Central forced-air HVAC systems with filters are commonly used in North American homes to reduce particle concentrations and occupant exposure. The filtration performance of these systems is strongly influenced by many system-, filter-, and building-specific parameters including system runtime, cycle on-time, recirculation rate, in-situ filter efficiency, and building air change rate. These parameters vary considerably across homes, and in the same home over time. Thus, characterization of these parameters is essential for filtration system performance evaluation. The goal of this dissertation is to improve the performance of residential filtration systems by advancing the understanding of their influencing parameters. This work first assesses the runtimes in both heating- and cooling-dominated climates. The results show that there were large variations in runtimes and the mean daily runtime was less than 20% in the studied homes. Secondly, this work evaluates the in-situ efficiency of four types of filters in 21 residences in Toronto, Ontario. The results show that the variations for the same type of filters across homes were greater than the differences between filter types. Comparisons between in-situ and lab-tested filter performance further confirm that the lab-tested results of a filter may not well represent its in-situ performance – they generally overestimate the efficiency and underestimate the pressure drops. Lastly, this work examines the range of effectiveness of the HVAC filtration systems with varying influencing parameters through a time-varying mass balance model. The results show that in addition to runtimes, when and how long (i.e., cycle on-time) the system operates also influences its performance. Based on this finding, four filtration operation strategies (concentration, source, pulsed, and concentration pulsed) are evaluated, and the results show that operating the system during periods with high indoor concentrations could achieve a comparable level of effectiveness as continuous operation but at runtimes as low as 80%. This strategy provides opportunities to reduce indoor particle concentration while conserving fan energy use. Overall, this dissertation provides a framework with tools and models to characterize the influencing parameters of residential filtration, evaluate filtration performance, and provide guidance on operation strategies for performance improvement.