There are two kinds of model-based proportional integral derivative (PID) design methods depending on whether the PID tuning algorithms have a tunable parameter. For the direct model-based PID design methods without resorting to a tunable parameter, the resulting control performance is usually not satisfactory for higher-order process dynamics owing to the inevitable modeling error. On the other end, while better control performance can usually be achieved by the internal model control (IMC)-PID or lambda-tuning method, the corresponding optimal tunable parameter is normally determined by trial-and-error procedure from the plant tests or simulation requiring prior information of process dynamics, resulting in an iterative tuning procedure and at the expense of considerable engineering efforts. To alleviate this drawback, a PID design method is proposed to design PID parameters directly using the process data collected from an off-line experiment in this paper. The optimization problem pertaining to the proposed design is derived, and the associated design issues are addressed. Extensive simulation results show that the proposed PID design outperforms the direct model-based PID design and gives better or comparable control performance than the respective best control performances attained by the IMC-PID and Maclaurin-PID designs that have been tuned by trial and error in the simulated closed-loop tests involving an a priori known process model. Therefore, the proposed method is able to retain the simple design procedure of direct model-based PID design methods while achieving comparable or better performance than the IMC-based PID design methods.