Traditional capacity sizing models based on production cost minimization do not consider the impact of facility construction and start-up time scale on the profitability of plant. For production facilities that link substantially different scales, such as biofuel production facilities that connect large numbers of small bioreactors to large scale processing equipment, the ramp-up to full production may be a significant consideration in the decision on facility scale. This paper presents a modified discounted cash flow method incorporating production growth during start-up, to determine the optimal production scale of systems that may operate at less than full capacity for a substantial period of time. For a case study of ethanol produced by blue-green algae, global sensitivity analysis using Sobols method indicates how exogenous and endogenous variables, including the ramp-up rate, affect the optimal scale and financial viability of the production system. Conditional-value-at-risk minimization shows how the optimal decision depends on the risk inclination of the decision-maker; a risk-neutral decision maker would choose a basic optimal production scale of 59 million liters (ML) of ethanol per year, while a risk-averse decision-maker might prefer to choose a capacity 30% less. (C) 2014 Elsevier Ltd. All rights reserved.