For sustainability of groundwater resources, especially in arid environments, this study developed Irrigation Water Quality Indices (IWQIs), employed Machine Learning Algorithms (MLA), and produced GIS maps for planning groundwater for irrigation. 103 groundwater samples were collected and analyzed to achieve the objective of this study. The diagrams created by Piper, Chadha, and USSL were designed to identify groundwater types and processes. Various IWQIs were assessed, including total dissolved solids (TDS), salinity index (SI), potential salinity (PS), magnesium hazard (MH), sodium percentage (Na%), permeability index (PI), kelly ratio (KR), soluble sodium percentage (SSP), chloro-alkaline index (CAI), chlorinity index (CI), corrosivity ratio (CR), sodium adsorption ratio (SAR), and the overall irrigation water quality index (IWQI). Three distinct machine learning models were developed to forecast IWQIs: an artificial neural network (ANN), a decision tree (DT), and a random forest (RF). The findings indicated a neutral to mildly alkaline nature of groundwater with three groups of facies: Na-Cl, Mg-Cl-SO4, and Ca-Mg-HCO3. The IWQIs showed that the groundwater is appropriate for irrigation, even though there are some high levels of PS, Na%, KR, CI, SSP, MH, and IWQI. The CR level suggested that the groundwater should not be transported through metal pipes. The IWQIs maps propose a distinct representation of groundwater quality for irrigation purposes, aimed to guide the development of sustainable groundwater management strategies. The ANN, RF, and DT models have shown a strong connection R2 values (from 0.65 to 1.00) alongside with the lowest mean squared error (MSE) values (from 0.001 to 37.35) in reliably predicting IWQIs.