Estimation of heterogeneous Poisson process parameters using the PSO algorithm compared with conventional methods

Abstract

In this paper, it was proposed to use the PSO algorithm to estimate the rate of occurrence of the Non-homogeneous Poisson process with rate described as a power law function. The results of the estimation method were compared with classical methods of maximum likelihood and least squares. The research included a realistic application in which the operating periods between the two stops per days of the Mosul Gas Station, and estimated the time of the station stop using the estimation methods proposed in the research. Among the applied results, it was observed that the PSO algorithm was better in estimating the rate of occurrence.