Dedicated　lane　(DL)　for　connected　and　autonomous　vehicles　(CAVs)　is　a　potential　solution　to　manage　mixed　traffic　flow　of　CAVs　and　human-driven　vehicles　(HVs)　in　the　near　future.　In　this　paper,　we　propose　a　modified　cell　transmission　model　(CTM)　to　evaluate　the　performance　and　efficacy　of　CAV　DLs　under　different　scenarios　in　mixed　traffic　environments.　We　integrate　the　mixed　traffic　flow　fundamental　diagram　and　logit　model　to　calculate　through　and　lane-change　flow.　The　model　specifically　considers　the　capacity　drop　phenomena　and　driving　rules　on　DLs.　Numerical　experiments　are　conducted　on　a　three-lane　circular　road　to　analyze　the　impacts　of　number,　length,　and　position　of　DLs　under　varying　CAV　penetration　rates　and　traffic　demands　on　the　mixed　traffic　flow.　The　experimental　results　suggest　that　the　greatest　benefits　are　achieved　when　DLs　are　implemented　under　conditions　of　high　traffic　demand　and　low　CAV　penetration　rates.　The　effectiveness　of　setting　different　numbers　of　DLs　varies　under　different　demand　levels　and　penetration　rates.　In　scenarios　where　lanes　have　different　speed　limits,　adjusting　the　horizontal　position　of　the　DLs　yields　more　significant　effects;　setting　DLs　sequentially　from　the　edge　and　in　adjacent　lanes　—　whether　on　high-speed　or　low-speed　lanes　—　is　an　effective　choice.　Subsequently,　based　on　the　overall　utility　of　the　road　segment,　the　conditions　for　implementing　partial　DLs　are　further　analyzed.　The　findings　provide　insights　for　maximizing　the　advantages　of　CAVs,　enhancing　the　orderliness　and　stability　of　mixed　traffic　flow,　and　offer　a　theoretical　basis　for　future　CAV　DL　deployment　and　traffic　management.　©　2025　Elsevier　B.V.