Production plan and maintenance task cooperation method and device

Abstract

The invention provides a production plan and maintenance task cooperation method and device, and the method comprises the steps: inputting the obtained to-be-tested time data and a to-be-tested monthly production plan into a target monthly daily productivity distribution interval prediction network, and obtaining a monthly daily productivity prediction distribution interval; inputting the monthly daily productivity prediction distribution interval and historical month data into a target daily productivity distribution interval probability prediction network to obtain a daily productivity distribution interval prediction probability; extracting features of the to-be-executed task from the to-be-executed maintenance task; and inputting the daily productivity distribution interval prediction probability, the to-be-executed task features and the task sorting result into a target task arrangement optimization network to obtain an optimal to-be-executed task arrangement strategy. And through monthly daily productivity distribution interval prediction and daily productivity distribution interval probability prediction, automatic arrangement of the maintenance plan is realized. Through collaborative operation of a production plan and a maintenance task, equipment maintenance is arranged under the condition that productivity is affected as little as possible.

Description

Technical Field

[0001] The present invention relates to the field of logistics technology, and in particular to a method and device for coordinating production planning and maintenance tasks. Background Art

[0002] Domestic enterprises such as professional coal transportation terminals have a large number of equipment. Equipment that works for a long time needs maintenance, which will affect production. At the same time, when the production volume is large, it is hoped that maintenance can be reduced as much as possible. Therefore, how to coordinate the operation time of production and maintenance has become a problem that needs to be solved when formulating production plans and maintenance plans. Summary of the invention

[0003] The present invention provides a method and device for coordinating production planning and maintenance tasks, which are used to arrange equipment maintenance while minimizing the impact on production capacity.

[0004] In a first aspect, the present invention provides a method for coordinating production planning and maintenance tasks, comprising:

[0005] Input the acquired time data to be tested and the monthly production plan to be tested into the target monthly daily capacity distribution interval prediction network to obtain the monthly daily capacity prediction distribution interval;

[0006] Input the monthly daily capacity forecast distribution interval and historical monthly data into the target daily capacity distribution interval probability forecast network to obtain the daily capacity distribution interval forecast probability;

[0007] Extracting features of tasks to be performed from the maintenance tasks to be performed;

[0008] Sorting the tasks to be performed according to the duration and priority of the maintenance tasks to be performed, and obtaining a task sorting result;

[0009] The predicted probability of daily capacity distribution interval, the characteristics of the tasks to be executed and the task sorting results are input into the target task scheduling optimization network to obtain the optimal scheduling strategy for the tasks to be executed.

[0010] Optionally, the predicted probability of the daily capacity distribution interval, the characteristics of the tasks to be executed and the task sorting results are input into a target task scheduling optimization network to obtain an optimal scheduling strategy for the tasks to be executed, including:

[0011] Using a Monte Carlo simulation method, combined with the predicted probability of the daily capacity distribution interval and the task sorting result, the optimal scheduling time is determined;

[0012] The characteristics of the tasks to be executed and the optimal scheduling time are input into a target task scheduling optimization network to obtain the optimal scheduling strategy for the tasks to be executed.

[0013] Optionally, using a Monte Carlo simulation method, combining the predicted probability of the daily capacity distribution interval and the task sorting result, to determine the optimal scheduling time includes:

[0014] Randomly sampling a set of generation scenarios constructed based on the predicted probability of the daily capacity distribution interval;

[0015] Calculate the frequency of the production volume being in a low range within the corresponding duration for each of the generation volume scenario sets;

[0016] The distribution results of the frequencies are combined, and the time period with the largest joint probability is selected as the optimal scheduling time.

[0017] Optionally, the steps of constructing the target monthly daily capacity distribution interval prediction network include:

[0018] Obtaining historical monthly daily capacity distribution data; the historical monthly daily capacity distribution data includes: historical monthly data, historical monthly production plan and historical monthly daily capacity range;

[0019] With the historical monthly daily capacity interval as a label, the historical monthly data and the historical monthly production plan as input, the constructed monthly daily capacity distribution interval prediction network is trained to obtain a target monthly daily capacity distribution interval prediction network.

