Printing order intelligent production scheduling optimization method and system based on equipment operation and maintenance state

Abstract

The invention discloses a printing order intelligent production scheduling optimization method and system based on an equipment operation and maintenance state, and relates to the technical field of the manufacturing industry, and the method comprises the steps: obtaining a multi-source operation and maintenance parameter set of equipment and work order attribute information during printing, and recognizing a same-type work order interval; on the basis of the same type of work order intervals, dividing loss levels to generate an adaptation result which is used for providing an adaptation basis for equipment and work order distribution; on the basis of the adaptation result and the work order attribute information, the interference influence of work order switching on equipment production is identified by analyzing the correlation strength of the same process work order and the equipment part loss, and disturbance strength data is obtained; identifying a to-be-distributed node through the disturbance intensity data, and obtaining a feasibility result through a redistribution operation of work order switching; and constructing a production scheduling network according to the feasibility result, and obtaining a distribution optimization result after path traversal and path re-planning so as to realize dynamic matching of production scheduling and equipment operation and maintenance states.

Description

Technical Field

[0001] This invention relates to the field of manufacturing technology, specifically to a method and system for intelligent scheduling optimization of printing orders based on equipment operation and maintenance status. Background Technology

[0002] With the development of manufacturing and intelligent production systems, traditional printing manufacturing is gradually adopting an intelligent production management model centered on production scheduling systems. In this type of production system, production orders typically enter the production process in the form of printing work orders, and equipment resources are allocated through a production scheduling system, allowing different work orders to be executed sequentially on printing equipment according to a predetermined order. Because the printing production process involves various process parameters such as paper thickness, ink type, printing roller specifications, cleaning operations, and plate changing operations, there are often significant process differences between different work orders. Therefore, in the specific production scheduling process, it is necessary to comprehensively consider multiple factors such as equipment operating capacity, process switching requirements, and production time constraints to ensure the continuity and stability of the production process.

[0003] In existing printing production scheduling technologies, the scheduling process typically focuses on optimizing indicators such as order delivery cycle, production efficiency, or equipment idle time, while paying less attention to equipment maintenance status and process switching relationships between work orders. In actual production environments, when multiple work orders with similar process parameters are executed consecutively, the equipment may operate under the same conditions for an extended period, leading to a continuous accumulation of load on critical components. This can cause fatigue phenomena such as increased temperature rise, enhanced vibration, or fluctuations in registration accuracy. However, traditional scheduling methods often prioritize work orders solely based on their execution time or order priority, without considering the fatigue changes of the equipment during continuous production to reasonably distribute the work orders. This can easily create consecutive intervals of similar work orders during production, causing the equipment's operating condition to gradually deteriorate.

[0004] On the other hand, switching between different printing work orders typically requires operations such as ink cleaning, plate roller replacement, or equipment adjustment. When adjacent work orders differ significantly in paper specifications, ink colors, or plate structures, the equipment often needs to perform complex cleaning and plate changing operations during the switchover process, thereby increasing equipment downtime and affecting equipment operational stability. Therefore, under current technological conditions, how to comprehensively consider equipment maintenance status and work order process differences during production scheduling, and rationally allocate work orders while ensuring the feasibility of the production sequence, remains a pressing technical problem to be solved in the field of printing production scheduling. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for intelligent scheduling and optimization of printing orders based on equipment operation and maintenance status, thus solving the problems mentioned in the background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The first aspect is a method for intelligent scheduling and optimization of printing orders based on equipment operation and maintenance status, which includes:

[0008] Obtain the multi-source operation and maintenance parameter set of the equipment and the work order attribute information during printing, and identify the range of work orders of the same type;

[0009] Based on the same type of work order range, the loss level is divided to generate adaptation results, which are used to provide an adaptation basis for equipment and work order allocation;

[0010] Based on the adaptation results and work order attribute information, by analyzing the correlation strength between work orders with the same process and equipment component losses, the interference impact of work order switching on equipment production is identified, and disturbance intensity data is obtained.

[0011] By using disturbance intensity data, nodes to be assigned are identified, and feasibility results are obtained through the reassignment of work orders.

[0012] Based on the feasibility results, a production scheduling network is constructed. After path traversal and path replanning, the allocation optimization results are obtained to achieve intelligent production scheduling optimization.

[0013] Preferably, the identification of work order ranges of the same type includes:

[0014] The system acquires multi-source operation and maintenance parameter sets and work order attribute information of the equipment, divides the operation behavior of the equipment in the continuous production stage into stages, identifies the time boundaries between the continuous operation interval and the downtime interval, obtains the equipment operation interval sequence, and extracts equipment operation and maintenance status segments.

[0015] Assign switching attributes to work orders to update work order attribute information. Switching attributes include cleaning level and change type.

[0016] Based on the multi-source operation and maintenance parameter set, the printing work order set within the current production scheduling cycle is read, and combined with the work order attribute information, the similarity between adjacent work orders is analyzed to identify work order intervals of the same type.

[0017] Preferably, based on the same type of work order range, loss levels are divided to generate adaptation results, including:

[0018] Retrieve equipment operation and maintenance status segments within the same type of work order interval, analyze the equipment's operating behavior during continuous work order operation, and obtain equipment fatigue value data;

[0019] Based on the trend of characteristic values ​​within the equipment fatigue value data, identify whether the equipment has intermittent anomalies and classify the wear level.

[0020] Preferably, based on the same type of work order range, loss levels are divided to generate adaptation results, which also includes:

[0021] By analyzing the overall operation and maintenance status of the equipment through a multi-source set of operation and maintenance parameters, a comprehensive health value can be obtained.

[0022] Based on the process characteristics of the work order, a process matching coefficient is set, and fatigue loss value is constructed by combining it with equipment fatigue value data;

[0023] The wear level is classified based on fatigue loss value and comprehensive health value, combined with intermittent abnormal behavior. The classification steps are as follows:

[0024] If the overall health value exceeds the upper limit of the preset health range, and the fatigue loss value does not exceed the lower limit of the preset loss range, and there are no intermittent abnormalities, then the loss level is the low loss level.

[0025] If the overall health value exceeds the lower limit of the preset health range but does not exceed its upper limit, and the fatigue loss value exceeds the lower limit of the preset loss range but does not exceed its upper limit, and there are no intermittent abnormalities, then the loss level is medium loss level.

[0026] If the overall health value does not exceed the lower limit of the preset health range, or the fatigue loss value exceeds the upper limit of the preset loss range, or there are intermittent abnormalities, then the loss level is high loss level.

[0027] Label the attribute constraints of the work orders that can be accepted for each level of equipment to generate adaptation results.

[0028] Preferably, based on the adaptation results and work order attribute information, by analyzing the correlation strength between work orders with the same process and equipment component losses, the interference impact of work order switching on equipment production is identified, and disturbance intensity data is obtained, including:

[0029] Using the adaptation results as prior adaptation constraints, and combining equipment fatigue value data and work order process combinations, the correlation strength between different process work orders and equipment component losses is calculated through Pearson correlation coefficient analysis to obtain the correlation coefficient.

[0030] Based on the correlation coefficient and work order attribute information, we analyze the differences in attributes between adjacent work orders and the changes in the status of switching nodes to obtain equipment replacement impact data that characterizes the impact of work order sequence on equipment wear and tear. This data is used to characterize the degree of fluctuation of work order switching on equipment.

[0031] Based on correlation coefficients and equipment changeover impact data, by analyzing the changes in production scheduling structure and equipment changeover increments caused by order insertion, we obtain disturbance intensity data characterizing the impact of order insertion on production stability.

[0032] Preferably, by using disturbance intensity data, nodes to be assigned are identified, and feasibility results are obtained through the reassignment operation of work order switching, including:

[0033] Based on disturbance intensity data, the distribution of disturbance intensity in the printing work order set in the equipment operation interval sequence is analyzed to identify disturbance sections and nodes to be assigned.

[0034] Based on the nodes to be assigned, the cleaning level and change type of the work order switching are analyzed by combination to construct the switching allocation constraints between work orders, obtain the conflict nodes, and realize the redistribution of work order switching operations.

