It is possible to reshape the building's post-control electricity consumption by introducing an electrical storage element. This paper presents a parametric stochastic optimization model, that combined with a heuristic decision-making method, allows to determine the most suitable electrical battery capacity to be installed in a Low Exergy (LowEx) multi-family house located in the temperate climate zone of Middle Europe. In contrast to most common battery models, this optimization is not formulated in terms of the battery payback period. This model does not explicitly consider the battery cost, but aims instead at a twofold objective, namely to minimize the normalized annual electricity cost and simultaneously peak-shave the 15-minute average electrical power. The sensitivity of the model to uncertainties is accounted for by introducing stochastic elements in several model components. The natural variability in the solar electricity production is modeled by using a historical 12-year-data period, and the estimated electricity consumption by occupants is formulated as a probability distribution. Multiple scenarios are hypothesized in order to evaluate the model performance under different conditions. This model assumes a market-based dynamic electricity-pricing scheme. The effect of the dynamic electricity prices is examined by evaluating the model's behavior when subjected to spot market prices from three different geographically close locations. (C) 2014 Elsevier B.V. All rights reserved.