Notebook computer production scheduling optimization method and system for intelligent factory

Abstract

The invention provides a notebook computer production scheduling optimization method and system for an intelligent factory, and relates to the field of intelligent manufacturing, and the method comprises the steps: calculating a process matching degree, constructing a propagation incidence matrix, evaluating a delay risk, recognizing a key blocking process, setting a buffer time length, and generating a scheduling scheme based on a process connection time sequence diagram. And a delay event is responded in real time for dynamic adjustment. Production delay propagation can be effectively prevented, the elastic adaptive capacity of the production line is improved, and precise scheduling and optimization of the notebook computer production process are achieved.

Description

Technical Field

[0001] This invention relates to intelligent manufacturing technology, and more particularly to a method and system for optimizing the production scheduling of laptops in smart factories. Background Technology

[0002] In modern manufacturing, laptop production involves complex processes and precise time management. Laptop production in a smart factory environment requires an efficient scheduling system to coordinate various process nodes and ensure the continuity and efficiency of the production flow. Traditional laptop production scheduling mainly relies on rules of thumb or simple algorithms for resource allocation. With the expansion of production scale and the increase in personalized demands, the scheduling complexity increases significantly. Production scheduling optimization has become a key link in achieving flexible manufacturing and improving production efficiency in smart factories.

[0003] Existing scheduling methods struggle to effectively handle complex dependencies between processes, especially when processes are delayed. They lack a quantitative assessment mechanism for the risk of delay propagation, meaning a single point of delay can disrupt the entire production plan, impacting overall production efficiency. Furthermore, they lack dynamic response capabilities to sudden delays in the production process. When a process is delayed, the system struggles to quickly assess the impact and make targeted adjustments, often resorting to global rescheduling, resulting in resource waste and increased scheduling costs. The lack of buffer designs for critical process nodes prevents the system from effectively absorbing and mitigating random disturbances during production while maintaining overall production efficiency, thus affecting the stable operation of the production line and on-time delivery. Summary of the Invention

[0004] This invention provides a method and system for optimizing notebook computer production scheduling in smart factories, which can solve the problems in the prior art.

[0005] A first aspect of this invention provides a method for optimizing notebook computer production scheduling in smart factories, comprising: Receive a set of laptop production tasks and equipment status information, decompose the set of production tasks into process nodes, construct process dependencies based on the processing sequence between processes, and calculate the matching degree between the capacity requirements of process nodes and the processing capacity of equipment. Extract process delay records from historical production data, calculate the impact duration between delayed processes and subsequent processes, construct an inter-process propagation correlation matrix, and obtain the process node delay propagation risk value through recursive calculation based on the matching degree and the propagation correlation matrix. Based on the analysis of process dependencies using delayed propagation risk values, the propagation impact coefficient of process nodes is calculated, and the process node with the largest propagation impact coefficient is selected as the blocking process node. The buffer time at the blocking process node is calculated based on the process delay record. The process nodes are assigned to equipment according to the matching degree. A buffer duration is inserted at the blocking process nodes. A process connection sequence diagram is constructed based on the material transfer time between processes, and a scheduling execution plan is generated. Collect production line execution data, and when process delays are detected, calculate the scope of impact based on the propagation correlation matrix, allocate the delay time to the buffer time of the blocked process node, optimize material transfer time based on the process connection sequence diagram, and update the laptop production scheduling plan.

[0006] The system receives a set of laptop production tasks and equipment status information, decomposes the production task set into process nodes, constructs process dependencies based on the processing sequence between processes, and calculates the matching degree between the capacity requirements of the process nodes and the processing capacity of the equipment, including: Receive a set of laptop production tasks and equipment status information, decompose the set of production tasks into a set of process nodes, and extract the processing requirements of each process node; Traverse the set of process nodes, identify the predecessor and successor process nodes of each process node, establish directed connections from predecessor process nodes to successor process nodes based on the processing order between processes, count the in-degree value of each process node, and generate process dependency relationships containing directed connections and in-degree values. Extract the capacity demand parameters of each process node from the process dependency relationship, and extract the processing capacity parameters of each equipment from the equipment status information; The absolute value of the difference between the capacity requirement parameter of each process node and the processing capacity parameter of each equipment is calculated and normalized to the initial matching degree value. The initial matching degree value of the corresponding process node is corrected according to the in-degree value of each process node in the process dependency relationship, so as to obtain the matching degree between the capacity requirement of the process node and the processing capacity of the equipment.

[0007] Extract process delay records from historical production data, calculate the impact duration between delayed processes and subsequent processes, construct an inter-process propagation correlation matrix, and obtain the process node delay propagation risk value through recursive calculation based on the matching degree and the propagation correlation matrix, including: Extract process delay records from historical production data and identify delayed processes and subsequent processes; The time difference between the delayed process and the subsequent process is calculated as the duration of the influence between the delayed process and the subsequent process. The number of times the same delayed process points to the same subsequent process is counted and normalized to generate the propagation weight coefficient. Establish a propagation association record by setting the duration of the impact between the delayed process and the subsequent process and the propagation weight coefficient. Then, perform attenuation processing on the propagation weight coefficient according to the duration of the impact to generate attenuation weight coefficient. Finally, associate the attenuation weight coefficient with the propagation association record to construct the inter-process propagation association matrix. Extract the matching degree values ​​of each process node from the matching degree as the initial risk value; Extract the propagation association records starting from the process node from the inter-process propagation association matrix, and extract the initial risk value and attenuation weight coefficient of subsequent process nodes in the propagation association records to generate risk increment; The in-degree value of the process node in the process dependency relationship is extracted as a dependency amplification factor and coupled with the risk increment to obtain the modulated risk increment. The modulated risk increment is superimposed on the initial risk value of the process node, and the process is recursively executed according to the propagation path of the propagation association record until the propagation path terminates to obtain the delayed propagation risk value of the process node.

[0008] The in-degree values ​​of process nodes in the process dependency relationship are extracted as dependency amplification factors and coupled with the risk increment to obtain the modulated risk increment, which includes: Extract the in-degree values ​​of process nodes from the process dependency relationship, construct a process dependency matrix by combining the in-degree values ​​of process nodes with the process node capacity parameters, and extract the correlation values ​​between process nodes in the process dependency matrix to generate a process node dependency amplification factor. Based on the process node dependency amplification factor, the influence duration and propagation weight coefficient between process nodes are extracted from the inter-process propagation correlation matrix. The process node capacity fluctuation sequence is extracted according to the influence duration to construct the attenuation feature matrix. The attenuation feature matrix is ​​mapped with the process node dependency amplification factor to generate the initial risk increment. Based on the initial risk increment, the propagation path of the process node is extracted from the inter-process propagation correlation matrix. The node propagation influence intensity is calculated according to the process node propagation path. The initial risk increment is divided into propagation component and attenuation component according to the node propagation influence intensity. The change of the in-degree value of the process node is mapped into the propagation component modulation vector and the attenuation component modulation vector. The propagation component is mapped to the propagation component using the modulation vector of the propagation component to generate the modulated propagation component. The attenuation component is mapped to the attenuation component using the modulation vector of the attenuation component to generate the modulated attenuation component. The modulated propagation component and the modulated attenuation component are combined to obtain the modulated risk increment.

