Production scheduling optimization method and system based on dynamic algorithm engine

Abstract

The invention relates to the technical field of computers, in particular to a production scheduling optimization method and system based on a dynamic algorithm engine, and the system comprises a data processing module which comprises a preprocessing unit used for preprocessing collected order basic data to output order features; the scheduling solving module comprises a function construction unit used for constructing a scheduling optimization objective function; the disturbance recalculation module comprises an evaluation unit used for determining an influence domain of scheduling execution abnormity according to real-time industrial production resource constraints; the schedule judgment module is used for determining whether the stability of the production schedule meets the requirement or not according to the fluctuation characterization value; the delay adjusting module is used for determining a transmission delay coefficient between scheduling processes according to the updating frequency of the recalculation scheduling scheme; and the penalty factor adjusting module is used for determining the dynamic load penalty factor of the high-load equipment in the scheduling optimization objective function solution according to the conflict rate of the production resources in the production scheduling plan. According to the invention, the stability of production scheduling is improved.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a production scheduling optimization method and system based on a dynamic algorithm engine. Background Technology

[0002] Against the backdrop of the continuous development of intelligent manufacturing and digital factories, production scheduling systems have become an important technical means to achieve rational allocation of production resources and ensure the smooth execution of production plans. Production scheduling typically needs to comprehensively consider various constraints such as equipment capacity, material supply status, and order priority. By rationally allocating and scheduling production tasks, it aims to maximize resource utilization efficiency while meeting delivery requirements. Existing production scheduling methods are mostly based on fixed rules or static optimization algorithms to build scheduling models, which are then executed according to a predetermined plan after the schedule is generated. In actual production environments, uncertainties such as equipment failures, fluctuations in material arrival times, and order changes frequently occur, requiring the recalculation of the overall scheduling plan during execution, affecting the continuity and stability of the production process. Therefore, there is an urgent need for a production scheduling optimization method and system based on a dynamic algorithm engine. By constructing an algorithm engine that can dynamically call and combine multiple optimization algorithms according to changes in production status, adaptive optimization of scheduling plans in complex production environments can be achieved, thereby improving the stability and execution efficiency of production scheduling.

[0003] Chinese Patent Publication No. CN117422212A discloses a process scheduling method, apparatus, electronic device, and storage medium. The method includes: acquiring at least one specification, production process, and equipment information pre-configured for a product to be produced, wherein the production process includes multiple processes for each specification of product, and the equipment information includes: at least one process performed by each production equipment and at least one specification of product parts produced by performing each process; obtaining corresponding process scheduling constraints based on the production process and equipment information, and sequentially setting process scheduling information for each product based on the process scheduling constraints to obtain an initial scheduling result, wherein each process scheduling information includes a set of production equipment and process execution time for the corresponding product; and performing the following operations for each switching time of each production equipment in the initial scheduling result to obtain an adjusted target scheduling result: adjusting the execution time of the corresponding process of the product parts produced after the switching time based on the preset switching time of the production equipment, wherein the preset switching time includes at least one of the following: switching time between producing product parts of different specifications, and switching time between performing different processes. It is evident that the process scheduling methods, devices, electronic equipment, and storage media have complex and dynamically changing time dependencies between processes, and process conflicts are prone to occur during the execution of the initial scheduling scheme, leading to chaotic and unpredictable production scheduling. Summary of the Invention

[0004] Therefore, the present invention provides a production scheduling optimization method and system based on a dynamic algorithm engine to overcome the problems in the prior art.

[0005] To achieve the above objectives, the present invention provides a production scheduling optimization system based on a dynamic algorithm engine, comprising: The data processing module includes a preprocessing unit for preprocessing the collected basic order data to output order features, and a vector construction unit connected to the preprocessing unit for quantifying the order features and constructing a standardized order feature vector. The scheduling solution module, which is connected to the data processing module, includes a function construction unit for constructing a scheduling optimization objective function based on production indicators and the standardized order feature vector, and a solution unit connected to the function construction unit for optimizing the objective function based on industrial production resource constraints and generating a production scheduling plan. The disturbance recalculation module, which is connected to the scheduling solution module, includes an evaluation unit for determining the impact domain of scheduling execution anomalies based on real-time industrial production resource constraints, and a recalculation unit connected to the evaluation unit for optimizing the scheduling processes within the impact domain to output a recalculated scheduling scheme. The scheduling determination module, which is connected to the disturbance recalculation module, is used to determine whether the stability of the production schedule meets the requirements based on the fluctuation characterization value determined by the number of scheduling processes optimized and adjusted per unit time and the total number of processes in the production scheduling plan. The delay adjustment module is connected to the disturbance recalculation module and the scheduling determination module respectively, and is used to determine the transmission delay coefficient between scheduling processes according to the update frequency of the recalculation scheduling scheme. The penalty factor adjustment module is connected to the scheduling solution module and the delay adjustment module respectively, and is used to determine the dynamic load penalty factor of high-load equipment in the scheduling optimization objective function solution based on the conflict rate of production resources in the production scheduling plan.

[0006] Furthermore, the scheduling determination module determines a fluctuation characterization value in response to the ratio of the number of scheduled processes optimized and adjusted per unit time to the total number of processes in the production scheduling plan, in order to determine whether the stability of the production schedule meets the requirements.

