Application of Grey Theory in Project Progress Prediction - A Case of Marine Construction

• Main Authors: Yuan-pin Kao, 高源斌
• Other Authors: Li-Chung Chao
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/07426783390087945390

碩士 === 國立高雄第一科技大學 === 營建工程研究所 === 100 === During a construction project, there are changes caused by a variety of disturbing factors, resulting in the actual progress deviating from the original plan and making it either ahead of or behind schedule. Hence, in the course of a project, it is required to perform frequent inspections and make predictions about whether or not the actual progress meets the schedule requirement, while analyzing the reasons for lagging behind schedule in order to take remedial measures or make adjustments. Therefore, if project progress control and progress trend forecast are done properly in advance, then timely detection prior to the occurrence of progress being behind schedule will make it possible to take measures and make adjustments to ensure successful project implementation so as to avoid or reduce losses that incur as a result of the progress being behind schedule. A study of application of the grey theory and regression analysis to developing progress prediction model has been carried out. Using actual progress data of marine construction projects for model verification, prediction accuracy of these two models has been compared, which can serve as a reference point for project owners, work inspectors, and construction contractors in progress control and prediction. It was found that both the grey theory model and the regression model achieved the best prediction accuracy when developed using progress data for four consecutive periods and that the two models obtained very close results in prediction error, but the grey theory model attained a slightly better accuracy. However, when developed using progress data comprising more periods, both models had greater prediction errors. Since the grey theory model is able to make numerical forecasts based on limited and incomplete data and it requires only progress data for four periods without any other constraints on prediction, it is suggested that, where no additional conditions are available, the grey model be adopted for predicting project progress.