Search

CN-122022243-A - Textile production man-machine efficiency optimization method and system integrating multiple AI algorithms

CN122022243ACN 122022243 ACN122022243 ACN 122022243ACN-122022243-A

Abstract

The invention provides a textile production man-machine efficiency optimization method and system integrating multiple AI algorithms in the technical field of textile production management, wherein the method comprises the steps of S1, collecting core textile production data, S2, carrying out anomaly detection and filtration on the core textile production data through an anomaly detection model, S3, inputting the core textile production data into a productivity prediction model to obtain a textile productivity prediction result, generating a textile production scheme list based on the textile productivity prediction result and a production configuration rule, S4, inputting the textile production scheme list into a production optimization model to obtain an optimal production scheme, S5, executing the optimal production scheme, and generating a textile production report based on the statistic indexes of the core textile production data generated after execution. The invention has the advantages of realizing intelligent production scheduling, automatic report generation and data closed-loop management, thereby effectively improving the utilization rate of equipment, shortening the report generation time and opening a data barrier.

Inventors

  • LIN XING
  • LU YIJIAN

Assignees

  • 福州市数字产业互联科技有限责任公司

Dates

Publication Date
20260512
Application Date
20251218

Claims (10)

  1. 1. A textile production man-machine efficiency optimization method integrating multiple AI algorithms is characterized by comprising the following steps: Step S1, acquiring core textile production data comprising textile order data, textile equipment data and employee data in real time, and preprocessing each core textile production data; S2, performing anomaly detection and filtration on the preprocessed core textile production data through a pre-trained anomaly detection model; s3, inputting the filtered core textile production data into a pre-trained productivity prediction model to obtain a textile productivity prediction result, and generating a textile production scheme list based on the textile productivity prediction result and a preset production arrangement rule; S4, inputting the textile production scheduling scheme list into a pre-trained production scheduling optimization model to obtain an optimal production scheduling scheme; And S5, executing the optimal production scheduling scheme, calculating statistical indexes based on the core textile production data generated after execution, and generating and displaying a textile production report based on each statistical index through a Word template engine.
  2. 2. The method for optimizing the man-machine efficiency of textile production by fusing the multi-AI algorithm as set forth in claim 1, wherein the step S1 is specifically: The method comprises the steps of collecting core textile production data comprising textile order data, textile equipment data and employee data in real time, wherein the textile order data at least comprises customer order information and product BOM data, the customer order information at least comprises product types, production quantity and delivery date, and the product BOM data at least comprises cotton yarn consumption; and preprocessing at least including data cleaning, data de-duplication and feature extraction is performed on each core textile production data.
  3. 3. The method for optimizing the human-computer efficiency of textile production by fusing the multi-AI algorithm as set forth in claim 1, wherein in the step S2, the anomaly detection model is created based on an isolated forest algorithm for identifying anomaly data exceeding a preset configuration range, marking the anomaly data as special operation data, and filtering.
  4. 4. The method for optimizing the human-machine efficiency of textile production by fusing the multi-AI algorithm as set forth in claim 1, wherein in the step S3, the productivity prediction model is constructed based on an LSTM network and is used for outputting a textile productivity prediction result carrying the machine output of 24 hours in the future and the equipment failure probability according to the input machine rotation speed, the actual operation duration, the material loss rate and the staff operation efficiency; the productivity prediction model is trained based on historical data of the past six months; The textile production scheduling scheme list comprises a preset number of textile production scheduling schemes, and each textile production scheduling scheme at least comprises a machine allocation scheme, a material demand list, a production time sequence and an employee binding relationship.
  5. 5. The method for optimizing the efficiency of textile production by fusing the multi-AI algorithm as set forth in claim 1, wherein in the step S4, the production scheduling optimization model is constructed based on a genetic algorithm; The formula of the fitness function of the scheduling optimization model is as follows: F=W1×F1+W2×F2+W3×F3; f1 =Σ (actual delivery time-required delivery time)/total amount of orders; f2 =1- Σ (single machine operation time length-average operation time length) 2/total machine number; f3 Σ (available stock quantity-production demand quantity)/total material category; Wherein F represents fitness, F1 represents order delay rate, F2 represents machine load balancing degree, F3 represents material matching degree, and W1, W2 and W3 are weight coefficients with values of 40%, 30% and 30% respectively; initializing the textile scheduling scheme list into a population, calculating fitness based on the fitness function, selecting particles with preset proportion from the population based on the fitness through a roulette method, sequentially executing cross operation and mutation operation on each selected particle to generate a new population, judging whether a preset termination condition is met, if yes, outputting an optimal scheduling scheme, otherwise, continuing population iteration; The optimal production scheduling scheme is the optimal textile production scheduling scheme; The termination condition is that the iteration times are more than or equal to 100 times, or the fitness is less than 0.12; the step S5 specifically comprises the following steps: Executing the optimal production scheduling scheme, calculating statistical indexes at least comprising the total daily yield, the machine efficiency, the production homonymous difference data and the production cyclic ratio difference data based on the core textile production data generated after execution, generating trend graphs and optimization suggestions based on the statistical indexes through a Word template engine, and generating and displaying a textile production report based on the statistical indexes, the trend graphs and the optimization suggestions.
