This research addresses a Multiple-Route Cyclic Vehicle Routing Problem under two routing strategies, namely unidirectional and bidirectional. In the unidirectional strategy, the route direction remains constant, while in the bidirectional strategy, the route direction alternates. Two mixed-integer linear programming (MILP) models are proposed to investigate the structural differences between these strategies. To solve the proposed problems, an Iterated Local Search (ILS) metaheuristic is developed.This approach allows us to experiment with larger-scale benchmark instances. Numerical results highlight the impact of routing strategies on solution quality.