<p>The idea behind proposing a robust model in this article is its relevance to real-world scenarios where uncertainty appears in optimization problems. This paper aims to analyze semi-infinite nonsmooth multiobjective fractional programming problem with data uncertainty. To address the nonsmoothness of functions, the functions involved are considered as directionally differentiable. Firstly, the robust necessary and sufficient optimality conditions are established under generalized univexity assumptions. For better comprehension of sufficiency results, an illustrative application is also provided. Further, a dual explored by a few authors: the Jagannathan type dual is formulated for the given primal problem. This dual has been explored due to its parametric form which can be solved with ease by a variety of computational approaches and hence, has potential applications in exploring various mathematical programming problems formulated in real life issues. Finally, the fundamental robust duality theorems namely weak, strong and strict converse duality are developed by employing the generalized univexity of the involved functions.</p>