Abstract: With the continuous development of financial markets and industrial policies, the number of investment fields and projects available to enterprises is increasing. For an enterprise, the budget for project investment is usually limited, and selecting the set of projects that can maximize the return on investment under financial constraints has been the focus of attention in previous project portfolio selection problems. However, due to the outbreak of sudden public events, a large number of projects have been damaged or even stalled due to a lack of labor and supply chain disruptions, and countless enterprises have suffered from broken capital chains or even collapsed due to failure to meet expected revenue targets on time. Before such events occur, business managers must adjust revenue targets based on the environment and market conditions and make decisions on the project portfolio most likely to achieve the expected goals. At the same time, because of the policy change and the lack of sufficient knowledge of project investment in emerging areas, managers’ project investment behavior often presents a certain degree of uncertainty. Corporate managers have a particular risk appetite when making portfolio investment decisions. Conservative decision-makers will reduce the investment amount to avoid risky losses; aggressive decision-makers will increase the investment amount to seek risky gains. Therefore, in the face of limited investment budgets and uncertain project returns, how enterprises make project portfolio choices based on risk expectations has become a significant challenge for managers and current research.
Based on the above issues and context, this study introduces the concept of the net present value bias coefficient to describe the risk preferences of managers. The information gap decision theory (IGDT) is employed to formulate project portfolio selection models based on risk preferences. Specifically, the envelope constraint is utilized to represent the uncertainty of the investment budget. Based on this, a robust model under the risk-averse strategy and an opportunity model under the opportunity-seeking strategy are constructed according to risk preferences. The objective of the robust model is to find the maximum uncertainty in the investment budget that satisfies the expected net present value; the objective of the opportunity model is to find the minimum uncertainty in the investment budget that satisfies the expected net present value. The constructed bi-level optimization model is transformed into a single-level optimization model for the solution by analyzing the relationship between the net present value and the investment budget. The model also considers critical factors such as active interruption, flexible periods, and financial constraints in project portfolio selection problems.
This paper utilizes a set of example data to solve the problem using the deterministic and IGDT models, obtaining the optimal project portfolio selection results for each model and conducting a comparative analysis. Additionally, an impact ability analysis is performed on the net present value bias coefficient. The results indicate that compared to the deterministic model, the robust model’s optimal solution reduces the number of selected projects to 9, decreasing the net present value of the portfolio selection by CNY 13,160. In contrast, the opportunity model’s optimal solution increases the number of selected projects to 11, increasing the net present value of the portfolio selection by CNY 15,340. Under the premise of meeting the expected net present value returns, the risk-averse strategy achieves a cost-saving ratio of 16.0%; the minimum increase in investment amount under the opportunity-seeking strategy is 11.2%. When the net present value bias coefficient varies within a small range, the best approach to meet the manager’s expected net present value requirements for project selection is to change the investment amount under the risk-averse strategy to retain as much investment cost as possible and under the opportunity-seeking strategy, to minimize investment cost expenditure. Once the bias coefficient exceeds a specific range, the reduction in investment cost does not significantly affect the net present value under the expected returns, and the increase in investment cost leads to a decrease in the benefit. This paper’s model can assist managers in formulating the optimal project portfolio investment strategy based on different risk preferences, providing a reference for managers in project risk control and capital management optimization.
For future research, there are many relevant issues that need to be explored in depth. For example: 1)the cost and value of the project itself fluctuate to a certain extent due to the influence of the environment, and there is a correlation between the two; 2)there is an inflow of external funds in different periods of the project implementation, and these factors directly affect the results of the project portfolio selection. Therefore, in the next stage, we can study the uncertainty of project investment cost, investment return, external funds at various stages, and other factors to extend the IGDT model of project portfolio selection. This would further refine the research on project portfolio selection, making the solution results more practical and applicable.

Key words: project portfolio selection, investment uncertainty, information gap decision theory, active interruption, flexible time horizon, capital constraints

摘要： 受政策的改变以及对新兴领域项目投资缺少足够认知的影响,管理者的项目投资行为往往呈现一定的不确定性。针对企业管理者在进行项目组合投资决策时存在一定风险偏好的情况,本文基于信息间隙决策理论(Information Gap Decision Theory,IGDT),引入净现值偏差系数对管理者的风险偏好进行描述,构建了考虑投资不确定性的项目组合选择新模型。采用包络约束对投资金额不确定性进行描述,给出风险规避与机会寻求两种策略下的项目组合选择IGDT优化模型,并将其转化成单层规划模型进行求解。通过算例仿真,验证了模型的有效性。结果表明:IGDT优化模型不仅满足了管理者投资的预期净现值要求,而且给出了最优的投资金额与项目选择结果。本文构建模型可以为管理者在项目风险管控和资金管理优化方面提供参考。

关键词: 项目组合选择, 投资不确定性, 信息间隙决策理论, 主动打断, 弹性时间段, 资金约束