|Ph.D Student||Goldschmidt Eitan|
|Subject||Development of a Model for Labor Inputs Control in|
Construction Based on Automatic Monitoring
|Department||Department of Civil and Environmental Engineering||Supervisor||Professor Ronie Navon|
|Full Thesis text - in Hebrew|
Previous research indicates the feasibility of developing a model of labor inputs control based on measuring workers' locations on site and converting them into labor inputs. The main goal of the current research was to deepen the understanding of the process of labor inputs control. This was achieved by the development of labor inputs control model, and experiments in an active construction site to test the control model and the main factors affecting its performance.
The control model receives two types of data: the workers’ locations, as measured on site during the workday and the planned data of the controlled project - e.g. physical and geometric data of the building elements, activities and their scheduled execution, data of the planned usage of inputs, etc. Feasibility of the application for the automatic measurement of workers locations at the construction site and the feasibility of developing a "Building Project Model" (BPM) for the planned data of the controlled project, where carried out in the research.
A computerized model which calculates the time invested by each worker, based on the workers' locations during the workday, was developed. The time Calculation is based on an association of the workers locations to activities, performed in three stages: "Geometric association to the building elements", "Geometric association to activities", and "Logical association".
In order to test the control model and analyze the main factors that influence its accuracy, an experimental model was developed and used on site. Changing "the values" of six of the model’s factors and a statistical analysis of the average deviations between the times calculated with the different values of these factors, to the actual invested time, permitted to identify the affecting factors and their values,
The main research conclusions are:
· A dynamic model for labor inputs control based on automatic monitoring can be developed.
· The findings of the experiments were:
o Existing Project Model is suitable to support the control model or even as a platform for the model itself.
o GPS - the only off-the-shelf available technology was found unsuitable for workers’ location measurement.
· The model is able to convert workers’ locations to times invested in performing vertical skeleton activities.
· “Activity duration” and "type of activity" are the two main factors affecting the accuracy of the model.