Traffic jams are an important problem both on an individual and on a societal level and much research has been done on trying to explain their emergence. The mainstream approach to road traffic monitoring is based on crowdsourcing roaming GPS devices such as cars or cell phones. These systems are expectedly able to deliver good results in reflecting the immediate present. To my knowledge there is as yet no system which offers advance notice on road conditions. Google Search intensity for the German word stau (i.e. traffic jam) peaks 2 hours ahead of the number of traffic jam reports as reported by the ADAC, a well known German automobile club and the largest of its kind in Europe. This is true both in the morning (7 am to 9 am) and in the evening (5 pm to 7 pm). I propose such searches as a way of forecasting road conditions. The main result of this paper is that after controlling for time of day and day of week effects we can still explain a significant portion of the variation of the number of traffic jam reports with Google Trends and we can thus explain well over 80% of the variation of road conditions using Google search activity. A one percent increase in Google stau searches implies a .4 percent increase of traffic jams. Our paper is a proof of concept that aggregate, timely delivered behavioural data can help fine tune modern societies.
stau traffic jams highways road conditions Google Trends prediction forecasting complexity endogeneity behaviour big data data science computational social science complex systems