The plan for the operation of trains on the Addis Ababa Light Rail Transit (AALRT) is based onrna fixed interval of riding between stations. Fixed riding time between stations will have the effectrnof inefficient energy consumption by the trains. Furthermore, the transportation capacity of thernnetwork cannot be optimal. By properly managing the reference trajectories the trains use, it isrnpossible to have optimum train operation with respect to energy consumption as well as networkrncapacity.rnIn this thesis, an optimization of train speed profiles is done. A multi-objective optimizationrnproblem has been formulated by making energy and time as the components of the two elementrnobjective vector function. A point mass model of the operation of trains has been developed byrnconsidering all the important force components acting on the train. The distance to travelrnbetween stations is discretized into 20 equal length elements where a two stage solutionrnprocedure has been applied to get to the final results. The first stage of the solution procedure isrnthe application of a multi-objective genetic algorithm based optimization technique taking vectorrnof riding modes as the decision variable. Using the developed algorithms for the calculation ofrncost functions for every type of riding mode, the MATLAB optimization toolbox determines arnPareto-optimal set of riding modes. The second stage of the solution process smoothes out thernresults found in the previous stage of the solution process without bringing about considerablernchange in the values of the cost functions.rnDifferent solutions for every section from Ayat station to Megenagna station are generated andrnthey are essentially tradeoff solutions. It has been observed that the fastest ride between stationsrncan be completed within a time of less than 3 minutes. This is equivalent to a 50% reduction inrnriding time over the plan. By shifting from the fastest to the slowest trajectories, it is possible tornsave up to 38.18% of energy, while 23.98% reduction in riding time can be achieved byrnpreferring the fastest profiles over the slowest ones.