[0007] Furthermore, the scheduling determination module determines that the stability of the production schedule meets the requirements when the fluctuation characterization value is less than or equal to a preset fluctuation characterization value. The scheduling determination module determines that the stability of the production schedule does not meet the requirements when the fluctuation characterization value is greater than the preset fluctuation characterization value.

[0008] Furthermore, in response to the condition that the stability of the production schedule does not meet the requirements, the delay adjustment module determines whether the scope of the impact domain assessment meets the requirements based on the update frequency of the recalculated scheduling scheme.

[0009] Furthermore, the delay adjustment module responds to the fact that the update frequency of the recalculation scheduling scheme is greater than the preset first update frequency and less than or equal to the preset second update frequency, thus determining that the scope of the impact domain assessment meets the requirements. The delay adjustment module determines that the scope of the impact domain assessment does not meet the requirements if the update frequency of the recalculation scheduling scheme is less than or equal to the preset first update frequency or greater than the preset second update frequency.

[0010] Furthermore, the delay adjustment module responds to the fact that the update frequency of the recalculated scheduling scheme is greater than the preset second update frequency, thereby reducing the transmission delay coefficient between scheduling processes; The delay adjustment module responds to the fact that the update frequency of the recalculation scheduling scheme is less than or equal to the preset first update frequency, and initially determines that the rationality of the allocation of industrial production resources does not meet the requirements. It then determines whether the rationality of the allocation of industrial production resources meets the requirements based on the conflict rate of production resources in the production scheduling plan.

[0011] Furthermore, the reduction in the transmission delay coefficient between scheduling processes is determined by the difference between the recalculated update frequency of the scheduling scheme and the preset second update frequency.

[0012] Furthermore, in response to the condition that the update frequency of the recalculation scheduling scheme is less than or equal to the preset first update frequency, the penalty factor adjustment module determines whether the rationality of the allocation of industrial production resources meets the requirements based on the conflict rate of production resources in the production scheduling plan.

[0013] Furthermore, the penalty factor adjustment module responds to the production scheduling plan where the conflict rate of production resources is less than or equal to a preset conflict rate, thus determining that the rationality of the allocation of industrial production resources meets the requirements. The penalty factor adjustment module responds to the fact that the conflict rate of production resources in the production scheduling plan is greater than the preset conflict rate, determines that the rationality of industrial production resource allocation does not meet the requirements, and increases the dynamic load penalty factor of high-load equipment in the solution of the scheduling optimization objective function. In the solution of the scheduling optimization objective function, the increase in the dynamic load penalty factor of high-load equipment is determined by the difference between the conflict rate of production resources in the production scheduling plan and the preset conflict rate.

[0014] This invention also provides a production scheduling optimization method based on a dynamic algorithm engine, comprising: The collected basic order data is sequentially cleaned, denoised, standardized, and feature extracted to output order features. The order features are then quantified and a standardized order feature vector is constructed. Based on production indicators and the standardized order feature vector, a scheduling optimization objective function is constructed. Based on industrial production resource constraints, the objective function is optimized and solved to generate a production scheduling plan. Based on real-time industrial production resource constraints, the impact domain of scheduling execution anomalies is determined, and the scheduling processes within the impact domain are optimized to output a recalculated scheduling scheme. The number of scheduling processes optimized and adjusted per unit time and the total number of processes in the production scheduling plan are obtained to determine the fluctuation characterization value, and the stability of the production scheduling is determined based on the fluctuation characterization value. If the stability of the production schedule does not meet the requirements, the update frequency of the recalculated scheduling scheme is obtained to determine whether the scope of the impact domain assessment meets the requirements. If the scope of the impact domain assessment does not meet the requirements, then determine whether to reduce the transmission delay coefficient between scheduling processes based on the update frequency of the recalculated scheduling scheme. If it is not necessary to reduce the transmission delay coefficient between scheduling processes, then the conflict rate of production resources in the production scheduling plan is obtained to determine the dynamic load penalty factor of high-load equipment in the solution of the scheduling optimization objective function.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: The system of this invention, by setting up a data processing module, a scheduling solution module, a disturbance recalculation module, a scheduling judgment module, a delay adjustment module, and a penalty factor adjustment module, determines the fluctuation characterization value based on the number of scheduled processes optimized and adjusted per unit time and the total number of processes in the production schedule plan to determine whether the stability of the production schedule meets the requirements. Since changes in equipment status and order priority adjustments during production execution may cause some processes in the original scheduling plan to need frequent adjustments, the production schedule is highly susceptible to external disturbances, resulting in insufficient stability. By determining the stability of the production schedule, it is possible to promptly identify whether there is excessive adjustment or scheduling imbalance during the scheduling execution process, ensuring the continuity and execution efficiency of the production plan. The transmission delay coefficient between scheduled processes is adjusted according to the update frequency of the recalculated scheduling plan. Since the impact of upstream process adjustments on downstream processes needs a certain time delay to be identified, a small number of directly related processes are included in the influence domain during the scheduling evaluation process. This leads to some actually affected subsequent processes not being included in the evaluation scope, resulting in incomplete scheduling adjustments and frequent subsequent adjustments. By reducing the process transmission delay coefficient, the time scale of influence transmission between processes can be shortened, allowing the scheduling system to include more potentially affected processes in the influence domain assessment. The dynamic load penalty factor of high-load equipment in the scheduling optimization objective function is adjusted according to the conflict rate of production resources in the production scheduling plan. Since different processes usually rely on multiple types of production resources such as equipment, tooling, and manpower during the production scheduling process, inaccurate modeling of resource relationships leads to multiple processes being simultaneously assigned to the same high-load equipment in the scheduling scheme, causing resource occupation conflicts and requiring frequent adjustments to the scheduling scheme. By increasing the dynamic load penalty factor of high-load production equipment, the cost of high-load equipment continuing to undertake new process tasks can be increased during the scheduling optimization calculation, so that production resources with lower load or alternative capabilities are selected during resource allocation, reducing the probability of resource conflicts and improving the stability of production scheduling.