[0020] Optionally, the steps of constructing the target daily capacity distribution interval probability prediction network include:

[0021] Inputting the historical monthly data and the historical monthly production plan into the target monthly daily capacity distribution interval prediction network to obtain the historical daily capacity prediction distribution interval;

[0022] Taking the historical daily capacity forecast distribution interval as a label, the historical ship plan data and the historical month data as input, the constructed daily capacity distribution interval probability forecasting network is trained to obtain the target daily capacity distribution interval probability forecasting network.

[0023] Optionally, the step of constructing the target task arrangement optimization network includes:

[0024] By defining the environment state, action space and reward mechanism, the constructed task scheduling optimization network is trained based on the characteristics of the tasks to be performed and the optimal scheduling time, so as to gradually achieve the maximum overall benefit;

[0025] The policy function under the maximum overall benefit is used as the target policy function, and the target task scheduling optimization network is constructed using the target policy function.

[0026] In a second aspect, the present invention provides a production plan and maintenance task coordination device, comprising:

[0027] The monthly daily capacity prediction module is used to input the acquired time data to be tested and the monthly production plan to be tested into the target monthly daily capacity distribution interval prediction network to obtain the monthly daily capacity prediction distribution interval;

[0028] A daily capacity distribution interval prediction module, which is used to input the monthly daily capacity prediction distribution interval and historical monthly data into the target daily capacity distribution interval probability prediction network to obtain the daily capacity distribution interval prediction probability;

[0029] A task feature extraction module is used to extract features of tasks to be performed from maintenance tasks to be performed;

[0030] A task sorting result determination module is used to sort the tasks to be performed according to the duration and priority of the maintenance tasks to be performed, and obtain the task sorting result;

[0031] The scheduling strategy determination module is used to input the predicted probability of the daily capacity distribution interval, the characteristics of the tasks to be executed and the task sorting results into the target task scheduling optimization network to obtain the optimal scheduling strategy for the tasks to be executed.

[0032] Optionally, the scheduling strategy determination module includes:

[0033] An optimal scheduling time determination submodule is used to determine the optimal scheduling time by using a Monte Carlo simulation method in combination with the predicted probability of the daily capacity distribution interval and the task sorting result;

[0034] The scheduling strategy determination submodule is used to input the characteristics of the tasks to be executed and the optimal scheduling time into the target task scheduling optimization network to obtain the optimal scheduling strategy for the tasks to be executed.

[0035] In a third aspect, the present invention provides an electronic device, comprising a processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in the method provided in the first aspect are performed.

[0036] In a fourth aspect, the present invention provides a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, runs the steps of the method provided in the first aspect.

[0037] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the steps in the method provided in the first aspect are performed.

[0038] It can be seen from the above technical solutions that the present invention has the following advantages:

[0039] The present invention provides a method and device for the coordination of production plan and maintenance task, the method comprising: inputting the acquired time data to be tested and the monthly production plan to be tested into the target monthly daily capacity distribution interval prediction network to obtain the monthly daily capacity prediction distribution interval; inputting the monthly daily capacity prediction distribution interval and the historical monthly data into the target daily capacity distribution interval probability prediction network to obtain the daily capacity distribution interval prediction probability; extracting the features of the tasks to be executed from the maintenance tasks to be executed; sorting the tasks to be executed according to the duration and priority of the maintenance tasks to be executed to obtain the task sorting result; inputting the daily capacity distribution interval prediction probability, the features of the tasks to be executed and the task sorting result into the target task arrangement optimization network to obtain the optimal task arrangement strategy to be executed. The maintenance plan is automatically arranged through the monthly daily capacity distribution interval prediction and the daily capacity distribution interval probability prediction. It is suitable for the formulation of production plan and maintenance tasks of professional coal transportation terminals. Through the collaborative operation of production plan and maintenance tasks, equipment maintenance can be arranged with the least impact on production capacity, and the additional damage to the equipment caused by the continuous operation time is minimized. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0041] Figure 1 A flowchart of a first embodiment of a method for coordinating production planning and maintenance tasks according to the present invention;

