Software sharing has increased in automotive embedded software development. lt provides acceleration to the development lifecycle. Distributed development of ECU s is trending in automotive industry. For this purpose, there is a need to automate model generation using information from the database. Model driven software development requires effective implementation techniques. This thesis explores and describes automation of Matlab/Simulink (ML/SL) model generation from the database and code generation by model transformation. The database forms the driving force for modeling and is made accessible through ADD API (Automotive Data Dictionary). Technical feasibility and usage costs are considered as evaluation criteria to solve the problem. The existing solution that connects the database to modeling tool (ASCET) is extended for ML/SL. The existing system is analyzed using reference examples. Different modeling possibilities in ML/SL are then evaluated for the considered criteria. Based on evaluation the best solution is decided for prototype implementation. The APis for remote invocation of Matlab and Simulink model generation are identified. The prototype is validated for correctness and compared with the result from ASCET to identify code artifacts if any and demonstrate equivalence.