Abstract <p>This paper presents an algorithm for calculating a previously proposed decoherence measure, which numerically describes the maximum deviation in the probability of obtaining a desired outcome. This algorithm is based on parameterization of operations and gradient descent for parameter optimization. The algorithm was tested for estimating the decoherence measure of a simple two-qubit operation. A comparison of the algorithm with grid-based search was conducted. The proposed algorithm is suitable for execution on a real quantum computer.</p>