Herein we describe the application of historical data design under response surface methodology to formulate the correlation of the fuel spray SMD (Sauter Mean Diameter) with factors (viscosity, fuel injection pressure, air-blast pressure). The Design Expert software was used to analyze the experimental data to obtain the equation between the factors and the response. The range of the values of variables used in this experiment were viscosity of the fuels 9.71 x 10(-4) Pa-1.207 x 10(-3) Pa, the fuel injection pressure 0.1-1.0 MPa, the air-blast pressure from 0 Pa to 2000 Pa and a total of 120 test points were conducted. The data were converted into logarithmic scale in order to utilize the linear fitting function of the DoE (Design of Experiments). According to the ANOVA (analysis of variance), the most influential factor to the fuel spray SMD is fuel injection pressure, followed by fuel viscosity and air-blast pressure. R-2 value of 0.9339 demonstrated that the regression equation obtained by historical data design fits the experimental results consistently. (C) 2015 Elsevier Ltd. All rights reserved.