Dynamic traffic assignment and signal setting for a network with nodal incident setting

• Main Authors: Castan, Russel Christopher, Cruz, Mylene Joyce, Hernandez, Lovelyn
• Format: text
• Language: English
• Published: Animo Repository 2013
• Subjects: Traffic signs and signals > Data processing; Traffic flow; Traffic assignment; Industrial Engineering
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/11054
• Institution: De La Salle University

Incidents are not uncommon in roads nowadays, most especially in urban areas. These do not cause hindrance to the passage of incoming vehicles, but it also intensifies the inherent traffic congestion in that specific vicinity. The delay brought by the blockage of the incident and the vehicle saturation creates for terrible travelling conditions for the road users. As such, past endeavors has been made by various researchers in search for an effective way in controlling incident-induced traffic congestion. In line with this, techniques have been formulated in search of a way improving the performance of incident-impacted road network. A simulation-based dynamic traffic assignment coupled with signal setting was used as a solution to address the mentioned gap of the research which is the lack of studies undertaking nodal incidents. Also despite the rise in numbers of studies that integrate dynamic traffic assignment and signal setting simultaneously, it has yet to be used in an incident-setting. The integration of solutions such as dynamic traffic assignment and signal setting to combat incident-induced traffic time penalties was tested against using solely dynamic traffic assignment. The DTA and signal setting simulation model was developed using the software Rockwell Solutions Arena 14.0. Several parameters were tested to see whether the model is sensitive to changes in value. These parameters are incident duration (10 minutes, 3 hours, 8 hours), incident magnitude (partial quadrant or entire quadrant blockage), and network demand (2304, 2880 and 3840 vehicles per hour). Average improvement is equivalent to 38.83% for ATP and 40.58% WTP. The results reaffirm that DTA with signal setting improve system performance better than using DTA solely. Also, results have shown the model is sensitive to all parameters that were tested. Results have also indicated that the model favors longer incident durations (e.g. lasting up to hours) compared to briefer ones. However, this is only true up to a certain incident time span. Also, partial lane blockage invokes more advantageous results compared to when the quadrant is fully blocked. Also, higher demand seems to make better reduction in time penalty as compared lower demand in the system. However, this may only be true up to a specific degree. Overall, the model yielded results that indicate the model is effective.