隨機森林及支持向量機應用於結構BIM建模及出圖人力成本預測能力之比較分析

COMPARISON OF RANDOM FOREST AND SUPPORT VECTOR MACHINE REGRESSION FOR PREDICTION OF BIM LABOR COST ON ARCHITECTURAL MODELING AND PLOTTING

10.6652/JoCICHE.202109_33(5).0006

黃建勳(Chien-Hsun Huang)；謝尚賢(Shang-Hsien Hsieh)

建築資訊塑模 ； 人力成本預測 ； 隨機森林 ； 支持向量機 ； building information modeling ； labor cost prediction ； random forest ； support vector machine

中國土木水利工程學刊

33卷5期（2021 / 09 / 01）

389 - 398

繁體中文

建築資訊塑模（BIM）近年來已廣泛應用於營建產業，因此執行相關作業所增加的人力成本便成為導入BIM作業時的重要議題。為評估機器學習技術能否更準確的預估BIM人力成本，本研究透過台灣某工程公司過去執行的21個案例所記錄的工時記錄資料為基礎，導入了隨機森林和支持向量機技術分別進行BIM執行人力成本預測，並與實務上常用的線型迴歸模型進行比較。結果顯示，在預測建置BIM結構模型人力成本時，隨機森林模型之MSE值為8.693，在所有模型中表現最佳。但在預測結構體施工圖產出的人力成本時，則以基於有效樓地板面積的線性迴歸模型表現最佳，其MSE值為2.186。

In recent years, Building Information Modeling (BIM) has been widely used in the construction industry. The increased labor cost of executing related BIM uses also become an important issue when BIM is adopted into building projects. At present, the BIM labor cost estimation is mainly based on simple linear regression such as the percentage of the total construction cost or multiplying the total floor area by a coefficient. In order to evaluate whether machine learning technologies can more accurately estimate BIM labor costs, this research adopts the timesheets data of BIM tasks recorded in 21 projects executed by an engineering company in Taiwan to build two machine learning models, which are based on Random Forest (RF) and Support Vector Machine (SVM) respectively. The research results show that, based on leave-one-out cross-validation (LOOCV), by comparing the mean absolute error (MAE) and mean square error (MSE) of the two models, the RF model which the MSE is 8.693 and the MAE is 2.307 performs better than others for predicting the BIM labor cost on building structural BIM model. However, for the production of construction working drawing from the BIM model, the best model is the linear regression model based on effective floor area which its MSE is 2.186 and MAE is 1.118. The performance of both RF and SVM models have no significant advantage over the commonly used linear regression models.

