2015 IEEE International Conference on Data Science and Data Intensive Systems (DSDIS) (2015)
Dec. 11, 2015 to Dec. 13, 2015
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/DSDIS.2015.96
Weather factors such as temperature and rainfall in residential areas and tourist destinations affect traffic flow on the surrounding roads. In this study, we attempt to find new knowledge between traffic congestion and weather by using big data processing technology. Changes in traffic congestion due to the weather are evaluated by using multiple linear regression analysis to create a prediction model and forecast traffic congestion on a daily basis. For the regression analysis, we use 48 weather forecasting factors and six dummy variables to express the days of the week. The final multiple linear regression model is then proposed based on the three analytical steps of (i) the creation of the full regression model, (ii) the removal of the variables, and (iii) residual analysis. We find that the Rsquared value of the proposed model has an explanatory power of 0.6555. To verify its predictability, the proposed model then evaluates traffic congestion in July and August 2014 by comparing predicted traffic congestion with actual traffic congestion. By using the mean absolute percentage error valuation method, we show that the final multiple linear regression model has a prediction accuracy of 84.8%.
Predictive models, Weather forecasting, Roads, Data models, Analytical models, Linear regression
J. Lee, B. Hong, K. Lee and Y. Jang, "A Prediction Model of Traffic Congestion Using Weather Data," 2015 IEEE International Conference on Data Science and Data Intensive Systems (DSDIS), Sydney, Australia, 2015, pp. 81-88.