Scuotto, Michele (2011) Quality, Availability and Lifecycle Cost of Transportation Systems. [Tesi di dottorato] (Unpublished)
Download (1MB) | Preview
|Item Type:||Tesi di dottorato|
|Uncontrolled Keywords:||Stochastic models - Parameters estimation - Non Markovian Processes|
|Date Deposited:||08 Dec 2011 19:13|
|Last Modified:||17 Jun 2014 06:03|
In recent years, the development of policies aiming at promoting the use of public transport assumed a continuously growing importance. Under this perspective, ensuring that passengers perceive high service Quality is of paramount importance. Thus, in order to ensure high service Quality levels, Transport Authorities should be able to measure quality perceived by customers via customer oriented, clear and easy to apply approaches. In addition, they should be able to assess the effects of organizational settings and maintenance management policies on service availability. On the other hand, in order to be competitive, companies delivering transportation systems should be able to assess the effects of reliability objectives on the system Lifecycle cost since the conceptual design phase. In such a context, this research work aims at giving a methodological contribution, based on a solid stochastic and statistical framework, for managing Quality, Availability and Lifecycle Cost of Transportation Systems. A methodology to evaluate transport service quality, relying on a set of innovative indexes based on the generic customer’s waiting time, is defined. It allows to “on-line” monitor Quality of high frequency bus services by means of commonly available data, avoiding expensive data collection activities. In fact, a specific strategy is formulated to obtain the waiting time distribution for a generic customer on the basis of commonly available data/information. The adequacy of the above strategy and of the main working hypotheses is checked by means of real data concerning the route 181 operated by A.N.M. The proposed approach enables to quantify the effects of service frequency and service regularity on the customers’ waiting time. Moreover, besides allowing to assess the current service quality level, it allows to forecast how changes of the service frequency and/or regularity can affect quality perceived by customers. Moreover, a simple approach to assess the effects of the inherent repair time distribution, number of spare vehicles and number of maintenance crews on the operational availability of a fleet of vehicles is defined. The considered approach allows to easily handle complex Non - Markovian stochastic processes, and, in accordance with results found in Queueing theory, reveals that fleet operational availability may be significantly influenced by the inherent repair time distribution in both transient and steady phase. In addition, it is shown that the proposed approach leads to operational availability evaluations that are more accurate than those usually performed in the industrial practice and that it may overcome some issues arising when Monte Carlo Simulation is used. At last, a lifecycle cost – reliability model based on a three-parameters Hyperbolic failure intensity is presented. Numerical procedures to obtain Maximum Likelihood and Minimum Chi-square estimates of the Hyperbolic model parameters are outlined. By means of reliability data concerning the first two years of operation of the Copenhagen Metro vehicles, it is shown that the Hyperbolic model may be adequate to explain the failure rate behavior of complex repairable systems. Then, relationships between reliability and the main cost components depending on reliability are developed. In this respect, the innovative concept of inertia pertaining to the reliability improvement process is used to model costs to be born, during the testing phase and the early stages of operation, to reach the planned reliability target.
Actions (login required)