CAMP (CAse Matching Prediction): A dynamic forecasting model for periodic expenditures of residential building projects

This dissertation deals with the development of a dynamic forecasting model named CAMP (CAse Matching Prediction). Dynamic and fragmented characteristics are two of the most significant factors that distinguish the construction industry from other industries. Previous forecasting techniques have failed to solve the problems derived from the above characteristics, and do not provide considerable support. CAMP is intended to provide a more precise forecasting by applying Case-based Reasoning (CBR). The CAMP based on CBR logic enables project managers to forecast monthly expenditures with less time and effort by retrieving and referring only projects of a similar nature, while filtering out irrelevant cases included in database. The proposed CAMP model combines and utilizes many of the advantageous features contained in a variety of existing models, namely, ease of use, active involvement of users and appreciation of the underlying logic of s-curve logic. For the purpose of accurate forecasting, (1) the choices of the numbers of referring projects and (2) the better selection among three levels—which include a 20-work package level, a 7-major work package level, and a total sum level analysis, were investigated in detail. It is concluded that selecting similar projects at 5∼6% out of the whole database will produce a more precise forecasting. Through CAMP, the improvement in forecasting monthly expenditures was significant and this confirms its effectiveness. The CAMP model, which suggests the predicted values based on previous projects, is more than just a forecasting methodology; it provides a bridge that enables current data collection techniques to be used within the context of the accumulated information. This will eventually help all the participants in the construction industry to build up the knowledge derived from invaluable experience.