A better project performance prediction model using fuzzy time series and data envelopment analysis
本文提出一个结合模糊时间序列和数据包络分析的理论框架，利用项目历史绩效数据预测未来表现，帮助项目经理更真实地评估项目前景并及时采取行动。
Earned value management (EVM) is a critical project management methodology that evaluates and predicts project performance from cost and schedule perspectives. The novel theoretical framework presented in this paper estimates future performance of a project based on the past performance data. The model benefits from a fuzzy time series forecasting model in the estimation process. Furthermore, fuzzy-based estimation is developed using linguistic terms to interpret different possible conditions of projects. Eventually, data envelopment analysis is applied to determine the superior model for forecasting of project performance. Multiple illustrative cases and simulated data have been used for comparative analysis and to illustrate the applicability of theoretical model to real situations. Contrary to EVM-based approach, which assumes the future performance is the same as the past, the proposed model can greatly assist project managers in more realistically assessing prospective performance of projects and thereby taking necessary and on-time appropriate actions.