Uncertainties exist inevitably in operating parameters of flow and temperature field in sinter cooling process, thus leading to the performances of a sinter cooling process uncertain. Traditional researches in sinter cooling use deterministic design optimization methods where the uncertainties of operating parameters are ignored, which may result in unreliable design and optimization results. In this paper, the uncertain operating parameters are fully considered by the interval model, and the interval uncertain optimization of the sinter cooler is carried out, which can provide an optimal exergy of waste heat recovery (EWHR) with the requirement of the uncertain final sinter temperature (FST) satisfied simultaneously. In interval optimization, the influences of parametric uncertainties on constraints can be fully considered, thus ensuring that the obtained optimal solution can meet the reliability request. To reduce the computational burden, the high-precision metamodels for the indicator parameters, namely, FST and EWHR are constructed. Based on this, a nonlinear interval optimization model for the sinter cooling process is established, in which both the objective and constraints are functions of interval parameters. By using the possibility degree of interval number, this interval uncertain optimization can be conveniently transformed to a deterministic optimization problem, which can be solved by the genetic algorithm (GA). The optimization results show that the performance of energy conservation can be improved significantly while the reliability of the sinter cooling process can be guaranteed simultaneously.