[0042] Figure 2 It is a flow chart of a second embodiment of a collaborative method for production planning and maintenance tasks of the present invention;

[0043] Figure 3 The present invention is a structural block diagram of an embodiment of a collaborative device for production planning and maintenance tasks according to the present invention. DETAILED DESCRIPTION

[0044] The embodiment of the present invention provides a method and device for coordinating production planning and maintenance tasks, which are used to arrange equipment maintenance while minimizing the impact on production capacity.

[0045] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] Embodiment 1

[0047] See also Figure 1 , Figure 1 This is a flow chart of a first embodiment of a method for coordinating production planning and maintenance tasks according to the present invention. The method includes:

[0048] Step S101, input the acquired time data to be tested and the monthly production plan to be tested into the target monthly daily capacity distribution interval prediction network to obtain the monthly daily capacity prediction distribution interval;

[0049] In an embodiment of the present application, the steps of constructing a target monthly daily capacity distribution interval prediction network include: obtaining historical monthly daily capacity distribution data; the historical monthly daily capacity distribution data includes: historical monthly data, historical monthly production plans and historical monthly daily capacity intervals; using the historical monthly daily capacity intervals as labels, the historical monthly data and the historical monthly production plans as inputs, training the constructed monthly daily capacity distribution interval prediction network, and obtaining a target monthly daily capacity distribution interval prediction network.

[0050] Step S102, inputting the monthly daily capacity forecast distribution interval and historical monthly data into the target daily capacity distribution interval probability forecast network to obtain the daily capacity distribution interval forecast probability;

[0051] In an embodiment of the present application, the steps of constructing a target daily capacity distribution interval probability prediction network include: inputting the historical monthly data and the historical monthly production plan into the target monthly daily capacity distribution interval prediction network to obtain the historical daily capacity prediction distribution interval; using the historical daily capacity prediction distribution interval as a label, the historical ship plan data and the historical monthly data as input, training the constructed daily capacity distribution interval probability prediction network to obtain the target daily capacity distribution interval probability prediction network.

[0052] Step S103, extracting features of tasks to be performed from the maintenance tasks to be performed;

[0053] In the embodiment of the present application, the maintenance characteristics of the task to be performed include: task complexity, required resources, historical task execution records, etc.

[0054] Step S104, sorting the tasks to be executed according to the acquired duration and priority of the maintenance tasks to be executed, and obtaining a task sorting result;

[0055] Step S105, inputting the predicted probability of the daily capacity distribution interval, the characteristics of the tasks to be executed and the task sorting results into a target task scheduling optimization network to obtain an optimal scheduling strategy for the tasks to be executed.

[0056] In an embodiment of the present application, a Monte Carlo simulation method is used to determine the optimal scheduling time in combination with the predicted probability of the daily capacity distribution interval and the task sorting result; the characteristics of the tasks to be executed and the optimal scheduling time are input into a target task scheduling optimization network to obtain the optimal scheduling strategy for the tasks to be executed.

[0057] The steps of constructing the target task scheduling optimization network include: by defining the environment state, action space and reward mechanism, based on the characteristics of the task to be executed and the optimal scheduling time, the constructed task scheduling optimization network is trained to gradually achieve the maximum overall benefit; using the strategy function under the maximum overall benefit as the target strategy function, and using the target strategy function to construct the target task scheduling optimization network.

