proposed.

[0007] In order to achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0008] The integrated operation management method for an intelligent factory based on multi-system integration comprises:

[0009] Data hub and integration: realize the collection, governance, storage, standardization and on-demand distribution of full-link data, provide data input for other steps;

[0010] Intelligent planning and robust decision-making: based on the full-domain data provided by the data hub, realize intelligent generation, dynamic optimization, and robustness enhancement of production planning, while connecting customer order demand;

[0011] Production execution and monitoring: interface with production site equipment and personnel, responsible for plan instruction issuance, execution process monitoring, disturbance response, and execution data feedback;

[0012] Equipment lifecycle management: Reduce failure risks and optimize equipment resource allocation through predictive maintenance, providing core data on equipment health index for robust control.

[0013] Preferably, the data hub and integration specifically include:

[0014] End-to-end data includes production site data, supply chain data, order and planning data, and quality and personnel data;

[0015] After standardizing and cleaning the data, the processed data is stored and distributed.

[0016] Preferably, the intelligent planning and robust decision-making specifically include:

[0017] Based on customer orders issued by ERP and equipment capacity, material inventory, and personnel scheduling data provided by the data hub, an initial master production schedule is automatically generated using a finite capacity scheduling algorithm, while meeting the objectives of prioritizing delivery time and balancing resource load.

[0018] Calling the supply chain risk index of the data center ( ), Equipment Health Index ( ), process stability index ( );

[0019] After receiving the three indices from the risk perception layer, dynamic time buffer calculations and contingency plan chain generation are performed.

[0020] Preferably, the dynamic time buffer calculation process includes:

[0021] Through the formula:

[0022]

[0023] in,

[0024] Let be the dynamic buffer time for the i-th critical path node;

[0025] Let be the baseline buffer time for the i-th critical path node;

[0026] , , These are the weighting coefficients;

[0027] , , The real-time index output by the risk perception layer.

[0028] Preferably, the contingency plan chain generation logic includes:

[0029] With the 3 indexes of the risk perception layer as trigger conditions, based on the factory resource library, process rule library and the 3 indexes of the risk perception layer, a preset number of backup schemes are automatically generated and prioritized according to the execution cost, impact range and success rate, forming a plan chain.

[0030] Preferably, the production execution and monitoring specifically includes:

[0031] The generated main production plan and plan chain are received, disassembled into process-level execution instructions, and pushed to the corresponding device operators.

[0032] Through the linkage of MES and device sensors, the actual start time, actual end time, processing quantity and device running state are collected in real time and compared with the planned time to calculate the progress deviation.

[0033] Visual monitoring is provided, and managers can intuitively view the progress of each process, device load and in-process product location, and abnormal conditions are displayed with a preset color for early warning.

[0034] A progress deviation threshold is preset, and the calculated progress deviation is compared with it, and corresponding processing is performed according to the comparison result.

[0035] The execution data is returned to the data hub for updating , , the index, while providing data support for subsequent optimization of the intelligent planning and robust decision-making steps.

[0036] Preferably, the device full life cycle management specifically includes:

[0037] A device digital archive is established to store device model, purchase date, installation location, technical parameters, maintenance history and backup device correlation;

[0038] Real-time sensor data of the device is collected, combined with historical fault data of the data hub, and through a machine learning model, the possible fault type and fault time of the device are predicted.

[0039] Based on the prediction result and the current running state of the device, the device health index is calculated.

[0040] Preferably, the method further includes:

[0041] When the device suddenly fails, the fault information is synchronized to the production execution and monitoring step and the intelligent planning and robust decision-making step.

[0042] Preferably, the supply chain collaboration is coordinated with material management: connecting factories with external suppliers, internal material flow, responsible for material demand forecasting, procurement management, logistics tracking, inventory optimization, incoming material quality inspection, material substitution, and reducing material delay risks through supply chain collaboration to provide supply chain risk index core data for robust control.