[0035] Based on the conflict nodes within the disturbance section, feasible results are allocated by analyzing the matching between equipment capacity and work order switching structure.

[0036] Preferably, based on the nodes to be assigned, by combining and analyzing the cleaning level and version change type of the work order switching, switching allocation constraints between work orders are constructed to obtain conflicting nodes, including:

[0037] Get the switching attributes of the node to be assigned; when the node to be assigned is both deep cleaning and full-size replacement, it is recorded as high-intensity switching; when the node to be assigned is only simple cleaning or small-size replacement, it is recorded as medium-intensity switching; when the node to be assigned is no cleaning and same-size replacement, it is recorded as low-intensity switching.

[0038] If it is a high-intensity switchover, the corresponding node to be assigned will be classified as a conflict node.

[0039] Preferably, based on the feasibility results, a production scheduling network is constructed. After path traversal and path replanning, the optimized allocation results are obtained, including:

[0040] Based on the feasibility results, a production scheduling network is constructed by limiting the equipment allocation and the adjacent conversion relationship of work orders. The nodes of the production scheduling network represent the running status of equipment executing a single work order in the corresponding time period, and the edges represent the sequential conversion relationship.

[0041] Starting from the initial node within the production scheduling network, traverse all paths, accumulating equipment fatigue value data, equipment changeover impact data, and disturbance intensity data node by node. If any data exceeds the limit, terminate the extension and mark it as an infeasible path. Only retain paths that have completely traversed all work orders and have not exceeded the limit throughout the entire process, and record them as candidate paths.

[0042] Based on candidate paths, local swaps are performed on adjacent work orders, followed by limit checks. If a limit is exceeded, the local swap is canceled; otherwise, the new order is retained. Through iterative cyclic switching to disperse high-difference switching, the allocation optimization result is obtained.

[0043] Secondly, the intelligent scheduling and optimization system for printing orders based on equipment operation and maintenance status includes:

[0044] The identification module is used to obtain the multi-source operation and maintenance parameter set of the equipment and the work order attribute information during printing, and to identify the range of work orders of the same type.

[0045] The first allocation module is used to divide the loss level based on the same type of work order range to generate the matching result, which is used to provide the matching basis for equipment and work order allocation;

[0046] The disturbance analysis module is used to identify the interference impact of work order switching on equipment production based on the adaptation results and work order attribute information by analyzing the correlation strength between work orders of the same process and equipment component losses, and to obtain disturbance intensity data.

[0047] The second allocation module is used to identify nodes to be allocated through disturbance intensity data and obtain feasibility results through the reassignment operation of work order switching;

[0048] The resource optimization module is used to construct a production scheduling network based on the feasibility results. After path traversal and path replanning, the allocation optimization results are obtained to achieve intelligent production scheduling optimization.

[0049] The above-described solution of the present invention has at least the following beneficial effects:

[0050] Based on the same type of work order range, and combined with equipment component wear data and work order process requirements, wear levels are classified and adaptation results are generated. This provides a scientific basis for the accurate allocation of equipment and work orders, avoiding increased changeover losses caused by mismatch between work orders and equipment. It also allows for the prediction of fatigue trends in equipment during continuous production, laying the foundation for subsequent analysis of the correlation between equipment wear and work order processes. This achieves accurate matching between equipment operation and maintenance status and work order processes, balancing equipment operation and maintenance safety with work order execution efficiency. It effectively reduces ineffective equipment wear and production resource waste caused by improper adaptation, while improving the rationality and relevance of work order allocation, providing solid data support and adaptation criteria for subsequent production scheduling optimization.

[0051] By analyzing the correlation strength between different process work orders and the wear and tear of equipment components such as rubber rollers and bearings using Pearson correlation coefficients, the differences in the impact of different process work orders on equipment component wear can be accurately identified. Simultaneously, the analysis focuses on the changes in production scheduling caused by work order insertion, the increase in equipment type, and the changes in wear and tear due to plate changing and cleaning during work order switching. This quantifies the degree of interference of work order switching on equipment production, obtaining disturbance intensity data characterizing the impact of work order insertion and switching on production stability. This process can accurately capture the interference patterns of work order switching on equipment production, clarify the aggravating effect of different process work order switching on equipment component wear, and quantify the production disturbances caused by events such as work order insertion and type changing. It provides accurate disturbance assessment basis for subsequent identification of nodes to be allocated and optimization of production scheduling, while achieving deep coupling between equipment operation and maintenance status and work order processes. This avoids excessive equipment wear caused by frequent work order switching, balances production efficiency and equipment operation and maintenance safety, and reduces the probability of unplanned downtime.

[0052] Based on disturbance intensity data, the system accurately identifies nodes to be allocated during the production scheduling process. Through work order switching and reassignment operations and feasibility verification, infeasible allocation schemes are eliminated, and feasible results that meet equipment maintenance constraints and work order production requirements are selected. Furthermore, through path traversal, path replanning, and iterative local work order exchanges, optimal production scheduling paths are selected. Simultaneously, by iteratively dispersing high-difference work order switching, equipment fatigue accumulation caused by continuous high-load process switching is avoided, ultimately resulting in optimized equipment and work order allocation. This process achieves intelligent production scheduling optimization for printing orders, ensuring the stability and feasibility of production scheduling, while balancing equipment maintenance safety and production efficiency. It reduces equipment downtime and changeover losses, extends equipment lifespan, ensures on-time order delivery, and improves the quality of printed products. This collaborative optimization of equipment maintenance and order scheduling drives the transformation of printing production towards intelligence, refinement, and efficiency. Attached Figure Description

[0053] Figure 1 This is a flowchart of the method of the present invention;

[0054] Figure 2 This is a system block diagram of the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] like Figure 1 As shown, embodiments of the present invention provide a method for intelligent scheduling optimization of printing orders based on equipment operation and maintenance status. The method includes:

[0057] Obtain the multi-source operation and maintenance parameter set of the equipment and the work order attribute information during printing, and identify the range of work orders of the same type;

[0058] Based on the same type of work order range, the loss level is divided to generate adaptation results, which are used to provide an adaptation basis for equipment and work order allocation;

[0059] Based on the adaptation results and work order attribute information, by analyzing the correlation strength between work orders with the same process and equipment component losses, the interference impact of work order switching on equipment production is identified, and disturbance intensity data is obtained.

[0060] By using disturbance intensity data, nodes to be assigned are identified, and feasibility results are obtained through the reassignment of work orders.

[0061] Based on the feasibility results, a production scheduling network is constructed. After path traversal and path replanning, the allocation optimization results are obtained to achieve intelligent production scheduling optimization.

[0062] In this embodiment, the matching and association between work orders and equipment can be accurately quantified, avoiding the concentrated superposition of high-loss switching. At the same time, through correlation analysis and path optimization, the accumulation of equipment fatigue and production disturbances are reduced, and the stability impact of order insertion and model change is mitigated. For example, when the equipment is at a low-loss level, it can handle high-load work orders such as high-speed printing and deep cleaning, while medium- and high-loss equipment only handles low-load work orders. This avoids excessive equipment wear and, through production scheduling network path optimization, disperses high-difference switching, preventing component damage caused by continuous high-load operation.

[0063] In addition, it can accurately identify disturbances caused by order insertion and work order switching, and avoid conflict nodes in advance. Compared with traditional production scheduling, it can reduce equipment failure downtime and reduce changeover losses through process adaptation constraints, and achieve dynamic matching between production scheduling and equipment operation and maintenance status. This solves the pain point of traditional production scheduling that only focuses on efficiency and ignores equipment losses and process adaptability.

