Title: A novel sequential ensemble approach and particle swarm optimisation algorithm for forecasting: applied for COVID-19 cases as case study

Authors: Nader A. Al Theeb; Hazem Jamil Smadi; Naser R. Al-Qaydeh

Addresses: Department of Industrial Engineering, Jordan University of Science and Technology, Irbid, 22110, Jordan ' Department of Industrial Engineering, Jordan University of Science and Technology, Irbid, 22110, Jordan ' Department of Industrial Engineering, Jordan University of Science and Technology, Irbid, 22110, Jordan

Abstract: COVID-19 virus has spread to most countries around the world, negatively affecting people livelihood. Providing accurate forecasts of COVID-19 cases can help governments to find the optimal combination of measures. In this study, a sequential ensemble forecasting approach that combine the gated recurrent unit (GRU) model with particle swarm optimisation (PSO) algorithm is proposed for forecasting of COVID-19 cases. The PSO method was used to select the best hyperparameters of the base predictor of the proposed model. The t-test was used to statistically compare the suggested model against a single optimised GRU, in addition to other benchmark models. Results revealed the superiority of the proposed method. Further, adding models sequentially improved the forecasting quality, compared to a single PSO-GRU model, the mean error was reduced by 15.52%, 16.05%, 16.53%, 16.39%, and 12.83% in terms of RMSE, MAP, MAPE, RMSPE, and RMSLE, respectively.

Keywords: deep learning; time series forecasting; ensemble model; particle swarm optimisation; PSO; long short-term memory; LSTM; gated recurrent unit; GRU.

DOI: 10.1504/IJLSM.2026.151734

International Journal of Logistics Systems and Management, 2026 Vol.53 No.2, pp.191 - 214

Received: 25 Feb 2023
Accepted: 11 Aug 2023
Published online: 18 Feb 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article