[0016] Furthermore, the system described in this invention determines whether the stability of the production schedule meets the requirements by setting a preset fluctuation characterization value. Since changes in equipment status and adjustments to order priorities may cause some processes in the original scheduling plan to be frequently adjusted during production execution, the production schedule is strongly affected by external disturbances and the scheduling plan is not stable enough. By determining the stability of the production schedule, it is possible to identify in a timely manner whether there is excessive adjustment or scheduling imbalance during the scheduling execution process, ensuring the continuity and execution efficiency of the production plan, and further improving the stability of the production schedule.

[0017] Furthermore, the system of the present invention adjusts the transmission delay coefficient between scheduling processes by setting a preset first update frequency and a preset second update frequency. Since the impact of upstream process adjustments on downstream processes needs a certain time delay to be identified, a small number of directly related processes are included in the influence domain during the scheduling evaluation process, resulting in some actually affected subsequent processes not being included in the evaluation scope, making the scheduling adjustment incomplete and triggering frequent subsequent adjustments. By reducing the process transmission delay coefficient, the time scale of the impact transmission between processes can be shortened, allowing the scheduling system to include more potentially affected processes in the influence domain scope when evaluating the influence domain, further improving the stability of production scheduling.

[0018] Furthermore, the system of the present invention adjusts the dynamic load penalty factor of high-load equipment in solving the scheduling optimization objective function by setting a preset conflict rate. Since different processes usually rely on multiple production resources such as equipment, tooling, and manpower during production scheduling, inaccurate modeling of resource relationships can lead to multiple processes being simultaneously assigned to the same high-load equipment in the scheduling scheme, causing resource occupation conflicts and requiring frequent adjustments to the scheduling scheme. By increasing the dynamic load penalty factor of high-load production equipment, the cost of high-load equipment continuing to undertake new process tasks can be increased during the scheduling optimization calculation, so that production resources with lower load or alternative capabilities are selected during resource allocation, reducing the probability of resource conflicts and further improving the stability of production scheduling. Attached Figure Description

[0019] Figure 1 This is a block diagram of the overall structure of the production scheduling optimization system based on a dynamic algorithm engine according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating the process by which the production scheduling optimization system based on a dynamic algorithm engine determines whether the stability of the production schedule meets the requirements, according to an embodiment of the present invention. Figure 3 This is a flowchart illustrating the process of determining the transmission delay coefficient between scheduling processes in the production scheduling optimization system based on a dynamic algorithm engine according to an embodiment of the present invention. Figure 4 This is an overall flowchart of the production scheduling optimization method based on a dynamic algorithm engine according to an embodiment of the present invention. Detailed Implementation

[0020] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0021] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0022] Please see Figure 1 As shown, it is an overall structural block diagram of the production scheduling optimization system based on a dynamic algorithm engine according to an embodiment of the present invention.

[0023] This invention discloses a production scheduling optimization system based on a dynamic algorithm engine, comprising: The data processing module includes a preprocessing unit for preprocessing the collected basic order data to output order features, and a vector construction unit connected to the preprocessing unit for quantifying the order features and constructing a standardized order feature vector. The scheduling solution module, which is connected to the data processing module, includes a function construction unit for constructing a scheduling optimization objective function based on production indicators and the standardized order feature vector, and a solution unit connected to the function construction unit for optimizing the objective function based on industrial production resource constraints and generating a production scheduling plan. The disturbance recalculation module, which is connected to the scheduling solution module, includes an evaluation unit for determining the impact domain of scheduling execution anomalies based on real-time industrial production resource constraints, and a recalculation unit connected to the evaluation unit for optimizing the scheduling processes within the impact domain to output a recalculated scheduling scheme. The scheduling determination module, which is connected to the disturbance recalculation module, is used to determine whether the stability of the production schedule meets the requirements based on the fluctuation characterization value determined by the number of scheduling processes optimized and adjusted per unit time and the total number of processes in the production scheduling plan. The delay adjustment module is connected to the disturbance recalculation module and the scheduling determination module respectively, and is used to determine the transmission delay coefficient between scheduling processes according to the update frequency of the recalculation scheduling scheme. The penalty factor adjustment module is connected to the scheduling solution module and the delay adjustment module respectively, and is used to determine the dynamic load penalty factor of high-load equipment in the scheduling optimization objective function solution based on the conflict rate of production resources in the production scheduling plan.

[0024] Specifically, order basic data includes order core data and order attribute data.

[0025] Specifically, the core order data includes the urgency of the order delivery, customer level, process complexity, and material lead time.

[0026] Specifically, order attribute data includes order demand data, order duration data, and process data.

[0027] Specifically, preprocessing includes cleaning, noise reduction, standardization, and feature extraction.

[0028] Specifically, order characteristics include delivery urgency, customer priority weight, process load characteristics, material preparation period, order demand, order processing time, and number of processes.

