First,　according　to　the　characters　of　tunnel　traffic　flow,　we　adopted　the　amount　of　traffic,　average　vehicle　speed,　occupancy　rate　and　other　indicators　as　the　identification　indicators　of　the　traffic　flow　to　apply　the　fuzzy　inference　method　to　determine　the　fuzzy　rules　and　membership　functions　of　three　factors.　Then,　according　to　the　traits　of　road　tunnel,　we　further　determined　the　membership　function　of　static　factors　of　the　traffic　safety.　Finally,　according　to　the　Fuzhou　tunnel　related　parameters,　we　used　of　AHP　and　fuzzy　evaluation　method　to　detect　the　state　of　traffic　safety　and　used　of　BP　neural　network　algorithm　to　improve　the　detection　accuracy.　The　experimental　results　show　that　the　proposed　detection　method　has　better　objectivity　and　accuracy.　©　2011　IEEE.