Full production process visual intelligent management and control method and system based on data analysis

Abstract

The invention relates to the technical field of intelligent management and control, and discloses a production whole process visual intelligent management and control method and system based on data analysis, and the method comprises the following steps: collecting real-time data and historical data of each production link; constructing a chaos prediction model and predicting a future product demand fluctuation condition; solving an optimal order replenishment amount and a corresponding replenishment time point by using a chaos optimization algorithm based on probability improvement; the storage condition of the product is automatically adjusted, and distribution management is carried out according to a set distribution path; and displaying product demand fluctuation prediction, a supply optimization result and a material distribution process in real time. According to the method, the chaotic characteristic of the production time sequence data can be fully considered, the prediction accuracy is improved, the warehousing cost and the service level can be effectively balanced, the complexity and uncertainty of warehousing management are reduced, and therefore the efficiency of warehousing management is improved, the cost is reduced, and the competitiveness of enterprises is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control, and specifically, to a method and system for visual intelligent control of the entire production process based on data analysis. Background Art

[0002] Traditional warehousing management and replenishment decisions mainly rely on manual experience and lack scientific data support, which often leads to excessive warehousing costs or insufficient service levels. In addition, uncertain factors in the production process, such as market demand fluctuations and supply chain disruptions, make warehousing management more complex. Under the traditional management mode, warehousing management and replenishment decisions mainly rely on manual experience, and this method has great limitations. First, manual experience is often affected by subjective factors, which easily leads to decision-making biases. Second, manual experience cannot process a large amount of data and is difficult to discover the patterns and trends in the data. Therefore, the traditional way of warehousing management and replenishment decisions often cannot achieve the best balance between warehousing costs and service levels.

[0003] At the same time, uncertain factors in the production process also pose greater challenges to warehousing management. Factors such as market demand fluctuations and supply chain disruptions make warehousing management more complex. In the case of market demand fluctuations, if warehousing management cannot be adjusted in a timely manner, it may lead to inventory backlogs or shortages. In the case of supply chain disruptions, if warehousing management cannot respond in a timely manner, it may lead to production interruptions or order delays. Therefore, the present invention proposes a method and system for visual intelligent control of the entire production process based on data analysis to improve the efficiency and accuracy of warehousing management. Through data analysis, the patterns and trends in the data can be mined to provide scientific support for warehousing management and replenishment decisions. At the same time, the intelligent control method can automatically adjust inventory and replenishment strategies to cope with the uncertain factors in the production process and achieve the best balance between warehousing costs and service levels. Summary of the Invention

[0004] In view of the problems in the related art, the present invention provides a method and system for visual intelligent control of the entire production process based on data analysis to overcome the above-mentioned technical problems existing in the existing related technologies.

[0005] To this end, the specific technical solutions adopted by the present invention are as follows:

[0006] According to one aspect of the present invention, there is provided a method for visual intelligent control of the entire production process based on data analysis, including the following steps:

[0007] S1. Collect real-time data and historical data of each production link, and preprocess the real-time data and historical data;

[0008] S2. Build a chaotic prediction model based on the preprocessed historical data, and combine the preprocessed real-time data to predict the future product demand fluctuations;

[0009] S3. According to the predicted future product demand fluctuations, use the chaos optimization algorithm based on probability improvement to solve the optimal order replenishment quantity and the corresponding replenishment time point;

[0010] S4. Automatically adjust the storage status of the product based on the optimal order replenishment quantity and the corresponding replenishment time point, and conduct distribution management according to the distribution path set by the order priority and delivery time;

[0011] S5. Use charts and dynamic interfaces to display the product demand fluctuation prediction, replenishment optimization results, and material distribution process in real time.

[0012] Preferably, the building of the chaotic prediction model based on the preprocessed historical data and combining the preprocessed real-time data to predict the future product demand fluctuations includes the following steps:

[0013] S21. Use the chaos detection method to identify the chaotic characteristics of the production time series data in the preprocessed historical data, calculate the fractal dimension of the production time series data, and evaluate the chaos degree of the production time series data;

[0014] S22. Based on the chaos characteristic detection results, select a chaos prediction algorithm to build a chaotic prediction model, and use the historical data to train the chaotic prediction model to obtain the trained chaotic prediction model;

[0015] S23. Collect real-time production time series data and preprocess it, and use the trained chaotic prediction model to combine the preprocessed real-time production time series data to predict the product demand fluctuations within a preset future time period.

[0016] Preferably, the use of the chaos detection method to identify the chaotic characteristics of the production time series data in the preprocessed historical data, calculate the fractal dimension of the production time series data, and evaluate the chaos degree of the production time series data includes the following steps:

[0017] S211. Based on the production time series data in the preprocessed historical data, draw the delay coordinate diagram and the Poincaré section diagram respectively, and analyze the chaos pattern according to the delay coordinate diagram and the Poincaré section diagram;

[0018] S212. Calculate the largest Lyapunov exponent and the fractal dimension of the production time series data in the historical data respectively, and comprehensively evaluate the chaos degree of the production time series data based on the results of the delay coordinate diagram, the Poincaré section, the largest Lyapunov exponent, and the fractal dimension.

