First,smart　card　data　and　the　P-space　representation　were　used　to　construct　the　weighted　and　directed　network.　The　improved　degree　centrality,　closeness　centrality,　the　betweenness　centrality　and　the　PageRank　were　employed　to　identify　important　nodes　and　edges.　Then　the　community　detection　algorithm,　Infomap　algorithm,　was　used　to　divide　the　weighted　network　into　several　communities　based　on　the　passenger　travel　demand　distribution　and　identify　the　interaction　of　passenger　flow　in　the　network.　Finally,taking　Beijing　rail　transit　network　as　a　case　study,　the　analysis　results　show　that　the　method　proposed　in　this　paper　can　effectively　identify　the　key　stations,　sections　and　community　structures　of　the　weighted　and　directed　network,　and　provide　support　for　the　operation　and　management　of　rail　transit　system.　©　2020,　Jilin　University　Press.　All　right　reserved.