[0029] Specifically, the process of quantifying order features and constructing a standardized order feature vector involves weighting and quantizing the order features corresponding to the core order data, horizontally combining the quantified core data features with the original attribute data features to form an ordered numerical sequence containing all feature dimensions, which is the order feature vector. The vector is then orthogonalized and standardized to unify the dimensions and value ranges of each feature dimension in order to generate a standardized order feature vector.

[0030] Specifically, production metrics include customer-first strategies, cost reduction and efficiency improvement strategies, and emergency order delivery strategies.

[0031] Specifically, the process of constructing the scheduling optimization objective function based on production indicators and standardized order feature vectors is to take minimizing the total production cycle of all orders, maximizing the on-time delivery rate of orders, and maximizing the overall utilization rate of global resources as the core optimization objectives, and to transform multiple objectives into a single comprehensive objective function through weight coefficients.

[0032] Specifically, the objective function for scheduling optimization is: ; Where: T represents the total production cycle for the entire order; D represents the on-time delivery rate of orders; U represents the overall resource utilization rate; ω1, ω2, and ω3 are dynamic weighting coefficients, and ω1, ω2, and ω3 exist: ; Preferably, the preferred embodiment of ω1 is 0.2, the preferred embodiment of ω2 is 0.5, and the preferred embodiment of ω3 is 0.3.

[0033] Specifically, the overall resource utilization rate is: ; Among them, U 设备 This is the ratio of the actual processing time to the available processing time for all production equipment. U 人力 This is the ratio of actual working hours to available working hours for all production personnel. α and β are dynamic weighting coefficients for resources, and α and β exist: ; Preferably, the preferred embodiment of α is 0.5, and the preferred embodiment of β is 0.5.

[0034] Specifically, the process of optimizing the objective function and generating a production schedule based on industrial production resource constraints involves taking industrial production resource constraints as input, constructing a mixed integer programming model, obtaining a feasible solution space that satisfies all industrial production resource constraints based on standardized order feature vectors, using this feasible solution space as the initial population, and using the objective function as the fitness function. Through selection, crossover, and iterative optimization using an improved genetic algorithm, the system converges to the globally optimal combination of decision variables to obtain the optimal solution, which is then translated into a production schedule.

[0035] Specifically, industrial production resource constraints include constraints on operators, processing equipment, material inventory, process sequence, and environmental protection-related production restrictions.

[0036] Specifically, production scheduling plans include resource allocation schemes, material demand sequences, and work order assignment instructions.

[0037] Specifically, the process of determining the impact domain of scheduling execution anomalies based on real-time industrial production resource constraints involves tracing all scheduled processes and their upstream and downstream dependent processes allocated to the abnormal resources, identifying strongly related alternative resources of the abnormal resources, and combining the time of the anomaly occurrence to identify the time impact boundary, order impact boundary, and resource impact boundary, thereby defining a local impact domain that accounts for 10%-20% of the global order.

[0038] Specifically, the impact domain of scheduling execution anomalies is the scope of the plan that needs to be rescheduled, determined through process dependency tracing and resource correlation analysis, when anomalies occur during the execution of the production scheduling plan.

[0039] Specifically, the process of optimizing the scheduling processes within the influence domain to output a recalculated scheduling scheme takes the local scope defined by the influence domain as input, uses the baseline work plan within the influence domain as the initial solution, and uses a local neighborhood search algorithm for incremental recalculation: remove the corresponding resource allocation relationships of the processes affected by the anomaly, and then, with the goal of maximizing the on-time delivery rate and minimizing the adjustment range, reallocate processes and resources in the available global resources and adjust the process time. After a limited number of iterations and verification based on local hard constraints of industrial production resources, the optimal local solution is selected. The compatibility between the local scheme and the global plan is verified. If there is a conflict, the constraint relaxation rule is triggered to resolve it, and the updated recalculated scheduling scheme is output.

[0040] Specifically, the recalculated scheduling scheme includes the updated resource allocation scheme, the revised material requirement sequence, and the re-arranged work order instructions.

[0041] Specifically, the dynamic load penalty factor for high-load equipment is a penalty weight coefficient added to the high-load equipment in the scheduling optimization objective function to reduce the usage tendency.

[0042] Specifically, the inter-process delay coefficient is the ratio of the delay time of the upstream process to the actual delay time of the downstream process.

[0043] In implementation, the system of this invention sets up a data processing module, a scheduling solution module, a disturbance recalculation module, a scheduling judgment module, a delay adjustment module, and a penalty factor adjustment module. It determines the fluctuation characterization value based on the number of scheduled processes optimized and adjusted per unit time and the total number of processes in the production schedule plan to determine whether the stability of the production schedule meets the requirements. Because changes in equipment status and order priority adjustments during production execution may cause frequent adjustments to some processes in the original scheduling plan, the production schedule is highly susceptible to external disturbances, resulting in insufficient stability. By determining the stability of the production schedule, it can promptly identify whether there is excessive adjustment or scheduling imbalance during the scheduling execution process, ensuring the continuity and execution efficiency of the production plan. The system adjusts the transmission delay coefficient between scheduled processes based on the update frequency of the recalculated scheduling plan. Since the impact of upstream process adjustments on downstream processes requires a certain time delay to be identified, a small number of directly related processes are included in the influence domain during the scheduling evaluation process, leading to some actual... The absence of affected subsequent processes in the evaluation scope leads to incomplete scheduling adjustments and frequent subsequent adjustments. By reducing the process propagation delay coefficient, the time scale of impact propagation between processes can be shortened, allowing the scheduling system to include more potentially affected processes in the impact domain assessment. The dynamic load penalty factor for high-load equipment in the scheduling optimization objective function is adjusted based on the conflict rate of production resources in the production scheduling plan. Since different processes typically rely on various production resources such as equipment, tooling, and manpower during production scheduling, inaccurate resource correlation modeling results in multiple processes being simultaneously assigned to the same high-load equipment in the scheduling scheme, causing resource occupation conflicts and requiring frequent adjustments to the scheduling scheme. By increasing the dynamic load penalty factor for high-load production equipment, the cost of high-load equipment continuing to undertake new process tasks can be increased during scheduling optimization calculations. This encourages the selection of lower-load or alternative production resources during resource allocation, reducing the probability of resource conflicts and improving the stability of production scheduling.

