Aircraft assembly production line productivity prediction method based on deep neural network

Abstract

The invention provides an aircraft final assembly production line productivity prediction method based on a deep neural network. A resource configuration scheme and historical capacity information areuniformly expressed in a data sequence form with time sequence information and are used as neural network input; a deep belief network (DBN) and a back propagation (BP) algorithm are used, and training of the aircraft final assembly production line productivity prediction model is completed by combining an unsupervised training mode and a supervised training mode, so that change prediction of theaircraft final assembly production line productivity in a period of time in the future is realized.

Description

technical field [0001] The invention belongs to the field of production capacity prediction of an aircraft assembly line, and specifically designs a method for verifying the production capacity of a resource allocation scheme based on a deep neural network. Background technique [0002] The aircraft assembly line is a multi-variety, small-batch production model, and the production capacity of the production line often needs to be adjusted according to the market order demand. If the actual production performance deviates from the production plan, it will lead to a surge in material storage costs, unbalanced tasks, idle personnel and other consequences, which may delay the delivery of orders and increase the cost of aircraft production and development. At the same time, compared with the resource cost ratio of the assembly link in the general manufacturing industry, which is generally less than 20%, the resource cost ratio of the aircraft assembly process is about 65%, which ...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problems of cumbersome operation, long time and inability to analyze the time delay of changes in resource allocation schemes and changes in production capacity in the process of model construction and parameter modification in the prior art, the present invention proposes a production capacity prediction method for aircraft assembly production lines based on deep neural networks , the resource allocation plan and historical production capacity information are uniformly expressed in the form of a data sequence with time series information and input as a neural network. Through the Deep Belief Network (DBN) and Back Propagation BP algorithm), combined with unsupervised and supervised training methods to jointly complete the training of the production capacity prediction model of the aircraft assembly line, and realize the prediction of the change in the capacity of the aircraft assembly line in the future

Images