One of the main problems of Addis Ababa light railway transit service is therncongestion and extended waiting line. The waiting time of passengers took onrnaverage 17 minutes, greatly reducing the satisfaction of customers. This is thernreason why the study set the main objective to develop an optimization model forrnimproving the service of AALRT station. As part of the approach to achieve thisrnobjective, the study first developed the characteristics of the data to investigaternthe congestion problem. second, an optimization model developed on the selectedrnstation. And finally, the researcher designed an alternative model throughrncomparing the performance and cost-effectiveness against the existing model. rnThe study adopted case study research methodology by taking the Torhailoch andrnLideta metro station using combined Queuing Theory and Monte-Carlo simulationrnmethod. Moreover, the case studies were selected for the most congested stationrnin the morningâ€™s and afternoonâ€™s peak hours. Based on this, primary data wererncollected from each station and secondary data were conducted from documentrnand through interview. rnThe best fit probability distributions of the passengers were found to be uniform,rnbinomial and negative binomial. These indicates that the congestion problemrncannot be solved through Queue Theory and Markov Chain. Hence, thernresearcher optimized the problem by combining Queuing Theory and Monte Carlornsimulation. The models output indicates that the congestion rate of AALRT inrnTorhailoch and Lideta are 109.13% - 115.76% and 116.18% -131.06%,rnrespectively. rnTherefore, it was found that adding two single tramcars could reduce the waitingrntime by seven minutes and the congestion by 95%. As well, the new modelrnimproved the profit by $4,152.25 and $3,623.22 per hour in the morning andrnafternoon, respectively.