[0009] Based on the analysis of process dependencies using delay propagation risk values, the propagation impact coefficient of each process node is calculated. The process node with the largest propagation impact coefficient is selected as the blocking process node. The buffer time at the blocking process node is calculated based on the process delay records, including: Calculate the risk increase of the delayed propagation risk value of subsequent process nodes relative to the delayed propagation risk value of process nodes, identify subsequent process nodes whose risk increase exceeds a preset increase threshold as risk amplification nodes, and count the number of risk amplification nodes corresponding to each process node as the process node propagation impact coefficient. The process node with the largest value among the process node propagation influence coefficients is selected as the blocking process node; Extract the historical delay duration records of blocking process nodes and the corresponding delay occurrence times from the process delay records. Extract the impact duration of blocking process nodes pointing to risk amplification nodes from the inter-process propagation correlation matrix. Construct a delay propagation time window based on the delay occurrence time and impact duration. By filtering historical delay records of actual delays at risk amplification nodes within the delayed propagation time window, a propagation-related delay set is constructed. Statistical processing of the delay durations in the propagation-related delay set yields the buffer duration at the blocking process node.

[0010] The process nodes are assigned to equipment according to their matching degree. A buffer duration is inserted at the blocking process nodes. A process connection sequence diagram is constructed based on the material transfer time between processes, and a scheduling execution plan is generated, including: The difference between the matching degree value of the process node and the matching degree value of the equipment is calculated based on the matching degree. The equipment with the smallest difference is selected as the assigned equipment of the process node, and the initial assignment relationship between the process node and the equipment is generated. Extract the upstream process node that blocks the process node from the process dependency relationship, calculate the risk gap between the delay propagation risk value of the upstream process node and the blocking process node, amplify and adjust the buffer duration based on the risk gap value to obtain the adjusted buffer duration, and insert the adjusted buffer duration after the execution time point of the blocking process node in the initial allocation relationship. Extract the predecessor and successor process nodes of each process node from the process dependency relationship, obtain the material transfer time, calculate the risk transmission gradient between the delay propagation risk values ​​of the predecessor process node and the successor process node, compress and adjust the material transfer time based on the risk transmission gradient, and construct the process connection sequence diagram based on the completion time of the predecessor process node and the compressed material transfer time. Extract the start and finish times of each process node from the process connection sequence diagram, extract the allocation equipment corresponding to each process node from the initial allocation relationship, and combine them to generate a scheduling execution plan.

[0011] Collect production line execution data. When a process delay is detected, calculate the impact range based on the propagation correlation matrix, allocate the delay time to the buffer duration of the blocked process node, optimize material transfer time based on the process connection sequence diagram, and update the laptop production scheduling scheme, including: Collect the time deviation between the actual completion time and the planned completion time of the process node in the production line execution data, and calculate the weighted combination value of the time deviation value and the capacity requirement of the process node; The weighted combination value is corrected based on the in-degree value in the process dependency relationship, and the process node whose weighted combination value exceeds the preset range after correction is identified as the delayed process node. The propagation weight coefficient and attenuation weight coefficient corresponding to the delayed process node are extracted from the propagation correlation matrix. The propagation intensity value is obtained by recursive calculation in combination with the propagation influence coefficient of the process node. The set of affected process nodes is determined from the propagation correlation matrix based on the propagation intensity value. Extract the impact duration records of the affected process nodes set, calculate the delay time allocation weight based on the propagation intensity value and the impact duration records, and allocate the delay time to the buffer duration of the blocked process nodes according to the allocation weight; Extract material transfer time from the process connection sequence diagram, calculate the material transfer time compression ratio based on the allocated buffer time of the blocking process node, and optimize the material transfer time in the process connection sequence diagram according to the compression ratio; Substitute the optimized material transfer time and the buffer time of the blocked process nodes after allocation into the process connection sequence diagram, recalculate the start and finish times of the process nodes, and update the laptop production scheduling scheme.

[0012] A second aspect of the present invention provides a notebook computer production scheduling optimization system for smart factories, comprising: The first unit is used to receive the set of laptop production tasks and equipment status information, decompose the set of production tasks into process nodes, construct process dependencies based on the processing sequence between processes, and calculate the matching degree between the capacity requirements of process nodes and the processing capacity of equipment. The second unit is used to extract process delay records from historical production data, calculate the impact duration between delayed processes and subsequent processes, construct the inter-process propagation correlation matrix, and obtain the process node delay propagation risk value through recursive calculation based on the matching degree and the propagation correlation matrix. The third unit is used to analyze process dependencies based on delay propagation risk values, calculate the propagation impact coefficient of process nodes, select the process node with the largest propagation impact coefficient as the blocking process node, and calculate the buffer time at the blocking process node based on the process delay record. The fourth unit is used to allocate process nodes to equipment according to matching degree, insert buffer time at blocking process nodes, construct process connection sequence diagram based on material transfer time between processes, and generate scheduling execution plan; The fifth unit is used to collect production line execution data. When a process delay is detected, the impact range is calculated based on the propagation correlation matrix, the delay time is allocated to the buffer time of the blocked process node, the material transfer time is optimized based on the process connection sequence diagram, and the laptop production scheduling scheme is updated.

[0013] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0015] In this embodiment, by constructing an inter-process propagation correlation matrix and recursively calculating the delay propagation risk value of process nodes, the key process nodes most prone to triggering cascading delays in the production process can be accurately identified, providing a scientific basis for scheduling decisions. Identifying blocking process nodes based on the propagation impact coefficient and setting buffer durations enables early blocking of production delays, preventing single-point delays from spreading to the entire production line and effectively improving overall production efficiency. By collecting production line execution data in real time and calculating the delay impact range using the propagation correlation matrix, the delay time is rationally allocated to a preset buffer, achieving dynamic optimization of the scheduling scheme and enhancing the adaptability of the production system to sudden delays. Resource allocation is performed based on the matching degree between the capacity demand of process nodes and the processing capacity of equipment, and material transfer time is optimized based on the process connection sequence diagram, improving the utilization rate of production resources and reducing waiting time and waste in the production process. Decomposing complex production tasks into process nodes and constructing dependencies simplifies the scheduling decision-making process, enabling the scheduling system to focus more on optimizing key links and improving the feasibility and execution efficiency of the scheduling scheme. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the laptop production scheduling optimization method for smart factories according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating the process scheduling of an embodiment of the present invention, which combines equipment matching and delay risk control. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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.

[0018] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0019] Figure 1This is a flowchart illustrating a method for optimizing notebook computer production scheduling in a smart factory, as described in an embodiment of the present invention. Figure 1 As shown, the method includes: Receive a set of laptop production tasks and equipment status information, decompose the set of production tasks into process nodes, construct process dependencies based on the processing sequence between processes, and calculate the matching degree between the capacity requirements of process nodes and the processing capacity of equipment. Extract process delay records from historical production data, calculate the impact duration between delayed processes and subsequent processes, construct an inter-process propagation correlation matrix, and obtain the process node delay propagation risk value through recursive calculation based on the matching degree and the propagation correlation matrix. Based on the analysis of process dependencies using delayed propagation risk values, the propagation impact coefficient of process nodes is calculated, and the process node with the largest propagation impact coefficient is selected as the blocking process node. The buffer time at the blocking process node is calculated based on the process delay record. The process nodes are assigned to equipment according to the matching degree. A buffer duration is inserted at the blocking process nodes. A process connection sequence diagram is constructed based on the material transfer time between processes, and a scheduling execution plan is generated. Collect production line execution data, and when process delays are detected, calculate the scope of impact based on the propagation correlation matrix, allocate the delay time to the buffer time of the blocked process node, optimize material transfer time based on the process connection sequence diagram, and update the laptop production scheduling plan.

[0020] This embodiment demonstrates a dynamic closed-loop mechanism for notebook computer production scheduling. Initially, a scheduling plan is generated through task decomposition, capacity matching, and risk assessment. During execution, data is collected in real time. If a process delay is detected, the impact is immediately calculated, and the plan is optimized and updated. If no anomalies are detected, continuous monitoring continues until the task is completed, achieving automatic prevention of production risks and dynamic adjustment of progress.

