Identifying water recycling strategy using multivariate statistical analysis for high-tech industries in Taiwan

Abstract

Multivariate statistical techniques of cluster analysis (CA) and discriminant analysis (DA) were applied in this study for the evaluation of water resource management strategies in high-tech industries, on the basis of the existing water use related data of 70 participating plants in Taiwan since 2011. The existing water use data were collected and transformed into detailed water balance charts, and the water use performance at individual plants was evaluated by three indices, namely the “processing water recovery rate”, the “plant water recovery rate”, and the “plant discharge rate”. Results from discriminant analysis showed that increase in the ratios of effluent recycled to pure water system (EPWR) and recycled to secondary water system (ESWR) had positive effects on achieving higher water use performance. On the other hand, process water consumption and ESWR were influential factors in discriminating samples with lower water use performance. The results also confirmed the finding from synergistic effect that improvement on both EPWR and ESWR contributed to the highest water use performance. Opportunities for water recycling in high-tech industries appears to be technically feasible, future efforts could usefully be undertaken to implement further investment on water-use efficiency and novel treatment techniques, and investigation on various reuse purposes.