While Imperial Oil continues to expand its heavy oil in situ thermal operations at Cold Lake (Fawcett et al., 2011), some cyclic steam stimulation (CSS) wells are evolving to a mature (late life) stage after decades of successful operation. One strategy that has been used in Cold Lake to improve recovery beyond CSS is to implement injector-only-infill (IOI) wells, targeting the cold reservoir region isolated from the CSS depleted zones. The cyclic IOI process can transition to a continuous, low pressure operation; i.e. infill steamflooding. A three year field trial at Cold Lake pads H01/H02 has demonstrated that an oil recovery level of 65% can be achieved by converting mature CSS areas to an infill steamflood (Stark, 2011). During the course of the trial operation and the subsequent commercial deployment of infill steamflooding, significant reservoir simulation studies have been conducted. Simulation has played an important role in providing a physics-based understanding of the infill steamflood process at Cold Lake by enhancing the understanding of how gravity drainage, inter-well communication, and out-of-pattern steam migration play key roles in the steamflooding recovery process. This paper focuses on validation of the reservoir simulation models through comparison to a variety of field performance data types (e.g. well production data, production well temperature logs). Optimization of the infill steamflood process remains a key focus area for maximizing oil recovery at Cold Lake with reservoir simulation providing guidance on various improvement opportunities including steam injection strategy, infill well and production well completion strategy and mitigation of out-of-pattern steam migration.