As　a　new　public　transport　operation　mode,　the　flexible　transit　system　integrates　the　high　efficiency　and　low　cost　of　conventional　public　transportation　with　the　flexibility　of　demand-responsive　public　transportation.　At　present,　it　has　been　successfully　employed　in　many　foreign　cities.　The　operation　results　show　that　flexible　public　transportation　is　an　effective　way　to　solve　the　problem　of　public　transport　service　in　low　travel　demand　density　zones.　This　paper　first　analyzes　five　factors　that　affect　the　planning　of　flexible　public　transportation　stops,　and　constructs　a　node　importance　model　based　on　these　five　factors,　using　K-means　clustering　algorithm　to　determine　the　nodes　suitable　for　planning　bus　stops.　The　results　show　that　the　method　is　feasible　and　practical,　and　can　provide　a　theoretical　basis　for　domestic　enterprises　to　plan　flexible　public　transportation.