[0044] Please continue reading. Figure 2 As shown, it is a logical flowchart of the process of determining whether the stability of the production schedule meets the requirements in the production scheduling optimization system based on the dynamic algorithm engine according to an embodiment of the present invention.

[0045] Specifically, the scheduling determination module determines a fluctuation characterization value based on the ratio of the number of scheduled processes optimized and adjusted per unit time to the total number of processes in the production scheduling plan, in order to determine whether the stability of the production schedule meets the requirements.

[0046] Specifically, the scheduling determination module determines that the stability of the production schedule meets the requirements when the fluctuation characterization value is less than or equal to a preset fluctuation characterization value. The scheduling determination module determines that the stability of the production schedule does not meet the requirements when the fluctuation characterization value is greater than the preset fluctuation characterization value.

[0047] Understandably, in a production scheduling optimization system based on a dynamic algorithm engine, the core logic of quantifying the stability of the production schedule using a preset fluctuation characterization value is to transform the abstract stability of the production schedule into a quantifiable adjustment range. By comparing this range with a preset fluctuation threshold, it is determined whether the schedule maintains the expected stability during actual execution. The preset fluctuation characterization value serves as a dividing line to distinguish whether the stability of the production schedule meets the requirements. The preset fluctuation characterization value can be set according to actual working conditions. The setting of the preset fluctuation characterization value aims to ensure the stability and practicality of production scheduling optimization. Optionally, the preset fluctuation characterization value is determined through a limited number of experiments by evaluating the optimization effect of different optimization adjustment ranges on the production schedule. The determined preset fluctuation characterization value should satisfy the condition that it is neither too small nor too large to cause excessive interference to the optimization process of the production schedule. For example, the preset fluctuation characterization value is generally selected in the range of [0.05, 0.15].

[0048] Preferably, the preferred embodiment of the preset fluctuation characterization value is 0.1.

[0049] In practice, the system described in this invention determines whether the stability of the production schedule meets the requirements by setting a preset fluctuation characterization value. Since changes in equipment status and adjustments to order priorities may cause some processes in the original scheduling plan to be frequently adjusted during production execution, the production schedule is highly susceptible to external disturbances and the scheduling plan is not stable enough. By determining the stability of the production schedule, it is possible to identify in a timely manner whether there is excessive adjustment or scheduling imbalance during the scheduling execution process, ensuring the continuity and execution efficiency of the production plan, and further improving the stability of the production schedule.

[0050] Please continue reading. Figure 3 As shown, it is a logical flowchart of the process of determining the transmission delay coefficient between scheduling processes in the production scheduling optimization system based on the dynamic algorithm engine according to an embodiment of the present invention.

[0051] Specifically, in response to the condition that the stability of the production schedule does not meet the requirements, the delay adjustment module determines whether the scope of the impact domain definition assessment meets the requirements based on the update frequency of the recalculated scheduling scheme.

[0052] Specifically, the delay adjustment module responds to the fact that the update frequency of the recalculation scheduling scheme is greater than the preset first update frequency and less than or equal to the preset second update frequency, and determines that the scope of the impact domain assessment meets the requirements. The delay adjustment module determines that the scope of the impact domain assessment does not meet the requirements if the update frequency of the recalculation scheduling scheme is less than or equal to the preset first update frequency or greater than the preset second update frequency.

[0053] Specifically, the delay adjustment module responds to the recalculation of the scheduling scheme when the update frequency is greater than the preset second update frequency, thereby reducing the transmission delay coefficient between scheduling processes; The delay adjustment module responds to the fact that the update frequency of the recalculation scheduling scheme is less than or equal to the preset first update frequency, and initially determines that the rationality of the allocation of industrial production resources does not meet the requirements. It then determines whether the rationality of the allocation of industrial production resources meets the requirements based on the conflict rate of production resources in the production scheduling plan.

[0054] It is understandable that the preset first update frequency is less than the preset second update frequency. The three intervals divided by the preset first update frequency and the preset second update frequency correspond to three different scenarios: The first interval is when the update frequency of the recalculated scheduling scheme is less than or equal to the preset first update frequency. The corresponding situation is: In the production scheduling process, different processes usually rely on multiple production resources such as equipment, tooling and manpower. Inaccurate modeling of resource association relationships leads to multiple processes being simultaneously assigned to the same high-load equipment in the scheduling scheme, causing resource occupation conflicts, and the scheduling scheme needs to be adjusted frequently. The second interval is when the update frequency of the recalculated scheduling scheme is greater than the preset first update frequency and less than or equal to the preset second update frequency. The corresponding situation is that the scope of the impact domain assessment meets the requirements. The third interval is when the update frequency of the recalculated scheduling scheme is greater than the preset second update frequency. The corresponding situation is: because the impact of upstream process adjustments on downstream processes needs to be identified after a certain time delay, a small number of directly related processes are included in the impact domain during the scheduling evaluation process, resulting in some actually affected subsequent processes not being included in the evaluation scope, making the scheduling adjustment incomplete and causing frequent subsequent adjustments. In this case, it is necessary to reduce the transmission delay coefficient between scheduling processes.

