Skip to content

Generating integrated bill of materials using mask R-CNN artificial intelligence model

Document Type
Publications · Journal Article
Registered DB
SCIEScopus
Journal Ranking
JCR Q1Scopus Q1
Journal
AUTOMATION IN CONSTRUCTION145 · 2023
Citation
Web of Science 29Scopus 31
FWCI (Scopus)
1.96
JCR IF
JCR 2022 10.3
ISSN
0926-5805

Authors

  1. Ayesha Munira ChowdhuryFirst authorPusan Natl Univ, Dept Civil & Environm Engn, Busan 46241, South Korea
  2. Moon, SungwooCorresponding authorPusan Natl Univ, Dept Civil & Environm Engn, Busan 46241, South Korea

Abstract

© 2022 Elsevier B.V.Concrete formwork is a crucial temporary structure for concrete placement not only in structural design but also in budget planning. Artificial intelligence (AI) possesses great potential in planning and managing this temporary structure and can automate quantity take-off and cost estimation. The objective of this paper is to integrate AI with a cost database for automatically generating formwork bills of materials (BoMs). A BoM-generation AI model (BoM-GAIM) is presented that can automate BoM generation based on Mask R-CNN and the image segmentation technique. The proposed model, trained using the images of 2D CAD formwork drawings, can recognize, classify, and extract formwork components. The mean average precision (mAP) of the trained Mask R-CNN was found to be 98% at intersection-over-union (IoU) = 50% and 85% at IoU = 75%. The optical character recognition (OCR) is also used to extract the numerical information from dimension objects. Finally, integrated with a cost database, the BoM-GAIM generated a BoM for concrete formwork in a user-interface environment. The application of BoM-GAIM streamlines the BoM generation and increases the concrete formwork design efficiency.

Keywords

Bill of material (BoM)Concrete formworkCost database integrationMask R-CNNObject detection and segmentation

Identifiers

Scopus EID
2-s2.0-85141426721
WoS UID
000977050700001