[0021] In one optional implementation, the process involves receiving a set of laptop production tasks and equipment status information, decomposing the production task set into process nodes, constructing process dependencies based on the processing sequence between processes, and calculating the matching degree between the capacity requirements of the process nodes and the processing capacity of the equipment, including: Receive a set of laptop production tasks and equipment status information, decompose the set of production tasks into a set of process nodes, and extract the processing requirements of each process node; Traverse the set of process nodes, identify the predecessor and successor process nodes of each process node, establish directed connections from predecessor process nodes to successor process nodes based on the processing order between processes, count the in-degree value of each process node, and generate process dependency relationships containing directed connections and in-degree values. Extract the capacity demand parameters of each process node from the process dependency relationship, and extract the processing capacity parameters of each equipment from the equipment status information; The absolute value of the difference between the capacity requirement parameter of each process node and the processing capacity parameter of each equipment is calculated and normalized to the initial matching degree value. The initial matching degree value of the corresponding process node is corrected according to the in-degree value of each process node in the process dependency relationship, so as to obtain the matching degree between the capacity requirement of the process node and the processing capacity of the equipment.

[0022] In the laptop manufacturing process, to achieve efficient production scheduling, it is necessary to establish a matching relationship between process nodes and equipment capabilities. First, a set of laptop production tasks and equipment status information are received. The production task set contains multiple production orders, each including information such as product model, quantity, and delivery time. The equipment status information includes parameters such as the current status, processing capacity, and available time of each piece of equipment in the factory. For example, a batch of production tasks containing three models of laptops (A, B, and C), totaling 500 units, is received, along with status information for 30 pieces of equipment in the factory.

[0023] Next, the production task set is decomposed into a set of process nodes. Based on the laptop manufacturing process, the production task for each product model is broken down into multiple process nodes, such as component preparation, motherboard assembly, screen installation, keyboard installation, casing assembly, software installation, and quality testing. Simultaneously, the specific processing requirements for each process node are extracted, including processing time, precision requirements, required tools, and materials. Taking model A laptop as an example, it can be decomposed into 8 process nodes, where the motherboard assembly process requires a processing precision of ±0.05mm and an average processing time of 15 minutes per unit.

[0024] Traverse the set of process nodes to identify the dependencies between them. Analyze the production process specifications to determine the predecessor and successor processes for each node. For example, the predecessor process for motherboard assembly is component preparation, and the successor processes include screen installation and keyboard installation. Based on these relationships, establish directed connections between predecessor process nodes and successor process nodes. For instance, construct a directed graph where the component preparation node points to the motherboard assembly node, and the motherboard assembly node points to the screen installation node and the keyboard installation node.

[0025] The in-degree value of each process node is calculated, indicating how many predecessor process nodes each process node has. For example, the in-degree of the motherboard assembly node is 1, while the in-degree of the final quality testing node might be 3, meaning that 3 predecessor processes need to be completed before quality testing can proceed. A complete process dependency graph is generated using directed connections and in-degree values.

[0026] Extract the capacity requirement parameters for each process node from the process dependency relationship. These parameters include the number of products to be processed per unit time, processing accuracy requirements, and special tool requirements. For example, for the motherboard assembly process of model A laptops, assuming a daily output of 100 units, the capacity requirement for this process is 100 units / day, with a processing accuracy requirement of ±0.05mm.

[0027] The processing capacity parameters of each device are extracted from the equipment status information. These parameters include the device's maximum processing speed, achievable machining accuracy, and supported tool types. For example, assembly equipment No. 1 has a processing capacity of 120 units / day and a machining accuracy of ±0.03mm.

[0028] Calculate the matching degree between the capacity requirements of each process node and the processing capacity of each piece of equipment. First, calculate the absolute value of the difference between the capacity requirement parameter and the processing capacity parameter of each piece of equipment, and then normalize the result to obtain an initial matching degree value. For example, for the motherboard assembly process, its capacity requirement is 100 units / day, which differs from the processing capacity of equipment No. 1 (120 units / day) by 20 units / day. After normalization, an initial matching degree of 0.85 can be obtained.

[0029] The initial matching degree of each process node in the process dependency relationship is corrected based on its in-degree value. Process nodes with higher in-degrees have more pre-existing constraints and require more precise matching to avoid production bottlenecks. The correction can be performed using a weighted coefficient method: Matching degree correction value = Initial matching degree × (1 + In-degree value × Adjustment factor). For example, if the in-degree of the motherboard assembly process is 1, and an adjustment factor of 0.05 is used, the corrected matching degree is 0.85 × (1 + 1 × 0.05) = 0.8925.

[0030] The final output is a matching degree matrix between each process node and the equipment. This matrix reflects the degree of compatibility between different process nodes and each piece of equipment, providing a basis for subsequent production scheduling. For example, for the motherboard assembly process of model A laptops, the matching degree with equipment 1 is 0.8925, and the matching degree with equipment 2 is 0.7650, indicating that equipment 1 is more suitable to perform this process task.

[0031] Using the above methods, the matching degree between processes and equipment can be scientifically calculated based on the capacity demand characteristics of process nodes and the processing capacity characteristics of equipment, combined with the dependencies between processes. This provides data support for task allocation and scheduling optimization of notebook computer production lines, thereby improving production efficiency and resource utilization.

[0032] In one optional implementation, process delay records are extracted from historical production data, the duration of the impact between the delayed process and subsequent processes is calculated, an inter-process propagation correlation matrix is ​​constructed, and the process node delay propagation risk value is obtained through recursive calculation based on the matching degree and the propagation correlation matrix, including: Extract process delay records from historical production data and identify delayed processes and subsequent processes; The time difference between the delayed process and the subsequent process is calculated as the duration of the influence between the delayed process and the subsequent process. The number of times the same delayed process points to the same subsequent process is counted and normalized to generate the propagation weight coefficient. Establish a propagation association record by setting the duration of the impact between the delayed process and the subsequent process and the propagation weight coefficient. Then, perform attenuation processing on the propagation weight coefficient according to the duration of the impact to generate attenuation weight coefficient. Finally, associate the attenuation weight coefficient with the propagation association record to construct the inter-process propagation association matrix. Extract the matching degree values ​​of each process node from the matching degree as the initial risk value; Extract the propagation association records starting from the process node from the inter-process propagation association matrix, and extract the initial risk value and attenuation weight coefficient of subsequent process nodes in the propagation association records to generate risk increment; The in-degree value of the process node in the process dependency relationship is extracted as a dependency amplification factor and coupled with the risk increment to obtain the modulated risk increment. The modulated risk increment is superimposed on the initial risk value of the process node, and the process is recursively executed according to the propagation path of the propagation association record until the propagation path terminates to obtain the delayed propagation risk value of the process node.

[0033] In one implementation scheme for calculating the risk value of process node delay propagation, process delay records are first extracted from the historical production database. These records contain information such as process ID, planned start time, actual start time, planned completion time, and actual completion time. Delayed processes are identified by comparing the difference between the actual completion time and the planned completion time. For example, if the actual completion time of process A is 2 hours later than the planned completion time, process A is marked as a delayed process.

[0034] For each delayed process, identify its directly affected subsequent processes based on process dependencies. Process dependencies can be obtained through a process route diagram and represented as a directed graph structure, where nodes represent processes and edges represent dependencies between processes. For example, if process B can only begin after process A is completed, then there is a dependency between process A and process B, and process A is a prerequisite process for process B.

[0035] Calculate the time difference between the delayed process and its subsequent process as the duration of the delay. Specifically, for delayed process A and its subsequent process B, calculate the difference between the actual start time and the planned start time of process B. If the difference is positive, it indicates that process B is affected by the delay of process A, and this difference is the duration of the delay.

