The effect of electricity peak demand reduction by electricity saving measures in the Japanese residential sector on the power system scale during summer was evaluated through the use of a simulation model developed by the, authors. In order to simulate the electricity peak demand on the power system scale, the model was improved so as to (1) represent the household distribution and residential stock on the power system scale and (2) improve the temporal resolution of the simulation. The proposed model is a bottom-up type model that simulates residential electricity demand based on occupant behavior considering numerous factors, such as family composition, residence floor area, and building insulation level. Therefore, the proposed model can be used to evaluate both occupant behavioral changes and energy conservation technologies. As a result, we determined that the most influential behavioral measure in reducing summer peak demand is turning off the lights. The peak demand reduction effect when 5% of households turned off the lights was 13 MW, which is equivalent to approximately 0.2% of the residential electricity demand during the daytime in summer in the Kansai region. The model also clarified differences in the electricity savings for each countermeasure among several family composition categories. (C) 2015 Elsevier B.V. All rights reserved.