Analysis of the Reliability and Availability of the Algiers Metro Traction System Based on the Pareto and Weibull Three Parameters Methods

Abstract

Rail transport by metro is a socio-economic form of transport, and the reliability of traction systems is considered to be one of the key factors required to ensure its smooth operation, safety and public comfort. In addition, it provides rapid mobility for citizens between different areas of a city and reduces the road congestion in urban areas by offering a fast, efficient transport alternative. This research is based on an in-depth analysis of the most frequent operating time data, failures and breakdowns (mechanical, electrical, etc.) of the ELDJAZAIR-Alger metro company's traction systems, with the aim of analysing the reliability and availability of these two series of traction systems, in order to minimize interruptions and maintain a high level of service to meet users' needs. Historical data was collected and thoroughly analyzed. Firstly, we visualized this data in the form of a Pareto chart known as the ‘20/80% rule’ (A, B, C) (Thomas E. Nisonger in The “80/20 Rule” and Core Journals. The Serials Librarian 55(1):62–84, in order to determine the highest priority areas. This makes it easier to process the data. Second, the reliability of vehicle traction systems is analyzed using the three-parameter Weibull probability criterion R (t, γ, η, β) (Vonta Ilia, Mangey Ram. Reliability Engineering: Theory and Applications. Publisher: CRC Press-Taylor & Francis Group ISBN: 9,780,815,355,175. (2018)), based on the calculation of the time-to-function and mean time between failures and the failure rate λ (t). The results obtained using the Pareto method show that the most critical faults are classified in zone A: where 20% corresponds to 80% of the total number of faults, representing five faults designated by codes as follows: [CLS, MON, CAJ, CAU, VAUX], which are more critical and take priority over emergency intervention by the maintenance department. The results of the analysis show that the correlatio n coefficient equals to 0.976, which indicates a very strong positive linear relationship between the percentage of variable failures and time and the reliability R (t, γ, η, β) of traction systems is almost in the approximate 50% with its parameters: γ = 0, β = 3.9776, η = 1067.62h and availability A(t) decreases over time, which requires measures to improve the latter by including new techniques for service management, control, preventive and corrective maintenance