In the present study, linear and nonlinear regression analysis for packed bed column adsorption of phenol onto corn cob activated carbon was investigated. The activation of the corn cob provided the activated carbon with enhanced surface area and micropore volume of 903.7m2/g and 0.389 cm3/g respectively. The analysis of the physical properties of the corn cob activated carbon (CCAC) revealed that it contained 33.47% of fixed carbon. SEM images indicated the presence of interspatial pores within the matrix of the adsorbent, while the FTIR analysis revealed that the major functional groups in CCAC were alkanol, alkanes, alkyls, carboxylic acids, ethers, esters, and nitro compounds. The effect of the process parameters influencing the dynamic adsorption process was investigated at flow rates (9 – 18mg/min), initial phenol concentration (100-300mg/l), bed height (5 – 10cm), and particle size (300-800µm). Breakthrough time and adsorption capacity increased with an increase in bed height but decreased with an increase in flow rate, initial phenol concentration, and particle size. At 9mg/min flow rate, 100mg/l initial phenol concentration, 10 cm bed height, and 300µm, the breakthrough and saturation points adsorption capacities were 2.143 and 8.570 mg/g respectively, the volume of effluent treated at saturation point was 12.96L, the length of mass transfer zone (MTZ) was 7.50cm, while 66.13% phenol removal efficiency was achieved. The linear and nonlinear regression analysis of the dynamic column adsorption models viz. Thomas, Adam Bohart, and Wolborska fitted better with the experimental data as compared to Yoon–Nelson. Generally, the nonlinear regression analysis proved to be a better tool for dynamic adsorption model analysis because the model parameters it predicted are in higher proximity to the experimental data when compared to those obtained via linear regression analysis. Conclusively, this study has shown that CCAC can successfully be used for the removal of phenol from aqueous solutions. It also provided experimental evidence that for a more accurate analysis of dynamic adsorption models nonlinear regression tool should be considered.