논문 상세보기
pp.79-89
This study proposes a quantitative and systematic evaluation framework for rationally determining investment priorities in maintenance projects for heterogeneous road infrastructures such as bridges and tunnels. In Korea, conventional maintenance decision-making relies significantly on empirical judgments and policy-driven preferences, thus resulting in inefficiencies, inconsistencies, difficulties across facility types, as well as limited transparency in budget allocation. Hence, a multicriteria decision-making model integrating three key indicators–defect (performance), economic value (asset-based benefit), and risk (safety)–is developed. In particular, the economic evaluation introduces the concept of asset-value recovery and employs artificial intelligence-based machine-learning models (i.e., random forest, light gradient boosting machine, and extreme gradient boosting) to estimate reasonable replacement costs and quantify benefits in monetary terms. The proposed model enables the correlation between these quantitative indicators with maintenance project types to prioritize investments by combining benefit scores and risk indices. The case study demonstrates that the proposed framework enhances the objectivity and efficiency of budget allocation and enables data-driven investment prioritization instead of policydependent decisions. Moreover, this approach provides a foundation for transitioning from experience-based decisions to data-driven infrastructure management. This methodology can be further expanded to include probabilistic risk assessment and life-cycle cost-based management frameworks, thus ultimately contributing to sustainable evidence-based decision support systems for national infrastructure asset management.
1. 서론
2. 투자우선순위 평가 방법론
3. 평가지표 정의
3.1. 결함 점수(성능 평가)
3.2. 경제성(자산) 평가
3.3. 위험도 평가
4. 사례 적용 결과
5. 결론
감사의 글
REFERENCES
