Lithium based ceramics (Li 2 TiO 3 , Li 4 SiO 4 etc.) in the packed pebble form, used in breeder blankets to breed tritium, are promising candidates for future fusion reactors. Neutron induced trap sites in the ceramic breeder affect the diffusion and inventory of tritium and thereby influence the quantity of tritium recovered. In the present work, a multi model framework has been employed to quantify the number of defects, which can be used to evaluate trap-sites, due to neutron irradiation in porous Li 2 TiO 3 pebbles. This framework is based on a three stage calculation viz. (i) radiation transport, (ii) obtaining the primary knock-on atom (PKA) energy spectrum and (iii) quantifying the ensuing damage by collision cascade simulations. The output of one stage is coupled to the next stage and finally the defect distribution is obtained. We use radiation transport calculations to obtain the neutron spectrum in ceramic pebbles. This is used as an input to SPECTER, a code that uses evaluated cross-sections, to obtain the Primary Knock-on Atom (PKA) energy spectrum. The PKA energy spectrum is then used to find the number and distribution of vacancies and displacements in porous Li 2 TiO 3 pebbles using binary collision approximation (BCA) based Monte Carlo simulations. The vacancy distribution and the number of vacancies in Li 2 TiO 3 have been evaluated for the ITER neutron source using the above frame work. The PKA energy spectrum of Li, Ti and O PKAs, have been obtained using SPECTER, and their average energies are 680, 120 and 320 keV respectively. The collision cascade simulations shows the expected Bragg peak for the number of defects created and has a range of 6-8 microns for O and Li PKA respectively and a fraction of a micron for Ti PKA. The total number of displaced atoms which is equal to the number of vacancies, at the end of the collision cascades, obtained using the above framework has also been used to estimate the displacements per atom (dpa) in Li 2 TiO 3 and this shows a good agreement with the dpa calculated using the NRT model.