[0064] In a preferred embodiment of the present invention, identifying work order intervals of the same type includes:

[0065] A standardized equipment operation and maintenance status acquisition process is established in the printing production system. Dedicated status acquisition units are deployed at key components such as printing equipment rollers, rubber rollers, and bearings to obtain multi-source operation and maintenance parameter sets and work order attribute information. The multi-source operation and maintenance parameter sets include operating data, historical operation and maintenance data, component wear data, and cleaning data. These parameter sets are then subjected to unified time axis alignment, outlier removal, missing data interpolation, and normalization processing. Furthermore, the operation behavior of the equipment during continuous production is divided into stages, identifying the time boundaries between continuous operation intervals and downtime intervals to obtain the equipment operation interval sequence and extract equipment operation and maintenance status segments. This step provides a time dimension basis for subsequent accurate extraction of equipment operation and maintenance status data for corresponding time periods, analysis of equipment fatigue changes during continuous production, and matching work order execution periods.

[0066] The operational data includes the temperature rise sequence of the rubber roller surface, the vibration spectral density sequence of the bearing, and the overprinting offset amplitude sequence;

[0067] Retrieve the equipment's historical operation and maintenance records for the past 18 months, and categorize and sort them by fault type, repaired parts, fault occurrence conditions, maintenance cycle, and maintenance execution details to form historical operation and maintenance data;

[0068] The device uses built-in high-precision sensors, including at least laser displacement sensors, eddy current sensors, visual image sensors, high-precision vibration sensors, and temperature sensors, to collect wear data on consumable components, such as the wear of rubber rollers, wear of doctor blades, surface precision deviation of printing rollers, and cleanliness of the inner wall of ink tanks. It also records the cumulative running time, cumulative number of cleaning cycles, and cumulative number of plate changes for each component to obtain cleaning data.

[0069] The work order attribute information includes the process requirements (such as paper thickness, printing speed, etc.) and production requirements (such as total print run, delivery cycle, ink type, printing plate type) of each work order; and converts the various parameters in the work order attribute information (such as paper thickness, ink type, printing speed, etc.) into standardized numerical vectors, so as to transform unstructured work order information into structured numerical data, which facilitates subsequent work order matching, changeover cost calculation, and production scheduling optimization through algorithms;

[0070] When the core parameters of the equipment, such as the roller speed fluctuation sequence and the rubber roller surface temperature rise sequence, are within the normal operating threshold and there are no downtime stamp records, it is marked as a continuous operating interval, which is the core period of effective equipment production. When a downtime stamp occurs or the core parameters are lower than the operating threshold (such as the roller speed approaching 0), it is marked as a downtime interval, which is the period when the equipment is in a stopped state, including planned downtime (such as plate changing and cleaning) and unplanned downtime (such as malfunction). The equipment operating interval sequence is the set of all continuous operating intervals and downtime intervals arranged in chronological order.

[0071] Equipment operation and maintenance status segments are collections of equipment operation and maintenance parameters corresponding to a certain continuous operating interval. They contain dynamic change data of all core evaluation parameters within that period, providing an accurate data source for subsequent analysis of equipment status changes and fatigue levels during specific production periods, and avoiding the inclusion of invalid data from downtime periods.

[0072] Assign switching attributes to work orders to update work order attribute information. Switching attributes include cleaning level and change type. This step is used to update work order attributes and clarify work order switching operation requirements.

[0073] Version change type and cleaning level are both classifications of process switching attributes for work orders, which are operational attribute classifications defined for process differences between adjacent work orders.

[0074] The steps for obtaining the cleaning level are as follows: First, establish a correspondence rule between ink color scheme and cleaning level: When the ink color schemes of adjacent work orders belong to the same color scheme (e.g., both are warm colors like red, orange, and yellow), and the color depth difference is ≤20%, it is judged as no cleaning; when the ink color schemes of adjacent work orders belong to similar color schemes (e.g., cool colors like blue and green), and the color depth difference is between 20% and 50%, it is judged as simple cleaning; when the ink color schemes of adjacent work orders belong to completely different color schemes (e.g., black → white, warm color → cool color), and the color depth difference is >50%, it is judged as deep cleaning. This is achieved through... The formula calculates the color depth difference, where This represents the percentage difference in color depth, i.e., the difference in color depth. and These are the lightness values ​​of the inks in the previous and current work orders (in the Lab color space). The channel value range is 0-100, where 0 is pure black and 100 is pure white.

[0075] It's important to note that color depth alone cannot accurately determine cleaning levels. For example, deep red and light red, even with significant differences in color depth, have similar ink compositions and are relatively easy to clean. Conversely, black and yellow, even with small differences in color depth, have significantly different ink compositions and are difficult to clean. Therefore, color matching is a fundamental constraint, while color depth difference is a more refined criterion. Only by combining both can the cleaning process be more accurately matched, avoiding over- or under-cleaning.

[0076] The steps for obtaining the printing plate change type are as follows: First, obtain the printing roller width specifications (e.g., A3, A4) of the current work order and the previous work order, and calculate the width difference rate: when the printing roller width specifications of adjacent work orders are exactly the same, it is determined to be a same-specification printing plate change; when the difference in printing roller width specifications between adjacent work orders is ≤30%, it is determined to be a small-width printing plate change; when the difference in printing roller width specifications between adjacent work orders is >30%, it is determined to be a full-specification printing plate change. The formula is used to... To obtain the differences in the printing roller width specifications, among which This represents the percentage difference in roller width specifications between adjacent work orders. and These are the roller width areas of the previous work order and the current work order, respectively. and For work order number, ;

[0077] Based on the multi-source operation and maintenance parameter set, the printing work order set within the current production scheduling cycle is read, and the similarity between adjacent work orders is analyzed in combination with the work order attribute information to identify the work order interval of the same type; this step is used to identify the work order interval of the same type, laying the groundwork for subsequent loss classification.

[0078] A printing work order set refers to the set of work orders assigned to the corresponding equipment, arranged in chronological order for execution.

[0079] Specifically through the formula Identify work order ranges of the same type, among which This is the similarity value, ranging from 0 to 1. The closer it is to 1, the more similar the adjacent work orders are. and The paper thickness of adjacent work orders. and The ink category code value is pre-established by the system to map different ink types to standardized integers, such as: CMYK four-color inks take 1, spot color red takes 2, spot color gold takes 3, UV inks take 4, etc., which is used to quantitatively distinguish different types of inks; and For the printing speed of adjacent work orders, This represents the permissible standard deviation of paper thickness variation. The permissible standard deviation of printing speed fluctuation, For ink category matching function, if If the result is positive, then take 1; otherwise, take 0. The formula uses an exponential function; it employs a Gaussian kernel function to quantify the similarity between paper thickness and printing speed, combined with strict matching of ink types, avoiding the rigidity of a single threshold judgment, and better reflecting the continuous change characteristics of parameters in actual production, thus achieving more accurate identification of the same type of work order intervals.

[0080] when When the similarity threshold is exceeded, it is determined to be a work order range of the same type;

[0081] By converting the ink types in actual printing production (such as CMYK four-color inks, special color inks, spot color inks, etc.) into standardized integers (or discrete values), the unstructured ink information can be processed in a structured and computable manner.

[0082] In this embodiment, by standardizing multi-source parameters and dividing the operating range, invalid data during downtime is accurately eliminated, avoiding subsequent analysis deviations caused by interference from maintenance data, and achieving accurate quantification of equipment fatigue status.

[0083] By employing a similarity algorithm that combines a Gaussian kernel function with strict matching of ink categories, the rigidity of single threshold judgment is avoided, accurately identifying work order ranges of the same type. Simultaneously, cleaning levels are categorized using both color scheme and color depth, and plate change types are categorized by sheet size difference rate. This achieves refined quantification of work order switching attributes, addressing the pain point of traditional, coarse-grained attribute classification. For example, adjacent work orders of the same color scheme, dark red and light red, are classified as requiring no cleaning even if there is a significant difference in color depth, avoiding waste of consumables and time caused by over-cleaning. Conversely, work orders of different color schemes, black and yellow, are classified as requiring deep cleaning even if there is a small difference in color depth, ensuring thorough cleaning.

[0084] Furthermore, by transforming unstructured ink information into standardized values, work order information can be calculable, providing accurate data support for subsequent correlation analysis and production scheduling optimization. This balances data accuracy with production practicality, thereby improving the scientific nature and reliability of production scheduling optimization.

