Extended abstract Within the framework of an international collaboration between VENUS, EOLE and PROTEUS experimental teams (VEP), project agreements in the fields of experimental reactor physics design and techniques are proposed to be shared between parties. In this context, the Experimental Physics Division (SPEx) of CEA is currently designing a new program devoted to core physics of GEN-III type lattices that requires higher enriched U5 PWR fuel pins. Those lattices are composed of a central 17x17 5wt% enriched $^{235}$U UO$_2$ assembly, surrounded by portions of 3,7% enriched $^{235}$U UO$_2$ PWR fuel pins. The EOLE stockpile requires manufacturing the central assembly, hence increasing the overall costs of the program. The PROTEUS situation is such that its 5% UO$_2$ pins stockpile could be used in those lattices, in place of new fuels to be manufactured. Nevertheless, due to their radial size, bigger than traditional PWR pins, a so-called representativity study must be started in order to verify that, under some constraints, those PROTEUS fuels could lead to adequate lattices characteristics, both in terms of critical mass, but also in terms of spectrum. To check this potential feasibility, several preliminary calculations were performed to determine adequate size and pitch of the mixed mock-up. Amongst them, representativity calculations were made, to optimize both overclad thickness and lattice pitch, compatible with the buffer zone. The targets being a 4.95% enriched UO$_2$ PWR cell, and then assembly, both at hot zero power condition, the process of representativity has been managed in two ways: -a traditional approach only based on conservation of the moderation ratio, and its analysis in terms of sensitivity profiles: the target moderation ratio being fixed once for all by adjusting the overclad thickness -a more innovative method based on optimization of the representativity itself, also based on similarities of the sensitivity profiles, and adding one or more degrees of freedom in technological parameter (clad thickness, pitch, etc…). The paper is split into 4 parts: after a summary of the experimental program needs and their design, the second chapter details the concept of representativity, as the tools required for this study. A section is devoted to the iterative method built to automatically optimize the representativity factor. The third part details the benchmark, the tools used by each party, as well as the covariance data. The last part makes a comparative analysis of results obtained by each participant, as well as the main feasibility conclusions and forthcoming studies.