Consider the following planning problem in PDDL 𝐼 𝑛 𝑖 𝑡 ( 𝐴 𝑡 ( 𝑃 1 , 𝐵 𝑙 𝑢 𝑒 ) , 𝐴 𝑡 ( 𝑃 2 , 𝐵 𝑙 𝑢 𝑒 ) , 𝐴 𝑡 ( 𝑃 3 , 𝑅 𝑒 𝑑 ) , 𝐴 𝑡 ( 𝑃 4 , 𝑅 𝑒 𝑑 ) , 𝐴 𝑡 ( 𝑃 5 , 𝐵 𝑙 𝑢 𝑒 ) , 𝐴 𝑡 ( 𝑃 6 , 𝑅 𝑒 𝑑 ) , 𝑁 𝑒 𝑥 𝑡 ( 𝑃 1 , 𝑃 2 ) , 𝑁 𝑒 𝑥 𝑡 ( 𝑃 2 , 𝑃 3 ) , 𝑁 𝑒 𝑥 𝑡 ( 𝑃 3 , 𝑃 4 ) , 𝑁 𝑒 𝑥 𝑡 ( 𝑃 4 , 𝑃 5 ) , 𝑁 𝑒 𝑥 𝑡 ( 𝑃 5 , 𝑃 6 ) , 𝑁 𝑒 𝑥 𝑡 ( 𝑃 6 , 𝑃 1 ) ) 𝐺 𝑜 𝑎 𝑙 ( 𝐴 𝑡 ( 𝑃 1 , 𝐵 𝑙 𝑢 𝑒 ) , 𝐴 𝑡 ( 𝑃 2 , 𝐵 𝑙 𝑢 𝑒 ) , 𝐴 𝑡 ( 𝑃 3 , 𝐵 𝑙 𝑢 𝑒 ) , 𝐴 𝑡 ( 𝑃 4 , 𝑅 𝑒 𝑑 ) , 𝐴 𝑡 ( 𝑃 5 , 𝑅 𝑒 𝑑 ) , 𝐴 𝑡 ( 𝑃 6 , 𝑅 𝑒 𝑑 ) ) 𝐴 𝑐 𝑡 𝑖 𝑜 𝑛 ( 𝑆 𝑤 𝑖 𝑡 𝑐 ℎ ( 𝑥 , 𝑦 , 𝑓 𝑟 𝑜 𝑚 , 𝑡 𝑜 ) )   𝑃 𝑅 𝐸 𝐶 𝑂 𝑁 𝐷 : 𝐴 𝑡 ( 𝑓 𝑟 𝑜 𝑚 , 𝑥 ) ∧ 𝐴 𝑡 ( 𝑡 𝑜 , 𝑦 ) ∧ 𝑁 𝑒 𝑥 𝑡 ( 𝑓 𝑟 𝑜 𝑚 , 𝑡 𝑜 ) 𝐸 𝐹 𝐹 𝐸 𝐶 𝑇 : 𝐴 𝑡 ( 𝑓 𝑟 𝑜 𝑚 , 𝑦 ) ∧ ¬ 𝐴 𝑡 ( 𝑓 𝑟 𝑜 𝑚 , 𝑥 ) ∧ 𝐴 𝑡 ( 𝑡 𝑜 , 𝑥 ) ∧ ¬ 𝐴 𝑡 ( 𝑡 𝑜 , 𝑦 ) ) Ignoring the precondition 𝑁 𝑒 𝑥 𝑡 ( 𝑓 𝑟 𝑜 𝑚 , 𝑡 𝑜 ) leads to defining a relaxed problem, whose solution can serve as a heuristic for solving the original problem via search. What holds for this heuristic?