Problem: a(a(x1)) -> b(b(x1)) c(c(b(x1))) -> d(c(a(x1))) a(x1) -> d(c(c(x1))) c(d(x1)) -> b(c(x1)) Proof: Bounds Processor: bound: 3 enrichment: match automaton: final states: {4,3} transitions: d3(70) -> 71* b1(17) -> 18* b1(23) -> 24* c3(85) -> 86* c3(77) -> 78* c3(69) -> 70* c3(91) -> 92* c3(68) -> 69* c1(15) -> 16* c1(5) -> 6* c1(26) -> 27* c1(6) -> 7* b3(92) -> 93* b3(86) -> 87* d1(27) -> 28* d1(7) -> 8* a1(35) -> 36* a1(25) -> 26* d2(57) -> 58* d2(39) -> 40* a0(2) -> 3* a0(1) -> 3* c2(65) -> 66* c2(47) -> 48* c2(37) -> 38* c2(49) -> 50* c2(56) -> 57* c2(38) -> 39* b0(2) -> 1* b0(1) -> 1* b2(50) -> 51* b2(66) -> 67* c0(2) -> 4* c0(1) -> 4* a2(55) -> 56* a2(59) -> 60* d0(2) -> 2* d0(1) -> 2* 1 -> 35,5 2 -> 25,15 6 -> 23* 8 -> 3* 16 -> 17,6 17 -> 55* 18 -> 38,16,17,6,23,4 23 -> 59* 24 -> 38,16,17,6,23,4 25 -> 37* 27 -> 65* 28 -> 69,39,7 35 -> 47* 36 -> 26* 39 -> 49* 40 -> 36,26 48 -> 38* 51 -> 27* 55 -> 77* 57 -> 91* 58 -> 70,50 59 -> 68* 60 -> 56* 67 -> 70,50 70 -> 85* 71 -> 60,56 78 -> 69* 87 -> 57* 93 -> 86* problem: Qed