Collecting construction equipment information such as the site equipment enter and exit date-time, driver’s name, type, and quantity is essential in construction management. Most construction projects use paper to record the equipment access history. However, manual recording is always labour-intensive and time-consuming. Therefore, this research aims to develop an assistant system, Site Equipment Management Assistant (SEMA), to automate the site equipment management processes. With the introduction of image recognition and multiple objects tracking technologies, the proposed system can extract equipment-related information from raw videos. SEMA is designed as a chatbot system that contains three major modules: data acquisition, information extraction, and information delivery. A deep learning-based model was first trained to automatically recognize and track construction equipment passing by the site monitor. Information such as equipment entering and exiting date-time, type, and quantity would be extracted and stored in a database. A chatbot interface was developed for users to obtain data from the database through an intuitive and easy-to-use interface. A system evaluation and usability test were conducted. The results showed that the system could effectively improve the construction equipment management process. SEMA can save 60.7% of users’ operation time on obtaining related information.
SEMA: A Site Equipment Management Assistant for Construction Management
KSCE J Civ Eng
KSCE Journal of Civil Engineering ; 26 , 3 ; 1144-1162
2022-03-01
19 pages
Article (Journal)
Electronic Resource
English
Ampersand - SEMA-show specials
Online Contents | 2000
|The Big Picture - SEMA Moment.
Online Contents | 2007
|Department Highlights - SEMA Show highlights
Online Contents | 2004
Gross - SEMA. Tricked Out in Vegas
Online Contents | 2000
OEMs Float Trial Balloons at SEMA
Online Contents | 1997
|