This study presents a customized predictive modeling framework employing ARMA, Holt-Winters, and ARIMA time series methods to estimate when space debris fragments will likely re-enter Earth’s atmosphere and potentially impact the surface. The analysis was conducted in RStudio using NASA’s historical orbital debris dataset spanning 2001 to 2021. Multiple evaluation criteria and diagnostic techniques were applied to identify the most suitable predictive model. Stationarity was verified using the Dickey-Fuller test. Model parameters were optimized using the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC). Goodness-of-fit was evaluated through several error metrics: Mean Error (ME), Mean Absolute Error (MAE), Mean Squared Error (MSE), Sum of Squared Errors (SSE), and Standard Deviation of Errors (SDE). In addition, Theil’s U inequality coefficients and their components ( \(U_m\) , \(U_s\) , \(U_c\) ) were computed to assess relative model performance and decompose forecast error sources. Results indicate that the ARIMA model consistently outperformed the alternatives, yielding the lowest values across all error metrics, particularly MAE. While Holt-Winters showed reasonable accuracy for trended and seasonal series, ARMA proved less suitable for this dataset. This research presents a validated modeling framework for predicting space debris re-entry, a comparative assessment of time series forecasting methods, and a quantitative analysis of error structure, providing a foundation for future risk mitigation efforts in orbital debris monitoring.