Intelligent Parameter Estimation for NHPP Models with Inverse Weibull Intensity Using a Hybrid ANN–ABC Approach
Keywords:
Inverse Weibull Process, Non-Homogeneous Poisson Process, Artificial Neural Network, Artificial Bee Colony, Reliability Modeling, Predictive Maintenance, Failure Data AnalyticsAbstract
This study proposes an intelligent hybrid parameter-estimation framework for modeling the time-dependent failure behavior of repairable systems using a Non-Homogeneous Poisson Process (NHPP) with an Inverse Weibull Process (IWP) intensity structure. Accurate estimation of NHPP parameters is essential for reliability analysis, predictive maintenance, and event forecasting; however, conventional maximum likelihood estimation (MLE) often encounters difficulties when the likelihood surface is highly nonlinear or multimodal. To address this limitation, a hybrid Artificial Neural Network–Artificial Bee Colony (ANN–ABC) estimator is developed, where the neural network captures complex nonlinear relationships while the Artificial Bee Colony algorithm performs global optimization of the network parameters to improve estimation accuracy and convergence stability. The proposed framework is evaluated through an extensive Monte Carlo simulation study using multiple sample sizes and is further validated using a real clinical dataset comprising 299 heart-failure patients. The performance of the proposed estimator is compared with classical MLE and Particle Swarm Optimization (PSO) using Root Mean Squared Error (RMSE) and Bayesian Information Criterion (BIC) as the primary evaluation metrics. The numerical results consistently demonstrate that the ANN–ABC estimator provides lower RMSE and BIC values than the competing methods, with the greatest improvements observed for medium and large sample sizes. These findings indicate that the proposed hybrid estimation strategy offers a robust, accurate, and computationally efficient alternative for NHPP parameter estimation. The proposed framework has promising applications in predictive maintenance, reliability engineering, healthcare event modeling, and intelligent decision-support systems. Future research will investigate Bayesian inference and imperfect repair models to further enhance uncertainty quantification and predictive performance.
References
References
[1] C. Lawson, “Statistical Methods for the Reliability of Repairable Systems,” Journal of Quality Technology, vol. 33, no. 4, pp. 527–528, Oct. 2001, doi: 10.1080/00224065.2001.11980114.
[2] K. Muralidharan, “Repairable System Modeling Using Power Law Process,” in System Reliability Management, CRC Press, 2018, pp. 79–101. doi: 10.1201/9781351117661-6.
[3] L. H. Crow, “An extended reliability growth model for managing and assessing corrective actions,” in Annual Symposium Reliability and Maintainability, 2004-RAMS, 2004, pp. 73–80.
[4] Karbasian M, “ESTIMATION OF PARAMETERS OF THE POWER-LAW-NON- HOMOGENOUS POISSON PROCESS IN THE CASE OF EXACT FAILURES DATA,” 2010. [Online]. Available: www.SID.ir
[5] Brito, É.S., Tomazella, V.L. and Ferreira, P.H., 2022. Statistical modeling and reliability analysis of multiple repairable systems with dependent failure times under perfect repair. Reliability Engineering & System Safety, 222, p.108375.
[6] Li, Q., Liu, L., Li, T. and Yao, K., 2023. Bayesian change-points detection assuming a power law process in the recurrent-event context. Communications in Statistics-Simulation and Computation, 52(12), pp.6011-6033.
[7] Saadoon, N., Salman, H., Az-Zo’bi, E. and Tashtoush, M., 2025. Survival Modelling of Breast and Brain Cancer Using Statistical Maximum Likelihood and SVM Techniques. Statistics, Optimization & Information Computing.
[8] J. van Dyck and T. Verdonck, “Precision of power-law NHPP estimates for multiple systems with known failure rate scaling,” Reliab Eng Syst Saf, vol. 126, pp. 143–152, 2014, doi: 10.1016/j.ress.2014.01.019.
[9] Hussain, A.S., Mahmood, K.B., Ibrahim, I.M., Jameel, A.F., Nawaz, S. and Tashtoush, M.A., 2025. Parameters Estimation of the Gompertz-Make ham Process in Non-Homogeneous Poisson Processes: Using Modified Maximum Likelihood Estimation and Artificial Intelligence Methods.
[10] Shinohara, S., Manome, N., Nakajima, Y., Gunji, Y.P., Moriyama, T., Okamoto, H., Mitsuyoshi, S. and Chung, U.I., 2021. Power laws derived from a bayesian decision-making model in non-stationary environments. Symmetry, 13(4), p.718. https://doi.org/10.3390/sym13040718.
[11] T. S. Kim, C. S. Park, and S. E. Ahn, “Determining the optimal maintenance action for a deteriorating repairable system,” Probabilistic Engineering Mechanics, vol. 23, no. 1, pp. 95–101, Jan. 2008, doi: 10.1016/j.probengmech.2007.12.003.
[12] M. D. de Oliveira, E. A. Colosimo, and G. L. Gilardoni, “Bayesian inference for power law processes with applications in repairable systems,” J Stat Plan Inference, vol. 142, no. 5, pp. 1151–1160, May 2012, doi: 10.1016/j.jspi.2011.11.016.
