Drinking water distribution networks are exposed to malicious or accidental contamination. Several levels of responses are conceivable. One of them consists of installing a sensor network to monitor the system in real time. Once a contamination has been detected, it is also important to take appropriate counter-measures. The SMaRT-OnlineWDN project relies on modelling to predict both hydraulics and water quality. An online model makes it possible to identify the contaminant source location and perform a simulation of the contaminated area. The sensor system is intended for detection by an early warning system and for online calibration of the transport model. The main objective of this deliverable report is to adapt specified offline sensor placement concepts to be tractable and optimal for large networks. Specification for offline sensor placement designs were given in the SMaRT-OnlineWDN D3.1 and eight objectives were selected for early warning plus one additional for online calibration. Firstly, the definition of contaminant scenarios is refined to concentrate on events with the biggest impact and the eight objectives for designing early warning detection systems are revised to incorporate imperfect sensors with false positive and false negative detections. Then, sensitivity analysis and expert knowledge from the water utility are proposed to find the best locations for self-learning of the hydraulic system state. Sensor technical restrictions such as minimum velocity and pipe diameter and installation costs are considered to refine the selection. Also, criteria that favour the source identification are discussed. Finally, graph decomposition is used to simplify the network to resources, reservoirs and supernodes. Supernodes are 'critical' nodes of degree greater than 2 obtained after removing the forest. It is suggested 1) to generate contaminant events only at supernodes, which reduces significantly the number of events to generate and 2) to seek for optimal sensor locations only at supernodes, which reduces significantly the feasible set. It was found in a further deliverable report D2.5 that the computational effort is substantially reduced with equivalent performance criterion scores.