A novel approach to extract significant time intervals of vehicles from superhighway Gantry Timestamp sequences

It is attractive to extract and determine the key features of traffic patterns for mitigating road congestion and predicting travel time of vehicles in traffic analysis. Based on previous works that is a scalable approach via Hadoop MapReduce programming model, and can extract maximal repeats from a huge amount of tagged sequences, this paper adapts that approach to extract significant patterns of travel time intervals of freeway traffics in Taiwan, consisting of gantry list with time intervals of vehicles passing those gantries. Experimental resource are the records of gantry timestamp sequences of each vehicle passed in three months (2016/11–12 and 2017/1), downloaded from the Traffic Data Collection System (TDCS, http://tisvcloud.freeway.gov.tw/history/TDCS/), one of Taiwan government open data platform. To focus on one gantry sequences for observation, in this study, the longest pattern on the trip within the Taiwan National Freeway No.5 is selected. Experimental results show that the statistics of vehicle travel time according to 24 hours/per day are provided, and can give clue to the drivers how to avoid rush hours. This work is expected to be capable to handle a huge amount of real data by extending the scale of time intervals from month to year, and to be promising for further traffic and transportation research in the future.