Optimal placement of water-level sensors to facilitate data-driven management of hydrological infrastructure assets in coastal mega-cities of developing nations
Management decisions, including real-time control of hydrological infrastructure assets such as drainage channels or waterways, floodgates, pumping stations, etc. are crucial for the sustainability of flood-prone coastal mega-cities. The veracity of such crucial flood control decisions depends heavily on the availability of city-wide, real-time water-level data which is often lacking in developing countries. Smart sensors can reliably provide the required data, but installing one of these devices in every single point in the hydrological network is not economically feasible. This study proposes a methodology for finding optimal locations for the placement of a limited number of water-level sensors, such that the acquired data are most relevant for facilitating informed decisions about management of the flood control infrastructure in different parts of a coastal city. The proposed methodology entails defining a set of optimisation objectives and constraint, which are then assessed computationally at each potential sensor location through the topological/connectivity analysis of a city-wide, graph-based hydrological infrastructure network. The computed values are then utilised in an optimisation algorithm (NSGA-II) to determine the optimal locations for the placement of a limited number of sensors. The usefulness of the proposed methodology is demonstrated in deploying water-level sensors in the city of Jakarta.