A statistical evaluation of traffic flow data obtained from the Sydney Coordinated Adaptive Traffic System (SCATS) prior to decision support system (DSS) implementation

• Main Authors: Bandera, Madeleine Louise B., Bibit, Joy Roxanne Q., Young, Milestill C., Cruz, Dennis E., Gobaco, Bryan O., Li, Richard C.
• Format: text
• Published: Animo Repository 2007
• Subjects: Decision support systems; Traffic engineering—Data processing; Operations Research, Systems Engineering and Industrial Engineering
• Online Access: https://animorepository.dlsu.edu.ph/faculty_research/6839
• Institution: De La Salle University

In a parallel study entitled "An Integrated Decision Support System for the Sydney Coordinated Adaptive Traffic System (SCATS) of the Metro Manila Development Authority - Traffic Engineering Center (MMDA-TEC)" by Bandera, Bibit, Young and Li (2006), a decision support system (DSS) was developed in order to provide consistency and structure to the strategic decision-making process of traffic engineers who are the ones responsible for providing strategic traffic control measures for SCATS to implement. The DSS utilizes correlation analysis in the generation of Strategic Approach (SA) and Strategic Input (SI). However, before the developed DSS rules could even be used on traffic data collected from SCATS, it is a requirement that these data be tested first for normality. This is done because parametric tests such as linear correlation and hypothesis testing for the equality of means and variances are used in processing data for the DSS. These tests require that collected data should follow the normal distribution. Hence, this study presents the statistical tests performed on the vehicle flow data obtained from SCATS. One week (five days) of vehicle flow traffic data were collected from the daily data recorded by the detectors of SCATS for the purpose of data evaluation. Data evaluation results show that data collected follow the normal distribution. Furthermore, majority of pair-wise combinations of data obtained from different days were shown to have statistical equality of means and variances. This means that performance comparison between the generated SCATS strategic controls with and without the application of DSS rules may be undertaken even if data collected come from different days.