[0019] Preferably, the calculation formula of the maximum Lyapunov exponent is as follows:

[0020]

[0021] The calculation formula of the fractal dimension is as follows:

[0022]

[0023] In the formula, λ represents the maximum Lyapunov exponent; ||δ(t)|| represents the distance between two very close trajectories at time t; ||δ(0)|| represents the distance between two very close trajectories at the initial time; D represents the fractal dimension; σ 2 represents the variance of the time series data; Δ represents the time interval.

[0024] Preferably, the steps of solving the optimal order replenishment quantity and the corresponding replenishment time point by using the chaos optimization algorithm based on probability improvement according to the predicted future product demand fluctuation are as follows:

[0025] S31. Respectively set the iteration flag, fine search flag of the chaotic variable, initialize the chaotic variable and the initial boundary of the search space;

[0026] S32. Determine the search space according to the predicted future product demand fluctuation, and map the chaotic variable to the search space of the current order replenishment quantity and replenishment time point by using probability p and 1 - p;

[0027] S33. For each mapped chaotic variable, calculate the cost of the order replenishment quantity and the replenishment time point, and update the current optimal solution according to the calculation result; at the same time, update the chaotic variable;

[0028] S34. Repeat S32 - S33 until the preset number of times is reached;

[0029] S35. Update the search space according to the current optimization result, and determine the search space boundary based on the updated search space;

[0030] S36. Return to S32 until the predetermined cost target or the number of iterations is reached, and output the optimal order replenishment quantity and the corresponding replenishment time point according to the final iteration result.

[0031] Preferably, the calculation formula of the cost of the order replenishment quantity and the replenishment time point is as follows:

[0032] f(x) = C h ×Q + C o ×T + C s ×(X - Q) + C t ×Q×d - R×Q×(1 - S)

[0033] In the formula, C h represents the annual holding cost per unit product; Q represents the order replenishment quantity; C o represents the fixed cost per order; T represents the replenishment times; C s represents the shortage cost per unit product; X represents the demand; C t represents the transportation cost per unit product; d represents the transportation distance; R represents the selling price per unit product; S represents the discount rate when the market demand is not met.

[0034] Preferably, the update formula of the chaotic variable is:

[0035]

[0036] In the formula, represents the chaotic variable of the i-th dimension after the (k + 1)-th iteration; represents the chaotic variable of the i-th dimension after the k-th iteration.

[0037] Preferably, the update formula of the search space is:

[0038]

[0039] In the formula, represents the left boundary of the i-th dimension search space after the (r + 1)-th iteration; represents the left boundary of the i-th dimension search space after the r-th iteration; represents the current optimal solution of the i-th dimension; r represents the fine search flag; represents the right boundary of the i-th dimension search space after the r-th iteration; represents the right boundary of the i-th dimension search space after the (r + 1)-th iteration.

[0040] Preferably, the formula for determining the search space boundary is:

[0041]

[0042] In the formula, represents based on the left boundary of the i-th dimension search space after the (r + 1)-th iteration determined; represents based on the right boundary of the i-th dimension search space after the (r + 1)-th iteration determined.

[0043] According to another embodiment of the present invention, a visual intelligent control system for the entire production process based on data analysis is provided. The visual intelligent control system for the entire production process based on data analysis includes a data acquisition and processing module, a product demand fluctuation prediction module, an optimal order strategy solving module, a storage condition adjustment module, and a data visualization module;

[0044] Among them, the data acquisition and processing module is used to collect real-time data and historical data of each production link, and preprocess the real-time data and historical data;

[0045] The product demand fluctuation prediction module is used to construct a chaos prediction model based on the preprocessed historical data, and predict the future product demand fluctuation situation in combination with the preprocessed real-time data;

[0046] The optimal order strategy solving module is used to solve the optimal order replenishment quantity and the corresponding replenishment time point according to the predicted future product demand fluctuation situation by using a chaos optimization algorithm improved based on probability;

[0047] The storage condition adjustment module is used to automatically adjust the storage condition of products based on the optimal order replenishment quantity and the corresponding replenishment time point, and perform distribution management according to the distribution path set by the order priority and the delivery time;

[0048] The data visualization module is used to display the product demand fluctuation prediction, the replenishment optimization result, and the material distribution process in real time by means of charts and dynamic interfaces.

[0049] The beneficial effects of the present invention are as follows:

[0050] 1) The present invention can not only predict the future product demand fluctuation situation by using the chaos prediction model, fully consider the chaos characteristics of production time series data, improve the prediction accuracy, but also solve the optimal order replenishment quantity and the corresponding replenishment time point by using a chaos optimization algorithm improved based on probability, effectively balancing the warehousing cost and service level, reducing the complexity and uncertainty of warehousing management, thereby helping to improve the efficiency of warehousing management, reduce costs, and enhance the competitiveness of enterprises.

