With impending economical concerns and fuel crisis, biodiesel produced from microalgae has been proposed as a potential substitute for petroleum fuel. But experimental characterisation of combustion, performance and emission behaviour of biofuel operating under different operating conditions would be resource consuming. Even numerical modelling using computational fluid mechanics methods would be computationally consuming when combinatorics of all operating conditions are considered. In this study, we append empirical modelling to numerical modelling to derive industry feasible solutions. An artificial neural network was trained with responses at limited operating conditions obtained from a software Diesel-RK. Various variables representing combustion, performance and emission behaviour of IC engine were predicted accurately with average r-value of 0.9801 +/- 0.0146 for operating conditions defined by blending, loading and fuel injection pressure. Redundancy amongst the system variables was also observed thus indicating to possible reduced empirical models.