| Title |
Investigation of Construction Duration Influence Factors and Comparison of Artificial Intelligence Methods for Estimating Appropriate Construction Duration in Public Sewerage Construction |
| Authors |
Lee, Kangyoung ; Ding, Zishan ; ? ? Im, Jin-bin |
| DOI |
https://dx.doi.org/10.6106/KJCEM.2024.25.6.066 |
| Keywords |
Public Sewerage Pipeline Construction; Construction Duration Estimation Factors; Guideline; Estimation Model |
| Abstract |
Due to the inability to allocate appropriate construction duration, issues such as construction quality defects and poor workmanship are arising. While general construction projects utilize guidelines for determining construction periods, there has been a lack of guidelines that reflect the specific characteristics of public sewerage pipeline construction, such as extensive outdoor work. This study proposes a guidelines for determining appropriate construction duration for public sewerage pipelines by examining perceptions of factors that delay construction timelines through surveys of experts, based on guidelines for general construction. Additionally, to predict appropriate construction duration, real construction data from 83 projects were used, setting the length of new pipelines, total construction cost, and number of months delayed as independent variables, and the construction period as the dependent variable. The prediction model utilized multiple linear regression, support vector regression, and long short-term memory models, and models were evaluated using the mean absolute error, mean absolute percentage error, and root mean square error. The results indicated that the multiple regression analysis model showed the lowest error in all categories, making it the optimal model for predicting public sewerage construction duration. |