Drug delivery system is an important concept in pharmaceutical field, it represents study the drug movement inside body as function of time. Extravascular routes are one of the important rout of drug administration, it consist of, intramuscular, transdermal, and oral routes like solid oral dosage forms. In this study100 mg Diclofenac Sodium (DS) retarded release tablets were used as an example of modified solid oral dosage forms, these tablets were studied, the basic concepts, data analysis, techniques and mathematical models for in-vitro dissolution test and in-vivo pharmacokinetics were done, The tests performed according to united state pharmacopoeia. The absorbance of each in-vitro dissoluted sample were obtained using Ultra Violet spectrophotometer, the percent released versus time profile obtained and five mathematical models were applied to this profile zero order, first order, Higuchi, Hixon-crowell and krosmyer-pepas to determine the release kinetic from the tablet. The drug introduced to seven healthy volunteers, blood samples were withdrawn from them according to specific sampling procedure, and then (DS) extracted from blood samples according to special extraction technique. The amount of (DS) in each sample determined using High Performance Liquid Chromatography (HPLC). The plasma concentration versus time obtained, compartmental and non compartmental analysis applied to in vivo release profile. Invitro–Invivo Correlation (IVIVC) plays a key role in pharmaceutical development of dosage forms. This tool hastens the drug development process and leads to improve the product quality. Deconvolution based IVIVC was applied using model dependant (Wagner-Nelson) and model independent (numerical deconvolution) techniques. The result from this study reveals that the in vitro release kinetic follows Higuchi model because it has the higher regression coefficient (0.986), which means that the release is diffusion controlled. In vivo the parameters obtained using compartmental and non compartmental analysis were comparable to each other. The in vivo release profile were fitted to single compartment model, the predicted values of plasma concentration exhibited a good correlation with observed values(0.909), then it founded that the in vitro-in vivo Correlation were fairer using numerical deconvolution than Wagner-Nelson method with correlation coefficient(0.93).