For control-oriented modeling of heat exchangers, the two predominant model types are the moving boundary (MB) approach and the finite volume (FV) method. In this contribution, we assess both approaches. As a case study, we present an organic Rankine cycle (ORC) for waste heat recovery (WHR), for which experimental data is available. For simulation of nominal operation, we observe that the MB approach requires less CPU time than the FV method, even for a low number of cells, which is in agreement with literature. In the start-up case, where only subcooled liquid is present, analysis of the MB model reveals that the model can exhibit an unphysical inverse response in outlet enthalpy due to its averaging assumption. This problem can be circumvented via the use of a hybrid MB-RI model or an adjusted averaging assumption. A simulation study based on experimental data shows that both proposed solution approaches can be successfully employed. (C) 2018 Elsevier Ltd. All rights reserved.