Introduction: Curve squeal is a loud tonal noise from railway traffic that contributes to environmental noise pollution. Squealing arises from friction-induced vibrations in the wheel-rail contact during vehicle curving under specific friction conditions. Environmental factors are known to influence these friction conditions and, consequently, the tendency of squeal occurrence. This study investigates whether environmental variables measured from the wayside of the track can be used to predict curve squeal using machine learning models.

Methods: Separate prediction models were developed for low-rail and high-rail squeal using a labeled dataset collected during a 19-month measurement campaign at a commuter railway curve in Sweden. Multiple machine learning algorithms were evaluated, and feature importance analysis was used to investigate the environmental predictors associated with low-rail and high-rail squeal. The practical usefulness of the prediction models was further assessed for adaptive top-of-rail friction modifier (TOR-FM) application.

Results: Environmental variables contained useful predictive information regarding curve squeal occurrence, although the overall predictive performance remained limited. The tree-based ensemble models generally achieved the highest performance, while the feature importance analysis indicated that low-rail and high-rail squeal relied on different feature patterns. The operational assessment demonstrated that the prediction models could reduce unnecessary TOR-FM triggering while maintaining high squeal detection rates.

Discussion: The results demonstrate the potential of combining environmental monitoring with machine learning to support site-specific predictive railway noise mitigation and maintenance decision-making. The differences between low-rail and high-rail prediction further indicate the value of treating the two squeal types as separate prediction tasks.

Railway curve squeal is a significant source of environmental noise, arising from friction-induced instabilities in the wheel–rail contact. These instabilities are influenced by prevailing friction conditions which are affected by environmental factors such as humidity and temperature. This study presents a statistical analysis of long-term curve squeal measurements from a curve operated by commuter trains in Sweden. The analysis focuses on the relationship between environmental variables and squeal occurrence, distinguishing between squeal generated on the low and high rail. The results reveal distinct differences in squeal tendencies. Low rail squeal is most likely during relative rail humidity 55%–75%, absolute humidity 7–9 g/m3, and temperatures 7–15 °C, with peak occurrence in the morning hours and during the autumn season. The probability decreases notably outside these ranges. A specific range of estimated friction coefficients is also associated with low rail squeal. Conversely, high rail squeal exhibits increased probability during dry conditions and elevated temperatures. Both low rail and high rail squeal probabilities are reduced during low temperatures, rail temperatures close to the dew point, high relative humidity, and during the winter season. The observed differences suggest that separate mechanisms may be responsible for squeal on the low and high rail, involving wheel tread contact on the low rail, and wheel flange or two-point contact on the high rail. The results provide new insights into the environmental dependencies of low rail and high rail generated curve squeal, which can support the development of targeted squeal noise mitigation strategies.