Among the many million train journeys taking place every day, not all of them are being measured or monitored for ride comfort. Improving ride comfort is important for railway companies to attract more passengers to their train services. Giving passengers the ability to measure ride comfort themselves using their smartphones allows railway companies to receive instant feedback from passengers regarding the ride quality on their trains. The purpose of this development is to investigate the feasibility of using smartphones to measure vibration-based ride comfort on trains. This can be accomplished by developing a smartphone application, analyzing the data recorded by the application, and verifying the data by comparing it to data from a track inspection vehicle or an accelerometer. A literature review was undertaken to examine the commonly used standards to evaluate ride comfort, such as the BS ISO 2631-1:1997 standard and Sperling's ride index as proposed by Sperling and Betzhold (1956). The literature review has also revealed some physical causes of ride discomfort such as vibrations induced by roughness and irregularities present at the wheel/rail interface. We are the first to use artificial neural networks to map data derived from smartphones in order to evaluate ride quality. Our work demonstrates the merits of using smartphones to measure ride comfort aboard trains and suggests recommendations for future technological improvement. Our data argue that the accelerometers found in modern smartphones are of sufficient quality to be used in evaluating ride comfort. The ride comfort levels predicted both by BS ISO 2631-1 and Sperling's index exhibit excellent agreement.