Enhancing reliability in an industrial LAN: design and performability evaluation

A fault-tolerant fiber-optic LAN (local area network) is presented and assessed. The reliability of this LAN has been increased by introducing redundancy into the physical transmission medium and by providing the network with a monitor process that is capable of detecting and isolating faults automatically. The network is modeled by applying Markov process theory, through which the mean time to failure (MTTF) and performability are assessed. The results obtained show that the network exhibits a significant increase in MTTF and performability, which is all the more evident if the coverage factor c is closer to 1 and if the number of stations is higher. Redundancy in the transmission medium and the monitor process enable the network to overcome the typical reliability limitations affecting ring topologies and allow it to be used in some critical industrial applications.<
>