Modelling the impacts of adverse weather conditions to traffic flow dynamics and the calibration and validation of a weather responsive estimation and prediction system of a macroscopic traffric flow model | |
| Author | Mejia, Hanzel Naval |
| Call Number | AIT Diss. no.TE-23-01 |
| Subject(s) | Traffic flow--Mathematical models Traffic flow--Forecasting Traffic safety |
| Note | A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Transportation Engineering |
| Publisher | Asian Institute of Technology |
| Abstract | The primary motivation for this dissertation is to enhance our understanding of traffic flow dynamics by considering the effects of adverse weather conditions. While numerous studies have focused on addressing the persistent problem of congestion, most have relied on data collected during normal weather conditions. This research specifically addresses two key topics in traffic flow theory.The first general topic is on the capacity drop phenomenon (CD), a subject that remains an active area of investigation due to the incomplete understanding of its underlying mechanisms. By considering weather factors, this study provided new insights into the capacity drop phenomenon, particularly how weather conditions influence its occurrence and potential control strategies. Studies indicate that the discharge coming out of a queue is low compared to the capacity after the queue during congestion. All previous research on this event has focused on data collected under clear weather conditions. This study is the first to empirically examine the relationship between queue discharge rates and varying weather conditions. Prior research suggests that the drop in capacity is caused by going beyond the critical density, thereby leading to a decrease in the discharge. Our findings demonstrate that this critical density is reduced in any adverse condition of the weather. Additionally, studies in the past have linked the capacity drop to congestion speed, though this relationship may not hold in inclement weather. Our findings reveal that the discharge coming out of the queue correlates with congestion speed regardless of weather conditions. For the first time, we also demonstrated a negative linear relationship between congestion speed and the percentage of the capacity drop.The second general topic is the traffic flow modeling principles considering the effects of weather factors. We have successfully calibrated and validated the macroscopic second-order traffic flow model METANET incorporating weather-specific considerations.The study revealed that incorporating traffic flow heterogeneity or the consideration of multiple fundamental diagrams affects the model's performance.The METANET model successfully replicated and tracked congestion occurrences under both normal and adverse weather conditions. However, a notable decrease in key traffic flow parameters, particularly free-flow speed and capacity, was observed. The model also effectively captured the scattering in the flow-density curve, demonstrating METANET's capability to account for complex traffic flow phenomena. Regarding the sensitivity of the model to its parameters, the analysis indicated that METANET exhibits minimal sensitivity to parameters such as travel time, delay, flow decay, minimum speed, and capacity drop. However, free-flow speed, capacity, and critical density have a significant impact on the performance of the model. This study has successfully highlighted the negative effects of bad weather conditions on key traffic flow parameters and the modeling of traffic noting that weather-specific modeling consistently outperforms models that did not account for weather factors. Weather specific models accurately captured both the onset and dissipation of congestion and effectively predicted spatiotemporal values. In contrast, models that did not consider weather factors failed to replicate the progression of congestion waves.These findings have important implications for control and operation strategies, particularly in integrating weather forecasts into active traffic management techniques. Rainfall, a significant climatic factor, especially during monsoon seasons, affects traffic capacity year-round in many countries. Unlike sudden events like road accidents that reduce traffic capacity, rainfall is predictable through weather forecasts. This predictability offers an opportunity to integrate weather data into traffic management strategies. By effectively utilizing the findings of this dissertation study, transportation systems can be optimized to mitigate the impact of reduced capacity due to rain, leading to higher queue discharge rates and reduced overall system delays. |
| Year | 2024 |
| Type | Dissertation |
| School | School of Engineering and Technology |
| Department | Department of Civil and Infrastucture Engineering (DCIE) |
| Academic Program/FoS | Transportation Engineering (TE) |
| Chairperson(s) | Kunnawee Kanitpong;Ampol Karoonsoontawong (Co-chairperson) |
| Examination Committee(s) | Hadikusumo, Bonaventura H.W.;Santoso, Djoen San |
| Scholarship Donor(s) | DOST-SEI Foreign Graduate Scholarships;Visayas State University, Leyte, Philippines;AIT Scholarship |
| Degree | Thesis (Ph. D.) - Asian Institute of Technology, 2024 |