WSEAS Transactions on Computers
Print ISSN: 1109-2750, E-ISSN: 2224-2872
Volume 13, 2014
Enabling Predictive Maintenance Strategy in Rail Sector: A Clustering Approach
Authors: ,
Abstract: One of the imperatives of predictive maintenance of assets is to analyze, understand and act upon the failure pattern hidden in the failure data which are represented, in general, by failure code, failure description and failure instance. A maintenance plan that will be in consonance with the failure trend inferred through data mining is bound to enhance the asset reliability. The paper shows how to integrate clustering approach into the realm of asset maintenance and particularly provides a road map to implement predictive maintenance strategy in rail sector. The proposed approach has been tested on actual failure data pertaining to passenger carrying vehicles of the trains. Finally, the performance of two fundamental approaches i.e. hard and soft clustering has been investigated on the data set and a recommendation made therein.
Search Articles
Keywords: Clustering, Data Analysis, Data Description, K Means algorithm, Railways, Pattern Analysis
Pages: 118-128
WSEAS Transactions on Computers, ISSN / E-ISSN: 1109-2750 / 2224-2872, Volume 13, 2014, Art. #11