On opportunistic techniques for solving decentralized Markov decision processes with temporal constraints