Logical analysis of survival data for remaining driving range prediction in battery electric bus operations

The remaining driving range (RDR) of a battery electric bus (BEB) is the predicted distance that the bus can travel before depleting usable battery energy. This paper proposes a probabilistic and interpretable model that combines Logical Analysis of Data (LAD) and Kaplan–Meier (KM) survival analysis to model the distance to cut-off state of charge (SOC) as a survival variable. LAD extracts interpretable patterns linking operating conditions to SOC-based range classes, while KM estimates nonparametric survival curves that capture the distribution of the distance to cut-off SOC and accommodate right-censored runs commonly observed in fleet operations. The model operates at distance checkpoints, where recent driving and environmental conditions are summarized over distance windows of 1 km and 5 km. At each checkpoint, LAD learns patterns based on SOC clustering using k-means, and pattern-specific KM survival curves are constructed from historical runs. For a new run, survival information from matched patterns is aggregated to predict the distance to cut-off SOC and the corresponding RDR. The model is evaluated on real-world BEB data from urban and suburban routes in Beijing across multiple seasons. Using the 0.3 quantile of the conditional survival distribution, the proposed LAD–KM model achieves a mean absolute error of 7.0 km, a mean absolute percentage error of 3.2%, and a predicted-before-actual-failure (PBAF) rate of 79%. Uncertainty is quantified using trip-level bootstrap confidence intervals. The results demonstrate accurate and reliable predictions while limiting late predictions, providing an interpretable decision-support tool for BEB dispatch and charging planning.