Journal of the Architectural Institute of Korea (대한건축학회논문집)

Architectural Institute of Korea (대한건축학회)
권호 표지
DOI QR코드

DOI QR Code

Strategies to Enhance Performance in Machine Learning-based Construction Duration Estimation Through Imputation of Missing Data in Actual Construction Duration Dataset

머신러닝기반의 공사기간 예측성능 향상을 위한 실적공기 학습데이터의 결측치 대체방안

  • Lee, Ha-Neul (Dept. of Architectural Eng., Gyeongsang National University) ;
  • Kang, Yun-Ho (Dept. of Architectural Eng., Gyeongsang National University) ;
  • Yun, Yeong-Chae (Dept. of Architectural Eng., Gyeongsang National University) ;
  • Yun, Seok-Heon (Dept. of Architectural Eng., Gyeongsang National University)
  • 이하늘 (경상국립대 건축공학과) ;
  • 강윤호 (경상국립대 건축공학과) ;
  • 윤영채 (경상국립대 건축공학과) ;
  • 윤석헌 (경상국립대 건축공학과)
  • Received : 2023.12.07
  • Accepted : 2024.01.03
  • Published : 2024.02.29
⟨ Previous Next ⟩

Abstract

In construction projects, the timeframe often relies on the project manager's experience or past construction records rather than a quantitative workload analysis. Accurate predictions necessitate estimating based on actual construction duration data, factoring in the workload. However, integrating construction duration predictions into machine learning models requires extensive big data, and missing data is a common challenge. This study aims to enhance the learning performance of construction duration prediction models by employing and comparing various imputation methods in the data preprocessing stage. Suitable imputation methods were proposed for machine learning model training based on the average error rate. Results showed that the median imputation method was the most fitting single imputation method, while the random forest regression imputation method stood out among multiple imputation methods. Additionally, with an increasing volume of data, regression imputation methods within multiple imputation proved more suitable than single imputation methods.

Keywords

Acknowledgement

이 연구는 2023년도 한국연구재단 연구비 지원에 의한 결과의 일부임. 과제번호:NRF-2019R1A2C1005833

References

  1. Cho, B. N., Kim, H. S., & Kang, L. S. (2015). Development of a Neural Network-Based Model for Estimating Working Periods for Each Construction Process, Journal of the Korea Academia-Industrial Cooperation Society, 16(5), 3477-3483. https://doi.org/10.5762/KAIS.2015.16.5.3477
  2. Cheng, M. Y., & Hoang, N. D. (2018). Estimating construction duration of diaphragm wall using firefly-tuned least squares support vector machine. Neural Computing and Applications, 30, 2489-2497.
  3. Hwang, H. S., Kim, K. R., Suh, S. W., Kim, C. D., & Shin, D. W. (2002). Comparative Analysis of Actual Air Conditioning by Factors Influencing Air Calculation - Focused on the Office Building Construction Air Conditioning Standards, Construction Management: Journal of the Korean Institute of Construction Management, 3(3), 84.
  4. Kim, J. S., & Kim, Y. S. (2021). Development of Construction Duration Estimation Model for Urban-type Residential Housing based on Multiple Regression Analysis. LHI Journal of Land, Housing, and Urban Affairs, 12(4), 93-101.
  5. Kim, S. W. (2021). A Study on the Prediction of Skeleton Construction Period using DNN. Master's Thesis, Hanyang University, 34
  6. Kim, Y. J. (2018). Development of a Prediction Model for the Construction Period of Composite Buildings in the Planning Stage of Construction Projects. Master's Thesis, Inha University, 22.
  7. Kim, Y. J. (2020). Comparison of Missing Value Imputation Methods in Meteorological Data. Master's Thesis, Inha University, 7.
  8. Lee, S. R. (2020). Performance Comparison of Missing Value Imputation Algorithms. Master's Thesis, Hankuk University of Foreign Studies, 1.
  9. Mahmoodzadeh, A., Nejati, H. R., & Mohammadi, M. (2022). Optimized machine learning modelling for predicting the construction cost and duration of tunnelling projects. Automation in Construction, 139.
  10. Sanni-Anibire, M. O., Zin, R. M., & Olatunji, S. O. (2021). Developing a machine learning model to predict the construction duration of tall building projects, Journal of Construction Engineering, 4(1), 022-036.
  11. Yoon, H. S. (2023). A Study on the Prediction of Construction Period for High-rise Curtain Wall using Monte Carlo Simulation and Machine Learning. Master's Thesis, Hanyang University.