In the current study, we conducted a numerical analysis of water-TiO2 nanofluid and entropy generation in a wavy channel under constant heat flow (q″) by using a two-phase mixture model. The analyses were carried under a laminar forced convection flow condition. Reynolds numbers (Re) are considered in the range of 50 ≤ Re ≤ 600 with the volume fraction of nanoparticles (φ) considered in the range of 1 % ≤ φ ≤ 5 %. The governing equations are solved by using ANSYS-Fluent software 14.5. The validation of the outcomes has demonstrated a strong consensus between the results and the literature data. The effect of various Re, φ and wavy wall amplitudes (α) on the flow behavior and heat transfer are examined. Moreover, the distribution of the static temperature, streamlines, total entropy generation (Ṡg,t), and Bejan number (Be) contours have been presented and discussed. Results showed that the heat transfer rate improves when φ, Re, and α increase. The heat transfer is enhanced when using a wavy wall compared to a straight wall. The performance evaluation criterion (PEC) increases with φ and Re. Therefore, it is recommended to use large values for each of φ and Re in the wavy channel from engineering and economics perspectives. Concerning Ṡg,t, Ṡg,t,h and Be, we conclude that these parameters decrease when φ and Re increase, except Ṡg,v is increased with φ and Re.