To　aim　at　the　travelling　salesman　problem　with　time　windows　(TSPTW),　an　ant　colony　optimization　algorithm　with　Mutation　Features　based　on　Magnetic　Field　(MFM-ACOMF)　was　put　forward.　It　improved　the　heuristic　function　in　the　traditional　ant　colony　optimization　(ACO)　algorithm,　to　meet　the　time　requirement　of　customers　and　reduce　the　probability　of　getting　a　local　optimal.　Moreover,　when　it　obtained　the　preliminary　solution　after　all　the　iterations,　a　mutation　strategy　was　used　to　optimize　the　customer　nodes　that　did　not　reach　the　time　window　limit.　The　simulation　results　show　that　the　MFM-ACOMF　algorithm　has　certain　improvement　on　both　the　optimal　solution　quality　and　customer　satisfaction,　compared　with　the　ACO　algorithm.