1. Ahmad, M. W.,Reynolds, J.,Rezgui, Y.(2018).Predictive modelling for solar thermal energy systems: A comparison of support vector regression, random forest, extra trees and regression trees.Journal of Cleaner Production,203,810-821.
2. Azhar, S.(2011).Building Information Modeling (BIM): Trends, benefits, risks, and challenges for the AEC industry.Leadership and Management in Engineering,11(3),241-252.
3. Breiman, L.(2001).Random forests.Machine Learning,45,5-32.
4. Bryde, D.,Broquetas, M.,Volm, J.(2013).The project benefits of Building Information Modelling (BIM).International Journal of Project Management,31(7),971-980.
5. Coates, P.,Arayici, Y.,Koskela, K.,Kagioglou, M.,Usher,C.,O'Reilly, K.(2010).The key performance indicators of the BIM implementation process.The International Conference on Computing in Civil and Building Engineering
6. Deb, A.(2011).Introduction to soft computing techniques: artificial neural networks, fuzzy logic and genetic algorithms.Soft Computing in Textile Engineering
7. Eastman, C.,Teicholz, P.,Sacks, R.,Liston, K.(2018).BIM Handbook: A Guide To Building Information Modeling For Owners, Designers, Engineers, Contractors And Facility Managers.Hoboken, NJ:Wiley.
8. Giel, B.,Issa, R.(2013).Return on investment analysis of using Building Information Modeling in construction.Journal of Computing in Civil Engineering,27(5),511-521.
9. Huang, C. H.,Hsieh, S. H.(2015).A case study on assessing the productivity of a BIM team in a construction company.The 6th International Conference on Construction Engineering and Project Management,Busan, Korea:
10. Hultquist, C.,Chen, G.,Zhao, K.(2014).A comparison of Gaussian process regression, random forests and support vector regression for burn severity assessment in diseased forests.Remote Sensing Letters,5(8),723-732.
11. JR, J.(2014).BIM Investment: Understanding value, return and models of assessment.38th AUBEA International Conference Website,Sydney:
12. Kononenko, I.,Kukar, M.(2007).Machine learning and data mining introduction to principles and algorithms.Oxford:Woodhead Publ..
13. Lei, C.,Deng, J.,Cao, K.,Xiao, Y.,Ma, L.,Wang, W.,Ma, T.,Shu, C.(2019).A comparison of random forest and support vector machine approaches to predict coal spontaneous combustion in gob.Fuel,239,297-311.
14. Liaw, A.,Wiener, M.(2002).Classification and regression by random forest.R News,2,18-22.
15. Lu, W.,Fung, A.,Peng, Y.,Liang, C.,Rowlinson, S.(2014).Cost-benefit analysis of Building Information Modeling implementation in building projects through demystification of time-effort distribution curves.Building and Environment,82,317-327.
16. Myrans, J.,Everson, R.,Kapelan, Z.(2018).Automated detection of faults in sewers using CCTV image sequences.Automation in Construction,95,64-71.
17. Neelamkavil, J.,Ahamed, S. S.(2012).,Canada:National Research Council of Canada.
18. Nitsche, P.,Stütz, R.,Kammer, M.,Maurer, P..Comparison of machine learning methods for evaluating pavement roughness based on vehicle response.Journal of Computing in Civil Engineering,28(4),04014015.
19. Poh, C.,Ubeynarayana, C.,Goh, Y.(2018).Safety leading indicators for construction sites: A machine learning approach.Automation in Construction,93,375-386.
20. Shin, M.,Lee, H.,Kim, H.(2018).Benefit–cost analysis of Building Information Modeling (BIM) in a railway site.Sustainability,10(11),4303.
21. Sugimoto, Y.,Seki, H.,Samo, T.,Nakamitsu, N.(2016).4D CAD-based evaluation system for crane deployment plans in construction of nuclear power plants.Automation in Construction,71,87-98.
22. Sun, W. Z.,Zhou, G. Q.(2010).KPIs: Analyzing the impact of Building Information Modeling on construction industry in China.2010 IEEE 17Th International Conference on Industrial Engineering and Engineering Management,Xiamen, China:
23. Sun, W.,Shi, M.,Zhang, C.,Zhao, J.,Song, X.(2018).Dynamic load prediction of tunnel boring machine (TBM) based on heterogeneous in-situ data.Automation in Construction,92,23-34.
24. Vapnik, V.(1995).The Nature of Statistical Learning Theory.New York:Springer.
25. Vapnik, V.,Golowich, S. E.,Smola, A.(1997).Support vectormethod for function approximation, regression estimation, and signal processing.Advances in Neural Information Processing Systems 9
26. Walasek, D.,Barszcz, A.(2017).Analysis of the adoption rate of Building Information Modeling.Procedia Engineering,172,1227-1234.
27. Witten, I. H.,Frank, E.,Hall, M. A.,Pal, C. J.(2017).Datamining: practical machine learning tools and techniques.Amsterdam:Morgan Kaufmann.
28. Yu, P.-S.,Yang, T.-C.,Chen, S.-Y.,Kuo, C.-M.,Tseng, H.-W.(2017).Comparison of random forests and support vector machine for real-time radar-derived rainfall forecasting.Journal of Hydrology,552,5292-104.
29. Zhang, W.,Wu, C.,Li, Y.,Wang, L.,Samui, P.(2019).Assessment of pile drivability using random forest regression and multivariate adaptive regression splines.Georisk: Assessment and Management of Risk for Engineered Systems and Geohazards
30. Zhang, W.,Zhang, R.,Wu, C.,Goh, A. T. C.,Lacasse, S.,Liu, Z.,Liu, H.(2020).State-of-the-art review of soft computing applications in underground excavations.Geoscience Frontiers,11(4),1095-1106.
31. Zhou, Y.,Li, S.,Zhou, C.,Luo, H.(2019).Intelligent approach based on random forest for safety risk prediction of deep foundation pit in subway stations.Journal of Computing in Civil Engineering,33(1),05018004.