The catalyzed diesel particulate filter (CDPF) is one of the best methods used to control the particulate matter from internal combustion engine. CeO₂ is used as the catalyst in the CDPF due to its good redox capacity, oxygen storage capacity and anti-oxidation capacity to NO_x. However, there are few studies on the effect of key factors on composite regeneration performance of CeO₂-based CDPF. Therefore, a single-channel numerical model was established in this study by coupling fluid flow, mass transfer, heat transfer and chemical mechanism. In order to save computing time and computing resources, a sample space of the simulated data from the model was established by optimal space-filling method to build a surrogate model by response surface methodology. A multi-objective optimization on the composite regeneration performance of CDPF was carried out, in which the inlet NO_x concentration, inlet temperature, catalyst coated amount and the N₂O concentration, soot regeneration rate, pressure drop, NO_x reduction rate, and peak temperature for coated catalyst are set as the design and optimization variables, respectively. Results show that the optimum inlet NO_x concentration, inlet temperature and catalyst coated amounts are 0.106 mol/m³, 761.798 K and 0.428 mm, respectively. The findings of this study give useful insights into the operation of CDPF and the thermal management of after-treatment systems.