Stochastic Train Domain Theory Framework

Abstract

The Axiomatic Safety-Critical Assessment Process (ASCAP) is illustrated as a Stochastic Train Domain Theory Framework "Plug & Play" large-scale Monte Carlo simulation vision. The framework supports the US traditional railway system component models as generalized operational train line taxonomy. The railway domain can be Direct Traffic Control (DTC), Centralized Traffic Control (CTC), Cab Signaling, Positive Train Control (PTC) or a Maglev system. The train line taxonomy is characterized as a stochastic train domain that provides either a design for safety or a risk assessment framework. The design for safety domain is the dual of the risk assessment domain. A significant property of the stochastic vision is that it handles the stochastic quantification of safety for processor-based train systems integrated with structural and electromechanical devices. The stochastic Train Domain Theory Framework vision is illustrated with a Maglev guideway reliability, availability, maintainability and safety (RAMS) Monte Carlo risk assessment component model. A Monte Carlo discrete event simulation method is used to illustrate and extend traditional guideway RAMS analysis beyond the traditional methods of a static Fault Tree Analysis (FTA) to a dynamic RAMS analysis that is based on the movement of all the vehicles and their intersection with unsafe guideway substructures, guideway switch conditions and/or electrical and propulsion segment "out of service" events. The need for a formalized validation, verification and certification vision is presented that includes individual component models, the overall Monte Carlo properties of the simulation and the statistical behavior of the Design for Safety and Risk Assessment methodologies. © 2004 Springer Science + Business Media, Inc.