One of the objectives of the ESA ”ROSETTA” mission to the comet 67P was to insert, in August 2014, an orbiter probe around the so-called nucleus of the comet, and to deposit the ”PHILAE” lander at the surface of the nucleus in November 2014. The selection of the landing site and the definition of the release point and initial descent velocity vector were made in the period August to October 2014 on the basis of simulations of the descent trajectory. This requested an assessment of the gravitational and aerodynamic forces on PHILAE. We here describe the so-called RZC model developed to predict the gas environment of 67P in November 2014 and compute the aerodynamic force. We first outline the unusual diffculties resulting from (1) the complexity of the nucleus surface on all scales, (2) the absence of direct measurements of the gas flux at the surface itself, (3) the time-dependence of the gas production induced by the fast nucleus rotation, (4) the need to perform the whole program within less than three months. Then we outline the physical approach adopted to overcome these diffculties, and describe the RZC model which included three differing tools: (1) a set of gasdynamic/gaskinetic codes to compute the vacuum outflow of a rarefied gas mixture from a highly aspherical rotating solid source; (2) an heuristic approach to deal with the solid/gas initial boundary conditions, and (3) an iterative procedure to derive the gas production parameters on the nucleus surface from the observational data acquired from the orbiter probe. The satisfactory operation of the RZC code in the weeks preceding the November 2014 PHILAE descent is shown, and the forecasted aerodynamic force during the PHILAE descent is compared to the gravitational force.