[0036] The propagation weight coefficient is calculated by statistically analyzing the number of times the same delayed process affects the same subsequent process in historical data. The calculation of the propagation weight coefficient uses normalization, which divides the number of times a delayed process affects a specific subsequent process by the total number of times that delayed process affects all subsequent processes. For example, if process A's delay affects process B 10 times and process C 5 times, then the propagation weight coefficient of process A to process B is 10 / (10+5) = 0.67, and the propagation weight coefficient to process C is 5 / (10+5) = 0.33.

[0037] Establish a propagation correlation record, recording delayed processes, subsequent processes, impact duration, and propagation weight coefficient. To account for time decay, an attenuation function is introduced to adjust the propagation weight coefficient, resulting in an attenuation weight coefficient. The attenuation function can adopt an exponential decay form: Attenuation weight coefficient = Propagation weight coefficient × exp(-λ × Impact duration), where λ is the attenuation coefficient, which can be adjusted according to actual production conditions, typically ranging from 0.01 to 0.1.

[0038] Organize all propagation correlation records into an inter-process propagation correlation matrix M, where each element M[i][j] represents the attenuation weight coefficient of process i on process j. If process i does not directly affect process j, then M[i][j] = 0.

[0039] The matching degree values ​​for each process node are extracted from the matching degree data and used as the initial risk value. The matching degree indicates the degree to which the actual execution of the process conforms to the plan; the lower the matching degree, the higher the initial risk value. The initial risk value can be expressed as: Initial Risk Value = 1 - Matching Degree.

[0040] The delayed propagation risk value of each process node is calculated recursively. For process node i, all propagation association records starting from i are extracted from the propagation association matrix. For each record, the initial risk value of the subsequent process node j and the attenuation weight coefficient from i to j are extracted, and the risk increment is calculated as: initial risk value of subsequent process j × attenuation weight coefficient.

[0041] To consider the impact of process dependencies on risk propagation, a dependency amplification factor is introduced. The dependency amplification factor is the in-degree value of a process node (i.e., the number of other processes that depend on it). The higher the in-degree, the wider the impact of process delays, and the greater the risk. Modulated risk increment = risk increment × (1 + α × dependency amplification factor), where α is an adjustment parameter that can be determined based on the actual production environment, typically ranging from 0.1 to 0.5.

[0042] The modulated risk increment is added to the initial risk value of process node i to obtain the updated risk value. Then, following the path of the propagation association record, the above process is recursively executed until all propagation paths terminate, i.e., there are no subsequent processes or the preset maximum recursion depth is reached. Finally, the delayed propagation risk value of each process node is obtained.

[0043] In practical applications, taking a manufacturing company's production line as an example, nearly six months of production data were extracted, containing 500 records of process delays across 30 process nodes. Using the method described above, the delay propagation risk value for each process node was calculated. The results show that the assembly process has the highest risk value, reaching 0.85, mainly because this process is the convergence point of multiple preceding processes, resulting in a large amplification factor. While the surface treatment process has a relatively high number of delays, its risk value is only 0.42 because it has fewer subsequent processes, limiting the propagation impact.

[0044] Based on the calculation results, production managers can develop preventative measures for high-risk process nodes, such as increasing resource allocation for critical processes, optimizing buffer times between processes, and adjusting production plans, effectively reducing the impact of process delays on overall production. Simultaneously, by continuously monitoring and updating the propagation correlation matrix, changes in the influence relationships between processes can be dynamically reflected, improving the accuracy of risk assessment.

[0045] In one optional implementation, the in-degree value of the process node in the process dependency relationship is extracted as a dependency amplification factor and coupled with the risk increment to obtain the modulated risk increment, which includes: Extract the in-degree values ​​of process nodes from the process dependency relationship, construct a process dependency matrix by combining the in-degree values ​​of process nodes with the process node capacity parameters, and extract the correlation values ​​between process nodes in the process dependency matrix to generate a process node dependency amplification factor. Based on the process node dependency amplification factor, the influence duration and propagation weight coefficient between process nodes are extracted from the inter-process propagation correlation matrix. The process node capacity fluctuation sequence is extracted according to the influence duration to construct the attenuation feature matrix. The attenuation feature matrix is ​​mapped with the process node dependency amplification factor to generate the initial risk increment. Based on the initial risk increment, the propagation path of the process node is extracted from the inter-process propagation correlation matrix. The node propagation influence intensity is calculated according to the process node propagation path. The initial risk increment is divided into propagation component and attenuation component according to the node propagation influence intensity. The change of the in-degree value of the process node is mapped into the propagation component modulation vector and the attenuation component modulation vector. The propagation component is mapped to the propagation component using the modulation vector of the propagation component to generate the modulated propagation component. The attenuation component is mapped to the attenuation component using the modulation vector of the attenuation component to generate the modulated attenuation component. The modulated propagation component and the modulated attenuation component are combined to obtain the modulated risk increment.

[0046] This implementation provides a method for modulating risk increments through process dependencies, enabling more accurate assessment of risk propagation characteristics between processes. First, the in-degree values ​​of process nodes are extracted from the process dependencies. For a production system containing n process nodes, a process dependency graph G=(V, E) is constructed, where V represents the set of process nodes and E represents the set of inter-process dependencies. For each process node v... i Calculate its in-degree value d i The in-degree value represents the number of other processes that directly depend on this process. For example, if process 1 is depended on by processes 2 and 3, then the in-degree value of process 1 is 2.

[0047] The process dependency matrix M is constructed by combining the in-degree values ​​of process nodes with their capacity parameters. The elements m in matrix M... ij This represents the degree of dependence of process i on process j, calculated as follows: if process j depends on process i, then m ij The in-degree value of process i is equal to its capacity parameter c. i The product of m; otherwise m ij The value is 0. A process node dependency amplification factor F is generated by extracting the correlation degree values ​​from the process dependency matrix M. For process node v... i Its dependence on amplification factor F i Calculate for all m related to it ij The weighted sum of values.

[0048] Based on the aforementioned process node dependency amplification factor, the influence duration t between process nodes is extracted from the inter-process propagation correlation matrix R. ij With propagation weight coefficient w ij Duration of impact (t) ij The propagation weighting coefficient w represents the time required for the capacity fluctuation of process i to affect process j. ij This indicates the intensity of the impact. Based on the duration of the impact, the production capacity fluctuation sequence {Δc} at each process node is extracted. i Construct the attenuation feature matrix D. The elements d in matrix D... ij This represents the decay characteristics of the influence of process i on process j over time, calculated as follows: d ij =Δc i (t ij )×e (-λ×tij) , where λ is the attenuation coefficient, reflecting the degree to which the influence weakens over time.

[0049] The initial risk increment I is generated by mapping the attenuation feature matrix D to the process node-dependent amplification factor F. For process node v i Its initial risk increment I i Calculated as: I i =F i×Σ(d ij ), which represents the initial value of the risk increment of this process node after considering dependencies.

[0050] Based on the initial risk increment, the propagation path P of each process node is extracted from the inter-process propagation correlation matrix R. For each pair of process nodes (v i v j If there exists from v i to v j If the path is specified, then record the path pij and its length l. ij The intensity S of node propagation influence is calculated based on the propagation path of the process node. For process node v i Its propagation influence intensity S i Calculated as: S i =Σ(w ij ×e (-β×lij) ), where β is the path attenuation coefficient, representing the degree to which the influence weakens as the path length increases.

[0051] The initial risk increment is split into a propagation component Ip and a decay component Id based on the intensity of the node propagation effect. For process node v i Its propagation component Ip i =I i ×S i Attenuation component Id i =I i ×(1-S i Then, the changes in the in-degree values ​​of the process nodes are mapped to the propagation component modulation vector Vp and the attenuation component modulation vector Vd.