[0043] In summary, due to the adoption of the above technical solutions, the beneficial effects of the present application are:

[0044] 1. The present application breaks through the limitations of traditional static planning through the synergy mechanism of multi-dimensional risk index linkage, dynamic time buffer, and pre-plan chain; on the one hand, DTB is based on real-time calculation of supply chain risk index, equipment health index, and process stability index; when the risk is low, it approaches the benchmark buffer to avoid resource waste; when the risk is high, it automatically expands to absorb the disturbance, ensuring the dynamic balance of efficient and anti-disturbance planning; on the other hand, the pre-generated pre-plan chain is quickly triggered when the disturbance exceeds the limit, without the need for global rearrangement, and only local adjustment can ensure production continuity.

[0045] 2. The present application creates full-process intelligent management and control around equipment, dynamically updates equipment status through the establishment of digital archives, predicts faults using machine learning, and generates preventive work orders to prevent problems; in the event of a sudden failure, rely on EHI to quickly capture and synchronize information, link production execution suspension instructions, intelligent plan decision-making to replace pre-plans, and can also automatically match idle personnel and spare parts; this system exerts force in all dimensions from equipment health monitoring, fault prevention to emergency response, shortens the length of equipment failure downtime, improves maintenance resource allocation efficiency, ensures the continuous and stable operation of the factory from the core production resource dimension, and solves the problem of production fluctuation caused by equipment failure. BRIEF DESCRIPTION OF DRAWINGS

[0046] In the following description of exemplary embodiments in conjunction with the accompanying drawings, more details, features and advantages of the present application are disclosed, in which:

[0047] Figure 1 The method flowchart of the present application. DETAILED DESCRIPTION

[0048] Several embodiments of the present application will be described in detail below with reference to the accompanying drawings to enable those skilled in the art to implement the present application. The present application can be embodied in many different forms and purposes and should not be limited to the embodiments described herein. These embodiments are provided to make the present application comprehensive and complete, and to fully convey the scope of the present application to those skilled in the art. The embodiments do not limit the present application.

[0049] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and the present disclosure and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[0050] Embodiment 1

[0051] DETAILED DESCRIPTION Figure 1 is described in detail.

[0052] APPENDIX Figure 1 The multi-system integration-based intelligent factory integrated operation management method flowchart provided in the embodiments of the present application shows the complete steps from the data hub to the integrated device full life cycle management.

[0053] In the embodiments, the following are included:

[0054] Data hub and integration: realize the collection, management, storage, standardization and on-demand distribution of full-link data, provide real-time, accurate and consistent data input for other steps, and are the basic support for integrated operation;

[0055] The data hub and integration specifically includes:

[0056] Full-link data includes production site data, supply chain data, order and plan data, and quality and personnel data;

[0057] Production site data: collect device sensor data (vibration, temperature, current) through industrial Ethernet (Profinet / EtherNet / IP), edge computing gateway, MES real-time progress data (process start / end time, work-in-process quantity);

[0058] Supply chain data: interface with SCM (supply chain management system), supplier platform, to obtain material order status, logistics tracking information, incoming material quality inspection results;

[0059] Order and plan data: interface with ERP (enterprise resource planning system), APS (advanced planning and scheduling system), to obtain customer order information (quantity, delivery period, process requirements), main production plan version;

[0060] Quality and personnel data: interface with QMS (quality management system) to obtain SPC process parameters and product unqualified rate, and interface with HR system to obtain operator scheduling and skill level data.

[0061] The collection frequency is dynamically adjusted according to the data type, the device sensor data is millisecond / second level, and the order and logistics data is minute / hour level.

[0062] After standardizing and cleaning the data, the processed data is stored and distributed.

[0063] The data standardization and cleaning includes: establishing a unified data dictionary and coding rules (such as device ID, material code, process code) to avoid data confusion caused by different names for the same thing; automatically cleaning abnormal data (such as sensor transient fluctuation value, repeated order information) through algorithm, using mean filling, trend prediction and other methods to complete missing data, to ensure data accuracy.

[0064] Data storage and distribution includes: using a hybrid storage architecture combining time series database and relational database: time series database stores high-frequency device data and progress data, and relational database stores structured data such as orders, materials, and personnel.

[0065] Based on data subscription and push mechanism, real-time distribution of required data to other steps (such as risk perception step, plan execution step) (such as pushing supply chain and device data required for SRI and EHI calculation to risk perception step).

[0066] Intelligent planning and robust decision-making: based on the global data provided by the data hub, intelligent generation, dynamic optimization, and robustness enhancement of production planning are realized, which is the core carrier of dynamic buffer (DTB) and contingency chain technology, and also connects customer order demand and factory execution ability to ensure that the plan is executable and anti-disturbance.

