We present a simulation-based risk model to analyse the impact of multiple risks on the cost performance of portfolios. The model considers the combined impact of risks affecting the work packages of portfolio's projects and the probabilistic occurrence of each risk. We test the model in a portfolio composed of four construction projects and we show that the model is able to: predict the effect of identified risks on the portfolio cost performance and aid the decision making process of responding to risks. The limitation of the proposed model is that it calculates the impact of risks at a specific date when each risk has a defined probabilistic distribution. In future work we will consider the dynamic nature of risks to enable the model to cope with the changing attributes of risks.