Keywords: safety performance, demobilization, early occupancy, fault tree analysis, minimal cut sets, zone overlap.

In the developing metropolis of Asia, due to its emerging economics, increasing business activities and swarming population, the space need of residents and offices has increased causing a boom in the construction of commercial and residential high-rise buildings. However, for reasons, such as investors' desire to accelerate their return on investment and alleviate financial burdens, or avoid fines due to delayed completion of construction, these buildings are opened, without being 100 percent complete, in the stage of demobilization.

Although these almost-finished buildings have passed various kinds of inspection, such as the inspection of fire alarm system, emergency exit, sprinkle system, electricity facility, plumbing and elevators, and have obtained an occupancy certificate for qualified usage, there are still some miscellaneous activities going on in the building. These completion activities include roof water-proofing, landscaping, entrance paving, hall way ceiling and material moving and cause space utilization conflict termed zone overlap.

In this almost-finished building, occupants and labourers are fall in two independent categories. Their activity behaviours and their concerns are totally different. The new occupants are in the stage of exploring the new environment and the facilities. They require convenience and security of the facilities. On the other hand, the workers are concerned the quality and schedule of the construction. Two independent groups acting in different ways in the same building leads to a conflict of space utilization and increases the probability of accident happening.

The issue of safety in the phase of demobilization and early stage occupancy seldom has been examined in the previous construction management research. The investigation of this topic is practically valuable for today's increasing use of commercial and residential high-rise buildings.

A fault tree model is developed to explain the causes of injury to occupants and labourers. A fault tree is a graphic model by which a particular top event or a failure can be expressed in terms of the parallel or sequential events. In this study, the top event is the failure of safety performance, which refers to occupants or labourers getting injured. Logic gates are used to connect the top event to its corresponding causes which are categorized as enabling, triggering, and support related causes. The causes can be developed into basic events which contribute to the failure.

The fault tree model developed by the authors shows the structure of causal relationships from the top event of failure, which is occupant or labourer injury, to the primary events or causes of the failure. It is comprised of three sub-models which represent injury to occupants due to unsafe stationary activity (unsafe site conditions), injury to occupants due to construction activity, and injury to labourers due to construction activity. The causes of each sub-model are expanded to their respective primary causes. The qualitative analysis of Minimal Cut Sets (MCS) is used to summarize the combinations of basic events that may contribute to the occurrence of the top event.

The authors found fault tree analysis to be a useful method of explaining the causal relationship of occupant or labourer injury in the demobilization and early occupancy stage. Enabling causes of occupant and labourer injury are related to faults in safety awareness programs that could be corrected by establishing sufficient safety programs. For example, orientation programs, activity announcement systems, and safety programs for occupants and labourers should be well managed. Zone overlap problems are an important issue to be considered. The authors recommend that space and time scheduling to avoid conflicts in location overlap and duration overlap be considered more carefully in a future study.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £95 +P&P)