This chapter presents a novel approach for the early identification of safety-critical events in time-series data. Safety-critical events, often challenging to define and mostly beyond our control, can, however, be detected at an early stage and mitigated. Presently, the prevailing method for identifying safety-critical events involves the analysis of real-time naturalistic driving data. Many accidents stem from lane-changing maneuvers, with delays in drivers’ reaction times and judgment errors being significant contributors to these incidents. This chapter delves into the ramifications of unsuccessful lane changes and abrupt accelerations in platooning traffic congestion. To address these issues, a time-series forecasting algorithm is introduced, which leverages statistical analysis tools rooted in kinematic triggers to identify and forecast failed lane changes, thereby aiding in the prevention of safety-critical events.