Improving Forecasting Accuracy of Project Earned Value Metrics: Linear Modeling Approach

Abstract

Accurately forecasting earned value (EV) metrics is a pivotal component of planning and controlling projects. Despite many approaches to forecasting EV metrics, most studies focus on improving the accuracy of estimating final cost and duration. Relatively few improve upon the use of planned value (PV) to predict EV and actual cost (AC). Thus, this paper takes a new approach to increasing the prediction accuracy of EV and AC by further linearly modeling PV. Data from 131 sample projects verify that a new data-transformation formula significantly improves the correlations between PV and EV and between PV and AC. A mathematical modeling procedure then develops EV and AC forecasting models based on PV for four sample projects. Finally, the study evaluates out-of-sample forecasting accuracy using mean absolute percentage error (MAPE). The results show that the proposed methodology improves forecasting accuracy by an average 13.00 and 19.93% for EV and AC, respectively.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 39 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.