A thesis study submitted in partial fulfilment of the requirements for the degree of master of engineering, School of Engineering and Technology

Thesis (M.Eng.) - Asian Institute of Technology, 1991

This study proposes an approach to evaluate reliability of a water supply system consisting of water source, water treatment plant , ground-level clear water storage , distribution pumping station, a nd a lumped demand node in series. The following stochastic factors are taken into account : influent water quality variation , pump failure and repair, a nd demand fluctuations. The expected unserved demand is used to measure the system reliability. The problem considered is that to find t h e optimal production rate of the treatment plant at each period so that the expected unserved demand is minimized throughout the finite time horizon . Under the restriction of the effluent water quality standard, the maximum plant flow is obtained using a computer software for a given influent water quality pattern. The semi-Markov process is assumed for the fluctuation of influent raw water quality pattern. For each influent water quality steady state , the treatment- storage-demand configuration is formulated using dynamic programming to compute the minimum expected unserved demand considering pump failure and repair , and demand variation.