Transcritical organic Rankine cycle is a great promising technology in the field of energy saving and environ" ment protection. Using trans-1,3,3,3-tetrafluoropropene (R1234ze(E)) as working fluid for low grade waste heat recovery was proposed and studied in this study. The sensitivity analysis was introduced to analyze the influence of multiple parameters on the thermal and exergy efficiencies of cycle. The results showed that the turbine efficiency and temperature of heat source had the most influence on the efficiencies of cycle. On the basis of sensitivity analysis, the parametric analysis and optimization were considered, and the results indicated that the effect of parameters on efficiencies of cycle was changed with the high pressure of cycle, and the best high pressure of cycle was affected by multiple parameters. Furthermore, to maximize the efficiencies of cycle, an accurate prediction model of the best high pressure of cycle considering multiple parameters was established using artificial neural network. Finally, the calculation model of efficiencies of cycle with the change of different parameters was developed based on artificial neural network. The results demonstrated that the current artificial neural network models were capable of predicting and calculating the best high pressure and efficiencies of cycle accurately.