Rice is the thirdly most valued cereal crops in Kenya after maize and wheat. The demand for rice in Kenya has increased greatly over the last few years while production has still remained low. This is because rice production is affected by serious constraints especially rice diseases of which the most threatening is rice blast. Rice blast infection and re-infection can occur in different stages of rice growth and therefore need to be controlled. This study aims to developed a mathematical model for rice blast re-infection. The model employs a system of nonlinear ordinary differential equations which is analysed in details for its stability properties. Basic reproduction number <i>R<sub>o</sub></i> for rice blast re-infection was found to be less than one. Numerical simulation of the model is done using Mathematica, and graphical profile of the main variables are depicted. We conclude that rice blast re-infection reduces rice yield and necessary remedy are needed.