Due to rapidly changing trends in the market, automotive manufacturers are always struggling to release new vehicles with drastically reduced timelines. Therefore, it is very important to constantly optimize the development phases, starting from concept initiation to the final testing of production ready vehicle. The real world tests conducted on vehicles take huge amount of time, since these tests are carried out for large kilometers to periodically analyze tire wear, clutch wear, brake failure etc. Collecting large kilometers of CAN data is also tedious and time consuming due to various unwanted variables which add up during real world tests. In this paper, a technique known as Rescale Range Analysis is adapted to abridge the collection of kilometers data from testing by nearly ten times. This analysis estimates a Hurst coefficient to correlate the entire data with its divided parts. The division factor of the entire data is very crucial for the analysis. The estimated Hurst coefficient varies between 0 and 1. If the conducted test is highly reproducible over a period of time, the Hurst coefficient is closer to 1. The parameters used for Hurst coefficient correlation are vehicle speed, engine speed and engine torque. A small portion derived from the entire data represents the testing trend with a high Hurst coefficient, thereby reducing the kilometers driven. If the correlation between the divided part and the entire data is accurate, the vehicle can be driven for the correlated kilometers making the testing process simpler and less time consuming. The analysis is used to develop a reproducibility procedure with a CAN data of 15000 kilometers. Three trials for each parameter are conducted to come up with a robust and accurate correlation.