Design of construction production systems: Representation, taxonomy, and design framework

Construction production has been confronted with chronic problems (budget overruns, project delay, and high worker turnover) for a long time. One of the main root-causes of these problems is misconception. Traditional engineers/managers lack the ability to systematically view a construction production/project, which is complex and has its own peculiarities. The construction industry does not have the kind of understanding or theory that allows engineers/managers to perceive and deal with this kind of complexity as effectively as other industries, such as the manufacturing industry. This study provides a road map for dealing with complexity in the construction industry. It is organized into three main modules: abstraction/representation, taxonomy, and system design. The abstraction/representation module allows engineers/managers to create a conceptual model of the construction production system more effectively by employing the basic idea of system theory with graphical and symbolical notations. The taxonomy module enables engineers/managers to classify construction production systems systematically. Some basic concepts of taxonomy and classification schemes are created for the construction production domain. The classification system can aid engineers/managers to store and organize their knowledge. In this study, a classification system is also developed for modeling/designing methods and tools selection systems. System design is the last module which includes a design framework, a menu of design variables/constraints, and a set of design strategies. This module uses the information of the first two modules to design and optimize a construction production system. These modules are illustrated by three examples: earthmoving, steel erection, and construction trades. The results shows that the earthmoving, steel erection, and construction trades. The results shows that the system approach provides an opportunity to design better working conditions than the traditional approach—which is an ad hoc way based on engineers/managers' experience—and the insight into behaviors in a construction production system particularly for a complex problem. The power of viewing construction production as a system also allows for the creation of customized modeling techniques for specific problems, such as a modeling method for the construction trade problem, called staircase scheduling. In summary, this dissertation brings a broad perspective to the construction production domain and opens the door to several fruitful research areas.