[0067] Specifically includes:

[0068] Based on customer orders (including urgent orders) issued by ERP, device capacity, material inventory, and personnel scheduling data provided by the data hub, using limited capacity scheduling algorithms (such as genetic algorithm, tabu search algorithm), an initial main production plan is automatically generated to meet the delivery period priority and resource load balancing target, and the plan start time, plan end time, assigned device, and required materials for each process are specified.

[0069] Order priority quantification rules: order priority is divided into three levels (P1 urgent order, P2 normal order, P3 backup order), and the corresponding priority coefficients are 1.5, 1.0, and 0.5 (which can be adjusted by the user).

[0070] Algorithm objective function refinement: the goal is to minimize the priority weighted delivery period deviation and maximize the resource load balancing degree, and the fitness function of the genetic algorithm is designed as:

[0071] ;

[0072] in, Let be the priority coefficient for the k-th order. , These are the weights (the sum is 1, the default values ​​are 0.6 and 0.4 respectively).

[0073] Example explanation: If order P1 (delivery period of 3 days) and order P2 (delivery period of 5 days) compete for the same equipment, the algorithm will prioritize allocating the equipment to order P1. If order P1 has insufficient material inventory, the non-critical process equipment of order P2 will be temporarily allocated to order P1 based on resource load balancing adjustment.

[0074] Input data association rules: The order priority coefficient (P1=1.5, P2=1.0, P3=0.5), equipment capacity limit (machine hours / day), material inventory availability (pieces), and personnel shift work hours (hours / person) are used as the core inputs of the algorithm to establish a mapping relationship between the input data and the scheduling target (such as priority coefficient × delivery period deviation as the core optimization item).

[0075] Algorithm execution steps:

[0076] Initialize the population (randomly generate 100 sets of process allocation schemes);

[0077] Fitness calculation ;

[0078] fitness function value, , Weighting coefficients The priority coefficient of the kth order. The actual delivery time of the kth order. The planned delivery time for the kth order;

[0079] Selection (using roulette wheel to select the best 50 groups), crossover (genetic recombination of process allocation schemes), and mutation (randomly adjusting the equipment allocation of 10% of the schemes).

[0080] After 50 iterations, the optimal solution, i.e., the initial master production schedule, is output.

[0081] Calling the supply chain risk index of the data center ( ), Equipment Health Index ( ), process stability index ( The methods for obtaining each index are as follows:

[0082] Supply chain risk index ( The index is calculated in real time through big data analytics. Input data includes supplier historical on-time delivery rates, real-time logistics tracking information, regional geopolitical risk levels, and extreme weather warnings. The index value ranges from [0,1]. The closer to 1, the more stable the supply chain represents; the closer to 0, the higher the risk of supply chain interruption, material delay.

[0083] Input data quantization: supplier historical on-time delivery rate (directly take the value, such as 95% = 0.95), logistics status (normal = 1.0, delay 1-3 days = 0.5, delay > 3 days = 0.1), regional geopolitical risk (no risk = 1.0, low risk = 0.8, medium risk = 0.5, high risk = 0.2), extreme weather warning (no warning = 1.0, with warning = 0.3);

[0084] Weighted calculation of original value: ; a + b + c + d = 1, a = 0.4 by default, b = 0.3, c = 0.2, d = 0.1;

[0085] Normalization method: Min-Max normalization, , , Determined by historical data statistics.

[0086] Equipment health index ( ): Based on the whole life cycle equipment management step, through the collection of real-time sensor data such as equipment vibration, temperature, current, running time, combined with predictive maintenance model (such as machine learning-based fault prediction algorithm) to evaluate the health status of equipment. The index value range is [0, 1], The closer to 1, the more stable the equipment represents; the closer to 0, the higher the risk of equipment failure, the processing efficiency may decline.

