Journal of the Korean Institute of Educational Facilities (교육시설 논문지)

Korean Institute of Educational Facilities (한국교육시설학회)
권호 표지
DOI QR코드

DOI QR Code

A Study on the Analysis and Estimation of the Construction Cost by Using Deep learning in the SMART Educational Facilities - Focused on Planning and Design Stage -

딥러닝을 이용한 스마트 교육시설 공사비 분석 및 예측 - 기획·설계단계를 중심으로 -

  • Jung, Seung-Hyun (Dept. of Architectural Engineering, Hongik Univ.) ;
  • Gwon, Oh-Bin (Dept. of Architectural Engineering, Hongik Univ.) ;
  • Son, Jae-Ho (Dept. of Architectural Engineering, Hongik Univ.)
  • Received : 2018.11.15
  • Accepted : 2018.11.29
  • Published : 2018.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

The purpose of this study is to predict more accurate construction costs and to support efficient decision making in the planning and design stages of smart education facilities. The higher the error in the projected cost, the more risk a project manager takes. If the manager can predict a more accurate construction cost in the early stages of a project, he/she can secure a decision period and support a more rational decision. During the planning and design stages, there is a limited amount of variables that can be selected for the estimating model. Moreover, since the number of completed smart schools is limited, there is little data. In this study, various artificial intelligence models were used to accurately predict the construction cost in the planning and design phase with limited variables and lack of performance data. A theoretical study on an artificial neural network and deep learning was carried out. As the artificial neural network has frequent problems of overfitting, it is found that there is a problem in practical application. In order to overcome the problem, this study suggests that the improved models of Deep Neural Network and Deep Belief Network are more effective in making accurate predictions. Deep Neural Network (DNN) and Deep Belief Network (DBN) models were constructed for the prediction of construction cost. Average Error Rate and Root Mean Square Error (RMSE) were calculated to compare the error and accuracy of those models. This study proposes a cost prediction model that can be used practically in the planning and design stages.

Keywords

GOSSCS_2018_v25n6_35_f0001.png 이미지

Fig. 1. Steps of the research process

GOSSCS_2018_v25n6_35_f0002.png 이미지

Fig. 2. Diagram of Single-Layer Perceptron(Rosenblatt, 1958)

GOSSCS_2018_v25n6_35_f0003.png 이미지

Fig. 4. Back Propagation Algorithm Process

GOSSCS_2018_v25n6_35_f0004.png 이미지

Fig. 3. Diagram of Multi-Layer Perceptron

GOSSCS_2018_v25n6_35_f0005.png 이미지

Fig. 5. ReLU & TanH

GOSSCS_2018_v25n6_35_f0006.png 이미지

Fig. 6. DBN Process

GOSSCS_2018_v25n6_35_f0007.png 이미지

Fig. 7. Cost factor in Design planning Step

GOSSCS_2018_v25n6_35_f0008.png 이미지

Fig. 8. Hyper Parameter in R and Java

GOSSCS_2018_v25n6_35_f0009.png 이미지

Fig. 9. ANN Modeling by Hyper Parameter

Table 1. Research using artificial intelligence

GOSSCS_2018_v25n6_35_t0001.png 이미지

Table 2. Selection of Variables in Prior Research

GOSSCS_2018_v25n6_35_t0002.png 이미지

Table 3. Example of Application (Step 2, 3)

GOSSCS_2018_v25n6_35_t0003.png 이미지

Table 4. Data Summary

GOSSCS_2018_v25n6_35_t0004.png 이미지

Table 5. Construction Cost Index by year

GOSSCS_2018_v25n6_35_t0005.png 이미지

Table 6. Hyper-Parameter results (ANN)

GOSSCS_2018_v25n6_35_t0006.png 이미지

Table 7. Error rate & RMSE of ANN model

GOSSCS_2018_v25n6_35_t0007.png 이미지

Table 8. Hyper-Parameter results (DNN)

