A Chance-Constrained Programming Model for the Optimal Stopping Decision of Power R&D Project

Abstract

A power R&D process can be regarded as a jump process of scientific knowledge full of exploration and complexity. Every jump represents a scientific breakthrough or a new discovery in this setting. In this paper, the inter-arrival times are treated as random variables observe arbitrary distributions. The ɑ-optimistic net return of project performance is proposed and a chance-constrained programming model is established to model the power R&D optimal stopping decision problem. Considering the complexity of the model, the stochastic simulation is designed to estimate the values of project return performance and the simultaneous perturbation stochastic approximation (SPSA) algorithm is employed to solve the proposed model. Finally, the effectiveness of the hybrid algorithm and the applicability of the model are illustrated by numerical examples.