This　work　focuses　on　multi-objective　scheduling　problems　of　automated　manufacturing　systems.　Such　an　automated　manufacturing　system　has　limited　resources　and　flexibility　of　processing　routes　of　jobs,　and　hence　is　prone　to　deadlock.　Its　scheduling　problem　includes　both　deadlock　avoidance　and　performance　optimization.　A　new　Pareto-based　genetic　algorithm　is　proposed　to　solve　multi-objective　scheduling　problems　of　automated　manufacturing　systems.　In　automated　manufacturing　systems,　scheduling　not　only　sets　up　a　routing　for　each　job　but　also　provides　a　feasible　sequence　of　job　operations.　Possible　solutions　are　expressed　as　individuals　containing　information　of　processing　routes　and　the　operation　sequence　of　all　jobs.　The　feasibility　of　individuals　is　checked　by　the　Petri　net　model　of　an　automated　manufacturing　system　and　its　deadlock　controller,　and　infeasible　individuals　are　amended　into　feasible　ones.　The　proposed　algorithm　has　been　tested　with　different　instances　and　compared　to　the　modified　non-dominated　sorting　genetic　algorithm　II.　The　experiment　results　show　the　feasibility　and　effectiveness　of　the　proposed　algorithm.