[0058] The collaborative method of production plan and maintenance task provided by the embodiment of the present invention includes: inputting the acquired time data to be tested and the monthly production plan to be tested into the target monthly daily capacity distribution interval prediction network to obtain the monthly daily capacity prediction distribution interval; inputting the monthly daily capacity prediction distribution interval and historical monthly data into the target daily capacity distribution interval probability prediction network to obtain the daily capacity distribution interval prediction probability; extracting the features of the tasks to be executed from the maintenance tasks to be executed; sorting the tasks to be executed according to the duration and priority of the maintenance tasks to be executed to obtain the task sorting result; inputting the daily capacity distribution interval prediction probability, the features of the tasks to be executed and the task sorting result into the target task arrangement optimization network to obtain the optimal task arrangement strategy to be executed. The maintenance plan is automatically arranged through the monthly daily capacity distribution interval prediction and the daily capacity distribution interval probability prediction. It is suitable for the formulation of production plans and maintenance tasks for professional coal transportation terminals. Through the collaborative operation of production plans and maintenance tasks, equipment maintenance can be arranged with minimal impact on production capacity, and the additional damage to the equipment caused by the continuous operation time is minimized.

[0059] Embodiment 2

[0060] See also Figure 2 , Figure 2 This is a flow chart of a second embodiment of a collaborative method for production planning and maintenance tasks of the present invention, the steps comprising:

[0061] Step S201, input the acquired time data to be tested and the monthly production plan to be tested into the target monthly daily capacity distribution interval prediction network to obtain the monthly daily capacity prediction distribution interval;

[0062] Maintenance should be arranged on working days with lower production capacity as much as possible. Therefore, the premise of arranging maintenance plans is to accurately predict the production capacity distribution of future working days. There is little current data, and it is unrealistic to accurately predict the daily production capacity for a period of time in the future. According to observations, the distribution of monthly daily production capacity intervals is correlated with factors such as monthly production plans and monthly days. As shown in the following table:

[0063]

[0064] The modeling idea of ​​the embodiment of the present application is: using historical monthly data (including historical monthly days and historical monthly data) and historical monthly production plans as input, using historical monthly daily capacity ranges as labels, using MATLAB as a modeling tool, and using polynomial fitting or shallow machine learning models (such as BP neural networks, forest models, etc.) to model, thereby predicting the daily capacity distribution in the future through the target monthly daily capacity distribution range prediction network.

[0065] Taking BP neural network as an example, its modeling steps are as follows:

[0066] 1. Data preparation: Ensure that the input feature matrix X contains historical monthly data, historical monthly days, and historical monthly production plans, and the output matrix y contains the actual number of days in each train quantity range, that is, the historical monthly daily capacity range.

[0067] 2. Data normalization: Use MATLAB's normalize function to standardize the input features.

[0068] 3. Data segmentation: Use the cvpartition function to divide the data into training set and test set.

[0069] 4. Neural network model: Use the fitnet function to create a BP neural network with one hidden layer. The hidden layer contains 10 neurons, and the training function uses the Levenberg-Marquardt algorithm.

[0070] 5. Model training and prediction: Use training data to train the network, and then use test data for prediction.

[0071] 6. Result output: Output the actual and predicted results of the training set and test set.

[0072] The output format of the prediction result data is shown in the table above.

[0073] Step S202, inputting the monthly daily capacity prediction distribution interval and historical monthly data into the target daily capacity distribution interval probability prediction network to obtain the daily capacity distribution interval prediction probability;

[0074] In the embodiment of the present application, since the subject of the study is a logistics enterprise, its daily production capacity data is jointly affected by multiple factors from upstream and downstream. The prediction accuracy of directly using limited historical data to model is poor, and the prediction results cannot be used as the basis for automatic production plan layout. Under the data conditions of the current database, the probability of the numerical range of the daily production capacity of each day in the monthly daily production capacity prediction distribution range can be predicted on the basis of obtaining the monthly daily production capacity prediction distribution range. The prediction result sample is shown in the following table:

[0075] month 170 columns 160 columns 150 columns 140 columns 130 columns X month X day 5％ 10％ 55％ 25％ 5％

[0076] The predicted probability distribution is more in line with the current data situation. This technical framework is more flexible, and the prediction effect can be continuously improved when new valid attributes are added. Based on the probability distribution of each day, the joint probability over a period of time can be calculated, laying the foundation for the automated layout of maintenance plans.