[0051] 2) The present invention can not only automatically adjust the storage condition of products based on the optimal order replenishment quantity and the corresponding replenishment time point, perform distribution management according to the distribution path set by the order priority and the delivery time, improve the automation level of warehousing management, reduce the possibility of human errors, but also display the product demand fluctuation prediction, the replenishment optimization result, and the material distribution process in real time, enabling managers to intuitively understand the status of the entire production process, improving the management efficiency and response speed, and enabling managers to make quick and wise decisions. Description of the Drawings

[0052] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0053] Figure 1 is a flowchart of a production full-process visualization intelligent control method based on data analysis according to an embodiment of the present invention. Detailed embodiments

[0054] To further illustrate the embodiments, the present invention provides drawings. These drawings are part of the disclosure of the present invention. They are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, those of ordinary skill in the art should be able to understand other possible implementation manners and the advantages of the present invention. The components in the drawings are not drawn to scale, and similar component symbols are usually used to represent similar components.

[0055] According to an embodiment of the present invention, there is provided a production full-process visualization intelligent control method and system based on data analysis.

[0056] Now, the present invention will be further described in conjunction with the drawings and specific embodiments. As Figure 1 shown, according to an embodiment of the present invention, there is provided a production full-process visualization intelligent control method based on data analysis, including the following steps:

[0057] S1. Collect real-time data and historical data of each production link, and preprocess the real-time data and historical data;

[0058] Specifically, collecting real-time data and historical data of each production link and preprocessing the real-time data and historical data are key steps in realizing intelligent management in industrial production. The following is a detailed explanation of this process:

[0059] 1) Data collection:

[0060] Real-time data collection: Through sensors, monitoring devices, and industrial control systems installed on the production line, various indicators in the production process are monitored in real time, such as the operating status of machines, quality parameters of products, energy consumption, etc.

[0061] Historical data collection: Extract historical production data from the enterprise's database, historical records, and log files, including production batch records, equipment maintenance records, fault reports, etc.

[0062] 2) Data preprocessing:

[0063] Data cleaning: Identify and correct errors, outliers, and missing values in the data to ensure data accuracy.

[0064] Data integration: Convert data from different sources and formats into a unified format for subsequent analysis.

[0065] Data transformation: Convert non-numeric data into numeric data, such as converting text-described faults into corresponding codes.

[0066] Data normalization / standardization: Adjust the scale of the data to make it comparable, which is crucial for subsequent analysis and model building.

[0067] Feature extraction: Extract features from the original data that are helpful for analysis, such as trends and periodicity of time series data.

[0068] Data dimensionality reduction: Reduce the data dimension through methods such as principal component analysis (PCA), remove redundant information, and improve analysis efficiency.

[0069] Through these steps, the data in the production process is converted into a format that can be used for further analysis and modeling, providing the necessary data foundation for building chaos prediction models, optimization algorithms, and other intelligent control tools. These preprocessing steps not only improve the quality of the data but also reduce the impact of noise and irrelevant information, thereby improving the accuracy and efficiency of subsequent analysis.

[0070] S2. Build a chaos prediction model based on the preprocessed historical data and predict future product demand fluctuations in combination with the preprocessed real-time data;

[0071] Among them, building a chaos prediction model based on the preprocessed historical data and predicting future product demand fluctuations in combination with the preprocessed real-time data includes the following steps:

[0072] S21. Use the chaos detection method to identify the chaotic characteristics of the production time series data in the preprocessed historical data, calculate the fractal dimension of the production time series data, and evaluate the chaos degree of the production time series data;

[0073] Specifically, using the chaos detection method to identify the chaotic characteristics of the production time series data in the preprocessed historical data, calculate the fractal dimension of the production time series data, and evaluate the chaos degree of the production time series data includes the following steps:

[0074] S211. Based on the production time series data in the preprocessed historical data, draw delay coordinate diagrams and Poincaré section diagrams respectively, and analyze the chaotic patterns according to the delay coordinate diagrams and Poincaré section diagrams;

[0075] Specifically, select an appropriate delay time, plot the delay coordinate graph, observe the distribution of data points in the delay coordinate graph, and search for possible chaotic patterns such as strange attractors; select an appropriate embedding dimension, plot the Poincaré section graph, and observe the point distribution in the Poincaré section graph to further identify chaotic patterns.

[0076] Plotting the delay coordinate graph and the Poincaré section graph are both based on time series data, but they differ in the way of data processing and presentation:

[0077] 1) Delay coordinate graph:

[0078] The delay coordinate graph is plotted based on a single time series data. When plotting the delay coordinate graph, select an appropriate delay time (or multiple delay times), and pair the values of the time series data at each time point with its values after the delay time to form a two-dimensional (or higher-dimensional) coordinate point. These coordinate points are plotted on the graph to help observe the evolution of the data over time and possible chaotic patterns.

