In this study, the effect of fuel injection timing on the performance of a diesel engine was investigated experimentally and analytically using diesel-biodiesel blends. Different experiments were carried out on the diesel engine at engine speeds of 1200, 1600, 2000 and 2400 rpm. The injection timing was regulated for 10, 15 and 20 degree crank angle before top dead centre. The experimental results of engine torque, brakespecific fuel consumption (BSFC), cylinder pressure, and exhaust gas temperature for fuel blends of B20, B40, and B100 at different engine speeds and injection timings were recorded. The results showed that advancing the fuel injection timing for fuel blends of B20, B40, and B100 increased engine torque by approximately 2.1%, 2.9% and 6.3%, respectively, and decreased maximum BSFC by approximately 2.7%, 3.3% and 6.6%, respectively. Then, an artificial neural network (ANN) was used to predict engine performance. The injection timing, engine speed, and fuel blends were used as input parameters whereas engine performance parameters such as engine torque, BSFC, peak cylinder pressure, and exhaust gas temperature were used as the output parameters. The results showed that an ANN is a good tool to predict engine performance.