The flow of alumina–water nanofluid across heated circular tubes arranged in inline and staggered arrays in a heat exchanger has been studied numerically using the finite volume method (FVM). For calculating the nanofluid’s thermophysical properties such as effective thermal conductivity and effective viscosity, Corcione’s correlations are utilized. Corcione’s correlations consider nanoparticles size, their Brownian motion, and operating temperature while calculating these effective properties of nanofluids. The impact of three parameters on heat transfer characteristics across inline and staggered arrays of heated circular cylinders has been examined. These parameters are nanoparticle diameter dp, which is varied between 10 nm and 50 nm, nanoparticle volume fraction ɸ varying from 0.01 to 0.05, and Reynolds number Re ranging from 10 to 200. It is observed that heat transfer augmentation across both inline and staggered arrays occurs when nanoparticle concentration is increased and smaller diameter nanoparticles are used. Mean Nusselt number NuM is increased by 31% when ɸ is increased from 0.01 to 0.05 at Re = 200 and dp = 10 nm in an inline array and by 25% in a staggered array. NuM is enhanced by 20% for the inline array and 16% for the staggering array when dp decreases from 50 nm to 10 nm at Re = 200 and ɸ = 0.05. At any given value of dp, ɸ, and Re, the mean Nusselt number is always higher for staggered array in comparison with the inline array. The results reported in the present study can be utilized for the optimal design of various heat exchange systems under the given operating conditions. The present results are extensively validated with the available experimental/numerical studies.