The perils of consuming contaminated groundwater are well documented. The present study deals with the quality assessment of the ground waters of Peenya industrial area, Bengaluru, India. The study has been carried out using machine learning techniques of python programming for 30 groundwater samples collected in and around the area, during two seasons (pre- and post-monsoon seasons for the years 2017, 2018 and 2019, and analyzed as per the protocols of American Public Health Association. The present work has been carried out in three phases. The first phase involves the coarse-grain level analysis of water quality components year wise for three years using Principal Component Analysis (PCA). The second phase involves fine-grain analysis for individual critical water quality components and the third phase involves forecasting of the water quality components based on the three- year analyses carried out in this study using time-series analysis models of machine learning. The analysis results revealed that the study area was highly polluted as seen by the non-potability of 76.67% of the samples due to the presence in excess of one or more parameters such as nitrates, hardness, total dissolved solids, calcium, magnesium, pH, fluoride, chloride, iron and chromium. The statistical findings revealed that there is an increasing tendency of the water quality parameter concentrations, though the variations are marginal. In extension with the traditional test, the statistical and predictive analysis approach has aided in revealing the parameters that influence water quality.