[0055] Understandably, in a production scheduling optimization system based on a dynamic algorithm engine, the preset first update frequency and preset second update frequency characterize the scope of the impact domain assessment. The core logic is to establish a quantitative correlation between the frequency of scheduling adjustments and the sufficiency of the impact domain coverage between processes by dividing the update frequency of the recalculated scheduling scheme into different intervals. The preset first update frequency and preset second update frequency serve as the dividing line to distinguish whether the scope of the impact domain meets the requirements. The preset first update frequency and preset second update frequency can be set according to actual working conditions. The setting of the preset first update frequency and preset second update frequency aims to ensure the stability and practicality of production scheduling optimization. Optionally, the preset first update frequency and preset second update frequency are determined through a limited number of experiments by evaluating the optimization effect of different scheme update frequencies on production scheduling. The determined preset first update frequency and preset second update frequency should satisfy the condition that they are neither too small nor cause excessive interference to the production scheduling optimization process. For example, the preset first update frequency is generally selected in the range of [0.18 times / hour, 0.22 times / hour], and the preset first update frequency and preset second update frequency are generally selected in the range of [0.48 times / hour, 0.52 times / hour].

[0056] Preferably, the first update frequency is 0.2 times / hour in a preferred embodiment, and the second update frequency is 0.5 times / hour in a preferred embodiment.

[0057] Specifically, the recalculation frequency of the scheduling scheme is the number of times the scheduling process is incrementally optimized and the scheduling scheme is recalculated per unit time.

[0058] Specifically, the reduction in the transmission delay coefficient between scheduling processes is determined by the difference between the recalculated update frequency of the scheduling scheme and the preset second update frequency.

[0059] Specifically, when the difference between the recalculation schedule update frequency and the preset second update frequency does not exceed 0.2 times / hour, the inter-process transmission delay coefficient is reduced to 0.9 times the original value. When the difference between the recalculation schedule update frequency and the preset second update frequency exceeds 0.1 times / hour, the inter-process transmission delay coefficient is reduced by 0.02 for every 0.1 times / hour exceeding the original value, in addition to being reduced to 0.9 times. For example, when the difference between the recalculation schedule update frequency and the preset second update frequency is 0.2 times / hour, the current inter-process transmission delay coefficient is 0.7, and the reduced inter-process transmission delay coefficient is 0.7×0.9-0.02×1=0.61.

[0060] In practice, the system of the present invention adjusts the transmission delay coefficient between scheduling processes by setting a preset first update frequency and a preset second update frequency. Since the impact of upstream process adjustments on downstream processes needs a certain time delay to be identified, a small number of directly related processes are included in the influence domain during the scheduling evaluation process, resulting in some actually affected subsequent processes not being included in the evaluation scope. This leads to incomplete scheduling adjustments and frequent subsequent adjustments. By reducing the process transmission delay coefficient, the time scale of the impact transmission between processes can be shortened, allowing the scheduling system to include more potentially affected processes in the influence domain when evaluating the influence domain, further improving the stability of production scheduling.

[0061] Specifically, the penalty factor adjustment module, in response to the condition that the update frequency of the recalculation scheduling scheme is less than or equal to the preset first update frequency, determines whether the rationality of the allocation of industrial production resources meets the requirements based on the conflict rate of production resources in the production scheduling plan.

[0062] Specifically, the penalty factor adjustment module responds to the production scheduling plan where the conflict rate of production resources is less than or equal to a preset conflict rate, thus determining that the rationality of the allocation of industrial production resources meets the requirements. The penalty factor adjustment module responds to the fact that the conflict rate of production resources in the production scheduling plan is greater than the preset conflict rate, determines that the rationality of the allocation of industrial production resources does not meet the requirements, and increases the dynamic load penalty factor of high-load equipment in the solution of the scheduling optimization objective function.

[0063] It is understandable that the two intervals of the preset conflict rate division correspond to two different scenarios: The first interval is when the conflict rate of production resources in the production scheduling plan is less than or equal to the preset conflict rate, which corresponds to the situation where the rationality of the allocation of industrial production resources is determined to meet the requirements. The second interval is when the conflict rate of production resources in the production scheduling plan is greater than the preset conflict rate. The corresponding situation is as follows: In the production scheduling process, different processes usually rely on multiple types of production resources such as equipment, tooling and manpower. Inaccurate modeling of resource association relationships leads to multiple processes being simultaneously assigned to the same high-load equipment in the scheduling plan, causing resource occupation conflicts. The scheduling plan needs to be adjusted frequently. At this time, it is necessary to increase the dynamic load penalty factor of high-load equipment in the solution of the scheduling optimization objective function.