[0079] 2) Poincaré section graph:

[0080] The Poincaré section graph is plotted based on multiple dimensions of the embedding space. When plotting the Poincaré section graph, first select an appropriate embedding dimension, which is usually determined by an embedding theorem (such as Takens' theorem). Then map the time series data into this embedding dimension to form a trajectory in the multi-dimensional space. The Poincaré section graph shows the dynamic behavior of the data by selecting a section (usually two-dimensional) in this multi-dimensional space and plotting the projection points of the trajectory on this section.

[0081] In practical applications, both the delay coordinate graph and the Poincaré section graph are important tools for analyzing the chaotic characteristics of time series data. They can help identify chaotic patterns, determine the embedding dimension, and provide guidance for constructing a chaotic prediction model. Generally, the delay coordinate graph is used for preliminary observation of chaotic characteristics, while the Poincaré section graph is used for more in-depth analysis and understanding of the dynamic behavior of the system.

[0082] S212. Calculate the largest Lyapunov exponent and the fractal dimension of the production time series data in the historical data respectively, and evaluate the chaotic degree of the production time series data by synthesizing the results of the delay coordinate graph, the Poincaré section, the largest Lyapunov exponent and the fractal dimension.

[0083] Specifically, the maximum Lyapunov exponent (LLE) is an important indicator to measure the degree of chaos of a system. A positive maximum Lyapunov exponent means that the system is chaotic. A commonly used method to calculate the maximum Lyapunov exponent is the Rosenstein algorithm, which is based on the local linearization estimation of time series data, and the calculation formula is:

[0084]

[0085] The calculation formula for the fractal dimension is:

[0086]

[0087] In the formula, λ represents the maximum Lyapunov exponent; ||δ(t)|| represents the distance between two very close trajectories at time t; ||δ(0)|| represents the distance between two very close trajectories at the initial time; D represents the fractal dimension; σ 2 represents the variance of the time series data; Δ represents the time interval.

[0088] Combining the results of the delay coordinate plot, Poincaré section, maximum Lyapunov exponent, and fractal dimension, when evaluating the degree of chaos of the production time series data, it includes: If the delay coordinate plot, Poincaré section plot, maximum Lyapunov exponent, and fractal dimension all indicate that the system has chaotic characteristics, then it can be considered that the production time series data is chaotic.

[0089] S22. Based on the chaos characteristic detection results, select a chaos prediction algorithm to construct a chaos prediction model, and use historical data to train the chaos prediction model to obtain a trained chaos prediction model;

[0090] Specifically, based on the chaos characteristic detection results, the specific steps to select a chaos prediction algorithm to construct a chaos prediction model and use historical data to train the chaos prediction model to obtain a trained chaos prediction model are as follows:

[0091] 1) Selection of chaos prediction algorithm: According to the chaos characteristic detection results, select a suitable chaos prediction algorithm, such as prediction based on the Lyapunov exponent, prediction based on phase space reconstruction, etc.

[0092] 2) Model construction: According to the selected algorithm, construct a chaos prediction model. This includes determining the parameters of the model, selecting a suitable model structure, etc.

[0093] 3) Preparation of historical data: Collect and organize the historical data for model training to ensure the quality and representativeness of the data.

[0094] 4) Model training: Use historical data to train the constructed chaotic prediction model. By comparing the predicted output of the model with the actual data, adjust the model parameters to improve the prediction accuracy.

[0095] 5) Model validation: During the training process, use a part of the historical data to validate the model to ensure that the model has good generalization ability.

[0096] 6) Model optimization: According to the validation results, optimize the model, such as adjusting parameters, improving algorithms, etc., to improve the prediction performance of the model.

[0097] 7) Model evaluation: Use evaluation metrics (such as mean squared error, coefficient of determination, etc.) to evaluate the trained model to ensure that it meets the prediction accuracy requirements.

[0098] 8) Model storage and update: Store the trained model in the system for subsequent use. Regularly update the model with newly collected data to adapt to possible changes in the production process.

[0099] Through these steps, a chaotic prediction model for specific production time series data can be obtained. This model can predict the future product demand fluctuations based on historical data, providing a scientific basis for warehouse management.

[0100] S23. Collect real-time production time series data, preprocess it, and use the trained chaotic prediction model to combine with the preprocessed real-time production time series data to predict the product demand fluctuations within a preset future time period.

[0101] Specifically, the specific steps of collecting real-time production time series data, preprocessing it, and using the trained chaotic prediction model to combine with the preprocessed real-time production time series data to predict the product demand fluctuations within a preset future time period are as follows:

[0102] 1) Real-time data collection: Real-time collect time series data during the production process through sensors, monitoring devices, and industrial control systems at the production site.

[0103] 2) Data preprocessing: Clean the collected real-time data to remove outliers and noise. Normalize or standardize the data to make it suitable for model input. Perform feature extraction to identify key factors affecting product demand fluctuations. If necessary, perform data dimensionality reduction to reduce the complexity of the model.

