In this paper we consider a two-level supply chain problem composed of a manufacturer and a third part logistic provider (3PL provider) at an operational level. The manufacturer workshop is a flow-shop and inventory costs are considered. The 3PL provider provides vehicles to the manufacturer and routing costs are considered. Both face tardiness penalty costs, paid for late delivery by the manufacturer to the customer and by the 3PL provider to the manufacturer. A scenario is proposed where the manufacturer dominates the negotiation with the 3PL provider. He imposes the composition of batches for vehicles, the vehicle departure dates and estimates a delivery time. If the 3PL provider respects the estimated delivery times, there is no penalty, otherwise, some tardiness penalty costs apply. Mixed Integer Linear Programming Model and metaheuristic algorithms are proposed for this specific scenario. Some random data sets are generated and the quality and the efficiency of the different methods are compared.