Traditionally, reliability of power systems has been an important measure of system performance and a key factor in system planning. Recently, the large-scale changes in the regulations governing the power industry have lead to a growing emphasis on distribution system reliability. Further, the shift towards a more technical and computerized society requires that power supply be increasingly reliable. Advanced models and methods are needed to obtain an improved understanding of the distribution system reliability. Monte Carlo simulation is one such method that can be used to find the statistical distribution of the reliability indices. This dissertation presents a computationally efficient Monte Carlo simulation algorithm for assessing the distribution reliability indices. Several state regulatory agencies have started to prescribe minimum reliability standards to be maintained by the distribution companies. The effect of these regulations has not been fully explored. In this work, a detailed analysis of the impact of various regulatory standards on a practical distribution system is presented. Storms cause a significant fraction of the distribution customer interruptions. While the impact of wind storms on distribution system reliability has been studied earlier, the effect of lightning storms on the reliability indices is not fully understood. Momentary interruptions caused by lightning storms may severely disrupt production at automated manufacturing facilities and other sensitive loads resulting in a loss of millions of dollars per incident. An analysis of lightning storm data is presented in this dissertation along with a method for calculating the impact of lightning storms on distribution system reliability. Finally, several topics for future research are discussed.