[0085] In a preferred embodiment of the present invention, loss levels are divided based on work order intervals of the same type to generate adaptation results, including:

[0086] Retrieve equipment operation and maintenance status segments within the same type of work order interval, analyze the equipment's operational behavior during continuous work order execution, and obtain equipment fatigue value data, including equipment fatigue values ​​for each continuous work order execution segment. This step is used to accurately quantify the degree of equipment fatigue, providing data support for subsequent loss analysis.

[0087] Specifically, the fatigue level of equipment is quantified by the rate of change of equipment operation and maintenance status. The calculation method is to obtain the rate of change of surface temperature of rubber roller, the rate of change of bearing vibration frequency band energy, and the rate of change of printing offset fluctuation amplitude through equipment operation and maintenance status segments, and obtain the equipment fatigue value through a weighted summation algorithm. Each weight can be obtained by entropy weight method, analytic hierarchy process, or multiple linear regression method. The role of equipment fatigue value is to quantify the degree of fatigue of equipment due to continuous production of similar process work orders within a certain work order interval of the same type. It is a pure state quantity of equipment, which only reflects the fatigue trend of the equipment itself in continuous operation. Each work order interval of the same type is labeled with fatigue intensity to distinguish the impact of different intervals on equipment wear and tear.

[0088] Based on the trend of characteristic values ​​within the equipment fatigue data, the system identifies intermittent anomalies and classifies wear levels. This step is used to promptly detect equipment anomalies and provides a basis for wear level classification.

[0089] Specifically, using equipment fatigue value data as a prerequisite trigger condition, when the equipment fatigue value data shows a continuous increase, continuous accumulation, or exceeds the preset fatigue benchmark, the system automatically generates an intermittent anomaly identification start command.

[0090] Upon receiving the command, the system extracts continuous operating segments corresponding to the fatigue accumulation period from the equipment operating interval sequence. The bearing vibration energy, overprinting offset amplitude, and rubber roller temperature rise within the segment are used as monitoring objects. The system uses a sliding time window to compare real-time data with the rated threshold point by point, and selects fluctuation segments that are momentarily out of tolerance and quickly self-heal as candidate intermittent abnormal segments.

[0091] After candidate intermittent abnormality segments are formed, correlation analysis is performed on the abnormal events within the candidate abnormality segments. By calculating the time interval, abnormal amplitude change and abnormal recovery time between abnormal events, the correlation between abnormal events is identified. When multiple abnormal events show periodic distribution characteristics on the time axis, they are identified as intermittent abnormal behavior of the equipment, and an abnormal state segment is established for the intermittent abnormal behavior.

[0092] By analyzing the overall operation and maintenance status of the equipment through a multi-source set of operation and maintenance parameters, a comprehensive health value is obtained. This value is used to transform the abstract overall status of the equipment into a calculable and comparable quantitative indicator. The higher the value, the better the overall operation and maintenance status of the equipment.

[0093] The following parameters are selected from the multi-source operation and maintenance parameter set: equipment real-time operation parameter compliance rate, component wear and tear rate, historical failure rate, and component cleaning compliance rate. The values ​​of each indicator are added together and then divided by the total number of selected indicators (arithmetic mean method) to obtain the comprehensive health value. Among them, the equipment real-time operation parameter compliance rate is the ratio of the number of parameters that meet the rated standards in real-time operation to the total number of parameters.

[0094] The component loss remaining rate is calculated by iterating through the difference between the design loss threshold and the actual loss value of all components, then comparing the difference with the component design loss threshold, and finally dividing the result of the ratio calculation for all components by the total number of components.

[0095] The historical failure rate is the ratio of the equipment's fault-free operating time within the statistical period to the total duration of the statistical period.

[0096] The component cleaning compliance rate is the ratio of the number of times all components meet the standard cleaning requirements to the total number of cleaning times, and is obtained by the arithmetic mean method.

[0097] Based on the process characteristics of the work order, a process matching coefficient is set, and combined with equipment fatigue value data, a fatigue loss value is constructed. The calculation formula is as follows: ,in The fatigue loss value is based on the equipment fatigue value, superimposed with the work order process intensity, load, temperature deviation, and process matching coefficient. It represents the equipment-work order coupling loss, reflecting the actual measurable loss caused to the equipment by continuously executing such work orders, and is used to determine whether the equipment has entered the fatigue deterioration warning stage. This step is used to achieve accurate matching between the equipment status and the work order process.

[0098] When calculating fatigue loss values, the equipment fatigue value obtained in step one is first used as the basic quantitative value, which only reflects the equipment's own fatigue accumulation state within the continuous operating range. Then, the process matching coefficient corresponding to the currently executed work order of the same type is retrieved. Combined with the local component load rate, the deviation between the real-time component temperature and the reference temperature during continuous operation, and the equipment's maximum allowable temperature parameter, the basic equipment fatigue value is multiplied by the process matching coefficient and the local component load rate in sequence according to the multi-factor coupled progressive calculation logic. Then, it is multiplied by the temperature correction term composed of temperature deviation and maximum allowable temperature. By layering the influence of work order process intensity, actual equipment load state and operating temperature deviation, the fatigue value that simply reflects the equipment's own state is transformed into a comprehensive quantitative result that can characterize the gradual loss and deterioration degree of the equipment caused by the continuous execution of a specific process work order. Finally, a fatigue loss value that can be used for fatigue early warning judgment and loss level classification is obtained.

[0099] This represents the equipment fatigue value. Load rate of local components of the equipment; The deviation between the real-time temperature of the component and the reference temperature. The maximum allowable temperature for the component; The process matching coefficient is specifically used to amplify or correct equipment fatigue based on the strength of the processes in the currently continuously executed work orders, so that the fatigue value more accurately reflects the degree of damage to the equipment caused by the work orders.

[0100] The local component load rate of equipment refers to the ratio of the actual operating load borne by key load-bearing components such as printing rollers, rubber rollers, and bearings during the continuous execution of the same type of work order to the rated maximum design load of the component. It is used to correct and improve the accuracy of fatigue wear quantification.

[0101] Specifically, the method for setting the process matching coefficient is as follows: First, read the process attributes of work orders within the current range of similar work orders, including printing speed, paper thickness, ink type, whether deep cleaning is required, and whether it is a high-pressure production condition; then, divide the process into normal conditions and high-pressure conditions. When the work order is a high-load process such as high-speed printing, dark ink printing, thick paper printing, or deep cleaning, it is determined to be a high-pressure condition, and the process will be... The value is assigned to 1.2; when the work order is a standard process with normal speed, ordinary paper, light-colored ink, and no or simple cleaning, it is judged as a normal working condition. Assign a value of 1.0; after the assignment is complete, It is substituted as a fixed coefficient into the fatigue loss value formula and used in the final calculation.

[0102] The wear level is classified based on fatigue loss value and comprehensive health value, combined with intermittent abnormal behavior. The classification steps are as follows:

[0103] If the overall health value exceeds the upper limit of the preset health range, and the fatigue loss value does not exceed the lower limit of the preset loss range, and there are no intermittent abnormalities, then the loss level is the low loss level.

[0104] If the overall health value exceeds the lower limit of the preset health range but does not exceed its upper limit, and the fatigue loss value exceeds the lower limit of the preset loss range but does not exceed its upper limit, and there are no intermittent abnormalities, then the loss level is medium loss level.

[0105] If the overall health value does not exceed the lower limit of the preset health range, or the fatigue loss value exceeds the upper limit of the preset loss range, or there are intermittent abnormalities, then the loss level is high loss level.

[0106] The attribute constraints for accepting work orders are marked for equipment of each loss level to generate adaptation results. These results are a complete, unique, and formal output of the equipment's operation and maintenance status and production adaptability. This step is used to clarify the equipment's adaptability scope and ensure the compliance and rationality of production scheduling.

