This paper assesses the feasibility of constructing a regional input/output table where only national input/output tables and limited regional data exist. A number of alternative simulation models are developed and applied to the UK table to obtain estimated Scottish tables. These simulated tables are then compared with the actual Scottish table by a variety of methods. The results of the present study are compared with those of previous works in this field. Finally, the effects of introducing more survey-based information into the simulation methods are analysed.