[0087] Predictive maintenance model design: LSTM neural network is adopted, input features are equipment vibration value (after standardization), temperature (after standardization), current (after standardization), running time (hours), maintenance interval (days); Model structure: input layer (5 features), hidden layer 1 (64 neurons), hidden layer 2 (32 neurons), output layer (1 neuron, output fault probability);

[0088] Original value calculation: ;

[0089] Process stability index ( ): Relying on intelligent quality control steps, through statistical process control (SPC) analysis of the fluctuation of key process parameters (such as temperature, pressure, machining precision), calculate the process capability index (Cp / Cpk). The index value range is [0, 1], The closer to 1, the smaller the process parameter fluctuation, the higher the product qualification rate; the closer to 0, the more unstable the process, the more likely to appear quality abnormalities.

[0090] Cp / Cpk calculation:

[0091] Sample collection: collect 1 set of key parameters (such as temperature) per hour, and continuously collect 25 sets as samples;

[0092] Statistical analysis: calculate sample mean μ and standard deviation σ;

[0093] Cp / Cpk calculation (USL is the upper specification limit, and LSL is the lower specification limit), ;

[0094] Normalization method: Min-Max normalization, where Cpk_min=0 (process completely out of control), Cpk_max=2 (process extremely stable), and the statistics are determined by nearly 1 year of historical data.

[0095] Key process parameter screening: through correlation analysis (such as Pearson correlation coefficient), parameters (such as temperature, pressure, machining precision) with a correlation coefficient ≥0.8 with product pass rate are screened.

[0096] After receiving the three indexes of the risk perception layer, dynamic time buffer calculation and preplan chain generation are performed.

[0097] The dynamic time buffer calculation process includes:

[0098] Focusing on the key path nodes in the main production plan (i.e. the process that will directly cause the entire order delivery delay, identified by the critical path method CPM);

[0099] CPM critical path node identification method:

[0100] Step 1: Traditional CPM calculation: calculate the earliest start time (ES), latest start time (LS), and total float (TF=LS-ES) of each process, and select the processes with TF≤0 as the basic key nodes (directly causing order delay);

[0101] Step 2: Risk correlation degree calculation: calculate the risk correlation degree of each process;

[0102] Step 3: Final key node screening: screen the processes with basic key nodes + TF≤threshold (such as 2 hours) and R≥0.5 to form a complete critical path node chain.

[0103] Through the formula:

[0104]

[0105] In the present application, only the dynamic time buffer The calculation formula needs to be de-dimensioned, and the values of the parameters in the formula are the results after de-dimensioning;

[0106] wherein,

[0107] is the dynamic buffer time for the ith key path node (unit: hour / minute, set according to the production cycle);

[0108] is the baseline buffer time for the ith key path node (based on historical execution data statistics of the node, such as 1.2 times the average delay time of the process in the past 12 months);

[0109] , , are weight coefficients, respectively, and , , the sum of which is 1; can be flexibly configured according to the type of factory, industry characteristics;

[0110] , , is the real-time index output by the risk perception layer (value range [0, 1]).

[0111] When the generation environment is stable ( , , ), , , are close to 1, avoiding buffer waste and ensuring production efficiency;

[0112] When a certain type of risk increases (such as deterioration of equipment health, , the proportion of the corresponding weight item increases, automatically expanding to reserve more buffer time for the node to resist equipment delay risks;

[0113] When the risk is eliminated (such as after equipment maintenance ), , automatically shrink to return to an efficient state.

[0114] Dynamic time buffer ( ) breaks through the limitations of traditional fixed buffer by precisely focusing on key paths, multi-risk index linkage, and self-adaptive flexible adjustment:

[0115] On the one hand, when the supply chain is stable (SRI≈1), the equipment is healthy (EHI≈1), and the process is controllable (PSI≈1), approaches the baseline buffer ( ), avoid resource idling and production cycle redundancy, and ensure efficiency; on the other hand, once a certain type of risk deteriorates (such as equipment health index EHI = 0.3, corresponding , automatically expand, reserve sufficient buffer time for key nodes, absorb disturbances, and prevent local delay from spreading to global delivery risk. Let the rigid schedule of production planning become more flexible and targeted.

[0116] The pre-plan chain generation logic includes:

[0117] With the three indexes of the risk perception layer as trigger conditions, based on the factory resource library (equipment, materials, cooperative manufacturers, etc.), process rule library and the three indexes of the risk perception layer, a preset number (for example, 2-3) of backup plans (pre-plans) are automatically generated, and are prioritized according to low execution cost, small impact range and high success rate, forming a pre-plan chain.

