<p>Nowadays, failure can occur in almost all systems often at the level of the individual subsystems. In the literature, mostly studies assumed that such failures occur independently. In practice, however, this assumption is often unrealistic, as subsystem failures are generally dependent on each other. Although some studies focused on it, but still many challenges remain open and requires further investigation. To address this issue, in this paper, we discuss the the multi-objective system cost and availability of series–parallel system by considering linear dependence, weak dependence, and strong dependence. The design allocations are the redundancy and repair teams at each subsystem. The Pareto front is generated using a new multi-objective enhanced cuckoo optimization algorithm (MOENCOA) with the respect of a minimum allowable system availability. The best compromise solution is identified through a fuzzy decision method to help the decision-maker and the results are compared to those obtained by the Non-dominated sorting genetic algorithm II (NSGA-II).</p>