Korean Journal of Construction Engineering and Management (한국건설관리학회논문집)

Korea Institute of Construction Engineering and Management (한국건설관리학회)
권호 표지
DOI QR코드

DOI QR Code

Cost Estimation of Case-Based Reasoning Using Hybrid Genetic Algorithm - Focusing on Local Search Method Using Correlation Analysis -

혼합형 유전자 알고리즘을 적용한 사례기반추론 공사비예측 - 상관분석을 이용한 지역탐색 기법을 중심으로 -

  • Jung, Sangsun (Department of Architectural Engineering, Seoul National University) ;
  • Park, Moonseo (Department of Architectural Engineering, Seoul National University) ;
  • Lee, Hyun-Soo (Department of Architectural Engineering, Seoul National University) ;
  • Yoon, Inseok (Department of Architectural Engineering, Seoul National University)
  • Received : 2019.10.11
  • Accepted : 2019.12.29
  • Published : 2020.01.31
Download PDF
⟨ Previous Next ⟩

Abstract

Estimates of project costs in the early stages of a construction project have a significant impact on the operator's decision-making in important matters, such as the site's decision or the construction period. However, it is difficult to carry out the initial stage with confidence because information such as design books and specifications is not available. In previous studies, case-based reasoning was used to predict initial construction costs, and genetic algorithms were used to calculate the weight of the inquiry phase among them. However, some say that it is difficult to perform better than the current year because existing genetic algorithms are calculated in random numbers. To overcome these limitations, correlation numbers using correlation analysis rather than random numbers are reflected in the genetic algorithm by method of local search, and weights are calculated using a hybrid genetic algorithm that combines local search and genetic algorithms. A case-based reasoning model was developed using the weights calculated and validated with the data. As a result, it was found that the hybrid GA-CBR applied with local search performed better than the existing GA-CBR.

건설프로젝트 초기 단계의 사업비 추정치는 사업지의 결정이나 공사 기간과 같은 중요한 사항에서 사업자의 의사결정에 상당한 영향을 미친다. 그러나 초기 단계는 설계도서나 시방서 등의 정보가 부족한 채로 진행되기 때문에 신뢰도 있게 수행하기 어렵다. 기존 연구에서는 초기 공사비를 예측하기 위해 사례기반추론을 사용했으며, 그 중 조회 단계의 가중치를 계산하는 방법으로 유전자 알고리즘을 사용했다. 그러나 기존 유전자 알고리즘은 임의의 수로 연산하기 때문에 현재해보다 좋은 성능을 내기 힘들다는 한게가 있다. 이러한 한계를 극복하기 위해 임의의 수가 아닌 상관분석을 이용한 상관계수를 지역탐색의 방법으로 유전자 알고리즘에 반영하고, 지역탐색과 유전자 알고리즘을 결합한 혼합형 유전자 알고리즘으로 가중치를 산정한다. 산정한 가중치를 사용하여 사례기반추론 모델을 개발하고 데이터를 통해 유효성을 검증하였다. 그 결과, 지역탐색을 적용한 혼합형 GA-CBR이 기존 GA-CBR보다 우수한 성능을 보인 것으로 확인되었다.

