An Optimization Model Using the Standard Deviation Method and Multiple Decision Making Statistics in Water Treatment Plants in Northeastern India

Water treatment plants provide an integral service to both households and industries by supplying an essential component for day to day living as well as for commercial processes. In order to maintain an efficient and sustainable water supply, the various components implicit in the processes in WTPs need to be aligned in an optimal configuration of settings that balances various input parameters. In order to identify these optimal allocations, this study aims to propose an indicator that represents the suitability of the instruments in surface water treatment plants. It utilizes a new adaptation of multi-criteria decision making techniques, the Standard Deviation Method alongside the well adapted Analytical Hierarchy Process for this purpose. In total six criteria, four sub-criteria and twelve alternatives were considered for the study with the global priorities being computed with the help of the AHP and STD methods.

The GMDH modelling framework was also used to evaluate the relation between various input parameters and the indicator. These methods in effect work to identify priority quality parameters of inputs with respect to the overall performance of the instruments. To refine and improve the MCDM methods used, this study developed twelve different models by adapting the number of inputs as well as the types of training algorithms to increase representative accuracy. Results suggest that daily changes in turbidity were the most significant parameter followed by pH affecting the efficiency of the WTP’s key processes.