Thermocline storage in a packed-bed is considered as a promising thermal energy storage (TES) method that achieves cost reduction with respect to current concentrated solar power (CSP) plants. Several parametric studies investigated thermal performances of different types of packed-bed thermocline TES systems, and cost analyses and dimension design studies were conducted based on this. However, parametric studies typically involve high computing costs for a dimension design, and results obtained by correlation fitting suffer from a relative lack of accuracy. The main objective of the present study involves directly determining the tank size of a packed-bed thermocline TES system to satisfy certain design requirements of a CSP plant without requiring parametric studies. In order to achieve this goal, a one-dimensional enthalpy-based dispersion-concentric (D-C) model is developed and validated to investigate the periodic thermal behavior of the system. An efficient tank size estimation strategy is proposed based on the periodic thermal performance of the system. The strategy is used to size four different packed-bed thermocline TES types under two exemplary operating conditions. The results reveal that tank size estimation strategy is independent of initial conditions, and that it is self-convergent, involves cost saving in terms of computing costs, is generally applicable, and can provide guidelines for the design of a TES system for CSP plants. (C) 2017 Elsevier Ltd. All rights reserved.