[0077] In the specific implementation, to predict the probability of daily production capacity being in different numerical ranges for each day, you can consider using multi-classification machine learning models, such as random forests, gradient boosting machines, and support vector machines. These models can handle complex feature relationships and provide probability predictions for each category (i.e., different ranges of the number of trains).

[0078] Taking random forest as an example, the modeling steps are as follows:

[0079] 1. Data collection and preprocessing:

[0080] (1) Data integration: Collect all relevant data, including historical monthly daily capacity distribution range, historical monthly days, historical monthly data, and historical monthly production plans.

[0081] (2) Handling missing values: Check the missing values ​​in the dataset and decide how to handle the missing values ​​(including filling, deleting, or interpolating).

[0082] (3) Feature encoding: Convert categorical data (such as timestamp if it is a month name) into numerical values, using one-hot encoding or label encoding.

[0083] (4) Normalization / Standardization: Normalize (or standardize) numerical features to ensure that the model does not favor certain features due to the dimension of the features.

[0084] 2. Feature Engineering Selection:

[0085] (1) Feature engineering: Create new features based on business understanding, such as seasonal factors extracted from dates, average production volume in previous days, etc.

[0086] (2) Dimensionality reduction: Use feature importance scores, principal component analysis (PCA), or other methods to select the most relevant features.

[0087] 3. Model building:

[0088] (1) Model selection: Random forest is used as the base model due to its excellent accuracy and ability to handle nonlinear relationships.

[0089] (2) Configure model parameters:

[0090] n_estimators: The number of trees in the forest. More trees increase the stability and accuracy of the model, but are also more computationally expensive.

[0091] max_features: The number of features to consider when finding the best split.

[0092] max_depth: Maximum depth of the tree. Deep trees may lead to overfitting.

[0093] min_samples_split: The minimum number of samples required to split an internal node.

[0094] min_samples_leaf: The minimum number of samples required for a leaf node.

[0095] 4. Training Model

[0096] (1) Data partitioning: Divide the data into training set and test set.

[0097] (2) Model training: Train the random forest model on the training set.

[0098] (3) Hyperparameter tuning: Use grid search (GridSearchCV) or random search (RandomizedSearchCV) to optimize the parameters of random forest.

[0099] 5. Model evaluation and validation:

[0100] (1) Cross-validation: Use cross-validation to evaluate the stability and predictive ability of the model.

[0101] (2) Performance indicators: The accuracy, recall, F1 score, and confusion matrix are calculated to analyze the prediction accuracy of the probability prediction network for the target daily capacity distribution interval.

[0102] Step S203, extracting features of tasks to be performed from the maintenance tasks to be performed;

[0103] Step S204, sorting the tasks to be executed according to the acquired duration and priority of the maintenance tasks to be executed, and obtaining a task sorting result;

[0104] Step S205, using a Monte Carlo simulation method, combining the predicted probability of the daily capacity distribution interval and the task sorting result, to determine the best scheduling time;

[0105] It should be noted that the Monte Carlo simulation method is a computational method that uses random sampling to solve complex problems and conduct risk analysis. It estimates the behavior of the system by simulating a large number of random events, and is particularly suitable for problems that are difficult to solve using analytical methods.

[0106] In the embodiment of the present application, the specific steps of implementing the joint probability calculation using the Monte Carlo simulation method include:

[0107] (1) Define scenarios: Based on the monthly daily production capacity forecast distribution range, construct multiple possible future production scenarios (such as different combinations of daily production volume).

[0108] (2) Random sampling: Based on the predicted probability of the daily capacity distribution range, random sampling is used to generate a set of production scenarios.

[0109] (3) Simulation operation: For each sampled production scenario, calculate the frequency of production being in the low range within a specific continuous time period (equal to the duration of each maintenance task).

