Abstarct: Reliability of electrical distribution networks is one of the vital issues that draws the attention of companies in distribution sector. The distribution automation systems (DAS) and distributed generators (DGs) are very important in distribution networks to improve reliability indices. The current thesis consists of two research projects. The first project is associated with the optimal placement of manual and automatic switches for DASs without DGs. In this project, the numbers of manual and automatic switches are also considered as the problem variables. The objective function in this project is to increase the economic savings amount by decreasing the sum of customer interruption costs (CIC) and switches purchasing and maintenance costs (SPMC). In this project, a method is developed which includes a set of detailed equation in order to estimate the index of CIC by considering all modes of fault in different branches with different fault rate for each branch and also various hourly, daily and monthly loading rates. In addition, the proposed algorithm employs an exact ladder method to determine the points of lost loads, the switches which must be switched and the fault time duration for each case. It is used the genetic algorithm (GA) as a tool to optimize the problem. The efficiency of offered method is shown by doing experiment in the extended sample network including 103 load points which is extracted from a real network. The second project presents a modified shuffled frog leaping algorithm (SFLA) for simultaneity optimal placement of DGs and line switches in DASs. The size of DGs and also the number of manual and automatic switches are determined by this algorithm. In the proposed method, a fuzzy multi-objective approach is used to calculate the general objective function, and a fuzzy membership function is defined for each term in the objective function. The primary objective is reliability improvement by minimizing CIC. The second objective is to minimize SPMC. Other objectives are to minimize the active power losses and improving voltage profiles. This paper also develops a method for calculating CIC and SPMC indices in a distribution network by considering DGs. The performance of the proposed approach is assessed and illustrated by studying on a real 11-KV distribution network including 95 load points. The results show a significant improvement in system reliability parameters, a substantial reduction of active power losses and a considerable voltage deviation reduction, by applying the proposed method to the case study.At the end, to demonstrate the efficiency of the proposed algorithm compared to the other exiting optimization method, the GA is also applied to solve this problem and then the results are compared and discussed. The comparison results demonstrate a better and faster convergence manner for the modified SFLA than GA.