The larger problem that I'm trying to solve is the following:

Given k $ lists of arrival times (all times less than or equal to an end time $ T $) derived from k $ non-homogeneous Poisson processes with unknown rate functions $ lambda_1 (t), ldots, lambda_k (t) $, I want to order the processes in descending order of $ lambda_1 (T), ldots, lambda_k (T) $

My current thinking was to predict the *following* time of arrival (perhaps the expected arrival time?) for each of the k $ PPSN and order them based on that, but this logic may be circular?

I found work on PSPS arrival time forecasting (eg Goulding *et al.*, 2016, Shen & Huang, 2008 and Weinberg *et al.*, 2006), but these documents seem to be mainly focused on the estimation of PPSN parameters, which does not interest me (directly?): I only care about the relative extent of their rate functions to an end ), which, I imagine, might not require as many calculations. Any pointers or tips would be greatly appreciated