This　paper　explores　the　effects　of　traffic　visibility　on　characteristics　of　vehicle　speed,　based　on　the　traffic　and　road　weather　data　of　detectors　positioned　on　the　road　segment　between　Jinshan　and　Wuhannan　interchange　of　Jinggangao　(G4)　expressway　in　Hubei　province.　Focusing　on　the　limited　visibility　weather　element,　this　paper　analyzes　the　effects　of　visibility　on　the　mean　speed　and　speed　variation　by　fog　or　non-fog　weather,　or　under　fog　weather　condition　with　different　visibility;　speed　characteristics　are　deeply　explored　from　aspects　of　different　factors　of　traffic　lane,　vehicle　types　and　temporal　hour,　utilizing　detailed　traffic　data　containing　traffic　lane　and　vehicle　type　information.　The　comprehensive　speed　prediction　model　is　built　under　fog　weather　condition　considering　several　factors　based　on　the　classic　traffic　flow　theory　of　Greenshield＇s　V-K　relation,　and　the　goodness-of-fit　of　the　speed　prediction　model　reaches　80%　approximately.　The　research　results　of　this　paper　can　be　useful　or　as　a　reference　to　impacts　analysis　of　limited　visibility　on　traffic　safety,　operation,　and　traffic　capacity　analysis,　or　decision　making　process　of　traffic　control　measures　such　as　variable　speed　limit　under　adverse　road　weather　conditions.　©,　2014,　Science　Press.　All　right　reserved.