Article Preview
TopIntroduction
Mathematical earned value management (EVM) is an efficient technique in analyzing and controlling the performance of a project in terms of cost and time performance. The EVM related researches in terms of schedule, cost and risk can be categorized as follow. For the cost and schedule aspects, initially, Fleming and Koppelman (1988) argued the criteria may influence cost/schedule control system. Considering planned cost into the progress control then discussed by Turner (2003). Lipke (1999) discussed a procedure for the management of the cost and schedule reserves in projects by developing new ratio. Strang (2009) created an measure called Time Performance Index that could be used for benchmarking project performance but unlike EVM approaches, TPI was accurate during and after completion (EVM indexes are valid only during projects since they become 1.0 when the work is complete). Anbari (2003) studied EVM regarding the effectiveness in implementation of the EVM. Moreover, Earned Schedule (ES) presented by Lipke (2003) as a new metric to evaluate the inefficiency and inaccurate measurement of SPI. Subsequently, Henderson (2004) developed ES and also the applicability of ES is discussed by Vandevoorde and Vanhoucke (2006) in another research project. A reliable forecasting method based on statistical modeling was discussed by Lipke et al.(2009).
For the risk aspect, following researches have been carried out in different branches. Boehm (1989) defined the risk management steps and the negative effect of risks on project development cost in software projects. Also, Ward (1999) considered the requirements for the efficiency of project risk management process. Four years later, Chapman and Ward(2003) discussed over insights of project risk management and its techniques and processes.
Currently, new aspect of risk attention in a form of decision support system(DSS) in project risk management have been presented by Fang et al.(2012) for the modeling and management of project risk interactions. So many authors have tried to integrate risk and EVM techniques, among them, Hayashi and Kataoka (2008) applied a risk management method by using EVM data to developing software projects. In this aspect, Na Yin and Jinlin Li (2007) suggested the ratio of earned value over the budget to make EVM more efficient in pre-event exceeding cost risk recognition. Kim (2010) studied common risk performance indexes in the literature and proposed the new cost risk performance index to integrate cost/schedule and risk for mega projects. With regard to Kim's last research, Kamyabniya (2012) proposed new risk performance index to improve the applicability of EVM metrics based on risk factors interactions. Also, since 2000, papers published in risk ranking and analysis has been increased, especially in applying multi-criteria decision making (MCDM) techniques. Strang (2010) demonstrated how MCDM approaches such as Analytical Hierarchy Procedure and Mathematical Programming could be used for decision making in project valuation. Software engineering and management highly faces with risk factors, however, no research was found which incorporated the score of risk factors into project time and cost performance indexes, especially when the sample size is small. Therefore, in this paper, non-parametric bootstrap technique is applied to obtain risk scores which EAC and EAC (t) be calculated based on.
The structure of this paper is arranged as follows: basic principles of the EVM are presented. Then available lacks in the existing metrics of EVM, non-parametric statistical bootstrap technique and relevant descriptions for the proposed model then are introduced. Also, a case study is presented and finally, conclusion is provided.