Abstract An optimization study for a steam-foam drive process in the Provost Upper Mannville B Pool (Bodo Reservoir) located in east-central Alberta was conducted using the Alberta Research Council (ARC) foam transport model. The optimization was done with respect to the foam injection strategy and the concentration of the foaming agent (surfactant). A "discounted net profit" due to foam injection was maximized. Within the scope of this study, oil recoveries of all steam-foam processes investigated were better than the steam-only baseline process. However, not all foam processes were economically viable, optimal processes were dependent on the heavy oil prices. Over a range of heavy oil prices of US$62.90/m3 to US$125.80/m3 [US$10/bbl to US$20/bbl], the single foam slug process with a total foam injection period of 2 to 3 months and a surfactant concentration 0.5% to 1% by mass in the liquid phase of the injected fluids appeared to be the most profitable. Introduction The process of adding foaming agent (surfactant) aqueous solution to the injected steam to form foam in situ has been recognized as a promising technique in recovering heavy oil from underground reservoirs. Foam is well known as a selective blocking agent which can optimize reservoir conformance and minimize steam channelling and gravity override. This paper will demonstrate the necessity of choosing a suitable foam injection strategy for steam-foam drive field applications. Using the foam transport model developed at the Alberta Research Council by Law et al.(1,2), an optimization study for a steam-foam drive process in the Provost Upper Mannville B Pool (Bodo Reservoir) located in east-central Alberta was conducted. In the optimization study, it was impossible to examine all of the operation variables. The objective of this study was to investigate the advantage of using a steam-foam drive process instead of a steam-only drive process. The search for an optimal steam-foam drive process was made by permuting operation variables such as foam injection schedule (single or multiple foam slug process), total length of the foam injection period and surfactant concentration. Other operation variables such as steam injection rate and steam quality were kept constant. Optimization of the steam-foam drive process was performed by maximizing a ""discounted net profit" due to surfactant injection. Foam Transport Model Detailed descriptions of the ARC foam transport model have been given by Law et al.(1,2). Only a very brief review of this model will be given in this paper. The foam transport model, which is based on the physical properties of an alkylaryl sulphonate surfactant (Sun Tech IV), is a fully implicit multicomponent thermal reservoir model having the ability to handle five components and three phases. The five components are water, heavy oil, surfactant and two additives (non-condensible gases or volatile hydrocarbon components). The three phases are the aqueous phase, the oil phase and the gaseous phase. Conservation of mass of all components is governed by Darcy's law and is solved implicitly together with conservation of energy.