No abstract

In a recently completed case-control epidemiological study, the New Jersey Department of Health and Senior Services (NJDHSS) with support from the Agency for Toxic Substances and Disease Registry (ATSDR) documented an association between prenatal exposure to a specific contaminated community water source and leukaemia in female children. An important and necessary step in the epidemiological study was the reconstruction of the historical water supply strategy of the water distribution system serving the Dover Township area, New Jersey. The sensitivity of solutions to: (1) pressure and pattern factor constraints, (2) allowable operational extremes of water levels in the storage tanks, and (3) the non-uniqueness of the water supply solution are analysed in detail. The computational results show that the proposed approach yields satisfactory results for the complete set of monthly simulations and sensitivity analyses, providing a consistent approach for identifying the historical water supply strategy of the water distribution system. Sensitivity analyses indicated that the alternative strategy obtained from the revised objective function and the variation of constraints did not yield significantly different water supply characteristics. The overall analysis demonstrates that the progressive optimality genetic algorithm (POGA) developed to solve the optimization problem is an effective and efficient algorithm for the reconstruction of water supply strategies in water distribution systems.

The New Jersey Department of Health and Senior Services (NJDHSS), with support from the Agency for Toxic Substances and Disease Registry (ATSDR) conducted an epidemiological study of childhood leukaemia and nervous system cancers that occurred in the period 1979 through 1996 in Dover Township, Ocean County, New Jersey. The epidemiological study explored a wide variety of possible risk factors, including environmental exposures. ATSDR and NJDHSS determined that completed human exposure pathways to groundwater contaminants occurred in the past through private and community water supplies (i.e. the water distribution system serving the area). To investigate this exposure, a model of the water distribution system was developed and calibrated through an extensive field investigation. The components of this water distribution system, such as number of pipes, number of tanks, and number of supply wells in the network, changed significantly over a 35-year period (1962--1996), the time frame established for the epidemiological study. Data on the historical management of this system was limited. Thus, it was necessary to investigate alternative ways to reconstruct the operation of the system and test the sensitivity of the system to various alternative operations. Manual reconstruction of the historical water supply to the system in order to provide this sensitivity analysis was time-consuming and labour intensive, given the complexity of the system and the time constraints imposed on the study. To address these issues, the problem was formulated as an optimization problem, where it was assumed that the water distribution system was operated in an optimum manner at all times to satisfy the constraints in the system. The solution to the optimization problem provided the historical water supply strategy in a consistent manner for each month of the study period. The non-uniqueness of the selected historical water supply strategy was addressed by the formulation of a second model, which was based on the first solution. Numerous other sensitivity analyses were also conducted using these two models. Both models are solved using a two-stage progressive optimality algorithm along with genetic algorithms (GAs) and the EPANET2 water distribution network solver. This process reduced the required solution time and generated a historically consistent water supply strategy for the water distribution system.