Advances in Information Technology have led to development of various Automatic Vehicle Location (AVL) systems. The automotive industry quickly adopted this system as it provides location services for vehicles through wireless technologies. But these systems are not explicitly optimized to take advantage of grid computing. Grid computing offers the infrastructure for developing coordinated, scalable resource sharing in a dynamic environment to maximize the utilization of the available resources in the network. Resources can be microprocessors, storage media, files, bandwidth, sensors etc. The problems like dynamic nature of resources, single point of failures, scalability, real time delivery of vehicle location information are mostly ignored in the design of such systems. The data grid concept, which addresses the above issues, can be used to improve the performance of Automatic Vehicle Location systems. We propose the architectural model for the vehicle transport system to be operated in grid environment to address the above mentioned problems, discuss implementation related issues, and benefits of our proposed model.