Transactions on Data Analysis in Social Science

Transactions on Data Analysis in Social Science

Estimation of the Change Point in a Poisson Process Using the Gravitational Search Algorithm Optimization Approach

Document Type : Original Article

Author
School of Mathematical Sciences, Universiti Sains Malaysia, Penang, Malaysia
10.22034/tdas.2026.248201
Abstract
Control charts constitute one of the most effective tools in Statistical Process Control (SPC) for monitoring process stability and detecting deviations caused by assignable factors. While these charts can successfully identify the occurrence of a process shift, they generally do not determine the exact time at which the change has occurred. Accurate estimation of the change point is therefore essential for identifying the root cause of the variation, implementing timely corrective actions, and preventing further deterioration of process performance. This study focuses on estimating the change point in a Poisson process using numerical optimization techniques. Specifically, the change point is estimated by maximizing the likelihood function and minimizing the Schwarz Information Criterion (SIC), both of which provide reliable statistical frameworks for model selection and parameter estimation. In addition, a novel optimization strategy based on the Gravitational Search Algorithm (GSA), a population-based metaheuristic inspired by Newtonian gravitational interactions, is investigated for solving the change-point estimation problem. The performance of the proposed approach is evaluated through extensive simulation experiments under different process conditions. The results demonstrate that the GSA-based estimator provides accurate and reliable estimates of the change point and exhibits satisfactory performance in identifying process shifts, indicating its potential applicability in statistical quality control and process monitoring.
Keywords

Volume 8, Issue 2
Spring 2026
Pages 1-10

  • Receive Date 03 December 2025
  • Revise Date 28 January 2026
  • Accept Date 25 April 2026