  6. 6. A textile production man-machine efficiency optimization system integrating multiple AI algorithms is characterized by comprising the following modules: the core textile production data acquisition module is used for acquiring core textile production data comprising textile order data, textile equipment data and employee data in real time and preprocessing each core textile production data; The abnormality detection and filtration module is used for carrying out abnormality detection and filtration on the preprocessed core textile production data through a pre-trained abnormality detection model; The textile production scheduling module is used for inputting the filtered core textile production data into a pre-trained productivity prediction model to obtain a textile productivity prediction result, and generating a textile production scheme list based on the textile productivity prediction result and a preset production scheduling configuration rule; The production scheduling optimization module is used for inputting the textile production scheduling scheme list into a pre-trained production scheduling optimization model to obtain an optimal production scheduling scheme; and the textile production report generation module is used for executing the optimal production scheduling scheme, calculating statistical indexes based on the core textile production data generated after execution, and generating and displaying a textile production report based on the statistical indexes through a Word template engine.
  7. 7. The textile production man-machine efficiency optimization system fusing multiple AI algorithms of claim 6, wherein said core textile production data acquisition module is specifically configured to: The method comprises the steps of collecting core textile production data comprising textile order data, textile equipment data and employee data in real time, wherein the textile order data at least comprises customer order information and product BOM data, the customer order information at least comprises product types, production quantity and delivery date, and the product BOM data at least comprises cotton yarn consumption; and preprocessing at least including data cleaning, data de-duplication and feature extraction is performed on each core textile production data.
  8. 8. The system for optimizing the efficiency of textile production man-machine integrating multiple AI algorithms as set forth in claim 6, wherein said anomaly detection model is created based on an isolated forest algorithm in said anomaly detection filter module for identifying anomaly data exceeding a predetermined configuration range, marking said anomaly data as special job data, and filtering.
  9. 9. The system for optimizing the efficiency of textile production man-machine integrating multiple AI algorithms as set forth in claim 6, wherein said productivity prediction model is constructed based on an LSTM network and is configured to output a textile productivity prediction result carrying a machine output of 24 hours in the future and a probability of a device failure according to an input machine rotation speed, an actual operation duration, a material loss rate, and an employee operation efficiency; the productivity prediction model is trained based on historical data of the past six months; The textile production scheduling scheme list comprises a preset number of textile production scheduling schemes, and each textile production scheduling scheme at least comprises a machine allocation scheme, a material demand list, a production time sequence and an employee binding relationship.
  10. 10. The system for optimizing the efficiency of textile production by fusing the multiple AI algorithms as set forth in claim 6, wherein said production scheduling optimization model is constructed based on a genetic algorithm; The formula of the fitness function of the scheduling optimization model is as follows: F=W1×F1+W2×F2+W3×F3; f1 =Σ (actual delivery time-required delivery time)/total amount of orders; f2 =1- Σ (single machine operation time length-average operation time length) 2/total machine number; f3 Σ (available stock quantity-production demand quantity)/total material category; Wherein F represents fitness, F1 represents order delay rate, F2 represents machine load balancing degree, F3 represents material matching degree, and W1, W2 and W3 are weight coefficients with values of 40%, 30% and 30% respectively; initializing the textile scheduling scheme list into a population, calculating fitness based on the fitness function, selecting particles with preset proportion from the population based on the fitness through a roulette method, sequentially executing cross operation and mutation operation on each selected particle to generate a new population, judging whether a preset termination condition is met, if yes, outputting an optimal scheduling scheme, otherwise, continuing population iteration; The optimal production scheduling scheme is the optimal textile production scheduling scheme; The termination condition is that the iteration times are more than or equal to 100 times, or the fitness is less than 0.12; The textile production report generation module is specifically used for: Executing the optimal production scheduling scheme, calculating statistical indexes at least comprising the total daily yield, the machine efficiency, the production homonymous difference data and the production cyclic ratio difference data based on the core textile production data generated after execution, generating trend graphs and optimization suggestions based on the statistical indexes through a Word template engine, and generating and displaying a textile production report based on the statistical indexes, the trend graphs and the optimization suggestions.