[0107] Specifically, after determining the equipment to be in low, medium, or high loss levels, the system uses historical equipment maintenance data, remaining lifespan of components, fatigue loss range, and intermittent anomaly indicators as the basis to sequentially mark work order attributes for equipment with different loss levels: For low loss level equipment, all process types are allowed to be accepted, allowing the acceptance of high-load, high-complexity work orders such as high-speed printing, thick paper printing, dark ink printing, deep cleaning, and full-specification plate change, without limiting the continuous production time or the number of consecutive work orders of the same type.

[0108] For medium-wear equipment, apply moderate constraints, prohibiting the acceptance of high-load work orders such as deep cleaning and full-size plate change. Only allow acceptance of regular work orders such as ordinary paper, regular printing speed, light-colored ink, simple cleaning or no cleaning, and same-size plate change. Also limit the maximum execution time of consecutive work orders of the same type to avoid further accumulation of fatigue.

[0109] For high-loss equipment, the strictest constraints are imposed, allowing only low-load work orders such as low-speed, thin paper, single-color ink, no cleaning, and same-specification plate change, and prohibiting the continuous execution of the same type of work orders;

[0110] After completing the above constraint annotation, the equipment number, comprehensive health value, fatigue loss value, intermittent abnormal working condition coefficient, operation and maintenance loss level, process range of work orders that can be accepted, types of work orders that cannot be accepted, continuous running time limit, cleaning level limit, and changeover type limit for each piece of equipment are uniformly compiled and summarized, and adaptation results are generated on a per-equipment basis, that is, the mapping relationship between loss level and process adaptation.

[0111] In this embodiment, multi-parameter coupling is used to quantify equipment fatigue and loss, combined with accurate identification of intermittent anomalies, breaking through the limitations of traditional methods that rely on a single indicator to determine equipment status. At the same time, fatigue loss values ​​are corrected by multiple factors such as process matching coefficients and local load rates, achieving accurate measurement of equipment-work order coupling loss. Furthermore, it can customize differentiated work order constraints for equipment with different loss levels, achieving a dynamic balance between loss and production capacity.

[0112] For example, low-wear equipment can handle high-load work orders such as high-speed and thick paper printing, while high-wear equipment can only handle low-load work orders. If a piece of equipment detects intermittent abnormalities in periodic bearing vibration, it is directly judged as a high-wear level, and continuous production is prohibited to avoid the escalation of the fault. In addition, the arithmetic average calculation method of multiple indicators of comprehensive health value takes into account both the real-time status of the equipment and its historical maintenance level. The adaptation result covers all dimensions of equipment information, providing a unique and accurate basis for subsequent production scheduling. This reduces excessive equipment wear and avoids production waste caused by work order mismatch, improving the precision and innovation of production scheduling adaptation.

[0113] In a preferred embodiment of the present invention, based on the adaptation results and work order attribute information, by analyzing the correlation strength between work orders of the same process and equipment component losses, the interference impact of work order switching on equipment production is identified, and disturbance intensity data is obtained, including:

[0114] Using the adaptation results as prior adaptation constraints, combined with equipment fatigue value data and work order process combinations, the correlation strength between different process work orders and equipment component losses is calculated through Pearson correlation coefficient analysis. The correlation coefficient is used to characterize the degree of influence of work orders on equipment component losses. The larger the absolute value, the stronger the aggravating effect of the process combination on the component losses.

[0115] This step is used to accurately quantify the correlation between process work orders and equipment component losses, reducing misjudgments of losses.

[0116] The equipment components include at least rubber rollers, bearings, and printing rollers;

[0117] In practical implementation, equipment operation and maintenance data, work order execution data, and equipment fatigue state sequences are used as calculation samples. The correlation coefficient between each process combination and the wear of each component is calculated using the Pearson correlation coefficient formula. The focus is on identifying the aggravating effect of high-pressure processes on the wear of key components. The coefficients are then filled into a matrix to form a quantitative coupling relationship between work orders and equipment wear. The relevant formula is: ,in For the first The combination of processes and the first The correlation coefficient of each component For process combination index, For component indexing, For sample index, The total number of samples; For the first In the nth sample, the nth The execution time of this process combination, i.e. The cumulative running time of a certain process work order on the equipment; For the first In the nth sample, the nth The wear value of each component is obtained by subtracting the component's remaining wear rate from 1, reflecting the degree of wear of the component during operation.

[0118] Work order process combination refers to the fixed combination of multiple process attributes of a work order into a set of calculable features. The combination includes paper thickness, printing speed, ink type, printing plate type, cleaning level, and plate change type.

[0119] Based on the correlation coefficient and work order attribute information, we analyze the differences in attributes between adjacent work orders and the changes in the status of switching nodes to obtain equipment replacement impact data that characterizes the impact of work order sequence on equipment wear and tear. This data is used to characterize the degree of fluctuation of work order switching on equipment.

[0120] This step is used to clarify the impact of work order switching on equipment, providing a basis for optimizing production scheduling.

[0121] In practice, the calculated correlation coefficient is first called and used as the loss weight of each attribute of the process difference, that is, the comprehensive correlation coefficient. Based on the work order attribute information and the equipment operation interval sequence, the paper thickness, ink color, printing plate type and other attributes of adjacent work orders in the printing work order set are extracted. The weight of each attribute is assigned in combination with the correlation coefficient. The process attribute with the larger absolute value of the correlation coefficient (that is, the attribute with the stronger impact on equipment loss, such as dark ink corresponding to a high correlation coefficient) is assigned a higher weight. Then, the comprehensive process difference of adjacent work orders is calculated based on the weight.

[0122] Subsequently, adjacent work orders with process differences exceeding a preset threshold are scanned and identified as high-difference process conversion work order pairs. The corresponding process switching nodes are located and marked. Then, the comprehensive process differences of the equipment before and after the switch are extracted as the changeover impact value to complete the loss quantification of the switchover process. Finally, equipment changeover impact data characterizing the impact of work order sequence on equipment loss is generated.

[0123] High-difference process changeover work orders are used to identify high-risk changeover points that must be avoided as much as possible, require buffering, or necessitate extended adjustment time; process changeover nodes are used to locate sudden changes in equipment status and extract vibration, temperature, and loss changes before and after that moment to quantify changeover losses.

[0124] For adjacent work orders, the difference percentage of each feature in the work order attribute information is obtained by calculating the difference percentage of color depth, and the comprehensive process difference is obtained by weighted summation.

[0125] The equipment changeover impact data is a set of quantified work order changeover losses arranged in chronological order. It contains multiple sets of changeover impact values. The changeover impact values ​​belong to the losses of a single changeover node caused by the changeover between adjacent work orders in the initial printing work order set analysis stage. It is static loss data at the equipment level and is used to set adjacent work order constraints, changeover cost constraints, and cleaning or plate change constraints.

[0126] Based on correlation coefficients and equipment changeover impact data, this step analyzes the production scheduling structure changes and equipment changeover increments caused by order insertions to obtain disturbance intensity data characterizing the impact of order insertions on production stability. This step is used to accurately assess the impact of order insertions, ensure production stability, and avoid exceeding disturbance limits.

[0127] In practice, based on the correlation coefficient between work order processes and equipment losses, as well as the data on the impact of equipment replacement, the insertion position of the inserted order within the established set of printing work orders is first determined. The adjacent work order combinations affected before and after the insertion are extracted. The sum of the original replacement impact value before the insertion and the sum of the newly added replacement impact value after the insertion are calculated separately. The difference between the two is taken as the replacement increment. Then, the replacement increment is multiplied by the absolute value of the correlation coefficient to obtain the disturbance intensity corresponding to a single insertion. This ultimately forms scheduling disturbance data characterizing the impact of the insertion on production stability, providing a complete basis for subsequent constraint setting. It should be noted that the correlation coefficient here is the comprehensive correlation coefficient corresponding to the work order process involved in the current insertion, i.e., the average value of the impact of this process on the losses of all core components.

[0128] The production scheduling disturbance data includes multiple sets of disturbance intensities. The disturbance intensity refers to the impact of order insertion behavior on the entire production scheduling structure during the order insertion decision evaluation stage, and is used to describe the impact of order insertion on system stability.