Keywords

References

  1. AACE Recommended Practice No. 17R-97 (1997). Cost Estimate Classification System, AACE, Inc.
  2. Aamodt, A., and Plaza, E. (1994). "Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches." AI Communications, 7(1), pp. 35-39.
  3. An, S.H., and Kang, J.I. (2005). "A Study on Predicting Construction Cost of Apartment Housing Using Experts̓ Knowledge at the Early Stage of Projects." JOURNAL OF THE ARCHITECTURAL INSTITUTE OF KOREA Structure & Construction, 21(6), pp. 81-88.
  4. Bruno, De Backer., Vincent, Furnon, Philip, Kilby, Patrick Prosser, and Paul, Shaw. (1997). "Local Search in Constraint Programming: Application to the Vehicle Routing Problem." Proc. CP-97 Workshop Indust. Constraint-Directed Scheduling. Schloss Hagenberg Austria.
  5. Christensen, P., and Dysert, L. R. (1997). "Cost estimate classification system." AACE international recommended practice 17R-97.
  6. Christos, Voudouris, and Tsang, E.P.K. (1998). "Solving the radio link frequency assignment problem using guided local search." Proceedings NATO Symposium on Radio Length Frequency Assignment, Sharing and Conservation Systems (Aerospace), Aalborg, Denmark.
  7. Chun, S.H., and Park, Y.J. (2006). "A new hybrid data mining technique using a regression case based reasoning: Application to financial forecasting." Korea Advanced Institute of Science and Technology, 31, pp. 329-336.
  8. E.K. Burke, T. Curtois, G. Post, R. Qu, and B, Veltman. (2007). "A Hybrid Heuristic Ordering and Variable Neighbourhood Search for the Nurse Rostering Problem." European Journal of Operational Research, 188(2), pp. 330-341. https://doi.org/10.1016/j.ejor.2007.04.030
  9. Goh, Y.M., and Chua, D.K.H. (2009). "Case-Based Reasoning for Construction Hazard Identification: Case Representation and Retrieval." Journal of Construction Engineering and Management, 135(11), pp. 1181-1189. https://doi.org/10.1061/(ASCE)CO.1943-7862.0000093
  10. Goldberg, David E., Bradley Korb, and Kalyanmoy Deb. (1989). "Messy genetic algorithms: Motivation, analysis, and first results." Complex systems 3.5, pp. 493-530.
  11. Heikki Maaranen., Kaisa Miettinen., and Antti Penttinen (2007). "On initial populations of a genetic algorithm for continuous optimization problems." Journal of Global Optimization, 37(3), p 405. https://doi.org/10.1007/s10898-006-9056-6
  12. Hwang, J.H., and Kim, S.Y. (2010). "Integer Programming-based Local Search Technique for Linear Constraint Satisfaction Optimization Problem." Journal of the Korea Society of Computer and Information, 15(9), pp. 47-55. https://doi.org/10.9708/jksci.2010.15.9.047
  13. Hwang, J.H. (2010). "An Integration of Local Search and Constraint Programming for Solving Constraint Satisfaction Optimization Problems." Journal of the Korea Society of Computer and Information, 15(5), pp. 39-47. https://doi.org/10.9708/jksci.2010.15.5.039
  14. Ji, S.H., Park, M.S., Lee, H.S., Seong, K.H., and Yoon, Y.S. (2008). "Method of Quantity Data Analysis for Building Construction Cost Estimation : Focusing on Finish Work of Public Apartment Project." Korean Journal of Construction Engineering and Management, KICEM, 9(6), pp. 235-243.
  15. Kang, M.G., Park, S.W., Im, S.J., and Kim, H.J. (2002). "Parameter Calibrations of a Daily Rainfall-Runoff Model Using Global Optimization Methods." Journal of Korea Water Resources Association, 35(5), pp. 541-552. https://doi.org/10.3741/JKWRA.2002.35.5.541
  16. Kim, G.H., An, S.H., and Cho, H.K. (2006). "Comparison of the Accuracy between Cost Prediction Models based on Neural Network and Genetic Algorithm - Focused on Apartment Housing Project Cost." JOURNAL OF THE ARCHITECTURAL INSTITUTE OF KOREA Structure & Construction, 22(3), pp. 111-118.