[0052] Specifically, let the in-degree values ​​of process node vi at times t and t-1 be d respectively. i (t) and d i (t-1), then the rate of change of in-degree is: Δd i =(d i (t)-d i (t-1)) / d i (t-1). Elements vp of the propagation component modulation vector. i Calculated as: vp i =1+α×Δd i Where α is the propagation modulation coefficient; the elements of the attenuation component modulation vector are vd. i Calculated as: vd i =1-γ×Δd i , where γ is the attenuation modulation coefficient.

[0053] The propagation component Ip is mapped using the modulation vector Vp to generate the modulated propagation component Ip', calculated as follows: Ip' i =Ipi ×vp i Similarly, the attenuation component Id is mapped using the attenuation component modulation vector Vd to generate the modulated attenuation component Id', calculated as follows: Id' i =Id i ×vd i .

[0054] Finally, the modulated propagation component Ip' and the modulated attenuation component Id' are combined to obtain the modulated risk increment I', which is calculated as follows: I' i =Ip' i +Id' i This modulated risk increment more accurately reflects the actual risk increment of process nodes after considering process dependencies and changes in in-degree.

[0055] In practical applications, if the in-degree value of a certain process increases from 3 to 5, it indicates that more subsequent processes are starting to rely on this process. In this case, the risk increment of this process will be reasonably amplified through the above modulation mechanism, reminding managers to pay more attention to the risk changes of this process, adjust production strategies in a timely manner, and avoid the amplification and spread of risk that will have a serious impact on the entire production line.

[0056] In one optional implementation, process dependencies are analyzed based on delay propagation risk values, the propagation impact coefficient of process nodes is calculated, and the process node with the largest propagation impact coefficient is selected as the blocking process node. The buffer duration at the blocking process node is calculated based on the process delay records, including: Calculate the risk increase of the delayed propagation risk value of subsequent process nodes relative to the delayed propagation risk value of process nodes, identify subsequent process nodes whose risk increase exceeds a preset increase threshold as risk amplification nodes, and count the number of risk amplification nodes corresponding to each process node as the process node propagation impact coefficient. The process node with the largest value among the process node propagation influence coefficients is selected as the blocking process node; Extract the historical delay duration records of blocking process nodes and the corresponding delay occurrence times from the process delay records. Extract the impact duration of blocking process nodes pointing to risk amplification nodes from the inter-process propagation correlation matrix. Construct a delay propagation time window based on the delay occurrence time and impact duration. By filtering historical delay records of actual delays at risk amplification nodes within the delayed propagation time window, a propagation-related delay set is constructed. Statistical processing of the delay durations in the propagation-related delay set yields the buffer duration at the blocking process node.

[0057] Analyzing process dependencies based on delay propagation risk values, identifying critical blocking process nodes, and determining reasonable buffer durations are effective means to solve the delay propagation problem in production scheduling. This implementation first collects inter-process dependency data and establishes a process network model, which includes each process node and its sequential relationships. Simultaneously, it constructs an inter-process propagation correlation matrix, which contains the impact duration between each pair of related process nodes; that is, how long a delay in one process node will affect downstream process nodes.

[0058] For each process node in the process network, calculate its delay propagation risk value. The delay propagation risk value can be obtained by comprehensively evaluating factors such as historical delay frequency, delay severity, and impact range. For example, the delay propagation risk value of a certain process node can be defined as the product of the node's historical average delay duration and historical delay occurrence frequency.

[0059] For each process node, analyze the delayed propagation risk value of all its directly subsequent nodes. Calculate the risk amplification for each subsequent node, defined as the ratio of the risk value of the subsequent node to the risk value of the current node. Set a risk amplification threshold, for example, 1.5, meaning that when the risk value of a subsequent node exceeds 1.5 times the risk value of the current node, that subsequent node is identified as a risk amplification node.

[0060] Assume that the delay propagation risk value of process node A is 10, and the delay propagation risk values ​​of its subsequent nodes B and C are 18 and 12, respectively. If the preset risk amplification threshold is 1.5, then the risk amplification of B is 18 / 10 = 1.8, which exceeds the threshold and is identified as a risk amplification node; while the risk amplification of C is 12 / 10 = 1.2, which does not exceed the threshold and is not a risk amplification node.

[0061] By performing similar analysis on each process node, the number of risk amplification nodes corresponding to each node is counted, and this number is defined as the propagation impact coefficient of that node. For example, if process node D connects to 4 risk amplification nodes and process node E connects to 2 risk amplification nodes, then the propagation impact coefficients of D and E are 4 and 2, respectively.

[0062] The process node with the largest propagation impact coefficient is selected as the blocking process node. This node is a critical point in the delay propagation network, and setting a reasonable buffer time here can effectively block delay propagation. If multiple nodes have the same and largest propagation impact coefficient, the risk of each node can be further considered to amplify the total risk value of the node, or a node located at the front end of the process network can be selected.

[0063] After identifying the blocking process node, extract the historical delay duration record and the corresponding delay occurrence time of the node from the process delay record database. Simultaneously, extract the impact duration of the blocking process node on each risk amplification node from the inter-process propagation correlation matrix.

[0064] For example, if a historical delay of the blocking process node P occurs at time t, and its impact on the downstream risk amplification node Q lasts for a duration of d, then the constructed delay propagation time window is [t, t+d], indicating that within this time window, node Q may be affected by the delay of node P.

[0065] For each risk amplification node, the actual delay records occurring within the corresponding delayed propagation time window are selected to form a propagation-related delay set. For example, if risk amplification node Q has 3 delay records within the time window [t, t+d], with delay durations of 5 minutes, 8 minutes, and 12 minutes respectively, then these three records constitute part of the propagation-related delay set.

[0066] Statistical processing is performed on the delay durations in all propagation-related delay sets to calculate their distribution characteristics. Various statistical methods can be used, such as the mean, median, or specific percentiles. For example, the 95th percentile of the delay duration can be taken as the buffer duration at the blocking process node to cover most delay propagation scenarios.

[0067] The buffer duration calculated using the above method, when applied to blocking process nodes in production scheduling, can effectively reduce delay propagation and improve overall production efficiency. This method is particularly suitable for delay propagation control in complex process networks, enabling precise setting of buffer times for the most critical propagation paths, thus avoiding the resource waste caused by the widespread use of buffers in traditional methods.

[0068] The implementation of this method can not only reduce the propagation of delays, but also provide in-depth insights into the risk propagation paths in the process network, which helps to continuously improve production processes, optimize resource allocation, and improve the stability and reliability of production systems.

[0069] In one optional implementation, process nodes are assigned to equipment according to matching degree, a buffer duration is inserted at the blocking process nodes, and a process connection sequence diagram is constructed based on the material transfer time between processes to generate a scheduling execution plan, including: The difference between the matching degree value of the process node and the matching degree value of the equipment is calculated based on the matching degree. The equipment with the smallest difference is selected as the assigned equipment of the process node, and the initial assignment relationship between the process node and the equipment is generated. Extract the upstream process node that blocks the process node from the process dependency relationship, calculate the risk gap between the delay propagation risk value of the upstream process node and the blocking process node, amplify and adjust the buffer duration based on the risk gap value to obtain the adjusted buffer duration, and insert the adjusted buffer duration after the execution time point of the blocking process node in the initial allocation relationship. Extract the predecessor and successor process nodes of each process node from the process dependency relationship, obtain the material transfer time, calculate the risk transmission gradient between the delay propagation risk values ​​of the predecessor process node and the successor process node, compress and adjust the material transfer time based on the risk transmission gradient, and construct the process connection sequence diagram based on the completion time of the predecessor process node and the compressed material transfer time. Extract the start and finish times of each process node from the process connection sequence diagram, extract the allocation equipment corresponding to each process node from the initial allocation relationship, and combine them to generate a scheduling execution plan.