[0118] Trigger condition definition:

[0119] Single index trigger threshold SRi < 0.3 (high risk of supply chain), EHI < 0.4 (high risk of equipment failure), PSI < 0.5 (process instability);

[0120] Combined trigger rule: meet any one of the single index thresholds, or any two indexes below the medium threshold to trigger the pre-plan chain;

[0121] Device resource library: stores the model, capacity, current state (running / standby / maintenance) of all devices in the factory, and the association relationship of backup devices;

[0122] Supplier library: stores the list of pre-audited outsourcing manufacturers, processing capacity, delivery cycle, and cooperation history;

[0123] Process library: stores the list of replacement materials, and replacement process parameters (such as processing temperature and feed rate corresponding to different materials);

[0124] and supports manual adjustment of pre-plan priority (such as when outsourcing costs are too high, the process replacement pre-plan priority can be raised above the outsourcing cooperation pre-plan), and version management of all pre-plans (records modification time, modifier, and applicable scenarios).

[0125] Production execution and monitoring: interfaces with production site equipment and personnel, responsible for plan instruction issuance, execution process monitoring, disturbance response, and execution data feedback, is the key link from decision to execution of robust control (integrates the core functions of MES and strengthens the linkage with robust decision-making);

[0126] Specifically includes:

[0127] Receive the generated master production plan and contingency chain, disassemble it into process-level execution instructions, and push it to the corresponding device operators through industrial tablets and SCADA systems;

[0128] The instruction content includes: process name, processing parameters (such as speed, temperature), material batch number, quality inspection standard, dynamic buffer time (for operator reference), pre-set contingency trigger conditions (such as delay exceeding 3 hours, starting backup equipment).

[0129] Through the linkage of MES and device sensors, real-time collection of actual start time, actual end time, processing quantity, and device running state (normal / fault / standby) is realized, and the progress deviation is calculated by comparing with the planned time;

[0130] Visual monitoring dashboards (such as digital twin map of the workshop, Gantt chart) are provided, and managers can visually view the progress of each process, device load, and in-process product location. Abnormal situations (such as device failure, progress delay) are displayed with pre-set colors (yellow: approaching buffer threshold; red: exceeding buffer threshold) for early warning;

[0131] The progress deviation threshold is pre-set, and the calculated progress deviation is compared with it, and corresponding processing is carried out according to the comparison result, including:

[0132] When the progress deviation does not exceed the dynamic buffer threshold: the system only records the deviation and does not intervene in the execution, ensuring the production rhythm;

[0133] When the deviation is expected to exceed the buffer threshold (such as current delay of 2 hours, remaining process requiring 2 hours, buffer only 2 hours): automatically trigger the pre-set contingency chain, such as pushing task transfer instructions to backup device operators, and update the planned progress of the monitoring dashboard to ensure transparency of adjustment;

[0134] Support for manual triggering of contingency by operators: when there are unforeseen small disturbances (such as slight material loss) on site, operators can submit contingency start application through industrial tablets, and the system will execute after review (to avoid misoperation).

[0135] Return the execution data (such as process completion rate, device fault duration, contingency execution effect) to the data hub for updating 、 、 exponent, and provide data support for subsequent optimization of intelligent planning and robust decision-making steps.

[0136] 、 、 Dynamic updating rules of exponent:

[0137] Data grouping: Group data into multiple groups (each group contains SRI / EHI / PSI real-time data during the execution of the plan) according to the planned execution cycle;

[0138] Plan effectiveness evaluation: Define effectiveness indicators;

[0139] Weighted update: Weight of each group of data , the updated index is:

[0140] ;

[0141] Through instruction disassembly and accurate pushing, the main production plan is refined into process-level instructions, covering all factors such as processing parameters, quality standards, dynamic buffer, etc. With the help of industrial tablets and SCADA systems, operators can directly connect to the operator, and production instructions can be changed from vague requirements to clear and executable operation manuals. At the same time, MES and device sensors work together to collect real-time data, and visualized dashboards can intuitively present progress and equipment status, with color-coded abnormality warnings (yellow near threshold, red beyond threshold), allowing managers to quickly identify risks and ensure production collaboration and efficiency from the source.