[0104] 3) Model input preparation: Convert the preprocessed real-time data into the input format required by the chaotic prediction model. Ensure the time series and integrity of the data so that the model can process it correctly.

[0105] 4) Model prediction: Use the trained chaotic prediction model to predict the preprocessed real-time data. Based on the chaotic characteristics in historical and real-time data, the model will predict the product demand fluctuations within a preset future time period.

[0106] 5) Analysis of prediction results: Analyze the prediction results of the model to identify potential product demand fluctuation trends and key time points. Interpret the prediction results to provide a basis for warehouse management decisions.

[0107] Through these steps, real-time monitoring and prediction of product demand fluctuations during the production process can be achieved, providing a scientific basis for the enterprise's warehouse management and decision-making, thereby improving the efficiency and response speed of warehouse management.

[0108] S3. According to the predicted future product demand fluctuations, use the chaos optimization algorithm improved based on probability to solve for the optimal order replenishment quantity and the corresponding replenishment time point;

[0109] Among them, the step of using the chaos optimization algorithm improved based on probability to solve for the optimal order replenishment quantity and the corresponding replenishment time point according to the predicted future product demand fluctuations includes the following steps:

[0110] S31. Respectively set the iteration flag, fine search flag of the chaotic variable, initialize the chaotic variable and the initial boundaries of the search space;

[0111] S32. Determine the search space according to the predicted future product demand fluctuations, and use probability p and 1 - p to map the chaotic variable to the search space of the current order replenishment quantity and replenishment time point;

[0112] Specifically, the step of determining the search space according to the predicted future product demand fluctuations and using probability p and 1 - p to map the chaotic variable to the search space of the current order replenishment quantity and replenishment time point includes the following steps:

[0113] 1) Analyze the prediction results: Analyze the product demand fluctuation prediction results output by the chaotic prediction model, including the magnitude, period, and trend of the fluctuations, etc.

[0114] 2) Set the dimension of the search space: Determine the dimension of the search space. For example, if the optimization objectives include the order replenishment quantity and the replenishment time point, the search space is two-dimensional.

[0115] 3) Define the boundaries of the search space: Set the boundaries of the search space according to the predicted product demand fluctuations and actual business requirements. For example, the search range of the order replenishment quantity is based on the minimum and maximum inventory levels, and the search range of the replenishment time point is based on the response time of the supply chain and the order cycle.

[0116] 4) Map chaotic variables using probabilities p and 1 - p: Map chaotic variables to the search space using probabilities p and 1 - p. Chaotic variables are random or pseudo - random numbers generated from a chaotic system. They are ergodic and can explore various parts of the search space. Probabilities p and 1 - p are used to control the distribution of chaotic variables in the search space. p is usually a number close to 0.5 to ensure the uniformity and randomness of the search.

[0117] 5) Map to the current order replenishment quantity and replenishment time point: Through the mapping function, convert chaotic variables into the actual order replenishment quantity and replenishment time point. The mapping function ensures that chaotic variables are within the boundaries of the search space and are evenly distributed.

[0118] Through these steps, the chaotic optimization algorithm can search for the optimal order replenishment quantity and replenishment time point within a well - defined search space, thus achieving the optimization of warehouse management in the case of predicted future product demand fluctuations.

[0119] S33. For each mapped chaotic variable, calculate the cost of the order replenishment quantity and replenishment time point, and update the current optimal solution according to the calculation result; at the same time, update the chaotic variable;

[0120] Specifically, in the chaotic optimization algorithm, for each chaotic variable mapped to the search space, calculating the cost of the order replenishment quantity and replenishment time point is a key step in evaluating candidate solutions. The following are the detailed steps of this process:

[0121] 1) Cost calculation: For the order replenishment quantity and replenishment time point obtained by mapping each chaotic variable, calculate the cost according to the cost function. The cost function usually includes holding cost (C h ), ordering cost (C o ), shortage cost (C s ), transportation cost (C t ), discount rate (S) when demand is not met, and other factors.

[0122] The formula for calculating the cost of the order replenishment quantity and replenishment time point is:

[0123] f(x) = C h ×Q + C o ×T + C s ×(X - Q)+C t ×Q×d - R×Q×(1 - S)

[0124] In the formula, C h represents the annual holding cost per unit product; Q represents the order replenishment quantity; C o represents the fixed cost per order; T represents the number of replenishments; C s represents the shortage cost per unit product; X represents the demand; Ct Let \(C\) represent the transportation cost per unit product; \(d\) represent the transportation distance; \(R\) represent the selling price per unit product; and \(S\) represent the discount rate when market demand is not met.

[0125] 2) Cost function evaluation: Use the cost function to evaluate the total cost of the replenishment strategy obtained from the current chaotic variable mapping. The cost function is a multi-factor function that reflects various aspects of warehouse management.