[0070] First, the difference between the matching degree values ​​of the process node and the equipment is calculated based on the matching degree. The equipment with the smallest difference is selected as the assigned equipment for the process node. Specifically, for each process node i, the matching degree difference D between it and all available equipment j is calculated. ij The formula for calculating the difference is D. ij =|MNi-MEj|, where MNi represents the matching degree value of process node i, and MEj represents the matching degree value of equipment j. The matching degree value can be determined based on the process requirements for equipment and the equipment's capability parameters. For example, for process nodes with high machining accuracy requirements, the matching degree value may be between 0.8 and 1.0, while for general processes, the matching degree value may be between 0.5 and 0.7. The equipment matching degree value is determined based on the equipment's performance parameters. For example, the matching degree value of high-precision equipment is 0.9, and the matching degree value of general-purpose equipment is 0.6. For each process node i, the equipment j with the smallest Dij is selected as its assigned equipment, forming the initial assignment relationship {(i, j)}.

[0071] Five process nodes and three pieces of equipment are set up. The matching degrees of the process nodes are N1=0.9, N2=0.7, N3=0.8, N4=0.6, and N5=0.5, respectively; the matching degrees of the equipment are E1=0.85, E2=0.65, and E3=0.5, respectively. After calculating the difference, N1 is assigned to E1 (difference 0.05), N2 is assigned to E2 (difference 0.05), N3 is assigned to E1 (difference 0.05), N4 is assigned to E2 (difference 0.05), and N5 is assigned to E3 (difference 0.0).

[0072] The process extracts upstream process nodes from the process dependency graph that block the process node, calculates the risk gap between the delay propagation risk value of the upstream process node and the blocking process node, and adjusts the buffer duration based on the risk gap value. A blocking process node is a critical node in the process dependency graph that has a significant impact on subsequent processes. First, all blocking process nodes are identified as a set B. For each blocking node b∈B, its upstream process node set Ub is extracted. For each pair (u, b), u∈Ub, its delay propagation risk values ​​RCu and RPb are calculated, and the risk gap value RDub=RPb-RPu. The delay propagation risk value can be determined based on factors such as the position of the process node in the process dependency graph, the number of successor nodes, and the process complexity. The basic buffer duration BCb is set as a certain proportion of the expected execution time of the blocking node, for example, 10%. The basic buffer duration is adjusted according to the risk gap value: ABCb=BCb*(1+α*RDub), where α is the adjustment coefficient, usually ranging from 0.1 to 0.3. The adjusted buffer duration ABCb is then inserted after the execution time of the blocking process node.

[0073] Assuming process node N3 is a blocking node, and its upstream node is N2, with a delay propagation risk value of 0.3 for N2 and 0.7 for N3, the risk drop value is 0.4. If the estimated execution time of N3 is 60 minutes, the basic buffer time is 6 minutes, and the adjustment coefficient α = 0.2, then the adjusted buffer time is 6 * (1 + 0.2 * 0.4) = 6.48 minutes, rounded to 6.5 minutes.

[0074] Extract the predecessor and successor process nodes from the process dependencies, obtain the material transfer time, calculate the risk propagation gradient between the delay propagation risk values ​​of the predecessor and successor process nodes, and compress and adjust the material transfer time based on the risk propagation gradient. For each pair of process nodes (p, s), where p is the predecessor node and s is the successor node, obtain the original material transfer time MTps. Calculate the risk propagation gradient RTGps = (RPs - RPp) / RPp, representing the relative rate of change of the risk value. Adjust the material transfer time according to the risk propagation gradient: AMTps = MTps * (1 - β * RTGps), where β is the compression coefficient, typically ranging from 0.05 to 0.15. Ensure that AMTps is not less than the minimum transfer time threshold. Based on the completion time Cp of the predecessor process node and the compressed material transfer time AMTps, calculate the earliest possible start time Ss = Cp + AMTps for the successor process node, and construct the process connection sequence diagram.

[0075] Assuming process N1 completes in 100 minutes, N2 is its subsequent node, the initial material transport time is 10 minutes, the delay propagation risk value for N1 is 0.2, and the delay propagation risk value for N2 is 0.5, then the risk propagation gradient is (0.5-0.2) / 0.2=1.5. If the compression coefficient β=0.1, then the compressed material transport time is 10*(1-0.1*1.5)=8.5 minutes, and the earliest possible start time for N2 is 100+8.5=108.5 minutes.

[0076] Finally, the start and finish times of each process node are extracted from the process sequence diagram, and the corresponding allocation devices for each process node are extracted from the initial allocation relationship. These are then combined to generate a scheduling execution plan. For each process node, its start and finish times are obtained from the process sequence diagram, and its allocation devices are obtained from the initial allocation relationship. These are combined to form a scheduling execution item, and all scheduling execution items constitute the final scheduling execution plan.

[0077] Continuing with the previous example, the complete scheduling execution plan is as follows: Operation N1 is on equipment E1, with a start time of 0 minutes and a completion time of 100 minutes; Operation N2 is on equipment E2, with a start time of 108.5 minutes and a completion time of 158.5 minutes; Operation N3 is on equipment E1, with a start time of 167 minutes (considering the material transfer time of 8.5 minutes from N2 to N3) and a completion time of 227 minutes; a 6.5-minute buffer time is inserted after N3 is completed; Operation N4 is on equipment E2, with a start time of 233.5 minutes and a completion time of 283.5 minutes; Operation N5 is on equipment E3, with a start time of 292 minutes (considering the material transfer time of 8.5 minutes from N4 to N5) and a completion time of 332 minutes.

[0078] Through the above steps, an allocation based on the matching degree between process nodes and equipment is achieved, and a buffer time is inserted at critical blocking process nodes. At the same time, the dynamic adjustment of material transfer time between processes is considered, and finally a scheduling execution scheme that balances efficiency and risk is generated.

[0079] In one optional implementation, production line execution data is collected. When a process delay is detected, the impact range is calculated based on the propagation correlation matrix. The delay time is allocated to the buffer duration of the blocked process node. Material transfer time is optimized based on the process connection sequence diagram. The updated laptop production scheduling scheme includes: Collect the time deviation between the actual completion time and the planned completion time of the process node in the production line execution data, and calculate the weighted combination value of the time deviation value and the capacity requirement of the process node; The weighted combination value is corrected based on the in-degree value in the process dependency relationship, and the process node whose weighted combination value exceeds the preset range after correction is identified as the delayed process node. The propagation weight coefficient and attenuation weight coefficient corresponding to the delayed process node are extracted from the propagation correlation matrix. The propagation intensity value is obtained by recursive calculation in combination with the propagation influence coefficient of the process node. The set of affected process nodes is determined from the propagation correlation matrix based on the propagation intensity value. Extract the impact duration records of the affected process nodes set, calculate the delay time allocation weight based on the propagation intensity value and the impact duration records, and allocate the delay time to the buffer duration of the blocked process nodes according to the allocation weight; Extract material transfer time from the process connection sequence diagram, calculate the material transfer time compression ratio based on the allocated buffer time of the blocking process node, and optimize the material transfer time in the process connection sequence diagram according to the compression ratio; Substitute the optimized material transfer time and the buffer time of the blocked process nodes after allocation into the process connection sequence diagram, recalculate the start and finish times of the process nodes, and update the laptop production scheduling scheme.

[0080] In optimizing the production scheduling scheme for laptops, the first step is to collect production line execution data and analyze the time deviation values ​​of each process node. By setting preset weighting coefficients, the time deviation values ​​are weighted and combined with the capacity requirements of each process node. For example, for the motherboard assembly process, if the actual completion time is 10:30 while the planned completion time is 10:20, resulting in a 10-minute time deviation, and the capacity requirement for this process is 30 units per hour, then a weighting coefficient of 0.6 for the time deviation value and 0.4 for the process capacity requirement can be set. The weighted combination value is calculated as 0.6 × 10 + 0.4 × (60 ÷ 30) = 6 + 0.8 = 6.8.

