In this paper, we consider a nonsmooth nonlinear multiobjective fractional bilevel programming problem (MFBP) in which the objective function of the upper-level is nonlinear fractional vector-valued, the objective function of the lower-level is nonlinear vector-valued and the common feasible region is determined by inequality constraints. Based on the optimistic approach and by using Karush-Kuhn-Tucker type conditions associated to the lower-level problem, we reformulate the bilevel programming problem into a generalized semi-infinite nonsmooth nonlinear multiobjective fractional single-level programming problem with an equality and an infinite number of inequality constraints (MFP). We prove that under appropriate constraint qualification, convexlikeness, and generalized \(d_{I}\) -V-invexity assumptions, the problems (MFBP) and (MFP) are equivalent in the sense that there is correspondance between their global (weakly) efficient solutions. We study the problem (MFP) by considering that each involved function is semidirectionally differentiable in its own direction instead of a same direction, and we establish necessary and sufficient efficiency conditions for a feasible point to be global (weakly) efficient under generalized \(d_{I}\) -V-invexity and a new infineness concept devoted to functions not necessarily locally Lipschitz. From this and the established relationships between the solutions of the problems (MFBP) and (MFP), necessary and sufficient efficiency conditions for the initial bilevel programming problem (MFBP) are derived. It is important to mention that our obtained results can be applied and extended also to other bilevel (multiobjective) optimization problems involving nonsmooth objective and constraints functions not necessarily locally Lipschitz. For illustration and verification purposes, an applicative example is synthesized and designed to illustrate the obtained results. The different steps of this example give insights and ideas on how to design a numerical method that can serve to solve similar problems to the studied one.