[0126] 3) Update the optimal solution: If the cost obtained from the current chaotic variable mapping is lower than the currently known optimal cost, update the current optimal solution. Updating the optimal solution includes recording the new order replenishment quantity and replenishment time point, as well as the corresponding optimal cost.

[0127] 4) Update of chaotic variables: Update the chaotic variables according to the rules of the chaotic optimization algorithm. This usually involves using the evolution rules of the chaotic system, such as the Logistic map or other chaotic systems, to generate new chaotic variables.

[0128] The update formula for the chaotic variables is as follows:

[0129]

[0130] where represents the \(i\)-th dimensional chaotic variable after the \((k + 1)\)-th iteration; represents the \(i\)-th dimensional chaotic variable after the \(k\)-th iteration.

[0131] S34. Repeat S32 - S33 until the preset number of times is reached;

[0132] S35. Update the search space according to the current optimization result and determine the search space boundary based on the updated search space.

[0133] Specifically, the update formula for the search space is as follows:

[0134]

[0135] where represents the left boundary of the \(i\)-th dimensional search space after the \((r + 1)\)-th iteration; represents the left boundary of the \(i\)-th dimensional search space after the \(r\)-th iteration; represents the \(i\)-th dimensional current optimal solution; \(r\) represents the fine search flag; represents the right boundary of the \(i\)-th dimensional search space after the \(r\)-th iteration; represents the right boundary of the \(i\)-th dimensional search space after the \((r + 1)\)-th iteration.

[0136] The formula for determining the search space boundary is:

[0137]

[0138] In the formula, represents the left boundary of the i-th dimensional search space after the (r + 1)-th iteration determined based on ; represents the right boundary of the i-th dimensional search space after the (r + 1)-th iteration determined based on ;

[0139] S36. Return to S32 until a predetermined cost target or the number of iterations is reached, and according to the final iteration result, output the optimal order replenishment quantity and the corresponding replenishment time point;

[0140] Specifically, in the probabilistic chaos optimization algorithm, after reaching the predetermined cost target or the number of iterations, the specific steps for outputting the optimal order replenishment quantity and the corresponding replenishment time point are as follows:

[0141] 1) Check the termination condition: Check whether the predetermined cost target or the number of iterations is met. If it is met, proceed to the next step; if not, continue with the chaos optimization iteration.

[0142] 2) Determine the optimal solution: Find the solution with the lowest cost or utility among all iterations, that is, the optimal order replenishment quantity and the optimal replenishment time point. This usually involves comparing the f(x) values in all iterations and selecting the solution corresponding to the minimum value.

[0143] 3) Output the optimal solution: Output the optimal order replenishment quantity and output the corresponding replenishment time point.

[0144] Through these steps, the probabilistic chaos optimization algorithm can provide data-driven decision support, helping enterprises optimize warehouse management, reduce costs, and improve service levels. In addition, more business rules and constraints may need to be considered in this embodiment.

[0145] S4. Automatically adjust the storage status of products based on the optimal order replenishment quantity and the corresponding replenishment time point, and conduct distribution management according to the distribution path set based on the order priority and delivery time;

[0146] Specifically, the specific steps for automatically adjusting the storage status of products based on the optimal order replenishment quantity and the corresponding replenishment time point and conducting distribution management according to the distribution path set based on the order priority and delivery time are as follows:

[0147] 1) Adjust the inventory status: Adjust the material inventory level according to the optimal order replenishment quantity and the replenishment time point. Ensure that the inventory level meets market demand while avoiding excessive inventory.

[0148] 2) Order processing: Collect all pending orders. Sort the orders according to the order priority (such as customer importance, order amount, delivery time urgency, etc.).

[0149] 3) Delivery route planning: Use heuristic algorithms (such as genetic algorithms, ant colony algorithms, etc.) or optimization software (such as ERP systems) to plan the delivery route. Consider factors such as delivery time, transportation cost, route efficiency, etc., to minimize the total delivery cost or maximize the delivery efficiency.

[0150] 4) Material delivery: Start material delivery according to the delivery route planning. Ensure that the delivery is carried out according to the order priority and delivery time.

[0151] 5) Inventory monitoring and adjustment: Monitor the material inventory level in real time to ensure that it meets the delivery requirements. Adjust the order replenishment quantity and replenishment time point according to the actual delivery situation.

[0152] 6) Order fulfillment and tracking: Ensure that orders are fulfilled on time and update the order status in a timely manner. Provide real-time tracking information to customers to improve customer satisfaction.

[0153] 7) Data analysis and optimization: Analyze the delivery and warehousing management data to identify potential problems and improvement opportunities. Adjust the warehousing management strategy and delivery route planning according to the analysis results.

[0154] 8) Reporting and communication: Regularly generate inventory and delivery reports for management and stakeholders to understand the operation status. Maintain communication with internal teams and external suppliers and customers to ensure the smoothness of the process.