[0110] (4) Joint probability estimation: Through a large number of sampling (such as thousands of times), the joint probability distribution of each time period is obtained, and the time period with the largest joint probability is selected as the optimal time to schedule the task.

[0111] Step S206: input the characteristics of the tasks to be executed and the optimal scheduling time into a target task scheduling optimization network to obtain the optimal scheduling strategy for the tasks to be executed.

[0112] In the specific implementation, the steps of building the reinforcement learning model for the task scheduling optimization process include:

[0113] (1) Status definition: Define the environment status, including information such as the current production plan, scheduled maintenance tasks, and remaining tasks.

[0114] (2) Action space definition: Define possible actions, such as selecting the next maintenance task to be scheduled and its time period.

[0115] (3) Reward mechanism definition: Set up a reward mechanism, for example, schedule tasks during low production time periods to get higher rewards, and avoid scheduling maintenance during peak production periods to get negative rewards.

[0116] (4) Model selection: Use Deep Q-Network (DQN) or policy gradient-based methods (such as PPO or A3C) to optimize the strategy through repeated training so that the task layout maximizes the overall benefit.

[0117] (5) Training process: By simulating multiple task scheduling processes, the policy function of the reinforcement learning model is updated, and the optimal task scheduling strategy is gradually approached. Then, the policy function under the optimal task scheduling strategy with the maximum overall benefit is used as the target policy function, and the target policy function is used to construct the target task scheduling optimization network.

[0118] (6) Model evaluation and optimization:

[0119] Model evaluation: Evaluate the performance of the model on historical data to examine the prediction accuracy and rationality of task arrangement.

[0120] Model optimization: Based on the evaluation results, adjust the sampling strategy of the Monte Carlo simulation and the hyperparameters of the reinforcement learning model to improve model performance.

[0121] The embodiment of the present invention discloses a collaborative method for production planning and maintenance tasks, including: inputting the acquired time data to be tested and the monthly production plan to be tested into the target monthly daily capacity distribution interval prediction network to obtain the monthly daily capacity prediction distribution interval; inputting the monthly daily capacity prediction distribution interval and historical monthly data into the target daily capacity distribution interval probability prediction network to obtain the daily capacity distribution interval prediction probability; extracting the features of the tasks to be executed from the maintenance tasks to be executed; sorting the tasks to be executed according to the duration and priority of the maintenance tasks to be executed to obtain the task sorting result; inputting the daily capacity distribution interval prediction probability, the features of the tasks to be executed and the task sorting result into the target task arrangement optimization network to obtain the optimal task arrangement strategy to be executed. The maintenance plan is automatically arranged through the monthly daily capacity distribution interval prediction and the daily capacity distribution interval probability prediction. It is suitable for the formulation of production plans and maintenance tasks for professional coal transportation terminals. Through the collaborative operation of production plans and maintenance tasks, equipment maintenance can be arranged with minimal impact on production capacity, and the additional damage to the equipment caused by the continuous operation time is minimized.

[0122] Embodiment 3

[0123] See also Figure 3 , Figure 3 The structure block diagram of a production planning and maintenance task coordination device of the present invention includes:

[0124] The monthly daily capacity prediction module 301 is used to input the acquired time data to be tested and the monthly production plan to be tested into the target monthly daily capacity distribution interval prediction network to obtain the monthly daily capacity prediction distribution interval;

[0125] The daily capacity distribution interval prediction module 302 is used to input the monthly daily capacity prediction distribution interval and historical monthly data into the target daily capacity distribution interval probability prediction network to obtain the daily capacity distribution interval prediction probability;

[0126] The task feature extraction module 303 is used to extract the features of the tasks to be performed from the maintenance tasks to be performed;

[0127] The task sorting result determination module 304 is used to sort the tasks to be performed according to the duration and priority of the maintenance tasks to be performed, and obtain the task sorting result;

[0128] The scheduling strategy determination module 305 is used to input the daily capacity distribution interval prediction probability, the characteristics of the tasks to be executed and the task sorting results into the target task scheduling optimization network to obtain the optimal scheduling strategy for the tasks to be executed.

