In a recent paper published in Optimization Letters, Montoya et al. (Optim Lett 8:1721-1734, 2014 proposed a branch-and-price approach for a multi-skill project scheduling problem. In that paper, an integer linear programming formulation was first introduced which, unfortunately, has a number of inconsistences. At the interest of mathematical rigor, in this note, we refine such formulation.Keywords Project scheduling · Multi-skilled resources · MILP models In a recent paper by Montoya et al. [2] that was published in Optimization Letters, an integer linear programming formulation was introduced for a multi-skill project scheduling problem. Because a number of inconsistencies could be identified in that formulation, in this note we refine such model at the interest of mathematical rigor. It should be noted that in our opinion, the issues found in the second section of the above mentioned paper do not influence the technical contents of Sections 3-7. In fact, in order to develop their branch-and-price approach, the authors started by developing a column generation formulation which is independent from the aspects we discuss in this note.The relevance of this note arises from the fact that models like the one proposed by Montoya et al. [2] are important for practitioners because, often, they do not have the necessary skills to make full use of integer and combinatorial optimization tools, thus I. Correia (B)