Natural disasters have been affecting the human life and causing the death of millions since the very first day the human being came into existence. Besides, it causes physical, financial, social and environmental losses and affects societies negatively by suspending daily life and human activities. In order to minimize losses, it is necessary to plan post disaster activities effectively. One of these activities is humanitarian relief logistics activities that aim to provide sufficient amount of humanitarian relief to disaster victims as soon as possible. In this paper, a multi depot vehicle routing problem with stochastic demand (SDMD_VRP) is taken into account. A mathematical model with chance constraint approach is developed for this rarely discussed problem. The proposed non-linear mathematical model is linearized with separable programming methods and examined on test problems. Lastly, a case study was carried out for Ankara-the capital city of Turkey.