Today, project-oriented companies often have difficulties completing projects according to schedule. Currently used decision-support solutions allow decision makers to estimate the cost of a new product development project, and compare the estimated cost to the target cost. The proposed approach provides a framework for searching for possible project completion scenarios, which allow us to reach the target cost. The focus of this article is a project prototyping problem described in terms of a constraint satisfaction problem. The proposed method uses artificial neural networks, to identify relationships between variables, and constraint programming, to search for project completion scenarios within the company's resources and project requirements. The results of an experiment indicate that constraint programming provides effective search strategies for finding admissible solutions. Consequently, the proposed approach allows decision makers to obtain alternative project completion scenarios within the limits imposed by project requirements.