In　order　to　balance　the　traffic　supply　to　meet　the　citizens　demand　for　public　transportation,　reduce　the　pressure　of　urban　traffic,　enhance　the　competitiveness　of　public　travel　and　improve　the　intelligent　transportation　system　service,　this　paper　proposes　a　bus　arrival　time　prediction　algorithm　based　on　Random　Forest.　In　this　paper,　traveling　data　of　the　607　bus　in　Beijing　is　analyzed,　the　data　are　pretreated　by　using　Space　Rectangular　Coordinate　System　instead　of　the　traditional　GPS　Geodetic　Coordinate　System.　The　traffic　junction　number,　travel　distance,　date　type,　time　period,　precipitation,　visibility　Six　kinds　of　influencing　factors　were　utilized　to　model　the　bus　arrival　time　prediction　model　using　Random　Forest.　The　experimental　results　demonstrate　that　the　mean　absolute　percentage　error　of　the　algorithm　is　20.43%　when　under　the　condition　of　setting　800　decision　trees.