Abstract—Projects face various risks and opportunities throughout their lifecycle. Integrating risk management into the project progression is a crucial factor for project success and an essential component of project management. This paper aims to evaluate the trade-off between Conditional Net Present Value at Risk (CNPVaR) and Net Present Value (NPV) in a stochastic environment. By assuming the activity duration and cash flow as random variables and setting activity start and execution modes as decision variables, a CNPVaR-NPV trade-off model is established with the objectives of minimizing CNPVaR and maximizing NPV, under priority relationships, execution mode, resource supply, and deadlines constraints. Based on the characteristics of the model, an algorithm combining nondominated sorting genetic algorithm II and Monte Carlo simulation is designed. The HVAC system project of Dongguan Jinyun Digital Park 5# Building is used as a case study to illustrate the effectiveness of the model and the algorithm. The results show that considering the CNPVaRNPV trade-off model can effectively mitigate potential risks and reduce project losses.