Abstract. Unstable approaches remain one of the most persistent safety risks in commercial aviation and are strongly associated with long landing and runway excursion events. While conventional safety monitoring practices primarily rely on threshold-based exceedance detection within stabilized approach criteria, considerably less attention has been given to the underlying energy management processes that precede such outcomes. This study presents an empirical analysis of Quick Access Recorder (QAR) data collected over a twelve-month operational period from a mixed fleet of Boeing 737 NG and MAX aircraft. Flights were classified into nominal and risk subsets, and systematic differences in key approach energy management parameters were examined across multiple altitude bands during the approach phase using a variability-oriented analytical framework. The results demonstrate that Fast on Approach should not be interpreted as an isolated speed exceedance, but rather as a progressive degradation of approach energy management developing well before stabilized approach criteria are formally violated. Fast on Approach was consistently associated with sustained speed deviations, increased thrust modulation, elevated pitch variability, and a higher likelihood of excess energy being carried into the landing phase. An association between Fast on Approach and long landing outcomes was observed, supporting the interpretation of excess approach speed as an intermediate undesired aircraft state linking early energy management deviations to adverse runway outcomes. The study proposes a risk-based, variability-oriented perspective on approach energy management that complements traditional threshold-based monitoring logic. The findings have direct implications for flight data monitoring systems, pilot training programs, and proactive safety management practices aimed at early identification of approach energy management degradation.

Keywords: unstable approach, energy management, fast on approach, long landing, flight data monitoring, threat and error management, runway excursion risk.