[0129] "Installation order" refers to an urgent order that is temporarily added during the production process. It is not an order in the initial production schedule and will forcibly disrupt the original work order order, increase the need for changeovers, and increase equipment wear and tear and disturbance.

[0130] In this embodiment, the impact of different process work orders on the wear and tear of equipment components is accurately quantified by the Pearson correlation coefficient, avoiding the error of traditional judgment based on experience. For example, the higher the correlation coefficient between the dark ink work order and the rubber roller component, the earlier the risk of wear and tear on the rubber roller can be predicted, and the component damage can be avoided in advance.

[0131] By innovatively constructing a quantitative coupling relationship between work orders and equipment wear, a calculable matrix data is formed, solving the pain point of traditional scheduling's difficulty in quantifying wear. Furthermore, it accurately captures the impact of interrupted orders on production stability. For example, if an urgent interruption leads to increased plate changeover frequency, the correlation coefficient can quickly pinpoint high-risk switchover points, allowing for advance adjustment time and preventing excessive equipment wear. For instance, when a high correlation coefficient is detected between a dark ink work order and the rubber roller component, the work order can be restricted from execution on the rubber roller in advance, preventing damage due to prolonged high loads. This ensures production continuity, achieves refined scheduling through quantitative data, accurately identifies high-risk switchover points, reduces equipment downtime, and balances production efficiency and equipment lifespan, achieving a dual improvement in scheduling stability and wear control.

[0132] In a preferred embodiment of the present invention, nodes to be assigned are identified using disturbance intensity data, and feasibility results are obtained through reassignment operations via work order switching, including:

[0133] Based on disturbance intensity data, the distribution of disturbance intensity in the printing work order set within the equipment operating interval sequence is analyzed to identify disturbance segments and nodes to be assigned in the production scheduling structure. This step is used to accurately locate disturbance segments and high-disturbance nodes, providing a basis for subsequent production scheduling optimization.

[0134] In practice, the production scheduling disturbance data is first read, and the equipment operating interval sequence is read simultaneously. Since the production scheduling disturbance sequence is generated based on the work order sequence, while the equipment operating interval sequence is divided based on the equipment operating time, the two types of data need to be time-aligned first, so that each disturbance intensity in the printing work order set can be mapped to a specific equipment operating interval. After time alignment, the production scheduling disturbance sequence is scanned segment by segment, using equipment operating intervals as units, and it is determined whether there is a disturbance intensity exceeding a preset disturbance threshold. When a disturbance intensity value exceeding the disturbance threshold is detected, the corresponding equipment operating interval is marked as a production scheduling structure disturbance segment, and all work orders involved in this segment and their sequential positions in the printing work order set are recorded.

[0135] After identifying the disturbance sections, the system further analyzes the work order switching structure within those sections. Specifically, it reads the work order attribute information and extracts the switching attributes between adjacent work orders in the printing work order set, including cleaning level and plate change type. These switching attributes are then constructed into a work order switching attribute sequence according to the production scheduling order, ensuring that each work order switching node in the printing work order set corresponds to a set of cleaning level and plate change type attributes. Next, this work order switching attribute sequence is matched point-by-point with the generated equipment changeover impact data, ensuring that each work order switching node simultaneously corresponds to a changeover impact value.

[0136] After completing the node attribute matching, a comprehensive judgment is made on each work order switching node: when the changeover impact value corresponding to a certain switching node exceeds the preset impact threshold, and the scheduling disturbance intensity corresponding to its location also exceeds the preset intensity threshold, the node is marked as a node to be assigned, i.e. a high disturbance node.

[0137] Based on the nodes to be assigned, the cleaning level and change type of the work order switching are analyzed by combination to construct switching allocation constraints between work orders, and conflict nodes are obtained to realize the reassignment of work order switching operations; this step is used to construct switching constraints to avoid work order switching conflicts and ensure smooth production scheduling.

[0138] Based on the conflict nodes within the disturbance zone, feasible results are allocated by analyzing the matching between equipment capacity and work order switching structure. This step is used to match the compatibility between equipment and work orders, and output feasible production scheduling basis that can be implemented.

[0139] In practice, the loss level and adaptation results are used as constraints on equipment capabilities. The equipment loss level is used to characterize the current operating status of the equipment, while the adaptation results are used to describe the types of work orders that different equipment can execute. Based on these data, the system initially selects a set of executable equipment for each work order.

[0140] Subsequently, the execution relationships of each work order in the printing work order set are analyzed one by one by using conflict nodes within the perturbation section. Specifically, for each work order in the printing work order set, its position in the set is first determined, and the numbers of the adjacent work orders before and after it are extracted. Then, the switching constraint information between the work order and its adjacent work orders is determined to determine whether conflict nodes will be formed when the work order is executed on different devices;

[0141] When a work order on a certain device is detected to cause a conflict node with an adjacent work order during execution, the device and the work order are determined to not meet the scheduling stability requirements, and the device is removed from the set of executable devices for that work order. When a device and the corresponding work order do not cause a structural conflict during execution, the device is retained as a candidate execution device for that work order. By screening all work orders in the printing work order set one by one, a set of device and work order feasibility is finally obtained, namely the feasibility result. Each work order corresponds to one or more candidate execution devices that meet the device state constraints, process adaptation constraints, and switching structure constraints, providing basic input data for subsequent scheduling optimization calculations. The feasibility result is used to output prohibited adjacent switching pairs and prohibited device assignments, i.e., which work orders cannot be adjacent and which devices cannot be connected.

[0142] In this embodiment, disturbance segments and nodes to be assigned in production scheduling are identified by disturbance intensity data. Accurate data matching is achieved by combining time axis alignment processing. Then, by analyzing work order switching attributes and constructing switching constraints, the identification of nodes to be assigned and the work order reassignment are completed. This can accurately locate switching nodes with high disturbance and high conflict, avoid production scheduling risks in advance, and achieve dynamic optimization of the production scheduling structure by screening conflict nodes and matching equipment capabilities. At the same time, it solves the problems of inaccurate disturbance identification and difficulty in avoiding switching conflicts in traditional production scheduling.

[0143] For example, when the disturbance intensity of a work order switching exceeds the threshold, it will be automatically marked as a high disturbance node and the allocation will be adjusted to avoid increased equipment wear due to switching conflicts. It can also accurately identify the stability impact of order insertion and switching, and reserve adjustment space in advance. For example, after a machine executes a high-speed printing work order, the system can quickly identify that its disturbance exceeds the standard and automatically adjust the work order sequence to avoid equipment failure caused by continuous high-load switching. This not only ensures the stability of production scheduling but also reduces equipment wear. At the same time, it achieves accurate adaptation in order insertion scenarios, solves the pain points of delayed disturbance identification and difficulty in predicting conflicts in traditional production scheduling, and improves the scientificity and stability of production scheduling.

[0144] In a preferred embodiment of the present invention, based on the nodes to be assigned, by combining and analyzing the cleaning level and version change type of the work order switching, switching assignment constraints between work orders are constructed to obtain conflicting nodes, including:

[0145] Obtain the switching attributes of the nodes to be assigned; when the switching of the node to be assigned belongs to both deep cleaning and full-size replacement, it is recorded as a high-intensity switching; when the switching of the node to be assigned belongs to only simple cleaning or small-size replacement, it is recorded as a medium-intensity switching; when the switching of the node to be assigned belongs to no cleaning and same-size replacement, it is recorded as a low-intensity switching; this step is used to identify the switching attributes of the nodes to be assigned, accurately classify the switching intensity, and provide a basis for subsequent constraint judgment.

[0146] It should be noted that if the node to be assigned is switched to a deep cleaning or a version change of the same specification, it is recorded as a medium intensity switch. Conversely, if the node to be assigned is switched to a version change without cleaning or a version change of the full specification, it is also recorded as a medium intensity switch.

[0147] If it is a high-intensity switchover, the corresponding node to be assigned will be classified as a conflict node. This step is used to screen high-risk switchover nodes and avoid equipment damage and production disruptions caused by high-intensity switchovers.

