The Pareto-efficient combustion (PEC) framework is extended for predicting transient ignition in turbulent flames. The PEC formulation utilizes a drift term to assess the compliance of specific combustion submodels with the underlying flow-field representation. This drift term is extended to consider autoignition delay as a process of interest, and the ignition delay time of a homogeneous isobaric reactor is utilized as surrogate. Combustion models considered are the steady-state flamelet/progress variable (FPV) model and a finite-rate chemistry (FRC) model. Large-eddy simulations are performed and the proposed PEC autoignition formulation (PEC-AI) is demonstrated to identify localized regions that require a finite rate chemistry description in order to accurately capture the transient ignition dynamics. The model is employed to examine the stochasticity of the autoignition in the turbulent environment by performing ensemble simulations to construct probability distributions of ignition time and ignition locations. Comparisons with measurements show the ability of PEC-AI in capturing the transient ignition dynamics and flame lift-off through drift-term adaptation with comparable accuracy to FRC simulations at reduced cost. These results illustrate the versatility of the PEC framework in targeting different combustion response functions that not only include species and emissions but also transient combustion processes. (C) 2020 The Combustion Institute. Published by Elsevier Inc. All rights reserved.