Description

Textile production man-machine efficiency optimization method and system integrating multiple AI algorithms Technical Field The invention relates to the technical field of textile production management, in particular to a textile production man-machine efficiency optimization method and system integrating multiple AI algorithms. Background As a traditional manufacturing field, the textile industry faces core management challenges such as order fragmentation, various equipment types, complicated production data and the like for a long time. Although the production management system in the prior art has basic data acquisition capability, the production management system still has obvious defects in key production scheduling and report generation links, and is specifically shown as follows: 1. The intelligent level of scheduling is insufficient, the current scheduling process mainly depends on manual experience, and the real-time running state (such as outage rate and running speed) of equipment and the dynamic change of material stock cannot be effectively integrated, so that phenomena such as idle high-yield equipment, core resources occupied by low-priority orders and the like frequently occur, unbalance of capacity configuration and delay of order delivery are caused, and customer satisfaction is further affected. 2. The report generation efficiency is low, the production report depends on manual screening data and is manually filled into a fixed template, the single month report takes up to 2 to 3 hours, errors are easily introduced due to manual operation, the accuracy of management decisions is easily affected, in addition, the report generation is seriously delayed (if the production daily report needs to be delayed to the next day to be completed), the analysis dimension is single, only basic data statistics can be provided, and real-time and dynamic production decisions are difficult to support. 3. The data value is not fully released, production traceability information, personnel efficiency data and the like accumulated in the system cannot be effectively linked with production scheduling and reporting links to form a data island, so that the construction of a closed-loop management mechanism of data acquisition, analysis decision-production execution is hindered, the excavation of the potential value of the data is limited, and the transformation and upgrading of enterprises to intelligent manufacturing are restricted. Therefore, how to provide a textile production man-machine efficiency optimization method and system integrating multiple AI algorithms, and achieve intelligent scheduling, automatic report generation and data closed-loop management, so that the utilization rate of equipment is effectively improved, the report generation time is shortened, and a data barrier is opened, and the method and system are the technical problems to be solved urgently. Disclosure of Invention The invention aims to solve the technical problem of providing a textile production man-machine efficiency optimization method and system integrating multiple AI algorithms, which realize intelligent production scheduling, automatic report generation and data closed-loop management, thereby effectively improving the equipment utilization rate, shortening the report generation time and opening a data barrier. In a first aspect, the invention provides a textile production man-machine efficiency optimization method fusing multiple AI algorithms, comprising the following steps: Step S1, acquiring core textile production data comprising textile order data, textile equipment data and employee data in real time, and preprocessing each core textile production data; S2, performing anomaly detection and filtration on the preprocessed core textile production data through a pre-trained anomaly detection model; s3, inputting the filtered core textile production data into a pre-trained productivity prediction model to obtain a textile productivity prediction result, and generating a textile production scheme list based on the textile productivity prediction result and a preset production arrangement rule; S4, inputting the textile production scheduling scheme list into a pre-trained production scheduling optimization model to obtain an optimal production scheduling scheme; And S5, executing the optimal production scheduling scheme, calculating statistical indexes based on the core textile production data generated after execution, and generating and displaying a textile production report based on each statistical index through a Word template engine. Further, the step S1 specifically includes: The method comprises the steps of collecting core textile production data comprising textile order data, textile equipment data and employee data in real time, wherein the textile order data at least comprises customer order information and product BOM data, the customer order information at least comprises product types, production quantity an