Recent Papers

Urban drainage systems are challenged by uncertain rainfall patterns in combination with urbanization and aging infrastructure. This papers investigates machine learning (ML) methods that are suitable for enhancing decision support and operational efficiency in urban drainage management and surpass traditional numerical models. The five reviewed machine learning methods include supervised learning, unsupervised learning, deep learning, reinforcement learning, and hybrid methods. Each method's advantages are discussed, together with applications, revealing reinforcement learning and deep learning techniques as common in academia. Various datasets utilized for the building of models, many of which are collected locally, are also discussed. Urban planners can, therefore, use ML for predicting floods better, optimizing stormwater systems, and developing sustainable urban drainage strategies as components of a resilient urban environment. The findings point to the transformative potential of ML in complex urban drainage issues that need immediate attention.

Copyright © 2024 Udit Kumar Sharma. This is an open access article distributed under the Creative Commons Attribution License.