id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-024870-79hf7q2r	Salierno, Giulio	An Architecture for Predictive Maintenance of Railway Points Based on Big Data Analytics	2020-04-29		.txt	text/plain	4028	218	52	In this paper, we propose a four-layers big data architecture with the goal of establishing a data management policy to manage massive amounts of data produced by railway switch points and perform analytical tasks efficiently. The goal of our work is to design a big data architecture for enabling analytical tasks typical required by the railway industry as well as enabling an effective data management policy to allows end-users to manage huge amounts of data coming from railway lines efficiently. As already mentioned, we considered predictive maintenance as the main task of our architecture; hence to show the effectiveness of the proposed architecture, we use real data collected from points placed over the Italian railway line (Milano -Monza -Chiasso). These log files are heterogeneous in type and contain different information resumed as: Data 3 and 4 are considered to train and evaluate the proposed model to estimate the health status of the points, thus to estimate its RUL (see Sect.	./cache/cord-024870-79hf7q2r.txt	./txt/cord-024870-79hf7q2r.txt