Abstarct: In this study, first double effect lithium bromide-water absorption chillers series and parallel flow were modeled by the EES software. Afterwards the two cycles were investigated in different performance coefficient circumstances by the input data provided by Viuna air conditioning Industry Company to assess the strengths and weaknesses of absorption chillers produced by this Company and compare the two series and parallel cycle in terms of their performance coefficient and the formation of crystals. In the first stage, the impact of the independent input variation on the performance coefficient and concentration ratios of the questioned point in the series and parallel cycles were examined by EES software. The variables influencing the performance coefficient and concentration ratio were selected from the obtained results. In the next stage, the best performance coefficient was calculated by Genetic Algorithm in MATLAB Software, provided that no lithium bromide crystals were formed. The results show that the non-crystal-formation ranges in the parallel cycles are wider than those in the series cycles. In the event of changing on value temperature of cold water entering the absorber that it is due variation of ambient temperature, in order to optimize performance coefficient in series and parallel flow absorption chillers, it is suggested in the same time the parameters value should be selected for temperature and mass flow rate of steam entering high temperature generator and mass flow rate of cold water entering the absorber, so that any changes in any of them would change the performance coefficient within a limited range. Since a large reduction in the temperature of the cold water entering the absorber leads to the formation of crystals and reduction of the mass flow rate of cold water entering the absorber also leads to reduction of density, to prevent the formation of crystals caused by the cold water of mass flow rate entering the absorber, it is recommended that the preventive measure be carried out through reduction of cold water mass flow rate entering the absorber. In parallel cycle, increase the ratio of mass flow rate entering number 2 dilution pump to mass flow rate outlet of low temperature heat exchanger has increasing and then decreasing effect on coefficient of performance that variation of parameters such as temperature and mass flow rate of steam entering to high temperature generator, temperature and mass flow rate of cold water entering absorber have effect on maximum amount of performance coefficient. The results of series cycles optimization by genetic algorithm show that the performance coefficient have improved 4% and 5% in the first and second cases respectively. The results of parallel cycle optimization by genetic algorithm show that the performance coefficient has improved 6% and 5% in the first and second cases respectively. By define rang for temperature of steam inlet to high temperature generator and temperature of the cold water entering the absorber, in terms of the performance coefficient and the non-crystal-formation ranges, the best cycle has been recommended.