  17. Kim, G.H., and Kang, K.I. (2003). "A Study on Model of Neural Networks Training by Genetic Algorithms for Predicting Cost Estimates of Apartment Projects at the Early Project Stage." JOURNAL OF THE ARCHITECTURAL INSTITUTE OF KOREA Structure & Construction, 19(10), pp. 133-142.
  18. Kim, G.H., and Kang, K.I. (2004). "A Study on Predicting Construction Cost of Apartment Housing Projects Based on Case Based Reasoning Technique at the Early Project Stage." JOURNAL OF THE ARCHITECTURAL INSTITUTE OF KOREA Structure & Construction, 20(5), pp. 83-92.
  19. Kim, H.H., and Choi, J.Y. (2009). "An Efficient Search Algorithm for Flexible Manufacturing Systems (FMS) Scheduling Problem with Finite Capacity." IE interfaces 22(1), pp. 10-16.
  20. Kim, S.G., Lee, U.K., Cho, H.H., Kang, K.I. (2006). "Decision Support System for Slab Form-work Selection of High-rise Building Construction." JOURNAL OF THE ARCHITECTURAL INSTITUTE OF KOREA Structure & Construction 22(11), pp. 207-214.
  21. Kolodner, J. (1993). Case-Based Reasoning. Morgan Kaufmann Publishers Inc.
  22. Lau, T.L., and Tsang, E.P.K. (1997). "Solving the Processor Configuration Problems with a Mutation-Based Genetic Algorithm." International Journal on Artificial Intelligence Tools, 6.04, pp. 567-585. https://doi.org/10.1142/S0218213097000281
  23. Lee, H.S., Kim, Euntai, and Kim, D. (2005). "Pattern Recognition System Combining KNN rules and New Feature Weighting algorithm." The Institute of Electronics Engineers of Korea - Computer and Information, 42(4), pp. 43-50.
  24. Lee, H.S., Kim, S.Y., Park, M.S., Ji, S.H., Seong, K.H., and Pyeon, J.H.(2011). "A Method of Assigning Weight Values for Qualitative Attributes in CBR Cost Model." Korean Journal of Construction Engineering and Management, 12(1), pp. 53-61. https://doi.org/10.6106/KJCEM.2011.12.1.53
  25. Lee, J.H. (2008). "A Study on the application of Case-based Reasoning with Feature Weighting using Self-Organizing Map." Korean Journal of Business Administration, 21(1), pp. 417-437.
  26. Oh, I.S., Lee, J.S., and Moon, B.R. (2004). "Hybrid genetic algorithms for feature selection." IEEE Tr.Pattern Analysis and Machine Intelligence, 26(11), pp. 1424-1437. https://doi.org/10.1109/TPAMI.2004.105
  27. Paredis. J. (1993). "Genetic State-Space Search for Constrained Optimization Problems." Proceedings of the 13th International Joint Conference on Artificial Intelligence, pp. 967-972.
  28. Park, M.S., Seong, K.H., Lee, H.S., Ji, S.H., and Kim, S.Y. (2010). "Schematic Cost Estimation Method using Case-Based Reasoning: Focusing on Determining Attribute Weight." Korean Journal of Construction Engineering and Management, KICEM, 11(4), pp. 22-31. https://doi.org/10.6106/KJCEM.2010.11.4.22
  29. Park, U.Y., and Kim, G.H. (2007) "A Study on Predicting Construction Cost of Apartment Housing Projects Based on Support Vector Regression at the Early Project Stage." JOURNAL OF THE ARCHITECTURAL INSTITUTE OF KOREA Structure & Construction, 23(4), pp. 165-172.
  30. Qu, R., and He, F. (2008). "A Hybrid Constraint Programming Approach for Nurse Rostering Problems." Applications and Innovations in Intelligent Systems XVI : Proceedings of AI-2008, pp. 211-224.
  31. Sevgi Zeynep Dogan., and David Arditi., and H. Murat Gunaydin. (2006). "Determining Attribute Weights in a CBR Model for Early Cost Prediction of Structural Systems." Journal of Construction Engineering and Management, 132(10), pp. 1092-1098. https://doi.org/10.1061/(ASCE)0733-9364(2006)132:10(1092)
  32. Song, K.R., Jung, E.K., and Im, C.S. (2006). "A Study on the Correlation Analysis of the Item of Expenditure using ISM Method." JOURNAL OF THE ARCHITECTURAL INSTITUTE OF KOREA Structure & Construction, 5, pp. 153-160.