GOSSCS_2018_v25n6_35_t0008.png 이미지

Table 9. Error rate & RMSE of DNN model

GOSSCS_2018_v25n6_35_t0009.png 이미지

Table 10. Hidden layer test of DBN model

GOSSCS_2018_v25n6_35_t0010.png 이미지

Table 11 Error rate & RMSE of DBN model

GOSSCS_2018_v25n6_35_t0011.png 이미지

Table 12. Comparison of DBN, DNN, ANN

GOSSCS_2018_v25n6_35_t0012.png 이미지

References

  1. AACE. (2016). " Cost Estimate Classification System" AACE International Recommended Practice, No. 18R-97.
  2. Chang, D. H. (2017). "Spacial Distinction and Site Planning of Newly-established Schools in Sejong City by Educational Curriculum and Administrative Clients" master's dissertation, Cheongju University.
  3. Chen, X. (2015) "Stock Price Prediction Via Deep Belief Networks" Doctoral dissertation, University of New Brunswick.
  4. Cho, H. G., and Kim, K. G. (2013). "A comparison of construction cost estimation using multiple regression analysis and neural network in elementary school project." Journal of the Korea Institute of Building Construction., 13(1), pp.66-74. https://doi.org/10.5345/JKIBC.2013.13.1.066
  5. Ferry, D. J., and Brandon, P. S. (1999). "Cost planning of buildings" Fifth Edition, GRANADA.
  6. Goki, S. (2016). "Deep Learning from Scratch"Hanbit Publishing Network Inc.
  7. Han, H. D., and Kim, J. H. and Yoon, J. H. and Seo, J. W. (2011). "Road Construction Cost Estimation Model in the Planning Phase Using Artificial Neural Network" Korean Society of Civil Engineers, 31(6D), pp.829-837.
  8. Hinton, G. E. (2009). "Deep belief networks" Scholarpedia, 4(5), pp.5947. https://doi.org/10.4249/scholarpedia.5947
  9. Hinton, G. E., and Osindero, S., and Teh, Y. W. (2006). "A fast learning algorithm for deep belief nets" Neural computation, 18(7), pp.1527-1554. https://doi.org/10.1162/neco.2006.18.7.1527
  10. Hira, N. A., and Walter, J. C. (1988). "Estimating: from Concept to Completion" Prentice Hall Inc.
  11. Jang, B. T. (2017). "Jang professor's Deep Learning" Hongrung Publishing Company.
  12. Josh, P., and Adam, G. (2017). "Deep Learning: A Practitioner's Approach" O'Reilly Media Inc.
  13. Kang, I. S., and Mun, J. W., and Park, J. C. (2017). "Recent Research Trends of Artificial Intelligent Machine Learning in Architectural Field - Review of Domestic and International Journal Papers -" Journal of the Architectural Institute of Korea Structure & Construction., 33(4), pp.63-68. https://doi.org/10.5659/JAIK_SC.2017.33.4.63
  14. KICT. (2004) "Development of Construction Cost Index" KOREA INSTITUTE of CIVIL ENGINEERING & BUILDING TECHNOLOGY.
  15. Kim, M. H. (2017). "Construct information system construction and open system development plan" Korean Public Procurement Service.
  16. Kim, S. K., and Son, J. H. (2008). "A Study on the Analysis and Estimation of the Construction Cost by Using Artificial Neural Network in the BTL Projects for Educational Facilities" Journal of the Architectural Institute of Korea Structure & Construction., 24(6), pp.135-142.
  17. Korea On-line E-procurement System, http://g2b.go.kr/index.jsp, Site accessed September 21, 2018
  18. Larochelle, H., and Bengio, Y., and Louradour, J., and Lamblin, P. (2009). "Exploring strategies for training deep neural networks" Journal of machine learning research, 10(Jan), pp.1-40.
  19. Park, H. Y,. and Song, Y. S. and Kim, S. K. (2003). "A Study on the Standard Database for Cost Modelling of Apartment Housing Projects" Journal of the Architectural Institute of Korea Structure & Construction, 19(6), pp.177-184.
  20. Rashid, T. (2016). "Make Your Own Neural Network" Createspace Independent Publishing Platform.
  21. Ribeiro, B., and Lopes, N. (2011). "Deep belief networks for financial prediction" In International Conference on Neural Information Processing., Springer Berlin Heidelberg, pp.766-773.
  22. Rosenblatt, F. (1958). "The perceptron: a probabilistic model for information storage and organization in the brain" Psychological review, 65(6), pp.386. https://doi.org/10.1037/h0042519
  23. Sejong City Office of Education, http://sje.go.kr/sje, Site accessed September 21, 2018
  24. Shin, J. M., and Kim, G. H. (2012). "A study on predicting construction cost of educational building project at early stage using support vector machine technique" The journal of Sustainable Design and Educational Environment Research, 11(3), pp.46-54. https://doi.org/10.7743/kiss.2012.11.3.046
  25. Timothy, M. (2015). "Deep Belief Nets in C++ and Cuda C, Volume 1, Restricted Boltzmann Machines and Supervised Feedforward Networks"CreateSpace Inc.