1. Introduction
PROBABILISTIC reliability concepts have been successfully applied to many areas, including generation capacity planning, operating reserve assessment, distribution systems, etc [1]. Blackouts in North American power system emphasis the importance of corrective measures based on probability based reliability analysis. The methods employed in composite reliability evaluation can be categorized as analytical and Monte Carlo simulation approaches [2]. Analytical techniques based on state enumeration are approximate and can be used to evaluate the expected value of reliability indices. Monte Carlo simulation techniques are universal than analytical approaches and can provide more reliability information such as probability distributions of the reliability indices in addition to the average values. Monte Carlo simulation techniques can be further divided into sequential and non-sequential simulation. The sequential approaches can simulate the chronological aspect of the system operation and provide the interruption duration of each system failure. Hence this approach able to produce accurate estimates of reliability indices. Sequential approaches require more computational effort than the non-sequential Monte Carlo and analytical methods, which may be infeasible for large systems. Non-sequential techniques have high computational efficiency but cannot simulate the chronological aspects of system operation. Chronological aspect of the system state is modeled in the sequential Monte Carlo simulation approach is done by state duration sampling [4]. The state sampling technique and state transition sampling technique [3] are used to sample the system state in non-sequential Monte Carlo Simulation.