Information retrieval in construction hazard identification

Hyunsoo Kim, Hyun Soo Lee, Moonseo Park, Sungjoo Hwang

Research output: Contribution to conferencePaperpeer-review

1 Scopus citations

Abstract

The repetitive occurrence of similar accidents is one of the most prevalent features of construction disasters. Similar accident cases provide direct information for determining the risks of scheduled activities and planning safety countermeasures. Researchers have developed many systems to retrieve and use past accident cases. However, this research has the limitations of performing too much retrieval to obtain results regarding construction site conditions, failing to reflect characteristics of safety planning steps, or both. To overcome these limitations, this study proposes an accident case retrieval system that can search similar accident cases. It also improves safety planning by using information retrieval and building information modeling. The retrieval system extracts BIM objects and composes a query set combining BIM objects with site information DB. DB compares a query set with past accident cases and seeks the most similar case. Results are then provided to safety managers. Based on the results of this study, safety managers can reduce excessive query generation. Furthermore, they can easily recognize construction site risks by obtaining coordination of objects where similar accidents occurred.

Original languageEnglish
Pages1320-1325
Number of pages6
StatePublished - 2011
Event28th International Symposium on Automation and Robotics in Construction, ISARC 2011 - Seoul, Korea, Republic of
Duration: 29 Jun 20112 Jul 2011

Conference

Conference28th International Symposium on Automation and Robotics in Construction, ISARC 2011
Country/TerritoryKorea, Republic of
CitySeoul
Period29/06/112/07/11

Keywords

  • Accident case
  • BIM
  • Information retrieval
  • Safety management

Fingerprint

Dive into the research topics of 'Information retrieval in construction hazard identification'. Together they form a unique fingerprint.

Cite this