Production plan management optimization method based on longitudinal federated learning industrial software docking

Abstract

The invention discloses a production plan management optimization method based on longitudinal federated learning industrial software docking, and the method comprises the steps: carrying out the joint modeling learning task confirmation of longitudinal federated based on the demands of an enterprise, carrying out the preprocessing of the data flow of federated learning participating enterprise samples, carrying out the sample alignment of the federated learning participating enterprises, and carrying out the optimization of the federated learning participating enterprises. Performing encrypted longitudinal federated learning of parameter exchange on federated learning participating enterprises, and finally performing enterprise production plan optimization management according to enterprise demand application federated learning prediction results. According to the method, the production plan optimal allocation condition of the next time period is obtained by applying a federal learning means, enterprise resources are comprehensively managed based on an industrial internet industrial software system, the enterprise productivity and production efficiency are improved, and a more powerful model is jointly established, so that the hysteresis quality of current enterprise production plan management is effectively improved.

Description

Technical field [0001] The present invention relates to the field of machine learning and industrial Internet technology, in particular to production plan management optimization methods based on longitudinal federal learning industrial software. Background technique [0002] The industrial Internet is a product of the new era information technology and manufacturing depth. It can lead the organic integration of new generation electronic information technology and advanced manufacturing technology to promote the informationization revolution of my country's manufacturing enterprises. Industrial Internet is an important in China's manufacturing industry upgrades. Infrastructure, has a comprehensive, deep and revolutionary impact on future development. [0003] In the manufacturing enterprise in information construction, the industrial data of the local data and the local data of the production equipment is low, the data acquisition capacity is poor, and based on the current produc...

AI Technical Summary

Problems solved by technology

[0003] At present, in manufacturing enterprises with poor informatization construction, the degree of overall management of edge modules and local data of production equipment is low, and the ability to obtain data is poor. The management of production plans is mainly based on current production needs. In this case There is an obvious hysteresis in the allocation of resources under the environment, which seriously affects production efficiency and enterprise productivity, and cannot complete the collaborative production of people, machines, and materials. Under the wave of global industrial production informatization, such enterprises urgently need to carry out informatization reform and industrial chain Upgrade, system management and application of production data at the edge layer

[0004] In the manufacturing enterprises with good information construction, the resource allocation in industrial production is mainly based on MES system, PLM system, ERP system and other industrial Internet production management industrial software to assist in setting production plans, resource allocation and plan design. This type of enterprise has a relatively systematic management of edge device data, but the use of these complex heterogeneous data is still insufficient, lacking scientific means to use these data for future production plan optimization, and different departments, upstream, midstream and downstream enterprises Due to the low level of data interoperability due to data confidentiality and conflicts of interest, enterprises and departments can only allocate resources based on their own production information, which has a lag and limits the optimal design of production plans.

[0005] Faced with higher requirements from customers and the market for product quality and production capacity, manufacturing companies have begun to realize that they should reform the most basic production management to enhance their competitiveness. At present, new production planning management methods are emerging in an endless stream, such as demand customization. The oriented BTO (Oriented Order Manufacturing) method and the high-quality, low-consumption production method JIT (Just-in-Time Production) method under the condition of multi-variety and small-batch mixed production aim to reduce inventory and optimize resource management. Although these methods have effectively improved The resource utilization rate and industrial efficiency of enterprises, but these production planning methods are essentially planning based on posterior knowledge, and the demand-oriented production method also has the hysteresis of production planning

[0006] In addition to the hysteresis of production planning, big data collaboration between upstream and downstream manufacturing companies in the same industry also has difficulties. Most manufacturing companies are still in the initial stage of using their industrial software data, that is, using their own data to optimize business and make decisions. management through the enterprise's industrial software system. This situation neither meets the needs of the enterprise's development strategy nor meets the relevant requirements in the "Guiding Opinions on Deepening the "Internet + Advanced Manufacturing" and Developing the Industrial Internet", so the enterprise has The motivation and desire to use external data to optimize business; however, for cross-enterprise industrial software docking, privacy issues are the biggest collaboration dilemma, resulting in the existence of physical and logical islands of data between enterprises. Therefore, the present invention proposes an industrial The production plan management optimization method of software docking to solve the problems existing in the existing technology

Images