Methods, techniques and algorithms for condition based maintenance of railway vehicles

Industrial Companies must avoid productivity losses due to low availability of equipment and machinery that cause high maintenance costs. In response to this concern, Companies try improving their maintenance policies over the years. Increasing the effectiveness of maintenance implies switching from simply reacting to machinery breakdowns (corrective maintenance - CM), to executing time-based preventive maintenance (PM). Today's emphasis is on the ability of detecting early degradation through condition based maintenance (CBM) practices. Each policy offers a different level of performances, and it requires a different amount of data for its actuation. This paper analyses methods to implement CBM for a fleet of locomotives. It presents the description of the Automatic Vehicle Inspection System (AVIS), to deliver a revolutionary integrated solution that could become a game changer in the rolling stock maintenance industry. A Company can implement a Condition- Based Maintenance solution with the usage of the AVIS System and a data conditioning and analysis software like OSIsoft PI System. The paper also highlights the reasons for choosing this type of maintenance policy and for using the PI System to collect operations and manufacturing data. The results have been achieved making use at the same time of the PI System (that manages the database) and Matlab (that manages data) applied to the fleet of railway vehicles. The possibilities with this system are endless, since with its flexibility, adding new features is very easy. Results show different aspects for a good maintenance policy: reliable and high quality measurements are important; a condition-based maintenance would provide longer life of the components compared to the current policy; at the end, a new maintenance policy would result in high savings for the Company.