YES We show the termination of the TRS R: g(x,|0|()) -> |0|() g(d(),s(x)) -> s(s(g(d(),x))) g(h(),s(|0|())) -> |0|() g(h(),s(s(x))) -> s(g(h(),x)) double(x) -> g(d(),x) half(x) -> g(h(),x) f(s(x),y) -> f(half(s(x)),double(y)) f(s(|0|()),y) -> y id(x) -> f(x,s(|0|())) -- SCC decomposition. Consider the dependency pair problem (P, R), where P consists of p1: g#(d(),s(x)) -> g#(d(),x) p2: g#(h(),s(s(x))) -> g#(h(),x) p3: double#(x) -> g#(d(),x) p4: half#(x) -> g#(h(),x) p5: f#(s(x),y) -> f#(half(s(x)),double(y)) p6: f#(s(x),y) -> half#(s(x)) p7: f#(s(x),y) -> double#(y) p8: id#(x) -> f#(x,s(|0|())) and R consists of: r1: g(x,|0|()) -> |0|() r2: g(d(),s(x)) -> s(s(g(d(),x))) r3: g(h(),s(|0|())) -> |0|() r4: g(h(),s(s(x))) -> s(g(h(),x)) r5: double(x) -> g(d(),x) r6: half(x) -> g(h(),x) r7: f(s(x),y) -> f(half(s(x)),double(y)) r8: f(s(|0|()),y) -> y r9: id(x) -> f(x,s(|0|())) The estimated dependency graph contains the following SCCs: {p5} {p1} {p2} -- Reduction pair. Consider the dependency pair problem (P, R), where P consists of p1: f#(s(x),y) -> f#(half(s(x)),double(y)) and R consists of: r1: g(x,|0|()) -> |0|() r2: g(d(),s(x)) -> s(s(g(d(),x))) r3: g(h(),s(|0|())) -> |0|() r4: g(h(),s(s(x))) -> s(g(h(),x)) r5: double(x) -> g(d(),x) r6: half(x) -> g(h(),x) r7: f(s(x),y) -> f(half(s(x)),double(y)) r8: f(s(|0|()),y) -> y r9: id(x) -> f(x,s(|0|())) The set of usable rules consists of r1, r3, r4, r6 Take the reduction pair: generalized weighted path order (>=MG, >MG) base order 1: max/plus interpretation on natural numbers: f#_A(x1,x2) = max{27, x1} s_A(x1) = x1 + 27 half_A(x1) = max{26, x1 - 11} double_A(x1) = max{13, x1 - 56} g_A(x1,x2) = max{14, x1 - 56, x2 - 12} |0|_A = 5 h_A = 6 base order 2: precedence: half > s = double = g > f# = h > |0| partial status: pi(f#) = [1] pi(s) = [1] pi(half) = [] pi(double) = [] pi(g) = [] pi(|0|) = [] pi(h) = [] The following rules are removed: p1 Then no dependency pair remains. -- Reduction pair. Consider the dependency pair problem (P, R), where P consists of p1: g#(d(),s(x)) -> g#(d(),x) and R consists of: r1: g(x,|0|()) -> |0|() r2: g(d(),s(x)) -> s(s(g(d(),x))) r3: g(h(),s(|0|())) -> |0|() r4: g(h(),s(s(x))) -> s(g(h(),x)) r5: double(x) -> g(d(),x) r6: half(x) -> g(h(),x) r7: f(s(x),y) -> f(half(s(x)),double(y)) r8: f(s(|0|()),y) -> y r9: id(x) -> f(x,s(|0|())) The set of usable rules consists of r1, r3 Take the reduction pair: generalized weighted path order (>=MG, >MG) base order 1: max/plus interpretation on natural numbers: g#_A(x1,x2) = max{6, x2 + 4} d_A = 14 s_A(x1) = max{17, x1 + 5} g_A(x1,x2) = 11 |0|_A = 8 h_A = 23 base order 2: precedence: h > s > g# > d > g > |0| partial status: pi(g#) = [2] pi(d) = [] pi(s) = [1] pi(g) = [] pi(|0|) = [] pi(h) = [] The following rules are removed: p1 Then no dependency pair remains. -- Reduction pair. Consider the dependency pair problem (P, R), where P consists of p1: g#(h(),s(s(x))) -> g#(h(),x) and R consists of: r1: g(x,|0|()) -> |0|() r2: g(d(),s(x)) -> s(s(g(d(),x))) r3: g(h(),s(|0|())) -> |0|() r4: g(h(),s(s(x))) -> s(g(h(),x)) r5: double(x) -> g(d(),x) r6: half(x) -> g(h(),x) r7: f(s(x),y) -> f(half(s(x)),double(y)) r8: f(s(|0|()),y) -> y r9: id(x) -> f(x,s(|0|())) The set of usable rules consists of r1, r3 Take the reduction pair: generalized weighted path order (>=MG, >MG) base order 1: max/plus interpretation on natural numbers: g#_A(x1,x2) = max{2, x2 - 1} h_A = 17 s_A(x1) = x1 + 12 g_A(x1,x2) = max{12, x1 - 37, x2 - 23} |0|_A = 8 base order 2: precedence: h > g# = s > g > |0| partial status: pi(g#) = [] pi(h) = [] pi(s) = [1] pi(g) = [] pi(|0|) = [] The following rules are removed: p1 Then no dependency pair remains.