Spare Parts Demand Prediction Research of Navigation Marks Based on the Method of Quadratic Exponential Smoothing

Abstract

This paper aims to research and provide a reasonable prediction method that can be applied on routine work of navigation marks. On one hand, the reasonable prediction method can help maintenance and management unit to reduce cost and relative risk. On the other hand, according to Yangtze river waterway practical condition and historical consumption data of navigation marks' spare parts, the study has shown the quadratic exponential smoothing method be adopted in actual consumption research can satisfy the requirement of demand prediction.

Land Subsidence Time Series Prediction Method Based on LSTM-AMSGD

“…In recent years, geological subsidence prediction methods include traditional prediction methods and machine learning methods. The traditional methods start from the physical evolution process within geological subsidence and obtain a complex set of physical parameters, including lithological and hydrological characteristics, through field visits and tests, and then predict geological subsidence [2][3][4]. However, due to the large number of parameters, in many cases, strict assumptions need to be made before applying such models, and these assumptions may sometimes fail, such as the difficulty in determining the trend and periodicity of the land subsidence process.…”