[0064] Understandably, in a production scheduling optimization system based on a dynamic algorithm engine, using a preset conflict rate to characterize the rationality of industrial production resource allocation can intuitively and quantitatively reflect the degree of overlap in the occupation of equipment, personnel, and other resources by production tasks. The size of the conflict rate directly corresponds to whether the resource allocation is balanced and whether the scheduling plan is feasible. The preset conflict rate can be set according to actual working conditions. The setting of the preset conflict rate aims to ensure the stability and practicality of production scheduling optimization. Optionally, the preset conflict rate is determined through a limited number of trials by evaluating the optimization effect of different resource conflict rates on production scheduling. The determined preset conflict rate should satisfy the condition that it is neither too small nor will it cause excessive interference to the production scheduling optimization process. For example, the preset conflict rate is generally selected in the range of [0.5%, 1.5%].

[0065] Preferably, the preset conflict rate is 1% in this preferred embodiment.

[0066] Specifically, the conflict rate of production resources in production scheduling is the ratio of the number of production resources that are simultaneously occupied or overloaded by multiple processes to the total number of production resources.

[0067] Specifically, the increase in the dynamic load penalty factor of high-load equipment in the solution of the scheduling optimization objective function is determined by the difference between the conflict rate of production resources in the production scheduling plan and the preset conflict rate.

[0068] Specifically, when the difference between the conflict rate of production resources in the production scheduling plan and the preset conflict rate is within 0.5%, the dynamic load penalty factor of high-load equipment in the scheduling optimization objective function solution is increased to 1.1 times the original value. When the difference between the conflict rate of production resources in the production scheduling plan and the preset conflict rate exceeds 0.5%, on the basis of increasing to 1.1 times the original value, for every 0.2% exceeding the original value, the dynamic load penalty factor of high-load equipment in the scheduling optimization objective function solution is increased by 0.03. For example, when the difference between the conflict rate of production resources in the production scheduling plan and the preset conflict rate is 0.9%, the current dynamic load penalty factor of high-load equipment in the scheduling optimization objective function solution is 0.4, and the increased dynamic load penalty factor of high-load equipment in the scheduling optimization objective function solution is 0.4×1.1+0.03×2=0.5.

[0069] In implementation, the system of the present invention adjusts the dynamic load penalty factor of high-load equipment in solving the scheduling optimization objective function by setting a preset conflict rate. Since different processes usually rely on multiple production resources such as equipment, tooling, and manpower during production scheduling, inaccurate modeling of resource association relationships can lead to multiple processes being simultaneously assigned to the same high-load equipment in the scheduling scheme, causing resource occupation conflicts. The scheduling scheme needs to be adjusted frequently. By increasing the dynamic load penalty factor of high-load production equipment, the cost of high-load equipment continuing to undertake new process tasks can be increased during the scheduling optimization calculation. This allows the allocation of resources to select production resources with lower load or alternative capabilities, reducing the probability of resource conflicts and further improving the stability of production scheduling.

[0070] Please continue reading. Figure 4 As shown, it is an overall flowchart of the production scheduling optimization method based on the dynamic algorithm engine in an embodiment of the present invention.

[0071] A production scheduling optimization method based on a dynamic algorithm engine includes: Step S1: The collected basic order data is cleaned, denoised, standardized, and feature extracted sequentially to output order features. The order features are then quantified and a standardized order feature vector is constructed. Step S2: Construct a scheduling optimization objective function based on production indicators and the standardized order feature vector; optimize and solve the objective function based on industrial production resource constraints and generate a production scheduling plan. Step S3: Determine the impact domain of scheduling execution anomalies based on real-time industrial production resource constraints, and optimize the scheduling processes within the impact domain to output a recalculated scheduling scheme. Step S4: Obtain the number of scheduling processes optimized and adjusted per unit time and the total number of processes in the production scheduling plan to determine the fluctuation characterization value, and determine whether the stability of the production schedule meets the requirements based on the fluctuation characterization value; Step S5: If the stability of the production schedule does not meet the requirements, obtain the update frequency of the recalculation schedule plan to determine whether the scope of the impact domain assessment meets the requirements. Step S6: If the scope of the impact domain assessment does not meet the requirements, determine whether to reduce the transmission delay coefficient between scheduling processes based on the update frequency of the recalculated scheduling scheme. Step S7: If it is not necessary to reduce the transmission delay coefficient between scheduling processes, then obtain the conflict rate of production resources in the production scheduling plan to determine the dynamic load penalty factor of high-load equipment in the scheduling optimization objective function solution.

[0072] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A production scheduling optimization system based on a dynamic algorithm engine, characterized in that, include: The data processing module includes a preprocessing unit for preprocessing the collected basic order data to output order features, and a vector construction unit connected to the preprocessing unit for quantifying the order features and constructing a standardized order feature vector. The scheduling solution module, which is connected to the data processing module, includes a function construction unit for constructing a scheduling optimization objective function based on production indicators and the standardized order feature vector, and a solution unit connected to the function construction unit for optimizing the objective function based on industrial production resource constraints and generating a production scheduling plan. The disturbance recalculation module, which is connected to the scheduling solution module, includes an evaluation unit for determining the impact domain of scheduling execution anomalies based on real-time industrial production resource constraints, and a recalculation unit connected to the evaluation unit for optimizing the scheduling processes within the impact domain to output a recalculated scheduling scheme. The scheduling determination module, which is connected to the disturbance recalculation module, is used to determine whether the stability of the production schedule meets the requirements based on the fluctuation characterization value determined by the number of scheduling processes optimized and adjusted per unit time and the total number of processes in the production scheduling plan. The delay adjustment module is connected to the disturbance recalculation module and the scheduling determination module respectively, and is used to determine the transmission delay coefficient between scheduling processes according to the update frequency of the recalculation scheduling scheme. The penalty factor adjustment module is connected to the scheduling solution module and the delay adjustment module respectively, and is used to determine the dynamic load penalty factor of high-load equipment in the scheduling optimization objective function solution based on the conflict rate of production resources in the production scheduling plan.

