This paper presents a mathematical model for solving a production planning (PP) problem in dynamic cell manufacturing system (DCMS) with uncertain demand. One of the characteristics of the DCMS is the reconfiguration, which is the number of machines in each manufacturing cell is allowed to change over a particular planning horizon. Consequently, the DCMS is more flexible, but complicated in determining an optimal solution, especially with the uncertain demand that confuses the accuracy of the optimization in traditional way. Therefore optimal planning time horizon is introduced as a distinguishing decision variable to handle the penalty from the uncertainty of the demand. The developed model is mixed integer non-linear equations. Numerical example is presented to show the performance of the method, where traditional DCMSs PP is employed for comparison. Improvement rate map has been drawn to illustrate the benefit zone for new method.