YES We show the termination of the relative TRS R/S: R: g(x,y) -> x g(x,y) -> y f(|0|(),|1|(),x) -> f(s(x),x,x) f(x,y,s(z)) -> s(f(|0|(),|1|(),z)) S: rand(x) -> x rand(x) -> rand(s(x)) -- SCC decomposition. Consider the non-minimal dependency pair problem (P, R), where P consists of p1: f#(|0|(),|1|(),x) -> f#(s(x),x,x) p2: f#(x,y,s(z)) -> f#(|0|(),|1|(),z) and R consists of: r1: g(x,y) -> x r2: g(x,y) -> y r3: f(|0|(),|1|(),x) -> f(s(x),x,x) r4: f(x,y,s(z)) -> s(f(|0|(),|1|(),z)) r5: rand(x) -> x r6: rand(x) -> rand(s(x)) The estimated dependency graph contains the following SCCs: {p1, p2} -- Reduction pair. Consider the non-minimal dependency pair problem (P, R), where P consists of p1: f#(|0|(),|1|(),x) -> f#(s(x),x,x) p2: f#(x,y,s(z)) -> f#(|0|(),|1|(),z) and R consists of: r1: g(x,y) -> x r2: g(x,y) -> y r3: f(|0|(),|1|(),x) -> f(s(x),x,x) r4: f(x,y,s(z)) -> s(f(|0|(),|1|(),z)) r5: rand(x) -> x r6: rand(x) -> rand(s(x)) The set of usable rules consists of r1, r2, r3, r4, r5, r6 Take the reduction pair: lexicographic combination of reduction pairs: 1. weighted path order base order: max/plus interpretations on natural numbers: f#_A(x1,x2,x3) = max{x1 - 1, x3} |0|_A = 2 |1|_A = 0 s_A(x1) = max{2, x1} g_A(x1,x2) = max{x1, x2 + 1} f_A(x1,x2,x3) = max{4, x1 + 3, x3 + 3} rand_A(x1) = max{6, x1 + 3} precedence: |1| = g = f > f# = |0| = s > rand partial status: pi(f#) = [3] pi(|0|) = [] pi(|1|) = [] pi(s) = [1] pi(g) = [1, 2] pi(f) = [3] pi(rand) = [] 2. weighted path order base order: max/plus interpretations on natural numbers: f#_A(x1,x2,x3) = x3 + 11 |0|_A = 13 |1|_A = 13 s_A(x1) = max{10, x1 + 9} g_A(x1,x2) = max{x1 + 1, x2 + 1} f_A(x1,x2,x3) = max{12, x3 + 11} rand_A(x1) = 1 precedence: f# > |0| = |1| = g = f > s > rand partial status: pi(f#) = [] pi(|0|) = [] pi(|1|) = [] pi(s) = [] pi(g) = [1, 2] pi(f) = [] pi(rand) = [] The next rules are strictly ordered: p2 We remove them from the problem. -- SCC decomposition. Consider the non-minimal dependency pair problem (P, R), where P consists of p1: f#(|0|(),|1|(),x) -> f#(s(x),x,x) and R consists of: r1: g(x,y) -> x r2: g(x,y) -> y r3: f(|0|(),|1|(),x) -> f(s(x),x,x) r4: f(x,y,s(z)) -> s(f(|0|(),|1|(),z)) r5: rand(x) -> x r6: rand(x) -> rand(s(x)) The estimated dependency graph contains the following SCCs: (no SCCs)