Although the Earned Value Management system (EVM) is common practice in the evaluation of project cost performance, this method runs into estimate errors early in the project, due to its small sample size and, besides, its outcomes are deterministic and do not consider that some errors could affect input. In order to give a new perspective, a forecasting method is here developed, based on the Kalman filter and the EVM frameworks: it is able to offer probabilistic outlooks of the project final cost and it may be applied since the very beginning without lower degrees of correctness. The algorithm is then applied in three oil and gas projects to evaluate its performances.
L’earned value management system (EVM) è comunemente usato nell’elaborazione di stime a finire relative ai costi di un progetto. Purtroppo il metodo presenta una scarsa accuratezza nella fase iniziale di esecuzione causata dai pochi dati a disposizione. Inoltre, i risultati forniti sono deterministici e non considerano la possibilità che gli input utilizzati possano essere corrotti da errori. In questo lavoro è stata sviluppata una nuova tecnica previsionale basata sui frameworks del filtro di Kalman e del EVM system che produce previsioni probabilistiche del costo finale di un progetto e che può essere usato sin dall’inizio evitando perdite significative in accuratezza. Al fine di valutarne le performances il metodo è applicato a tre progetti nel settore dell’oil & gas.
Probabilistic estimate at completion of a project through the integration of the Kalman filter with the earned value management system methodology
GARZANTI, FABIO
2017/2018
Abstract
Although the Earned Value Management system (EVM) is common practice in the evaluation of project cost performance, this method runs into estimate errors early in the project, due to its small sample size and, besides, its outcomes are deterministic and do not consider that some errors could affect input. In order to give a new perspective, a forecasting method is here developed, based on the Kalman filter and the EVM frameworks: it is able to offer probabilistic outlooks of the project final cost and it may be applied since the very beginning without lower degrees of correctness. The algorithm is then applied in three oil and gas projects to evaluate its performances.File | Dimensione | Formato | |
---|---|---|---|
elaborato_tesi.pdf
accessibile in internet per tutti
Descrizione: Testo Della Tesi
Dimensione
3.5 MB
Formato
Adobe PDF
|
3.5 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/142864