This work presents a general and systematic methodology for the techno-economic optimization of Rankine cycles. The proposed superstructure for Rankine cycles allows to reproduce a wide range of cycle configurations, such as cycles with/without regenerator, cycles with single or multiple pressure levels, and cycles integrated with multiple heat sources. The model is integrated with a recently developed methodology capable of optimizing also the arrangement and sizing of the heat exchangers of the plant (heat exchanger network synthesis). This allows to perform a full techno-economic optimization of the entire system. The resulting problem is a challenging Mixed Integer Non Linear Problem (MINLP) which is solved with an ad hoc algorithm. The methodology is applied to two case studies for power cycles with single and multiple heat sources. This work can help engineers identify the right thermodynamic cycle to integrate with an industrial process and design techno-economically optimal Rankine cycles for waste heat recovery from single or multiple heat sources, by considering heat integration and cycle design optimization simultaneously. (C) 2017 The Authors. Published by Elsevier Ltd.