Abstract: To resolve the low accuracy and low efficiency problems of mid-term conflict detection algorithms under the situation of aircrafts with multi-route in high flight density airspace, a novel algorithm is proposed. In this paper, the aircrafts’ routes are divided into a sequence of route segments with no heading and speed changes. Based on Prandini’s instantaneous conflict probability theory, the maximum value of all the route segments’ maximum instantaneous conflict probability is a criticality measure of the two aircrafts’ conflict probability in their entire routes. A novel algorithm based on the discretization of predicted position space is proposed to calculate instantaneous conflict probability of route segments. Discretizing two aircrafts’ predicted position space into a certain number of small rectangles of the same size, calculating the centroid of every rectangle and the probability that aircraft’s position is within the rectangle according to the position prediction probability density function, the instantaneous conflict probability is calculated by cumulating the probability that an aircraft’s rectangle is in conflict with another aircraft’s rectangle. The simulation results show that the accuracy and efficiency of the algorithm proposed in this paper are much better than Prandini’s randomized algorithm and the results of the algorithm in the paper is much more stable. It is able to meet the requirement of real-time conflict detection for aircrafts with multi-route in high flight density airspace.