<p>Although many studies have been conducted on scheduling problems in a flow-shop environment, only a few papers have investigated flow-shop problems with dedicated machines. In addition, in the most studies transportation times between machines or stages are neglected. This paper considers a scheduling problem in a two-stage flow-shop with dedicated machines and transportation times. Indeed, all jobs should be processed first on a single common machine at stage 1. Then, these semi-finished jobs are transferred from stage 1 to stage 2 by a robot of unitary capacity. At the second stage, each job has to be processed by one of the two dedicated machines according to its type. This problem with makespan minimization is known to be strongly NP-hard. We present a reduced mixed integer programming model for the problem, which is solved by CPLEX. We propose two meta-heuristics: a variable neighborhood search (VNS) and an iterated local search (ILS) to solve approximately the problem. Computational experiments are carried out to evaluate the proposed methods. The experimental results show that the two algorithms generate efficient solutions in a reasonable running time.</p>