[0129] In an optional embodiment, the scheduling strategy determination module includes:

[0130] An optimal scheduling time determination submodule is used to determine the optimal scheduling time by using a Monte Carlo simulation method in combination with the predicted probability of the daily capacity distribution interval and the task sorting result;

[0131] The scheduling strategy determination submodule is used to input the characteristics of the tasks to be executed and the optimal scheduling time into the target task scheduling optimization network to obtain the optimal scheduling strategy for the tasks to be executed.

[0132] In an optional embodiment, the optimal scheduling time determination submodule includes:

[0133] A random sampling unit, used for randomly sampling a set of generation scenarios constructed based on the predicted probability of the daily capacity distribution interval;

[0134] A calculation unit, used for calculating the frequency of the production volume being in a low range within the corresponding duration of each of the generation volume scene sets;

[0135] The optimal scheduling time determination unit is used to combine the distribution results of the frequencies and select the time period with the maximum joint probability as the optimal scheduling time.

[0136] In an optional embodiment, the target monthly daily capacity distribution interval prediction network construction device includes:

[0137] A historical data acquisition module is used to acquire historical monthly daily capacity distribution data; the historical monthly daily capacity distribution data includes: historical monthly data, historical monthly production plan and historical monthly daily capacity range;

[0138] The monthly daily capacity distribution interval prediction network training module is used to train the constructed monthly daily capacity distribution interval prediction network with the historical monthly daily capacity interval as a label, the historical monthly data and the historical monthly production plan as input, and obtain the target monthly daily capacity distribution interval prediction network.

[0139] In an optional embodiment, the construction module of the target daily capacity distribution interval probability prediction network includes:

[0140] An input module, used for inputting the historical monthly data and the historical monthly production plan into the target monthly daily capacity distribution interval prediction network to obtain the historical daily capacity prediction distribution interval;

[0141] The daily capacity distribution interval probability prediction network training module is used to train the constructed daily capacity distribution interval probability prediction network with the historical daily capacity distribution interval as a label, the historical ship planning data and the historical month data as input, and obtain the target daily capacity distribution interval probability prediction network.

[0142] In an optional embodiment, the target task scheduling optimization network construction device includes:

[0143] A task scheduling optimization network training module is used to train the constructed task scheduling optimization network based on the characteristics of the tasks to be performed and the optimal scheduling time by defining the environment state, action space and reward mechanism, so as to gradually achieve the maximum overall benefit;

[0144] The target task scheduling optimization network generation module is used to use the policy function under the maximum overall benefit as the target policy function, and use the target policy function to construct the target task scheduling optimization network.

[0145] Embodiment 4

[0146] An embodiment of the present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of a method for coordinating production planning and maintenance tasks in any embodiment.

[0147] Embodiment 5

[0148] An embodiment of the present invention further provides a computer storage medium having a computer program stored thereon, wherein when the computer program is executed by the processor, the steps of a method for coordinating production planning and maintenance tasks of any embodiment are implemented.

[0149] Embodiment 6

[0150] An embodiment of the present invention further provides a computer program product having a computer program stored thereon, wherein when the computer program is executed by the processor, the steps of a method for coordinating production planning and maintenance tasks of any embodiment are implemented.

[0151] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0152] In the several embodiments provided in the present application, it should be understood that the methods, devices, electronic devices and storage media disclosed in the present invention can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0153] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0154] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0155] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a readable storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned readable storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0156] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for coordinating production planning and maintenance tasks, characterized in that: include: Input the acquired time data to be tested and the monthly production plan to be tested into the target monthly daily capacity distribution interval prediction network to obtain the monthly daily capacity prediction distribution interval; Input the monthly daily capacity forecast distribution interval and historical monthly data into the target daily capacity distribution interval probability forecast network to obtain the daily capacity distribution interval forecast probability; Extracting features of tasks to be performed from the maintenance tasks to be performed; Sorting the tasks to be performed according to the duration and priority of the maintenance tasks to be performed, and obtaining a task sorting result; The predicted probability of daily capacity distribution interval, the characteristics of the tasks to be executed and the task sorting results are input into the target task scheduling optimization network to obtain the optimal scheduling strategy for the tasks to be executed.

