The invention relates to a method for survivability evaluation of an Internet of Vehicles link. The method comprises the following steps of constructing a continuous Markov chain model for the Internet of Vehicles link; utilizing a continuous time random logic formula to define two survivability-related attributes, namely a maximum probability and a minimum probability of network damage when a failure occurs; respectively analyzing the two attributes for two failure types in probability model detection software, and respectively outputting analysis charts; observing the analysis charts to obtain parameters with the practical guide significance, wherein the parameters are an upper limit value of failure intensity and a safety value of failure intensity; and evaluating the survivability of the current Internet of Vehicles link according to the running state of the Internet of Vehicles link. The practice indicates that a probability model detection research method has the characteristics of simplicity and high efficiency on survivability analysis of a communication network of Internet of Vehicles.