A mixed integer linear programming model for the optimal synthesis of a biorefinery and its supply chain for producing ethanol from agricultural residue in the Indian context is developed. A superstructure considering multiple options for each processing stage is developed. Mass and energy balance equations along with fuel demand and feedstock availability equations constitute the model constraints. Process selection decisions and the mass flows are the decision variables, and minimization ethanol production cost for a specific production target is the objective function. The model determined the minimum ethanol cost to be Rs. 70/l ($1.05/l) for a biorefinery producing 100 Mg/month of ethanol. Dilute acid pretreatment and simultaneous saccharification and fermentation (SSCF) were the optimal process selections. Pretreatment selection had a significant impact on the ethanol cost. The ethanol cost reduced to Rs. 43.9/l ($0.65/l) for a biorefinery producing 1000 Mg/month of ethanol. (C) 2018 Elsevier Ltd. All rights reserved.