2. The production scheduling optimization system based on a dynamic algorithm engine according to claim 1, characterized in that, The scheduling determination module determines the fluctuation characterization value based on the ratio of the number of scheduled processes optimized and adjusted per unit time to the total number of processes in the production scheduling plan, in order to determine whether the stability of the production schedule meets the requirements.

3. The production scheduling optimization system based on a dynamic algorithm engine according to claim 2, characterized in that, The scheduling determination module determines that the stability of the production schedule meets the requirements when the fluctuation characterization value is less than or equal to the preset fluctuation characterization value. The scheduling determination module determines that the stability of the production schedule does not meet the requirements when the fluctuation characterization value is greater than the preset fluctuation characterization value.

4. The production scheduling optimization system based on a dynamic algorithm engine according to claim 3, characterized in that, In response to the condition that the stability of the production schedule does not meet the requirements, the delay adjustment module determines whether the scope of the impact domain assessment meets the requirements based on the update frequency of the recalculated scheduling scheme.

5. The production scheduling optimization system based on a dynamic algorithm engine according to claim 4, characterized in that, The delay adjustment module responds to the fact that the update frequency of the recalculation scheduling scheme is greater than the preset first update frequency and less than or equal to the preset second update frequency, and determines that the scope of the impact domain assessment meets the requirements. The delay adjustment module determines that the scope of the impact domain assessment does not meet the requirements if the update frequency of the recalculation scheduling scheme is less than or equal to the preset first update frequency or greater than the preset second update frequency.

6. The production scheduling optimization system based on a dynamic algorithm engine according to claim 5, characterized in that, The delay adjustment module responds to the fact that the update frequency of the recalculation scheduling scheme is greater than the preset second update frequency, thereby reducing the transmission delay coefficient between scheduling processes; The delay adjustment module responds to the fact that the update frequency of the recalculation scheduling scheme is less than or equal to the preset first update frequency, and initially determines that the rationality of the allocation of industrial production resources does not meet the requirements. It then determines whether the rationality of the allocation of industrial production resources meets the requirements based on the conflict rate of production resources in the production scheduling plan.

7. The production scheduling optimization system based on a dynamic algorithm engine according to claim 6, characterized in that, The reduction in the transmission delay coefficient between scheduling processes is determined by the difference between the update frequency of the recalculated scheduling scheme and the preset second update frequency.

8. The production scheduling optimization system based on a dynamic algorithm engine according to claim 7, characterized in that, The penalty factor adjustment module, in response to the condition that the update frequency of the recalculation scheduling scheme is less than or equal to the preset first update frequency, determines whether the rationality of the allocation of industrial production resources meets the requirements based on the conflict rate of production resources in the production scheduling plan.

9. The production scheduling optimization system based on a dynamic algorithm engine according to claim 8, characterized in that, The penalty factor adjustment module responds to the production scheduling plan if the conflict rate of production resources is less than or equal to the preset conflict rate, and determines that the rationality of the allocation of industrial production resources meets the requirements. The penalty factor adjustment module responds to the fact that the conflict rate of production resources in the production scheduling plan is greater than the preset conflict rate, determines that the rationality of industrial production resource allocation does not meet the requirements, and increases the dynamic load penalty factor of high-load equipment in the solution of the scheduling optimization objective function. In the solution of the scheduling optimization objective function, the increase in the dynamic load penalty factor of high-load equipment is determined by the difference between the conflict rate of production resources in the production scheduling plan and the preset conflict rate.

10. A production scheduling optimization method applied to the production scheduling optimization system based on a dynamic algorithm engine as described in any one of claims 1-9, characterized in that, include: The collected basic order data is sequentially cleaned, denoised, standardized, and feature extracted to output order features. The order features are then quantified and a standardized order feature vector is constructed. Based on production indicators and the standardized order feature vector, a scheduling optimization objective function is constructed. Based on industrial production resource constraints, the objective function is optimized and solved to generate a production scheduling plan. Based on real-time industrial production resource constraints, the impact domain of scheduling execution anomalies is determined, and the scheduling processes within the impact domain are optimized to output a recalculated scheduling scheme. The number of scheduling processes optimized and adjusted per unit time and the total number of processes in the production scheduling plan are obtained to determine the fluctuation characterization value, and the stability of the production scheduling is determined based on the fluctuation characterization value. If the stability of the production schedule does not meet the requirements, the update frequency of the recalculated scheduling scheme is obtained to determine whether the scope of the impact domain assessment meets the requirements. If the scope of the impact domain assessment does not meet the requirements, then determine whether to reduce the transmission delay coefficient between scheduling processes based on the update frequency of the recalculated scheduling scheme. If it is not necessary to reduce the transmission delay coefficient between scheduling processes, then the conflict rate of production resources in the production scheduling plan is obtained to determine the dynamic load penalty factor of high-load equipment in the solution of the scheduling optimization objective function.

Images