中文摘要：

企业资源规划系统（ERP）项目投资具有很大的风险和不确定性，投资决策一直是困扰决策者的难题．ERP项目投资决策是一个带有潜在随机过程和约束条件的多阶段投资决策问题，包含大量内在关联的投资机会．多段随机规划方法可以较好地解决带有潜在随机过程和约束条件的多阶段决策问题，克服了二项式方法和有限微分方法难以求解多段关联复合期权的弊端．运用多段随机整数规划方法结合ERP系统的投资特点建立了基于实物期权的ERP项目投资决策分析模型，设计了合理的模型求解算法．模型很好地考虑了项目投资过程中未来收益和投人成本的不确定性，相对于传统决策评价方法更加适合于ERP投资决策．

英文摘要：

There are lots of risks and uncertainties in the process of the enterprise resources planning（ERP） project investment. Therefore, it is very difficult for managers to make good investment decision. ERP project investment is a multistage complex investment decision with underlying stochastic processes and constraints, including a lot of investment options. On the base of risk and uncertain analysis, in this research, a multistage stochastic integer programming is employed to build an ERP decision analytical model based on real option and a stochastic programming algorithm. In the past, a few researchers analyzed the investment of ERP project base on real option. On the other hand, many researchers used the Lattice simulation and finite difference method to compute the value of real option in the process of ERP investment. Multistage stochastic integer programming is an approach aimed at finding Optimal decision strategies with respect to given constraints and objective function in complex decision problems. It is much more appropriate to evaluate the compound real options than Lattice simulation and finite difference method, compared with traditional net present value, it is demonstrated that the model is an effective approach to solve the complex ERP investment decision problem.