2. The method for coordinating production planning and maintenance tasks according to claim 1, characterized in that: The predicted probability of the daily capacity distribution interval, the characteristics of the tasks to be executed and the task sorting results are input into the target task scheduling optimization network to obtain the optimal scheduling strategy for the tasks to be executed, including: Using a Monte Carlo simulation method, combined with the predicted probability of the daily capacity distribution interval and the task sorting result, the optimal scheduling time is determined; The characteristics of the tasks to be executed and the optimal scheduling time are input into a target task scheduling optimization network to obtain the optimal scheduling strategy for the tasks to be executed.

3. The method for coordinating production planning and maintenance tasks according to claim 2, characterized in that: Using the Monte Carlo simulation method, combined with the predicted probability of the daily capacity distribution interval and the task sorting result, the optimal scheduling time is determined, including: Randomly sampling a set of generation scenarios constructed based on the predicted probability of the daily capacity distribution interval; Calculate the frequency of the production volume being in a low range within the corresponding duration for each of the generation volume scenario sets; The distribution results of the frequencies are combined, and the time period with the largest joint probability is selected as the optimal scheduling time.

4. The method for coordinating production planning and maintenance tasks according to claim 1, characterized in that: The steps of constructing the target monthly daily capacity distribution interval prediction network include: Obtaining historical monthly daily capacity distribution data; the historical monthly daily capacity distribution data includes: historical monthly data, historical monthly production plan and historical monthly daily capacity range; With the historical monthly daily capacity interval as a label, the historical monthly data and the historical monthly production plan as input, the constructed monthly daily capacity distribution interval prediction network is trained to obtain a target monthly daily capacity distribution interval prediction network.

5. The method for coordinating production planning and maintenance tasks according to claim 4, characterized in that: The steps of constructing the target daily capacity distribution interval probability prediction network include: Inputting the historical monthly data and the historical monthly production plan into the target monthly daily capacity distribution interval prediction network to obtain the historical daily capacity prediction distribution interval; Taking the historical daily capacity forecast distribution interval as a label, the historical ship plan data and the historical month data as input, the constructed daily capacity distribution interval probability forecasting network is trained to obtain the target daily capacity distribution interval probability forecasting network.

6. The method for coordinating production planning and maintenance tasks according to claim 2, characterized in that: The steps of constructing the target task arrangement optimization network include: By defining the environment state, action space and reward mechanism, the constructed task scheduling optimization network is trained based on the characteristics of the tasks to be performed and the optimal scheduling time, so as to gradually achieve the maximum overall benefit; The policy function under the maximum overall benefit is used as the target policy function, and the target task scheduling optimization network is constructed using the target policy function.

7. A collaborative device for production planning and maintenance tasks, characterized in that: include: The monthly daily capacity prediction module is used to input the acquired time data to be tested and the monthly production plan to be tested into the target monthly daily capacity distribution interval prediction network to obtain the monthly daily capacity prediction distribution interval; A daily capacity distribution interval prediction module, which is used to input the monthly daily capacity prediction distribution interval and historical monthly data into the target daily capacity distribution interval probability prediction network to obtain the daily capacity distribution interval prediction probability; A task feature extraction module is used to extract features of tasks to be performed from maintenance tasks to be performed; A task sorting result determination module is used to sort the tasks to be performed according to the duration and priority of the maintenance tasks to be performed, and obtain the task sorting result; The scheduling strategy determination module is used to input the predicted probability of the daily capacity distribution interval, the characteristics of the tasks to be executed and the task sorting results into the target task scheduling optimization network to obtain the optimal scheduling strategy for the tasks to be executed.

8. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the method according to any one of claims 1 to 6 is executed.

9. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is performed.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images