BIM log mining: learning and predicting design commands

This paper develops a framework to learn and predict design commands based upon building information modeling (BIM) event log data stored in Autodesk Revit journal files, which has the potential to improve the modeling efficiency. BIM design logs, which automatically keep detailed records on the mod...

Full description

Saved in:
Bibliographic Details
Main Authors: Pan, Yue, Zhang, Limao
Other Authors: School of Civil and Environmental Engineering
Format: Article
Language:English
Published: 2022
Subjects:
Online Access:https://hdl.handle.net/10356/161070
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-161070
record_format dspace
spelling sg-ntu-dr.10356-1610702022-08-15T01:03:47Z BIM log mining: learning and predicting design commands Pan, Yue Zhang, Limao School of Civil and Environmental Engineering Engineering::Civil engineering Log Data Mining Design Command Prediction This paper develops a framework to learn and predict design commands based upon building information modeling (BIM) event log data stored in Autodesk Revit journal files, which has the potential to improve the modeling efficiency. BIM design logs, which automatically keep detailed records on the modeling process, are the basis of data acquisition and data mining. Long Short-Term Memory Neural Network (LSTM NN), as a probabilistic deep learning model for learning sequential data with varying lengths from logs, is established to provide designers with predictions about the possible design command class in the next step. To demonstrate the feasibility of this method, a case study runs at large design logs over 4 GB from an international design firm for command class prediction. To begin with, useful data retrieved from logs is cleaned and saved in a 320 MB Comma Separated Values (CSV) file with totally 352,056 lines of commands over 289 projects. Subsequently, various design commands are categorized into 14 classes according to their effects and given numerical labels, which are then fed into LSTM NN for training and testing. As a result, the overall accuracy of this particular case study can reach 70.5% in the test set, which outperforms some classical machine learning methods, like k nearest neighbor, random forest and support vector machine. This research contributes to applying a probabilistic LSTM NN with optimal parameters to learn features from designers' subjective behaviors effectively and predict the next possible design command class intelligently towards automation of the design process. Moreover, the three most possible command classes will be offered as the recommendations under the assumption that the correct class tends to appear owning the top three highest probabilities, which can possibly enhance the reliability of predictions. Ministry of Education (MOE) Nanyang Technological University The Start-Up Grant at Nanyang Technological University, Singapore (No. M4082160.030) and the Ministry of Education Tier 1 Grant, Singapore (No. M4011971.030) are acknowledged for their financial support of this research. 2022-08-15T01:03:47Z 2022-08-15T01:03:47Z 2020 Journal Article Pan, Y. & Zhang, L. (2020). BIM log mining: learning and predicting design commands. Automation in Construction, 112, 103107-. https://dx.doi.org/10.1016/j.autcon.2020.103107 0926-5805 https://hdl.handle.net/10356/161070 10.1016/j.autcon.2020.103107 2-s2.0-85079126973 112 103107 en M4082160.030 M4011971.030 Automation in Construction © 2020 Elsevier B.V. All rights reserved.
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Civil engineering
Log Data Mining
Design Command Prediction
spellingShingle Engineering::Civil engineering
Log Data Mining
Design Command Prediction
Pan, Yue
Zhang, Limao
BIM log mining: learning and predicting design commands
description This paper develops a framework to learn and predict design commands based upon building information modeling (BIM) event log data stored in Autodesk Revit journal files, which has the potential to improve the modeling efficiency. BIM design logs, which automatically keep detailed records on the modeling process, are the basis of data acquisition and data mining. Long Short-Term Memory Neural Network (LSTM NN), as a probabilistic deep learning model for learning sequential data with varying lengths from logs, is established to provide designers with predictions about the possible design command class in the next step. To demonstrate the feasibility of this method, a case study runs at large design logs over 4 GB from an international design firm for command class prediction. To begin with, useful data retrieved from logs is cleaned and saved in a 320 MB Comma Separated Values (CSV) file with totally 352,056 lines of commands over 289 projects. Subsequently, various design commands are categorized into 14 classes according to their effects and given numerical labels, which are then fed into LSTM NN for training and testing. As a result, the overall accuracy of this particular case study can reach 70.5% in the test set, which outperforms some classical machine learning methods, like k nearest neighbor, random forest and support vector machine. This research contributes to applying a probabilistic LSTM NN with optimal parameters to learn features from designers' subjective behaviors effectively and predict the next possible design command class intelligently towards automation of the design process. Moreover, the three most possible command classes will be offered as the recommendations under the assumption that the correct class tends to appear owning the top three highest probabilities, which can possibly enhance the reliability of predictions.
author2 School of Civil and Environmental Engineering
author_facet School of Civil and Environmental Engineering
Pan, Yue
Zhang, Limao
format Article
author Pan, Yue
Zhang, Limao
author_sort Pan, Yue
title BIM log mining: learning and predicting design commands
title_short BIM log mining: learning and predicting design commands
title_full BIM log mining: learning and predicting design commands
title_fullStr BIM log mining: learning and predicting design commands
title_full_unstemmed BIM log mining: learning and predicting design commands
title_sort bim log mining: learning and predicting design commands
publishDate 2022
url https://hdl.handle.net/10356/161070
_version_ 1743119593831923712