[0081] Based on process dependencies, the weighted combination value is adjusted using the in-degree value. The in-degree represents the number of preceding processes a process depends on; a higher in-degree means a greater impact from process delays. The adjustment formula is calculated by multiplying the weighted combination value by (1 + in-degree × 0.1). For example, if the motherboard installation process has an in-degree of 3, the adjusted weighted combination value is 6.8 × (1 + 3 × 0.1) = 8.84. When the adjusted weighted combination value exceeds a preset threshold of 8, the process is identified as a delayed process node.

[0082] From the propagation correlation matrix, the propagation weight coefficients and attenuation weight coefficients corresponding to the delayed process nodes are extracted. The propagation correlation matrix is ​​an n×n matrix, where n is the number of process nodes, and the matrix element aij represents the degree of influence of process i on process j. Combining the propagation influence coefficient of each process node, a recursive calculation is performed to obtain the propagation strength value. During the recursive calculation, starting from the delayed process, the influence is propagated to its dependent processes. With each layer of propagation, the influence strength decreases according to the attenuation weight coefficient. For example, the initial propagation strength of the delayed motherboard installation process on the memory installation process is 0.8, and the propagation attenuation coefficient is 0.7. Therefore, the propagation strength of the second layer is 0.8 × 0.7 = 0.56.

[0083] When the propagation intensity value exceeds a preset threshold of 0.3, the corresponding process is included in the set of affected process nodes. For the identified set of affected process nodes, their impact duration records are extracted; these records are derived from historical production data statistics. Delay time allocation weights are calculated based on the propagation intensity value and the impact duration records. For example, for a memory installation process with a propagation intensity of 0.6 and a hard disk installation process with a propagation intensity of 0.4, if the original delay time is 20 minutes, then the memory installation process is allocated a delay time of 20 × (0.6 ÷ 1.0) = 12 minutes, and the hard disk installation process is allocated a delay time of 8 minutes.

[0084] The calculated delay time is allocated to the buffer time of the blocking process node. The blocking process node is a critical node in the process dependency chain that can block the propagation of delay, and it usually has a certain buffer time. For example, the original buffer time of the memory installation process is 15 minutes. After allocating a 12-minute delay, the remaining buffer time is 3 minutes.

[0085] Material transfer time is extracted from the process sequence diagram, which shows the time relationship of material transfer between each process. Based on the allocated buffer time for blocking process nodes, the material transfer time compression ratio is calculated. The compression ratio calculation takes into account the buffer time consumption; for example, if the buffer time consumption is 80%, the corresponding material transfer time can be compressed by 20%. The material transfer time in the process sequence diagram is optimized, such as compressing the original 5-minute transfer time to 4 minutes.

[0086] Finally, the optimized material transfer time and the allocated buffer time for blocking process nodes are substituted into the process connection sequence diagram to recalculate the start and finish times of each process node. For example, the motherboard installation process was originally scheduled to finish at 10:20, but was delayed by 10 minutes and adjusted to 10:30. However, through material transfer time optimization, the start time of the subsequent memory installation process was adjusted from the original 10:25 to 10:34, reducing the propagation of delays.

[0087] This optimization method is suitable for rapid response and adjustment to sudden delays on laptop production lines. By identifying delayed processes, analyzing the scope of impact, rationally allocating delay time, and optimizing material handling, it effectively reduces the impact of delays on the overall production plan. In practical applications, the weights and thresholds of various parameters can be adjusted according to the actual conditions of the production line to achieve the best scheduling effect. This invention is not only applicable to laptop manufacturing but can also be extended to other electronic product manufacturing fields with similar process dependencies.

[0088] A second aspect of this invention provides a notebook computer production scheduling optimization system for smart factories, the system comprising: The first unit is used to receive the set of laptop production tasks and equipment status information, decompose the set of production tasks into process nodes, construct process dependencies based on the processing sequence between processes, and calculate the matching degree between the capacity requirements of process nodes and the processing capacity of equipment. The second unit is used to extract process delay records from historical production data, calculate the impact duration between delayed processes and subsequent processes, construct the inter-process propagation correlation matrix, and obtain the process node delay propagation risk value through recursive calculation based on the matching degree and the propagation correlation matrix. The third unit is used to analyze process dependencies based on delay propagation risk values, calculate the propagation impact coefficient of process nodes, select the process node with the largest propagation impact coefficient as the blocking process node, and calculate the buffer time at the blocking process node based on the process delay record. The fourth unit is used to allocate process nodes to equipment according to matching degree, insert buffer time at blocking process nodes, construct process connection sequence diagram based on material transfer time between processes, and generate scheduling execution plan; The fifth unit is used to collect production line execution data. When a process delay is detected, the impact range is calculated based on the propagation correlation matrix, the delay time is allocated to the buffer time of the blocked process node, the material transfer time is optimized based on the process connection sequence diagram, and the laptop production scheduling scheme is updated.

[0089] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0090] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0091] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimizing notebook computer production scheduling in smart factories, characterized in that, include: Receive a set of laptop production tasks and equipment status information, decompose the set of production tasks into process nodes, construct process dependencies based on the processing sequence between processes, and calculate the matching degree between the capacity requirements of process nodes and the processing capacity of equipment. Extract process delay records from historical production data, calculate the impact duration between delayed processes and subsequent processes, construct an inter-process propagation correlation matrix, and obtain the process node delay propagation risk value through recursive calculation based on the matching degree and the propagation correlation matrix. Based on the analysis of process dependencies using delayed propagation risk values, the propagation impact coefficient of process nodes is calculated, and the process node with the largest propagation impact coefficient is selected as the blocking process node. The buffer time at the blocking process node is calculated based on the process delay record. The process nodes are assigned to equipment according to the matching degree. A buffer duration is inserted at the blocking process nodes. A process connection sequence diagram is constructed based on the material transfer time between processes, and a scheduling execution plan is generated. Collect production line execution data, and when process delays are detected, calculate the scope of impact based on the propagation correlation matrix, allocate the delay time to the buffer time of the blocked process node, optimize material transfer time based on the process connection sequence diagram, and update the laptop production scheduling plan.

2. The method according to claim 1, characterized in that, The system receives a set of laptop production tasks and equipment status information, decomposes the production task set into process nodes, constructs process dependencies based on the processing sequence between processes, and calculates the matching degree between the capacity requirements of the process nodes and the processing capacity of the equipment, including: Receive a set of laptop production tasks and equipment status information, decompose the set of production tasks into a set of process nodes, and extract the processing requirements of each process node; Traverse the set of process nodes, identify the predecessor and successor process nodes of each process node, establish directed connections between predecessor process nodes and successor process nodes based on the processing order between processes, count the in-degree value of each process node, and generate process dependency relationships containing directed connections and in-degree values. Extract the capacity demand parameters of each process node from the process dependency relationship, and extract the processing capacity parameters of each equipment from the equipment status information; The absolute value of the difference between the capacity requirement parameter of each process node and the processing capacity parameter of each equipment is calculated and normalized to the initial matching degree value. The initial matching degree value of the corresponding process node is corrected according to the in-degree value of each process node in the process dependency relationship, so as to obtain the matching degree between the capacity requirement of the process node and the processing capacity of the equipment.

