Assume that an LR(0)-grammar is given. Prove that each string has at most one rightmost derivation. Give an algorithm that checks whether the input string is derivable.
Assume that an arbitrary input string is given. We construct an LR process on that string stepwise. Assume that the current stack of the LR-process is S. We have to decide whether a shift or reduce action is needed (and which rule should be used in the reduction case). The definition of LR(0) grammar guarantees that only one action is possible, and all the information needed to make the decision is contained in State(S). Therefore, we can find the (only possible) next step of the LR-process.