| The rapid development of intelligent transportation systems (ITS) has generated large amounts of data for transportation professionals. Currently, operators, planners, researchers, air quality analysts, transit providers, consultants, media, and others are using archived data. Benefits generated from these systems have already been accounted for in many large cities in the United States. Benefits include more detailed temporal data; alternative data to the existing data allowing the costs of data collection to reduce; data with greater geographic coverage; data that meets unmet data gaps in the past; and data that are on electronic media allowing the expedition of data analysis and the dissemination of information. However, analysis of archived ITS generated data can provide additional benefits for highway users. Additional research of the data generated from intelligent transportation systems will provide transportation professionals the ability to make better decisions. Nonetheless, the better utilization of archived data will take time, but the more experimentation with data will allow greater benefits.;In this research, a set of archived traffic data from Las Americas Expressway (PR-18) in the San Juan Metropolitan Region (SJMR) in Puerto Rico was examined. The case study is a facility that has gone through many improvements in the last ten years with the purpose of reducing traffic congestion. However, it has been a major challenge for the Puerto Rico Highway and Transportation Authority (PRHTA) since, even with the installation of a moveable barrier on a large portion of PR-18, congestion remains a problem. Traffic flow data, accident data, and work zone data from this facility were studied by means of data mining.;The methodology used was a combination of association mining and the knowledge discovery in databases (KDD) process. Association mining was used to learn about the hidden patterns within each model created and the KDD process was used as the framework that guided the entire process. The KDD process used consisted of seven steps: building the data mining database, examining and preparing the data, evaluating the data mining application, building the models, evaluating the models, preparing a list of the conclusions/knowledge gained, and the decisions. A total of six specific studies were developed in which different variables were studied using the association mining tools of the IBM Intelligent Miner for Data software package.;The objective was to gain knowledge from the data about interrelationship between the variables. Thus the results obtained from the mining tool used were excellent for the purpose of this research. The approach was found to be a source of valuable information that could not have been detected by the use of traditional statistical analysis alone. The approach allowed the identification of: "red flags" during work zone operations; similar patterns in levels of service (LOS) between Tuesdays and Wednesdays and similar patterns in LOS between Mondays, Thursdays, and Fridays; and it allowed the analysis of LOS over time. The major benefit learned from applying data mining to ITS generated data was that it allowed the analysis of numerous variables from multiple levels of information.;The new knowledge provides the basis for more advanced studies to be developed. In addition, the methodology could be used at other locations to increase the quality of information available for decision-making on similar facilities. |
