Special　vehicles　(SVs)　are　vehicles　which　conduct　tasks　such　as　the　maintenance　of　urban　roads　and　are　typically　characterized　by　travelling　at　a　lower　speed　at　a　constant　rate　of　speed　within　the　same　lane.　In　order　to　reduce　the　influence　of　SVs,　guidance　zone　is　designed　and　provides　traffic　guidance　suggestions　(TGS)　for　human-driven　vehicles　(HVs)　helping　drivers　for　better　decision　between　car-following　(CF)　and　lane-changing　(LC).　To　verify　the　effectiveness　of　TGS,　an　improved　Dogit-agent-based　model　is　established　to　simulate　the　captive　and　not　captive　choice　of　CF　and　LC　for　different　driver　types　under　TGS,　and　build　the　rules　for　mixed　traffic　flow　of　SV　and　HVs.　Finally,　a　numerical　simulation　with　a　three-lane　system　is　conducted　to　analyze　the　traffic　efficiency　through　a　set　of　indicators,　and　the　results　show　that　the　TGS　can　reduce　the　influence　of　SVs　on　traffic　flow　in　a　specific　occupancy　rates　range,　increase　the　cross-section　traffic　volume　by　about　5%.　The　TGS　also　can　increase　the　average　speed　of　HVs　in　the　lane　behind　SV　by　about　5%　to　30%,　and　increase　traffic　density　to　200%　on　the　underutilized　lane　in　the　raw　space　in　front　of　the　SV.　Special　vehicles　(SVs)　are　vehicles　which　conduct　tasks　with　special　purposes,　such　as　the　maintenance　of　urban　roads.　SVs　typically　move　at　a　low　constant　speed　and　stay　within　the　same　lane,　which　constitute　its　representative　characteristics.　Traffic　guidance　suggestions　(TGS)　provide　drivers　with　road　information　and　route　suggestions,　helping　vehicles　to　leave　the　congested　road　with　the　highest　efficiency.　In　this　paper,　TGS　issued　for　human-driven　vehicles　drivers　and　an　improved　Dogit-agent-based　model　are　established　to　verify　the　effectiveness　of　TGS.　Finally,　a　numerical　simulation　with　a　three-lane　system　is　conducted　to　analyze　the　traffic　efficiency　through　a　set　of　traffic　indicators,　and　the　results　show　that　the　TGS　can　reduce　the　influence　of　SV　on　traffic　flow　in　a　specific　traffic　occupancy　rates　range.　image