Evaluation of project cost and schedule performance using fuzzy theory-based polynomial function

Abstract

Earned value method (EVM) is used for cost performance forecasting. But, it has not been used for time performance forecasting. Also, the extended methods to EVM for improving the time forecasting have some limitations and deterministic evaluation. Therefore, this research introduces an integrated tool called the fuzzy project performance evaluation (FPPE) system, which is mainly depending on fuzzy theory and polynomial function to evaluate the project performance forecasting (PPF) under uncertainties. Two actual projects are provided to demonstrate the advantages of FPPE system over traditional time forecasting methods (TTFMs). The goal to use the polynomial function is to obtain the best-fit curve that simulates the planned cumulative progress curve (PCPC) of a project. The best two degrees were 11 and 10 for the two Projects A and B respectively. Then, the best-fit curve is updated in the light of new work performed percent to obtain time forecasts, which have been compared against that produced by the critical path method. Such a comparative study reveals that the best-fit provided the most accurate time forecast with the lowest mean absolute percentage of error (MAPE_a = 0.73% and MAPE_b = 0.41%). Then, the next step is to incorporate the best-fit procedures into the fuzzy theory to develop the FPPE system, which can be used to help the contractors through detecting early warning points of project performance deterioration and then making corrective decisions. The new integrated system FPPE has been programmed in M-file for MATLAB software and it can be applied to all kind of projects because it works depending on the three EVM curves at the summary of project level.

Keywords:

• Earned value
• evaluation
• fuzzy sets
• forecasting
• system
• polynomial function
• planned duration

Disclosure statement

This research is a part from Ph.D. thesis (Abdelazeem 2019). It is prepared by the first author and supervised by the second author.

Data availability statement

Data generated or analysed during the study are available from the corresponding author by request.