Occupant-controlled heating, ventilating and air conditioning (HVAC) systems. allow inhabitants of open-plan spaces some degree of control over their immediate microclimate. Typically, cooled air is supplied at floor or desktop levels. The amount and direction of air flow is under occupant control. Productivity increases have been attributed to this form of control. This paper proposes a simplified model of the thermal environment created by an occupant-controlled HVAC system and the behavior of the occupants within it. The thermal environment is characterized by individual nodes representing sub-areas of the conditioned space and a single well-mixed ceiling space above the occupied zone. Random processes are used to simulate the comings and goings of individual occupants and their HVAC control behavior. The model is used to identify the parameters which have the largest influence on the energy efficiency. Energy use of task HVAC with occupant sensors is found to be less than that of a conventional HVAC system by 13%. Individual HVAC control requires about 10% more energy than uniform temperature conditions.