This paper presents the parametric optimization and performance analysis of ORC (organic Rankine cycle) system using GA (genetic algorithm) for recovering exhaust waste heat of a diesel engine. First, the effects of three key parameters, including evaporation pressure, superheat degree and condensation temperature on the system performances are conducted with POPA (net power output per unit heat transfer area) and EDR (exergy destruction rate) as objective functions. The optimal evaporation pressure, superheat degree and condensation temperature corresponding to the maximum POPA and the minimum EDR are provided under engine various operating conditions. Subsequently, as a key component of the ORC system, the performance of a fin-and-tube evaporator used in this ORC system is evaluated. The results show that the optimal evaporation pressures are mainly influenced by the engine operating conditions, whereas the superheat degree and the condensation temperature present a slight variation over the whole operating range. Furthermore, the PPTD (pinch point temperature difference) in the evaporator may occur at different positions with the variation of engine operating conditions. At engine maximum rated power point, the ORC system can achieve maximum POPA of 0.74 kW/m(2), and the ratio of maximum effective heat transfer area to actual area of the evaporator is 69.19%. (C) 2015 Elsevier Ltd. All rights reserved.