Concerns about environmental protection from harmful chemicals are augmented due to continuously increasing industrialization. To make environment clean and reduce the effects of these chemicals, adsorption technology has got prevalence over other methods in terms of simple experimental design and low cost. In present work, adsorption of metals and dyes was performed on selected adsorbents. Initially, five sorbents (Ficus carica, Aerva javinica, Saccharum arundinaceum, Citrus sinensis and Pongamia pinnata) were taken in account to check their adsorption potential. After initial screening, three sorbents (Ficus carica, Aerva javinica, Saccharum arundinaceum) were selected, which showed significant adsorption (more than 75%) for metals as well as dyes. Experimental work was divided into two steps. In first step, Saccharum arundinaceum was used for adsorption of Cd2+ ions from aqueous solution employing 24 full factorial design experiment. Base-treated Saccharum arundinaceum provided the best adsorption of Cd2+ (98.6055%) at pH 12, 90 minutes shaking time, 0.5 g sorbent amount and 40 ppm initial concentration of sorbate. Equilibrium, kinetic and thermodynamic studies were performed for adsorption of Cd2+ ions onto Saccharum arundinaceum. Equilibrium studies showed good fit to Freundlich, Langmuir, Temkin and Elovich models with R2 values 0.9446, 0.9958, 0.9852 and 0.9433, respectively. Sorption energy calculated by DubininRadushkevich models was found below 1, which is indication of physical nature of sorption of Cd2+ on sorbent surface; as E < 8 kJ/mol is representative of physical sorption and E > 8-16 kJ/mol is due to chemical sorption. Error analysis for equilibrium isotherms was made in order to check the fit of adsorption model, which revealed viability of these models. In thermodynamic analysis, Gibbs free energy (∆Gº), Enthalpy (∆Hº), Entropy (∆Sº) and Sorption Energy were determined and negative values of ∆Gº (–633.44 kJ/mol at 313 K) indicated a spontaneous nature and feasibility of adsorption process. Positive values for enthalpy and entropy revealed the endothermic nature of process. Adsorption kinetics analysis, employing linear and non-linear forms of pseudo-first and pseudo second order equations, was performed. Non-linear forms of pseudo-second order equation showed a good predictive ability, as it shows results close to experimental results (Qe (mg/g) = 18.1707). In second step, adsorption of 13 selected metal ions and 14 dyes was performed on Ficus carica and Aerva javinica. The batch adsorption experiment of metal ions and dyes was achieved onto Ficus carica and Aerva javinica adsorbents by employing best suited conditions determined after parameters optimization by Response Surface Methodology. The Quantitative Structure Property Relationship (QSPR) was developed for determining percent adsorption of different material onto agrowastes. To generate QSPR models for adsorption of metal ions, percent sorption of metal ions onto Ficus carica, adsorption capacity (Q) from Langhmuir and Freundlich (KF) isotherms was taken as dependent variable and properties of metal ions were independent variables. Regression analysis using Microsoft Excel® 2010 showed that most significant descriptors affecting the metal adsorption onto Ficus carica were covalent index, standard reduction potential and covalent radius (R2 > 0.90). Correlation of adsorption capacity, calculated through Langmuir and Freundlich isotherms, also revealed covalent index as the most significant descriptor for the adsorption phenomenon. Metal ions with high value of covalent index (Cu+2 = 5.5) give better sorption on a particular sorbent as compared to metal ions with low covalent index (K+1 = 1.45). QSPR model generated for adsorption of selected metal ions onto Aerva javinica indicated covalent index, polarizability and size of ions as the significant descriptors (R2 = 0.9179). The generated models were validated by leave-one-out-cross-validation (LOOCV) method (Q2 > 0.90). In QSPR modeling for adsorption of dyes onto Ficus carica and Aerva javinica, initially 1666 descriptors were calculated by employing E-DRAGON software. The large pool of descriptors was reduced rationally by heuristic method. The QSPR models were then generated by Stepwise Multiple Linear Regression (SMLR) analysis on the reduced datasets. The QSPR models for adsorption onto Ficus carica identified 4 descriptors and for adsorption onto Aerva javinica identified 5 descriptors. Both the models showed good statistical significance (R2 > 0.9). The descriptors identified in both cases are built on basic molecular properties, which play important role in physical and chemical behavior of molecules. Some of these descriptors are nHAcc, HATS5p and DISPm, which are based on functional group present on the surface of dyes, polarizability and geometry of the molecules respectively, these characteristics have prime importance in attachment of dyes on the surface of any sorbent material.The models were validated by LOOCV method (Q2 = 0.881 and 0.921for Ficus carica and Aerva javinica, respectively). The descriptors in QSPR models were then used to generate Artificial Neural Network (ANN) models. The ANN model was found superior to regression model with high R2 value (0.9) in case of Aerva javinica. The generated QSPR models, by different techniques, showed high statistical significance, robustness and good predictive abilities and thus can be used to analyze and predict the adsorption potential of different pollutants onto selected agrowastes. Studied sorbents were applied for the removal of metal ions (Cr+3, Fe+2, Pb+2, Cu+2, Co+2 and Ni+2) from real samples obtained from different tanneries industries of Sialkot as well. Results reveal the efficient removal of metal ions (up to 85%) from these samples. So studied sorbents are cheaper alternative of the costly treatment techniques used in industries for the removal of these pollutants from industrial effluents. Isothermal, kinetic and thermodynamic investigations as well as analysis of sorption data through prediction technique revealed that proposed sorbent materials have very promising potential for the removal of metal ions as well as dyes from industrial effluents.