[0148] Specifically, after obtaining the nodes to be assigned, the system first reads all nodes to be assigned and locates the adjacent work order pairs corresponding to each node based on the printing work order set. Then, it extracts the cleaning level and plate change type attributes involved in the switching process of that work order pair. Since different types of cleaning and plate change operations have varying degrees of impact on equipment operational stability during printing production, it is necessary to perform joint analysis of the cleaning level and plate change type.

[0149] In the joint analysis process, the cleaning level is first classified and identified, such as deep cleaning, simple cleaning and no cleaning, and the changeover type is also classified and identified, such as full-size changeover, small-format changeover or same-size changeover, and then the two types of attributes are combined and analyzed.

[0150] Based on the feasibility results, a production scheduling network is constructed. After path traversal and path replanning, the optimized allocation results are obtained, including:

[0151] Based on the feasibility results, a production scheduling network is constructed by defining the adjacent transformation relationship between equipment allocation and work orders. The nodes of the production scheduling network represent the running status of equipment executing a single work order in the corresponding time period, and the edges represent the sequential transformation relationship. This step is used to construct the production scheduling network, clarify the work order connection logic, and provide a foundation for path traversal and optimization.

[0152] After removing the incompatible devices corresponding to the conflicting nodes, the work order combination corresponding to the conflicting nodes will be accurately assigned to other candidate devices that can execute the combination without conflict and meet the adaptation requirements. After re-verification, the new devices and work order combinations will be included in the production scheduling status network, and the appropriate execution path will be found through path search.

[0153] In practical implementation, a production scheduling status diagram is constructed using the order of work order execution as the time progression dimension and equipment number as the node distribution dimension. Each node represents the operational status of a single piece of equipment executing a single work order in a corresponding time period. The connecting edges between nodes represent the sequential transformation relationship between work orders. Equipment changeover impact data, production scheduling disturbance data, and equipment fatigue value data are loaded into the transformation edge attributes. At the same time, the cleaning level and changeover type corresponding to each transformation edge are recorded, ensuring that the status network only includes equipment allocation and adjacent switching relationships allowed by feasible results. Among them, the operational status includes equipment changeover impact data, production scheduling disturbance data, and equipment fatigue value data.

[0154] The sequential conversion relationship refers to the connection relationship where the execution of the previous work order is completed and the execution of the next work order is carried out in sequence. It represents the sequential switching of work orders on the timeline and carries switching attributes such as cleaning level, change type, change impact and disturbance intensity.

[0155] Starting from the initial node within the production scheduling network, all paths are traversed, and equipment fatigue value data, equipment replacement impact data, and disturbance intensity data are accumulated node by node. That is, the equipment fatigue value, replacement impact value, and disturbance intensity are accumulated node by node. If any data (accumulated equipment fatigue value, replacement impact value, and disturbance intensity at each node) exceeds the limit (exceeds the corresponding operation and maintenance constraint threshold), the extension is terminated and marked as an infeasible path. Only paths that have completely traversed all work orders and have not exceeded the limit throughout are retained and recorded as candidate paths. This step is used to accumulate and verify data node by node to ensure the reliability of the path and to select feasible production scheduling paths.

[0156] The starting node is the status node of the first executable work order for all devices in the earliest time slice of the production schedule;

[0157] Traversing all work orders means assigning all work orders in this production schedule to the corresponding equipment and completing the timing sequence without omission or loss.

[0158] If no candidate path exists, the constraint threshold flexible adjustment mechanism is automatically triggered. The maintenance constraint threshold is slightly adjusted to avoid excessive relaxation leading to increased equipment wear. The adjustment range is controlled within 10%-20% of the original threshold. After adjustment, the scheduling status network path traversal process is restarted to screen for feasible scheduling paths again. If no feasible path is found, the adjustment can be repeated 1-2 times (the cumulative adjustment range should not exceed 30% of the original threshold) until a feasible path is found. If a feasible path still cannot be generated, the system automatically outputs an exception prompt and pushes it synchronously to the production scheduling end, prompting maintenance personnel to intervene and adjust. Maintenance personnel can choose to suspend some non-urgent work orders, temporarily schedule backup equipment to supplement capacity, or adjust the delivery cycle of some work orders according to actual production needs. After the adjustment is completed, the system re-executes the scheduling status network construction and path traversal process until a feasible scheduling path is generated.

[0159] In this invention, all parameters are dimensionless through dimensionless processing technology to remove their dimensions; all thresholds in this invention can be obtained using the mean-standard deviation method.

[0160] Based on candidate paths, adjacent work orders are partially swapped, followed by an over-limit check (the cumulative equipment fatigue value, replacement impact value, and disturbance intensity at each node are compared with the corresponding maintenance constraint thresholds). If the limits are exceeded, the partial swap is cancelled; otherwise, the new order, i.e., the path after the partial swap, is retained. High-difference switching is dispersed through iterative iteration to obtain the optimized allocation result. This step is used to optimize the work order order through iterative swapping, disperse high-difference switching, and ensure stable production.

[0161] In practice, the system reads the work order sequence in the candidate path one by one and performs local swap operations on adjacent work orders in the sequence. Each swap operation recalculates the corresponding changeover impact value, equipment fatigue value, and production scheduling disturbance value, and compares them with the corresponding indicators in the original sequence. If none of the swapped indicators exceed the constraint threshold, the new work order sequence is retained; if any indicator exceeds the limit, the swap operation is canceled and the original sequence is restored. After completing one local swap, the system continues to perform the same swap calculation on the next group of adjacent work orders, and repeats this operation throughout the entire sequence. Through multiple rounds of iterative traversal, work orders with large changeover differences in the sequence are gradually dispersed, thereby reducing the frequency of continuous high-difference switching.

[0162] The allocation optimization result is the final determination of each device, the order in which work orders are executed, and the time period for execution. By exchanging and distributing, high conflicts are no longer continuous, and cumulative indicators will not rise linearly.

[0163] The purpose of exchanging adjacent work orders is to disperse the high-difference process switching that occurs continuously in a concentrated set of printing work orders, so as to avoid the rapid increase in the changeover impact value, equipment fatigue accumulation value and disturbance intensity caused by the concentrated superposition of high-intensity switching. By adjusting the local sequence to distribute the high-difference switching intervals, the cumulative operation and maintenance indicators of the whole process can be effectively reduced, making the production scheduling structure more stable and the constraint satisfaction higher.

[0164] In this embodiment, the switching attributes of the nodes to be assigned are first accurately identified to distinguish between high-intensity, medium-intensity, and low-intensity switching types, and a scientific switching constraint system is constructed to avoid the concentrated superposition of high-intensity switching. Then, by classifying and marking switching types of different intensities, high-risk switching nodes with high disturbance and high loss are accurately located, solving the problems of chaotic switching and excessive loss in traditional scheduling.

[0165] Subsequently, a production scheduling network containing equipment operating status is constructed to distribute high-differentiation work orders in a decentralized manner, avoiding continuous high-intensity switching. For example, work orders for dark ink and light paper that were originally continuous are now concentrated and overlapped. After local exchange and adjustment, continuous high-load switching no longer occurs, which reduces equipment fatigue and wear and tear and avoids resource waste caused by excessive cleaning.

[0166] Through iterative optimization, highly variable work orders are no longer concentrated and piled up, which ensures production continuity and balances equipment wear and production efficiency, effectively improving the stability and adaptability of production scheduling.

[0167] like Figure 2 As shown, embodiments of the present invention also provide a printing order intelligent scheduling optimization system based on equipment operation and maintenance status, including:

[0168] The identification module is used to obtain the multi-source operation and maintenance parameter set of the equipment and the work order attribute information during printing, and to identify the range of work orders of the same type.

[0169] The first allocation module is used to divide the loss level based on the same type of work order range to generate the matching result, which is used to provide the matching basis for equipment and work order allocation;

[0170] The disturbance analysis module is used to identify the interference impact of work order switching on equipment production based on the adaptation results and work order attribute information by analyzing the correlation strength between work orders of the same process and equipment component losses, and to obtain disturbance intensity data.