3. The method according to claim 1, characterized in that, Extract process delay records from historical production data, calculate the impact duration between delayed processes and subsequent processes, construct an inter-process propagation correlation matrix, and obtain the process node delay propagation risk value through recursive calculation based on the matching degree and the propagation correlation matrix, including: Extract process delay records from historical production data and identify delayed processes and subsequent processes; The time difference between the delayed process and the subsequent process is calculated as the duration of the influence between the delayed process and the subsequent process. The number of times the same delayed process points to the same subsequent process is counted and normalized to generate the propagation weight coefficient. Establish a propagation association record by setting the duration of the impact between the delayed process and the subsequent process and the propagation weight coefficient. Then, perform attenuation processing on the propagation weight coefficient according to the duration of the impact to generate attenuation weight coefficient. Finally, associate the attenuation weight coefficient with the propagation association record to construct the inter-process propagation association matrix. Extract the matching degree values ​​of each process node from the matching degree as the initial risk value; Extract the propagation association records starting from the process node from the inter-process propagation association matrix, and extract the initial risk value and attenuation weight coefficient of subsequent process nodes in the propagation association records to generate risk increment; The in-degree value of the process node in the process dependency relationship is extracted as a dependency amplification factor and coupled with the risk increment to obtain the modulated risk increment. The modulated risk increment is superimposed on the initial risk value of the process node, and the process is recursively executed according to the propagation path of the propagation association record until the propagation path terminates to obtain the delayed propagation risk value of the process node.

4. The method according to claim 3, characterized in that, The in-degree values ​​of process nodes in the process dependency relationship are extracted as dependency amplification factors and coupled with the risk increment to obtain the modulated risk increment, which includes: Extract the in-degree values ​​of process nodes from the process dependency relationship, construct a process dependency matrix by combining the in-degree values ​​of process nodes with the process node capacity parameters, and extract the correlation values ​​between process nodes in the process dependency matrix to generate a process node dependency amplification factor. Based on the process node dependency amplification factor, the influence duration and propagation weight coefficient between process nodes are extracted from the inter-process propagation correlation matrix. The process node capacity fluctuation sequence is extracted according to the influence duration to construct the attenuation feature matrix. The attenuation feature matrix is ​​mapped with the process node dependency amplification factor to generate the initial risk increment. Based on the initial risk increment, the propagation path of the process node is extracted from the inter-process propagation correlation matrix. The node propagation influence intensity is calculated according to the process node propagation path. The initial risk increment is divided into propagation component and attenuation component according to the node propagation influence intensity. The change of the in-degree value of the process node is mapped into the propagation component modulation vector and the attenuation component modulation vector. The propagation component is mapped to the propagation component using the modulation vector of the propagation component to generate the modulated propagation component. The attenuation component is mapped to the attenuation component using the modulation vector of the attenuation component to generate the modulated attenuation component. The modulated propagation component and the modulated attenuation component are combined to obtain the modulated risk increment.

5. The method according to claim 1, characterized in that, Based on the analysis of process dependencies using delay propagation risk values, the propagation impact coefficient of each process node is calculated. The process node with the largest propagation impact coefficient is selected as the blocking process node. The buffer time at the blocking process node is calculated based on the process delay records, including: Calculate the risk increase of the delayed propagation risk value of subsequent process nodes relative to the delayed propagation risk value of process nodes, identify subsequent process nodes whose risk increase exceeds a preset increase threshold as risk amplification nodes, and count the number of risk amplification nodes corresponding to each process node as the process node propagation impact coefficient. The process node with the largest value among the process node propagation influence coefficients is selected as the blocking process node; Extract the historical delay duration records of blocking process nodes and the corresponding delay occurrence times from the process delay records. Extract the impact duration of blocking process nodes pointing to risk amplification nodes from the inter-process propagation correlation matrix. Construct a delay propagation time window based on the delay occurrence time and impact duration. By filtering historical delay records of actual delays at risk amplification nodes within the delayed propagation time window, a propagation-related delay set is constructed. Statistical processing of the delay durations in the propagation-related delay set yields the buffer duration at the blocking process node.

6. The method according to claim 1, characterized in that, The process nodes are assigned to equipment according to their matching degree. A buffer duration is inserted at the blocking process nodes. A process connection sequence diagram is constructed based on the material transfer time between processes, and a scheduling execution plan is generated, including: The difference between the matching degree value of the process node and the matching degree value of the equipment is calculated based on the matching degree. The equipment with the smallest difference is selected as the assigned equipment of the process node, and the initial assignment relationship between the process node and the equipment is generated. Extract the upstream process node that blocks the process node from the process dependency relationship, calculate the risk gap between the delay propagation risk value of the upstream process node and the blocking process node, amplify and adjust the buffer duration based on the risk gap value to obtain the adjusted buffer duration, and insert the adjusted buffer duration after the execution time point of the blocking process node in the initial allocation relationship. Extract the predecessor and successor process nodes of each process node from the process dependency relationship, obtain the material transfer time, calculate the risk transmission gradient between the delay propagation risk values ​​of the predecessor process node and the successor process node, compress and adjust the material transfer time based on the risk transmission gradient, and construct the process connection sequence diagram based on the completion time of the predecessor process node and the compressed material transfer time. Extract the start and finish times of each process node from the process connection sequence diagram, extract the allocation equipment corresponding to each process node from the initial allocation relationship, and combine them to generate a scheduling execution plan.

7. The method according to claim 1, characterized in that, Collect production line execution data. When a process delay is detected, calculate the impact range based on the propagation correlation matrix, allocate the delay time to the buffer duration of the blocked process node, optimize material transfer time based on the process connection sequence diagram, and update the laptop production scheduling scheme, including: Collect the time deviation between the actual completion time and the planned completion time of the process node in the production line execution data, and calculate the weighted combination value of the time deviation value and the capacity requirement of the process node; The weighted combination value is corrected based on the in-degree value in the process dependency relationship, and the process node whose weighted combination value exceeds the preset range after correction is identified as the delayed process node. The propagation weight coefficient and attenuation weight coefficient corresponding to the delayed process node are extracted from the propagation correlation matrix. The propagation intensity value is obtained by recursive calculation in combination with the propagation influence coefficient of the process node. The set of affected process nodes is determined from the propagation correlation matrix based on the propagation intensity value. Extract the impact duration records of the affected process nodes set, calculate the delay time allocation weight based on the propagation intensity value and the impact duration records, and allocate the delay time to the buffer duration of the blocked process nodes according to the allocation weight; Extract material transfer time from the process connection sequence diagram, calculate the material transfer time compression ratio based on the allocated buffer time of the blocking process node, and optimize the material transfer time in the process connection sequence diagram according to the compression ratio; Substitute the optimized material transfer time and the buffer time of the blocked process nodes after allocation into the process connection sequence diagram, recalculate the start and finish times of the process nodes, and update the laptop production scheduling scheme.

8. A notebook computer production scheduling optimization system for smart factories, used to implement the method of any one of claims 1-7, characterized in that, include: The first unit is used to receive the set of laptop production tasks and equipment status information, decompose the set of production tasks into process nodes, construct process dependencies based on the processing sequence between processes, and calculate the matching degree between the capacity requirements of process nodes and the processing capacity of equipment. The second unit is used to extract process delay records from historical production data, calculate the impact duration between delayed processes and subsequent processes, construct the inter-process propagation correlation matrix, and obtain the process node delay propagation risk value through recursive calculation based on the matching degree and the propagation correlation matrix. The third unit is used to analyze process dependencies based on delay propagation risk values, calculate the propagation impact coefficient of process nodes, select the process node with the largest propagation impact coefficient as the blocking process node, and calculate the buffer time at the blocking process node based on the process delay record. The fourth unit is used to allocate process nodes to equipment according to matching degree, insert buffer time at blocking process nodes, construct process connection sequence diagram based on material transfer time between processes, and generate scheduling execution plan; The fifth unit is used to collect production line execution data. When a process delay is detected, the impact range is calculated based on the propagation correlation matrix, the delay time is allocated to the buffer time of the blocked process node, the material transfer time is optimized based on the process connection sequence diagram, and the laptop production scheduling scheme is updated.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

Images