[0155] Through these steps, enterprises can effectively manage material inventory, optimize the delivery process, improve customer satisfaction, and reduce operating costs.

[0156] S5. Use charts and dynamic interfaces to display the product demand fluctuation prediction, replenishment optimization results, and material delivery process in real time.

[0157] According to another embodiment of the present invention, a visual intelligent control system for the entire production process based on data analysis is provided. The visual intelligent control system for the entire production process based on data analysis includes a data acquisition and processing module, a product demand fluctuation prediction module, an optimal ordering strategy solving module, a storage status adjustment module, and a data visualization module;

[0158] Among them, the data acquisition and processing module is used to collect real-time data and historical data of each production link and preprocess the real-time data and historical data;

[0159] The product demand fluctuation prediction module is used to construct a chaos prediction model based on the preprocessed historical data and predict the future product demand fluctuation situation in combination with the preprocessed real-time data;

[0160] The optimal order strategy solving module is used to solve the optimal order replenishment quantity and the corresponding replenishment time point according to the predicted future product demand fluctuation situation by using the chaos optimization algorithm improved based on probability;

[0161] The storage status adjustment module is used to automatically adjust the storage status of products based on the optimal order replenishment quantity and the corresponding replenishment time point, and conduct distribution management according to the distribution path set by the order priority and delivery time;

[0162] The data visualization module is used to display the product demand fluctuation prediction, replenishment optimization results and material distribution process in real time by means of charts and dynamic interfaces.

[0163] In summary, by means of the above technical solutions of the present invention, the present invention can not only predict the future product demand fluctuation situation by using the chaos prediction model, fully consider the chaos characteristics of production time series data, improve the prediction accuracy, but also solve the optimal order replenishment quantity and the corresponding replenishment time point based on the chaos optimization algorithm improved based on probability, effectively balance the warehousing cost and service level, reduce the complexity and uncertainty of warehousing management, thereby helping to improve the efficiency of warehousing management, reduce costs and enhance the competitiveness of enterprises. In addition, the present invention can not only automatically adjust the storage status of products based on the optimal order replenishment quantity and the corresponding replenishment time point, conduct distribution management according to the distribution path set by the order priority and delivery time, improve the automation level of warehousing management, reduce the possibility of human errors, but also display the product demand fluctuation prediction, replenishment optimization results and material distribution process in real time, enabling managers to intuitively understand the status of the entire production process, improve the management efficiency and response speed, and enabling managers to quickly make wise decisions.

[0164] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification. Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method can be completed by instructing relevant hardware through a program. The program can be stored in a computer-readable storage medium. When the program is executed, it includes the steps described in the above method. The storage medium, such as: ROM / RAM, magnetic disk, optical disk, etc.

[0165] The above-described embodiments merely represent several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all fall within the protection scope of the present invention. Therefore, the protection scope of the present invention patent shall be subject to the appended claims.

Claims

1. A visualization intelligent control method for the entire production process based on data analysis, characterized in that, The visualization intelligent control method for the entire production process based on data analysis includes the following steps: S1. Collect the real-time data and historical data of each production link, and preprocess the real-time data and historical data; S2. Construct a chaotic prediction model based on the preprocessed historical data, and combine the preprocessed real-time data to predict the future product demand fluctuation; S3. According to the predicted future product demand fluctuation, use the chaos optimization algorithm improved based on probability to solve the optimal order replenishment quantity and the corresponding replenishment time point; S4. Automatically adjust the storage status of products based on the optimal order replenishment quantity and the corresponding replenishment time point, and conduct distribution management according to the distribution path set by the order priority and delivery time; S5. Use charts and dynamic interfaces to display the product demand fluctuation prediction, replenishment optimization results, and material distribution process in real time.

2. The method for visual intelligent control of the entire production process based on data analysis according to claim 1, wherein The step of constructing a chaotic prediction model based on the preprocessed historical data and combining the preprocessed real-time data to predict the future product demand fluctuation includes the following steps: S21. Use the chaos detection method to identify the chaotic characteristics of the production time series data in the preprocessed historical data, calculate the fractal dimension of the production time series data, and evaluate the chaos degree of the production time series data; S22. Based on the chaos characteristic detection results, select a chaos prediction algorithm to construct a chaotic prediction model, and use the historical data to train the chaotic prediction model to obtain a trained chaotic prediction model; S23. Collect and preprocess the real-time production time series data, and use the trained chaotic prediction model to combine the preprocessed real-time production time series data to predict the product demand fluctuation within a preset future time period.

3. The production full-process visualization intelligent control method based on data analysis according to claim 2, wherein The step of using the chaos detection method to identify the chaotic characteristics of the production time series data in the preprocessed historical data, calculate the fractal dimension of the production time series data, and evaluate the chaos degree of the production time series data includes the following steps: S211. Based on the production time series data in the preprocessed historical data, draw the delay coordinate diagram and Poincaré section diagram respectively, and analyze the chaos pattern according to the delay coordinate diagram and Poincaré section diagram; S212. Calculate the maximum Lyapunov exponent and fractal dimension of the production time series data in the historical data respectively, and comprehensively evaluate the chaos degree of the production time series data based on the results of the delay coordinate diagram, Poincaré section, maximum Lyapunov exponent, and fractal dimension.

