Productivity prediction in aircraft final assembly lines: Comparisons and insights in different productivity ranges

This paper investigates the productivity prediction for aircraft final assembly lines. Various approaches are proposed and performance of each approach is compared discussed. The predictive models established in this can be divided into three groups. first group a simulation model, built by practitioners with extensive insights plant layout, material handling, manufacturing process resource allocation. second consists representative regression models, including linear, polynomial exponential regression. last one based on machine learning, multilayer perception, gradient boost tree random forest. Compared typical light industrial products, product has large number modified types order to meet customers’ personalized requirements. In addition, management lines depends different ranges managers often need verify within given tolerance which not covered previous studies. above, provides comprehensive comparison among modelling approaches. A real line adopted illustrate feasibility approach. advantages limitations terms their efficiency, precision generalization ability presented. aim provide practical guidance choice suitable datasets, as well data processing techniques. conclusion drawn applicable other customer dependent multi-product problems.