[0142] Equipment life cycle management: Focus on the core production resources (equipment) of the factory, covering the entire life cycle of equipment procurement, installation, operation, maintenance, and scrap. Through predictive maintenance, reduce the risk of failure, optimize equipment resource allocation, and provide equipment health index (EHI) core data for robust control, while supporting equipment replacement options in the plan chain (integrate CMMS system functions and strengthen the linkage with production execution);

[0143] Plan effectiveness indicators : Define three core indicators and calculate the value by weighting:

[0144] Delivery period guarantee rate: ;

[0145] Cost control rate: ;

[0146] Production continuity: ;

[0147] ;

[0148] Specifically includes:

[0149] Establish a digital archive of equipment, storing equipment model, purchase date, installation location, technical parameters (such as rated capacity, precision level), maintenance history (fault reason, maintenance parts, maintenance duration), and backup equipment correlation (such as equipment A can be replaced by equipment B when equipment A fails);

[0150] Support dynamic update of equipment account (such as adding new equipment, scrapping equipment), and real-time synchronization with the equipment repository in the intelligent plan and robust decision-making step, to ensure the accuracy of equipment information when generating the plan;

[0151] Collect real-time sensor data of equipment (vibration, temperature, current, lubricating oil status), combine with historical fault data of data hub, and predict the type and time of possible equipment failure through machine learning models (such as LSTM, random forest);

[0152] Based on the prediction results and the current operating status of the equipment, calculate the equipment health index (EHI) (value 0-1): The better the health status (failure probability less than 5%), the closer the EHI to 1; the higher the risk of failure (failure probability higher than 30%), the closer the EHI to 0;

[0153] When EHI is lower than the threshold (such as 0.4), automatically generate a preventive maintenance work order and push it to the maintenance personnel to avoid production interruption caused by equipment failure (reduce disturbance risk from the source);

[0154] When the equipment fails suddenly (EHI decreases by more than the preset threshold within the preset time interval), synchronize the fault information (equipment ID, fault type, estimated repair time) to the production execution and monitoring step and the intelligent plan and robust decision-making step:

[0155] Push the equipment downtime warning to the production execution and monitoring step, and suspend the execution instructions for that equipment;

[0156] Feedback the fault data to the intelligent plan and robust decision-making step to support it to determine whether to start the equipment replacement plan;

[0157] When the failed equipment is a critical equipment, automatically match idle maintenance personnel and spare parts inventory to shorten the repair time (such as equipment A failure, automatically dispatch maintenance personnel Zhang Gong, and confirm that the spare parts warehouse has corresponding bearings).

[0158] When the equipment fails suddenly, accurately capture the failure by the EHI decrease amplitude, synchronize the fault information to the production execution and intelligent plan step in the first time, realize the fault response. Push the downtime warning to the production execution and suspend the instructions to avoid more losses or quality problems caused by the failure of the equipment running; feedback the data to the intelligent plan to let the plan decision-making layer quickly determine whether to start the replacement plan, break the fault discovery lag and information isolation problem, form a collaborative response link from the equipment end to the execution layer and decision-making layer, lock the impact of the fault on production at the source and handle it quickly;

[0159] For critical equipment failure, automatically match idle maintenance personnel, spare parts inventory, say goodbye to the traditional manual one by one search inefficient mode. System intelligent identification of resource idle state, rapid scheduling personnel, confirm spare parts, shorten the maintenance preparation time, let the fault equipment can faster recovery operation.

[0160] For example: when equipment A fails, automatically dispatch Zhang Gong who understands equipment A maintenance, and confirm that the spare parts library has corresponding bearings, reduce waiting and coordination time loss, improve maintenance efficiency from resource allocation dimension, ensure production continuity, and compress the impact range and length of equipment failure to the minimum.

[0161] Supply chain coordination and material management: link factories and external suppliers, internal material flow, responsible for material demand forecasting, procurement management, logistics tracking, inventory optimization, incoming quality inspection, material substitution, reduce material delay risk through supply chain coordination, provide supply chain risk index (SRI) for robustness control Core data, support material substitution scheme in pre-plan chain (integrate SCM, WMS, procurement management system core functions).