[13] R. Calabria and G. Pulcini, “Bayes inference for the modulated power law process,” Commun Stat Theory Methods, vol. 26, no. 10, pp. 2421–2438, 1997, doi: 10.1080/03610929708832057.
[14] A. Sen, “Bayesian estimation and prediction of the intensity of the power law process,” J Stat Comput Simul, vol. 72, no. 8, pp. 613–631, 2002.
[15] Pozza, F., Durante, D. and Szabo, B., 2024. Skew-symmetric approximations of posterior distributions. arXiv preprint arXiv:2409.14167.
[16] S. K. Bar-Lev, I. Lavl, and B. Reiser, “BAYESIAN INFERENCE FOR THE POWER LAW PROCESS,” 1992.
[17] Z. F. Jaheen, “A Bayesian analysis of record statistics from the Gompertz model,” Appl Math Comput, vol. 145, no. 2–3, pp. 307–320, 2003.
[18] U. Khair, H. Fahmi, S. al Hakim, and R. Rahim, “Forecasting error calculation with mean absolute deviation and mean absolute percentage error,” in Journal of Physics: Conference Series, 2017, vol. 930, no. 1, p. 012002.
[19] Hussain, A.S., Sulaiman, M.S., Hussein, S.M., Az-Zo’bi, E.A. and Tashtoush, M., 2025. Advanced Parameter Estimation for the Gompertz-Makeham Process: A Comparative Study of MMLE, PSO, CS, and Bayesian Methods. Statistics, Optimization & Information Computing, 13(6), pp.2316-2338.
[20] Hussain, A.S., Saadoon, N.Q., Abdulghafour, A.S. and Abduirazzaq, N.T., 2025, April. Parameters estimation of a suggested model non-homogeneous poisson process: A comparison of conventional and artificially intelligent approaches. In AIP Conference Proceedings (Vol. 3282, No. 1, p. 020023). AIP Publishing LLC.
[21] Hussain, A.S., Jameel, A.F., Az-Zo’Bi, E.A. and Tashtoush, M.A., 2025. Enhanced Parameter Estimation for the Modified Gompertz-Makeham Model in Nonhomogeneous Poisson Processes Using Modified Likelihood and Swarm Intelligence Approaches. Babylonian Journal of Mathematics, 2025, pp.32-43.
[22] HUSSAIN, A.S., 2019. STABILITY OF RICCATI STOCHASTIC DIFFERENTIAL EQUATION. Journal of Duhok University, 22(1), pp.11-16.
[23] Hussain, A., Oraibi, Y., Mashikhin, Z., Jameel, A., Tashtoush, M. and Az-Zo’Bi, E.A., 2025. New Software Reliability Growth Model: Piratical Swarm Optimization-Based Parameter Estimation in Environments with Uncertainty and Dependent Failures. Statistics, Optimization & Information Computing, 13(1), pp.209-221.
[24] Tashtoush, M.A., Ibrahim, I.A., Taha, W.M., Dawi, M.H., Jameel, A.F. and Az-Zo’bi, E.A., 2024. Various closed-form solitonic wave solutions of conformable higher-dimensional Fokas model in fluids and plasma physics. Iraqi Journal for Computer Science and Mathematics, 5(3), p.18.
[25] Veza, I., Afzal, A., Mujtaba, M.A., Hoang, A.T., Balasubramanian, D., Sekar, M., Fattah, I.M.R., Soudagar, M.E.M., EL-Seesy, A.I., Djamari, D.W. and Hananto, A.L., 2022. Review of artificial neural networks for gasoline, diesel and homogeneous charge compression ignition engine. Alexandria Engineering Journal, 61(11), pp.8363-8391.
[26] Ahmad, T., Munir, A., Bhatti, S.H., Aftab, M. and Raza, M.A., 2017. Survival analysis of heart failure patients: A case study. PloS one, 12(7), p.e0181001.
[27] Alonso-Gonzalez, R., Swan, L., Uebing, A., Diller, G.P., Gatzoulis, M.A. and Dimopoulos, K., New York Heart Association (NYHA) classification in adults with congenital heart disease: relation to objective measures of exercise and outcome.
[28] Hacilar, H., Dedeturk, B.K., Ozmen, M., Celik, M.E. and Gungor, V.C., 2025. Accelerated Artificial Bee Colony Optimization for Cost‐Sensitive Neural Networks in Multi‐Class Problems. Expert Systems, 42(5), p.e70045.
Downloads
Published
Data Availability Statement
000
Issue
Section
License
Copyright (c) 2026 Adel S. Hussain, Ala Amourah, Abdullah Alsoboh, Hasanain Jalil Neamah Alsaedi, Mohammad A. Tashtoush, Rana A. Almuttalibi (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors retain copyright and grant the Sigma Journal of Mathematical and Statistical Sciences (SJMSS) the right of first publication. Articles are published under the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits use, distribution, and reproduction in any medium, provided the original author(s) and source are properly credited.
