Peanuts are grown around the world and in United States where most peanuts are consumed after roasting. Peanuts are roasted to a specified color on Lab* scale as it is correlated with quality and acceptability. Two batches of Virginia type peanuts were acquired, one normal and other a high oleic variety. A surface response model using the Box-Behnken design was developed for Behmor 2000AB and GeneCafe coffee roasters, for normal and high oleic peanuts respectively with sample size, roast time and power/temperature as dependent variables and L* as a response variable. The model for Behmor was not significant (p>0.05 and R2 =0.87) but with effect contribution of roast time while the GeneCafe model was significant (p<0.05 and R2=0.98) with multiple first and second order effect contributions from temperature and roast time. Each model was validated and Behmor was found to be more consistent and predictable compared to GeneCafe. Both varieties of peanuts were roasted on each roaster and tested for volatile analysis using SPME GC/MS with high variation observed within samples which may be caused by uneven roasting. The volatile results showed similar trends for seventeen compounds between normal and high oleic samples. The Behmor roaster was more effective at predictable roasting for 50 to 100 g sample and more validation is needed on GeneCafe to improve its model. The results can help with quality testing of new varieties of Virginia type peanuts quickly without relying on large sample size typically used in other lab scale studies.