[0162] Specifically includes:

[0163] Based on the main production plan, historical material consumption data, order fluctuation trend of intelligent planning and robustness decision steps, generate material demand plan (clear the demand quantity and demand time of each material) through demand forecasting algorithm (such as ARIMA, machine learning regression model), which also includes deviation correction and manual review mechanism to avoid prediction deviation:

[0164] Prediction deviation statistics: the system real-time statistics of the prediction deviation rate of the last 3 months, and calculates the average deviation rate;

[0165] Demand plan correction: after the prediction algorithm output, automatically correct to ;

[0166] The average deviation rate of the last 3 months of material demand prediction reflects the deviation degree between the predicted value and the actual consumption;

[0167] Purchasing trigger condition: only when the corrected demand quantity - existing inventory - in-transit order quantity ≥ safety stock, generate a draft purchase order;

[0168] Manual review link: the draft purchase order is pushed to the procurement management personnel, which supports to modify the demand quantity, supplier, delivery period, and the formal order is issued to the supplier platform after the audit is passed.

[0169] Automatic docking with supplier platform, generate purchase order, synchronize to qualified suppliers, and track order status (confirmed / delivered / in-transit / signed) in real time, avoid material shortage leading to process stagnation;

[0170] Collect multi-dimensional supply chain data: supplier historical on-time delivery rate, logistics transportation status (such as GPS tracking information), regional risk (geopolitics, extreme weather), supplier capacity fluctuation (such as supplier factory power outage);

[0171] Calculate the supply chain risk index (SRI) based on the above data (value 0-1): when the supplier on-time rate is high (>95%), logistics is smooth, and there is no regional risk, the SRI is close to 1; when the supplier delay rate is high (>20%), the logistics is blocked, and the SRI is close to 0;

[0172] When the SRI is lower than the threshold (such as 0.3), automatically send material delay warning to procurement personnel, plan management personnel, and start response measures (such as switching to backup suppliers) in advance;

[0173] Interface with WMS (warehouse management system) to track material entry, inspection, exit, and production station in real time:

[0174] After the material arrives at the factory, the inspection process is automatically triggered, and the qualified material is recorded in the inventory, and the unqualified material starts the supplier return and replacement process (to avoid unqualified material flowing into production);

[0175] Set the safety stock of each material through the safety stock model (such as economic order quantity EOQ), and automatically trigger the replenishment reminder when the inventory is lower than the safety threshold; at the same time, allocate materials based on production plan priority (such as emergency order priority material allocation);

[0176] Record the source, flow path, and use process of each batch of materials through barcode / RFID technology, quickly locate the material batch when quality problems occur, and reduce the loss range.

[0177] Establish a material substitution list: for commonly used materials, pre-authenticate substitute materials (such as material A can be used when material A is out of stock, and the processing temperature needs to be adjusted by 10°C), and synchronize with the process library in the intelligent plan and robust decision-making steps;

[0178] When the material delay risk is high (SRI is low), automatically push the material substitution suggestion to support the generation of material substitution plan; at the same time, push the process parameter adjustment instructions of the substitute material to the production personnel to ensure production continuity.

[0179] The above formulas are dimensionless to calculate their numerical values, and the formulas are obtained by software simulation of a large amount of data to obtain the most recent real situation, and the preset parameters in the formula are set by the person skilled in the art according to the actual situation.

[0180] It is apparent that for the person skilled in the art, many modifications and changes can be made to the embodiments described without departing from the spirit and scope of the application. It is therefore understood that the above-described drawings and descriptions are illustrative in nature and should not be considered limiting the scope of the present application.

[0181] It should be noted that, in this text, relationship terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply there is any such actual relationship or order between these entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment. Without more limitations, the element defined by the statement "including a" does not exclude the presence of another identical element in the process, method, article or equipment including the element.

[0182] It should be understood that in various embodiments of the present application, the size of the sequence number of each process described above does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0183] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0184] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0185] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, i.e. they can be located in one place or distributed on multiple network units. Part or all of the units can be selected to achieve the purpose of the present embodiment according to actual needs.

[0186] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit.

[0187] The above merely describes some exemplary embodiments of the present application, but the protection scope of the present application is not limited thereto, any modification or replacement within the technical range disclosed by the present application can be easily thought by those skilled in the art, and should be covered in 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.

[0188] The above only describes some exemplary embodiments of the present application by way of illustration, and it is needless to say that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present application. Therefore, the above drawings and descriptions are illustrative in nature, and should not be understood as limiting the protection scope of the claims of the present application.

