This work emphasises the modelling capabilities of the unsteady flamelet/reaction progress variable approach to implement diesel spray flames for capturing the auto-ignition and flame lift-off phenomena. The droplet size distribution based on the moment scheme characterises the poly-disperse spray model [1] employed in this work. The flamelet progress variable solutions embedded in a Reynolds-averaged Navier-Stokes (RANS) framework, together with the probability density function (PDF) approach, signify the turbulence-chemistry interaction. All thermochemical scalars are represented as a function of mean mixture fraction, mixture fraction variance, reaction progress variable and scalar dissipation rate. Mixture fraction is assumed to follow a beta-PDF distribution, because the reaction progress variable and scalar dissipation rate distributions are assumed to be a delta-PDF. In order to assess the capability of this developed model, the predicted results are compared with experimental data [2]. The developed model gives a reasonably good overall prediction performance in terms of auto-ignition, flame development and flame lift-off length. The flame temperature distributions are comparable with the formations of luminous flames. The predicted flame growth rate is consistent with the experimental results but there is a small over-prediction. Therefore, the present approach can accurately and efficiently capture the auto-ignition and flame lift-off phenomena of diesel spray flame. © 2012 Elsevier B.V. All rights reserved.