Good layout plan leads to in improve machine utilization, part demand quality, efficient setup time, less work-in-process inventory and material handling cost. Cellular Manufacturing (CM) is an application of GTCM is the combination of job shop and/or flow shop. Facility Layout Problem (FLP) for CMS includes both inter-cell layout and intra-cell layout. A bi-level mixed-integer non-linear programming continuous model has been formulated to fully define the problem and the relationship between intra-cell and inter-cell layout design. Facilities are assumed unequal size; operation sequences, part demands, overlap elimination, aisle are considered. The problem is NP-hard; hence, a simulated annealing meta-heuristic employing a novel constructive radial-based heuristic for initialization have been designed and implemented. For the first time, a novel heuristic algorithm has been designed to allocate and displace facilities in radial direction. In order to improve the search efficiency of the developed SA algorithm, the cell size used in the initialization heuristic algorithm is assumed twice as that of the original size of the cells. A real case study from the metal cutting inserts industry has been used. Results demonstrate the superiority of the developed SA algorithm against rival comparable meta-heuristics and algorithms from the literature.