A generalised likelihood uncertainty estimation mixed-integer programming model: Application to a water resource distribution network

s: A framework for incorporating uncertainty in water distribution models that uses generalised likelihood unbiased estimation (GLUE) and mixed-integer programming (MIP) is proposed and applied to a small water distribution system. Model parameters with uncertainty are first modelled in GLUEWIN and the mean estimates are used instead of single-point estimates. The MIP is solved in GAMS using CPlex solver with single-point estimates and then GLUE-generated estimates. There is a large difference between the results from GLUEMIP and those of the general MIP. It is therefore recommended that GLUEMIP framework be used so as to avoid penalties associated with failure to meet demand and for better planning. Subjects: Advanced Mathematics; Discrete Mathematics; Mathematics & Statistics; Operational Mathematics; Science