Traffic congestion is not uniform across the metropolitan cities like Bengaluru, in India. Road travel time becomes unpredictable many a times. In this aspect, several prediction models have been built. All these models require data on a daily basis for training. IoT based model to collect the input data using sensors can be the best choice. The model can be placed on all vehicles easily with minimum or no modification. This paper presents four varieties of approaches to measure the time taken for the journey. The timestamp is calculated based on engine run time. The collected data is stored in a file as well as synchronized to cloud storage. In the case of the problem with internet connectivity, the data written down in the file would serve as a backup. Travel time modelling is constructed using various sensors. This also helps advanced traveller information systems. Inevitably there is need to combine and make the best out of the available. GPS can be used to identify the source, destination, and few cardinal values even in the absence of internet connectivity.