Clarification on malfunction generation

Hello!

Thank you for your work on this project. I am currently using the flatland framework in a course at the University of Potsdam, to schedule trains using Answer Set Programming (ASP). We are tackling malfunctions currently, and while looking up the documentation I saw the malfunctions are supposed to follow a Poisson process. However, when I checked the current implementation of malfunction generation, I gathered that was is happening is for any agent, at each time point we have the same fixed probability that a malfunction is triggered, that probability being calculated as 1 - np.exp(-rate) in the code. Is my understanding correct, and if so isn't this then a Bernoulli process with the aforementioned probability? If so, is this being used to approximate the Poisson process? In that case what method is this approximation using, and why is the rate being used as such in the formula above? Thank you from me and the students for the clarification!

Best,

Amade