4. The method for visual intelligent control and management of the entire production process based on data analysis according to claim 3, wherein The calculation formula for the maximum Lyapunov exponent is: The calculation formula for the fractal dimension is: In the formula, λ represents the maximum Lyapunov exponent; ||δ(t)|| represents the distance between two very close trajectories at time t; ||δ(0)|| represents the distance between two very close trajectories at the initial time; D represents the fractal dimension; σ 2 represents the variance of time series data; Δ represents the time interval.

5. The production full-process visualization intelligent control method based on data analysis according to claim 1, characterized in that The step of using the chaos optimization algorithm improved based on probability to solve the optimal order replenishment quantity and the corresponding replenishment time point according to the predicted future product demand fluctuation includes the following steps: S31. Set the iteration flag, fine search flag of the chaotic variable respectively, initialize the chaotic variable and the initial boundaries of the search space; S32. Determine the search space according to the predicted future product demand fluctuation situation, and map the chaotic variable to the search space of the current order replenishment quantity and replenishment time point using probability p and 1 - p; S33. For each mapped chaotic variable, calculate the costs of the order replenishment quantity and the replenishment time point, and update the current optimal solution according to the calculation results; meanwhile, update the chaotic variable; S34. Repeat S32 - S33 until the preset number of times is reached; S35. Update the search space according to the current optimization result, and determine the search space boundary based on the updated search space; S36. Return to S32 until the predetermined cost target or the number of iterations is reached, and output the best order replenishment quantity and the corresponding replenishment time point according to the final iteration result.

6. The method for visual intelligent control of the entire production process based on data analysis according to claim 5, characterized in that The calculation formula for the costs of the order replenishment quantity and the replenishment time point is: f(x) = C h ×Q + C o ×T + C s ×(X - Q) + C t ×Q×d - R×Q×(1 - S) where C h represents the annual holding cost per unit product; Q represents the order replenishment quantity; C o represents the fixed cost per order; T represents the number of replenishments; C s represents the shortage cost of a unit product; X represents the demand; C t represents the transportation cost of a unit product; d represents the transportation distance; R represents the selling price per unit product; S represents the discount rate when the market demand is not met.

7. The method for visual intelligent control of the entire production process based on data analysis according to claim 5, wherein The update formula for the chaotic variable is: wherein, represents the chaotic variable of the i-th dimension after the (k + 1)-th iteration; Denote the chaotic variable of the $i$-th dimension after the $k$-th iteration.

8. The method for visual intelligent control of the entire production process based on data analysis according to claim 7, characterized in that The update formula for the search space is: wherein, represents the left boundary of the i-th dimensional search space after the (r + 1)-th iteration; Denote the left boundary of the $i$-th dimensional search space after the $r$-th iteration; Denote the current optimal solution of the i-th dimension; r represents the fine search flag; Denote the right boundary of the $i$-th dimensional search space after the $r$-th iteration; Denote the right boundary of the $i$-th dimensional search space after the $(r + 1)$-th iteration.

9. The method for visual intelligent control of the entire production process based on data analysis according to claim 8, wherein The formula for determining the search space boundary is: wherein, represents the left boundary of the $i$-th dimensional search space after the $(r + 1)$-th iteration determined based on ; the right boundary of the $i$-th dimensional search space after the $(r + 1)$-th iteration determined based on Indicates the right boundary of the i-th dimensional search space after the (r + 1)-th iteration determined based on ​ 10. A production full-process visualization intelligent control system based on data analysis is used to implement the steps of the production full-process visualization intelligent control method based on data analysis described in any one of claims 1-9, and is characterized in that, This production full - process visualization intelligent control system based on data analysis includes a data acquisition and processing module, a product demand fluctuation prediction module, an optimal order strategy solving module, a storage status adjustment module, and a data visualization module; Among them, the data acquisition and processing module is used to collect the real - time data and historical data of each production link, and pre - process the real - time data and historical data; The product demand fluctuation prediction module is used to construct a chaotic prediction model based on the pre - processed historical data, and predict the future product demand fluctuation situation in combination with the pre - processed real - time data; The optimal order strategy solving module is used to solve the best order replenishment quantity and the corresponding replenishment time point according to the predicted future product demand fluctuation situation by using a chaotic optimization algorithm improved based on probability; The storage status adjustment module is used to automatically adjust the product storage status based on the best order replenishment quantity and the corresponding replenishment time point, and conduct distribution management according to the distribution path set by the order priority and delivery time; The data visualization module is used to display the product demand fluctuation prediction, replenishment optimization result, and material distribution process in real time in the form of charts and dynamic interfaces.

Images