<p>In reality, many investors use the surplus capital remaining after investment in the previous year and the reinvestment by profits from projects already in operation, to secure more investment capital under a limited budget. In the uncertain environment, the investment outlay and the investment profit are uncertain variable, so the surplus capital and the reinvestment also become uncertain. This paper proposes a mean–variance project selection model that considers uncertain surplus capital and reinvestment, with the two objectives of maximizing the expected value (mean) of uncertain net present value (NPV) and minimizing the risk (variance) of uncertain NPV, in the framework of uncertainty theory. The model also takes into account mutual exclusiveness and complementariness between projects. The effects of uncertain surplus capital on project selection are analyzed theoretically. The proposed model is solved using the modified binMOJaya algorithm which is a specific parameter-less meta-heuristic algorithm. Through comparison with several other meta-heuristic algorithms, the performance of the proposed algorithm is evaluated. The numerical experiment and sensitivity analysis are also conducted to demonstrate the application of the proposed model.</p>