[0171] The second allocation module is used to identify nodes to be allocated through disturbance intensity data and obtain feasibility results through the reassignment operation of work order switching;

[0172] The resource optimization module is used to construct a production scheduling network based on the feasibility results. After path traversal and path replanning, the allocation optimization results are obtained to achieve intelligent production scheduling optimization.

[0173] It should be noted that this system is a system corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0174] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent scheduling optimization of a printing order based on a device operation state, characterized in that, The method includes: Obtain the multi-source operation and maintenance parameter set of the equipment and the work order attribute information during printing, and identify the range of work orders of the same type; Based on the same type of work order range, the loss level is divided to generate adaptation results, which are used to provide an adaptation basis for equipment and work order allocation; Based on the adaptation results and work order attribute information, by analyzing the correlation strength between work orders with the same process and equipment component losses, the interference impact of work order switching on equipment production is identified, and disturbance intensity data is obtained. By using disturbance intensity data, nodes to be assigned are identified, and feasibility results are obtained through the reassignment of work orders. Based on the feasibility results, a production scheduling network is constructed. After path traversal and path replanning, the allocation optimization results are obtained to achieve intelligent production scheduling resource optimization.

2. The printing order intelligent scheduling optimization method based on the equipment operation state according to claim 1, characterized in that, Identify work order ranges of the same type, including: The system acquires multi-source operation and maintenance parameter sets and work order attribute information of the equipment, divides the operation behavior of the equipment in the continuous production stage into stages, identifies the time boundaries between the continuous operation interval and the downtime interval, obtains the equipment operation interval sequence, and extracts equipment operation and maintenance status segments. Assign switching attributes to work orders to update work order attribute information. Switching attributes include cleaning level and change type. Based on the multi-source operation and maintenance parameter set, the printing work order set within the current production scheduling cycle is read, and combined with the work order attribute information, the similarity between adjacent work orders is analyzed to identify work order intervals of the same type. 3.The method of claim 2, wherein, Based on the same type of work order range, loss levels are divided to generate adaptation results, including: Retrieve equipment operation and maintenance status segments within the same type of work order interval, analyze the equipment's operating behavior during continuous work order operation, and obtain equipment fatigue value data; Based on the trend of characteristic values ​​within the equipment fatigue value data, identify whether the equipment has intermittent anomalies and classify the wear level.

4. The printing order intelligent scheduling optimization method based on equipment operation status according to claim 3, characterized in that, Based on the same type of work order range, loss levels are divided to generate adaptation results, which also includes: By analyzing the overall operation and maintenance status of the equipment through a multi-source set of operation and maintenance parameters, a comprehensive health value can be obtained. Based on the process characteristics of the work order, a process matching coefficient is set, and fatigue loss value is constructed by combining it with equipment fatigue value data; The wear level is classified based on fatigue loss value and comprehensive health value, combined with intermittent abnormal behavior. The classification steps are as follows: If the overall health value exceeds the upper limit of the preset health range, and the fatigue loss value does not exceed the lower limit of the preset loss range, and there are no intermittent abnormalities, then the loss level is the low loss level. If the overall health value exceeds the lower limit of the preset health range but does not exceed its upper limit, and the fatigue loss value exceeds the lower limit of the preset loss range but does not exceed its upper limit, and there are no intermittent abnormalities, then the loss level is medium loss level. If the overall health value does not exceed the lower limit of the preset health range, or the fatigue loss value exceeds the upper limit of the preset loss range, or there are intermittent abnormalities, then the loss level is high loss level. Label the attribute constraints of the work orders that can be accepted for each level of equipment to generate adaptation results.

5. The printing order intelligent scheduling optimization method based on equipment operation status according to claim 4, characterized in that, Based on the adaptation results and work order attribute information, by analyzing the correlation strength between work orders of the same process and equipment component losses, the interference impact of work order switching on equipment production is identified, and disturbance intensity data is obtained, including: Using the adaptation results as prior adaptation constraints, and combining equipment fatigue value data and work order process combinations, the correlation strength between different process work orders and equipment component losses is calculated through Pearson correlation coefficient analysis to obtain the correlation coefficient. Based on the correlation coefficient and work order attribute information, we analyze the differences in attributes between adjacent work orders and the changes in the status of switching nodes to obtain equipment replacement impact data that characterizes the impact of work order sequence on equipment wear and tear. This data is used to characterize the degree of fluctuation of work order switching on equipment. Based on correlation coefficients and equipment changeover impact data, by analyzing the changes in production scheduling structure and equipment changeover increments caused by order insertion, we obtain disturbance intensity data characterizing the impact of order insertion on production stability.

6. The intelligent scheduling optimization method for printing orders based on equipment operation and maintenance status according to claim 5, characterized in that, By analyzing disturbance intensity data, nodes to be assigned are identified, and feasibility results are obtained through work order switching and reassignment operations, including: Based on disturbance intensity data, the distribution of disturbance intensity in the printing work order set in the equipment operation interval sequence is analyzed to identify disturbance sections and nodes to be assigned. Based on the nodes to be assigned, the cleaning level and change type of the work order switching are analyzed by combination to construct the switching allocation constraints between work orders, obtain the conflict nodes, and realize the redistribution of work order switching operations. Based on the conflict nodes within the disturbance section, feasible results are allocated by analyzing the matching between equipment capacity and work order switching structure.

7. The intelligent scheduling optimization method for printing orders based on equipment operation and maintenance status according to claim 6, characterized in that, Based on the nodes to be assigned, by combining the cleaning level and version change type of the work order switching analysis, switching assignment constraints between work orders are constructed to obtain conflicting nodes, including: Get the switching attributes of the node to be assigned; when the node to be assigned is both deep cleaning and full-size replacement, it is recorded as high-intensity switching; when the node to be assigned is only simple cleaning or small-size replacement, it is recorded as medium-intensity switching; when the node to be assigned is no cleaning and same-size replacement, it is recorded as low-intensity switching. If it is a high-intensity switchover, the corresponding node to be assigned will be classified as a conflict node.

8. The intelligent scheduling optimization method for printing orders based on equipment operation and maintenance status according to claim 7, characterized in that, Based on the feasibility results, a production scheduling network is constructed. After path traversal and path replanning, the optimized allocation results are obtained, including: Based on the feasibility results, a production scheduling network is constructed by limiting the equipment allocation and the adjacent conversion relationship of work orders. The nodes of the production scheduling network represent the running status of equipment executing a single work order in the corresponding time period, and the edges represent the sequential conversion relationship. Starting from the initial node within the production scheduling network, traverse all paths, accumulating equipment fatigue value data, equipment changeover impact data, and disturbance intensity data node by node. If any data exceeds the limit, terminate the extension and mark it as an infeasible path. Only retain paths that have completely traversed all work orders and have not exceeded the limit throughout the entire process, and record them as candidate paths. Based on candidate paths, local swaps are performed on adjacent work orders, followed by limit checks. If a limit is exceeded, the local swap is canceled; otherwise, the new order is retained. Through iterative cyclic switching to disperse high-difference switching, the allocation optimization result is obtained.

9. A printing order intelligent scheduling optimization system based on equipment operation and maintenance status, used to implement the printing order intelligent scheduling optimization method based on equipment operation and maintenance status as described in any one of claims 1 to 8, characterized in that, include: The identification module is used to obtain the multi-source operation and maintenance parameter set of the equipment and the work order attribute information during printing, and to identify the range of work orders of the same type. The first allocation module is used to divide the loss level based on the same type of work order range to generate the matching result, which is used to provide the matching basis for equipment and work order allocation; The disturbance analysis module is used to identify the interference impact of work order switching on equipment production based on the adaptation results and work order attribute information by analyzing the correlation strength between work orders of the same process and equipment component losses, and to obtain disturbance intensity data. The second allocation module is used to identify nodes to be allocated through disturbance intensity data and obtain feasibility results through the reassignment operation of work order switching; The resource optimization module is used to construct a production scheduling network based on the feasibility results. After path traversal and path replanning, the allocation optimization results are obtained to achieve intelligent production scheduling resource optimization.

Images