The quality of road infrastructure significantly influences road safety, vehicle performance, and the overall driving experience. Traditional methods of assessing road quality, such as manual inspections, often lack the efficiency and accuracy needed to address modern transportation challenges. To overcome these limitations, this project focuses on developing an innovative model to assess road roughness using sensor data. The model leverages Android sensor technologies, primarily utilizing two types of sensors: accelerometers (inertial sensors) and GNSS (Global Navigation Satellite System) sensors. Given resource constraints, data was collected using an Android phone mounted on bicycles, which provided valuable insights despite some challenges and errors encountered during data collection. At the core of our model is the analysis of the International Roughness Index (IRI), which has been widely recognized as a reliable indicator for assessing road roughness on a quantitative scale. By deriving the parameters associated with IRI and applying the proposed formulae, we were able to recognize and categorize road surface irregularities such as potholes and humps. Our approach was further validated through the application of statistical methods, including the Kolmogorov-Smirnov (KS) test and Q-Q (Quantile-Quantile) plots. These methods demonstrated that the IRI is indeed a robust metric for indicating road roughness and low-cost sensors can be used for estimating road roughness. The metrics established in this study can serve as the foundation for developing more sophisticated algorithms that assess road roughness based on accelerometer data, ultimately contributing to enhanced transportation efficiency and road safety.