Claims

1. An integrated operation management method for an intelligent factory based on multi-system integration, characterized by, Comprise: Data hub and integration: realize the collection, governance, storage, standardization and on-demand distribution of full-link data, provide data input for other steps; Intelligent planning and robust decision-making: based on the global data provided by the data hub, realize the intelligent generation, dynamic optimization and robust enhancement of production planning, while connecting customer order demand; Production execution and monitoring: interface with production site equipment and personnel, responsible for plan instruction issuance, execution process monitoring, disturbance response, and execution data feedback; Equipment full life cycle management: reduce failure risk through predictive maintenance, optimize equipment resource allocation, and provide equipment health index core data for robust control.

2. The intelligent factory integrated operation management method based on multi-system integration according to claim 1, characterized in that, Data hub and integration specifically includes: Full-link data includes production site data, supply chain data, order and plan data, and quality and personnel data; And after standardization and cleaning of the data, the processed data is stored and distributed. 3.The smart factory integrated operation management method based on multi-system integration of claim 1, wherein, Intelligent planning and robust decision-making, specifically includes: Based on the customer order issued by ERP, the equipment capacity provided by the data hub, the material inventory, the personnel scheduling data, the limited capacity scheduling algorithm is adopted to automatically generate the initial main production plan under the conditions of meeting the delivery period priority and resource load balancing target; a supply chain risk index (SRI) of the data hub is called , a device health index (DHI) of the data hub is called , a process stability index (PSI) of the data hub is called ; After receiving the three indexes of the risk perception layer, dynamic time buffer calculation and contingency chain generation are performed.

4. The intelligent factory integrated operation management method based on multi-system integration according to claim 3, characterized in that, The dynamic time buffer calculation process includes: Through the formula: Where, dynamic buffer time for the ith critical path node; reference buffer time for the ith critical path node; , , are weight coefficients, respectively; , , Real-time index output for the risk perception layer.

5. The intelligent factory integrated operation management method based on multi-system integration according to claim 4, characterized in that, The contingency chain generation logic includes: Taking the three indexes of the risk perception layer as the trigger condition, based on the factory resource library, process rule library and three indexes of the risk perception layer, a preset number of backup solutions are automatically generated and sorted by execution cost, impact range and success rate to form a contingency chain.

6. The intelligent factory integrated operation management method based on multi-system integration according to claim 5, characterized in that, Production execution and monitoring, specifically includes: Receive the generated main production plan and contingency chain, decompose it into process-level execution instructions, and push it to the corresponding device operators; Through the linkage of MES and device sensors, real-time collection of process actual start time, actual end time, processing quantity and device running state is realized, and the progress deviation is calculated by comparing with the planned time; Provide visual monitoring, management personnel can intuitively view the progress of each process, device load and in-process position, and abnormal conditions are displayed with preset colors for early warning; Preset the progress deviation threshold, compare the calculated progress deviation with it, and perform corresponding processing according to the comparison result; The execution data is transmitted back to the data hub for updating , , the index, while providing data support for the subsequent optimization of the plan by the intelligent planning and robust decision steps. 7.The smart factory integrated operation management method based on multi-system integration of claim 1, wherein, Equipment full life cycle management, specifically includes: Establish a digital archive of the equipment, store the equipment model, purchase date, installation location, technical parameters, maintenance history, and backup device correlation; Collect real-time sensor data of the equipment, combine with historical fault data of the data hub, and predict the possible fault type and fault time of the equipment through a machine learning model; Based on the prediction result and the current running state of the equipment, calculate the equipment health index. 8.The smart factory integrated operation management method based on multi-system integration of claim 7, wherein, Also includes: When the equipment suddenly fails, the fault information is synchronized to the production execution and monitoring step and the intelligent planning and robust decision-making step. 9.The smart factory integrated operation management method based on multi-system integration of claim 1, wherein, Supply Chain Collaboration and Material Management: Connect factories with external suppliers and internal material flow, responsible for material demand forecasting, procurement management, logistics tracking, inventory optimization, incoming quality inspection, material substitution, and reduce material delay risk through supply chain collaboration to provide core data for robustness control and supply chain risk index.

Images