The production of forward jets separated by a large rapidity gap at LHC, the so-called Mueller-Navelet jets, is a fundamental testfield for perturbative QCD in the high-energy limit. Several analyses have already provided us with evidence about the compatibility of theoretical predictions, based on collinear factorization and BFKL resummation of energy logarithms in the next-to-leading approximation, with the CMS experimental data at 7 TeV of centerof-mass energy. However, the question if the same data can be described also by fixed-order perturbative approaches has not yet been fully answered. In this paper we provide numerical evidence that the mere use of partially asymmetric cuts in the transverse momenta of the detected jets allows for a clear separation between BFKL-resummed and fixed-order predictions in some observables related with the MuellerNavelet jet production process.

The agreement between calculations inspired by the resummation of energy logarithms, known as BFKL approach, and experimental data in the semi-hard sector of QCD has become manifest after a wealthy series of phenomenological analyses. However, the contingency that the same data could be concurrently portrayed at the hand of fixed-order, DGLAP-based calculations, has been pointed out recently, but not yet punctually addressed. Taking advantage of the richness of configurations gained by combining the acceptances of CMS and CASTOR detectors, we give results in the full next-to-leading logarithmic approximation of cross sections, azimuthal correlations and azimuthal distributions for three distinct semi-hard processes, each of them featuring a peculiar final-state exclusiveness. Then, making use of disjoint intervals for the transverse momenta of the emitted objects, i.e. $$\kappa $$ κ -windows, we clearly highlight how high-energy resummed and fixed-order driven predictions for semi-hard sensitive observables can be decisively discriminated in the kinematic ranges typical of current and forthcoming analyses at the LHC. The scale-optimization issue is also introduced and explored, while the uncertainty coming from the use of different PDF and FF set is deservedly handled. Finally, a brief overview of , a numerical tool recently developed, suited for the computation of inclusive semi-hard reactions is provided.