In the 21st century, developing countries are still struggling with unreliable power systems that are not smart and automated. Many researchers show that reliability of the power system can be improved by using distribution automation technologies since the distribution system accounts for more than 80% of the unreliability of the entire power system. This unreliability has an economic impact on both the power system utility and its customers. Utilities are losing money because of unsold power to customers as a result of long-term power outages and equipment damages. This also affects operation and production of industrial, commercial, and residential customers. This research aims to improve reliability of the distribution system by designing and simulating a supervisory control and data acquisition (SCADA) system on Bahir Dar city distribution feeder known as R6-G2. This feeder has about 82 distribution transformers that are connected to industrial, commercial, and residential customers with about 1362 installed power meters. The yearly interruption data of this feeder, having a closing and opening date and time information, is collected from Ethiopian electric utility, Bahir Dar district, for reliability assessment of the existing distribution feeder using standard reliability indices. Results of these reliability indices are compared with reliability indices of the automated system simulated in MATLAB. This distribution feeder is modeled in MATLAB Simulink based on parameters collected from the distribution authority. Current and voltage sensors are placed on both the primary and secondary sides of all distribution transformers to monitor the status of the feeder on the SCADA system designed using Java programming language. This human machine interface (HMI) is connected to MATLAB Simulink using a serial communication protocol to send and receive sensor and actuator data. Circuit breakers are also added to the Simulink model to give remote access for system operators to isolate and reconnect sections of the feeder when a fault condition is detected. The SCADA system gives summarized fault and status reports in addition to real-time monitoring of the feeder with color-coded status indicators and alarming signals. It also gives geographical location of distribution transformers on mobile phones of field operators. Based on results gained from the simulation, new reliability indices are calculated and compared with reliability indices of the existing system (the base case). The system average interruption duration index (SAIDI) of the feeder has improved from xii 158:28 hours per customer per year to 61:16 hours per customer per year. The customer average interruption duration index (CAIDI) of the feeder is also improved, from 01:03 hours per interruption to 00:24 hours per interruption. The average service availability index (ASAI) of the feeder has also improved from 98.19% to 99.30%. This is a 97 hours improvement on the hours of service delivered to the customer in the year 2021. Keywords: SCADA system; distribution automation; feeder reliability; GPS-location; real-time monitoring; Java HMI; condition monitoring.