Research On The Design Methodology Of Traffic Management Measures To Reduce Vehicular Emissions

In recent years, transportation pollution emissions issue has caused negative impacton public health and environment. Consequently, how to reduce traffic emissions isbecoming the paramount of daily transportation management. This research proposes acombined traffic management measure system that involves emission pricing duringtraffic peak hour and speed limit during non-peak hour. Taking lowering vehicularemissions as the target, the author investigates the optimization methodology of thesemeasures through network modeling, algorithm design and numerical example analysis.First of all, this research analyzed the mechanism of traffic management measures.On that basis, the model framework of traffic management measure design to reducetraffic emissions was established. However, most of current emission calculatingmethods used in optimization models were oversimplified, which weakened thereliability and applicability of these models. To improve it, this paper employed themost advanced traffic emission estimating model, MOVES2010, and presented anemission optimizing process based on MOVES. This process laid the foundation forsubsequent studies.After that, the author explored how to determine the optimal emission chargingscheme for traffic peak hour. In a network where all links are tollable, the first-bestlink-based emission charging design model was formulated to meet emission budgets inemission regulation areas while minimizing the total emissions across the network.Besides, this paper proved that a non-negative charging scheme will exist if the linkemission function is strictly monotonically increasing. However, some links were nottollable in practice. In this case, the author proposed second-best link-based emissioncharging design model with complementarity constraints， and applied ManifoldSub-optimization algorithm to solve it. Considering that area-based emission pricingmanner is more practical, this paper formulated a mathematical program withcomplementarity constraints to determine optimal area-based emission charges with thesame objective as link-based emission charging. A numerical example showed thatcompared to the network performance under UE flow distribution, the area-basedcharging scheme can ensure to comply with zonal emission budgets while alleviating traffic congestion and reducing network-wide emissions. In order to use MOVES toestimate vehicular emissions more accurately, a bi-level programming model warformulated to design area-based emission toll. Then the solution framework wasimplemented based on the SID-PSM algorithm. A numerical example was demonstratedthat tight emission budgets can affect emission reduction across road network.Furthermore, this research conducted a comparative analysis between emissionpricing and speed limit setting. The result indicated that imposing speed limit can adjustnetwork flow distribution only during non-peak hour with relatively low traffic demand.Another numerical example verified that setting appropriate speed limit schemecontributed to lowering traffic emissions and improving the network performance. Sincethat the traffic demand during non-peak hour is uncertain, the author proposed a robustspeed limit design model against the worst feasible demand scenario so as to minimizethe sum social cost of system travel time and vehicular emissions. The cutting plane andManifold Sub-optimization algorithms were integrated to solve this model. Nevertheless,robust speed limit scheme couldn’t response to the changes of external environment. Toachieve better safety, efficiency and environment sustainability on a traffic network, thispaper envisioned a variable speed limit system where speed limits vary strategically andperiodically in accordance with external environment, and proposed a multi-objectivebi-level model to optimize speed limits schemes under arbitrary strategic scenario.MOVES was employed to estimate emissions and a solving framework based onNSGA-II algorithm was established. The numerical example was demonstrated thatweather conditions have impact on the optimal road speed limit scheme.Finally, the road network within the2ndring road in Beijing was used todemonstrate the proposed models and solution algorithms in this research. Resultsshowed that the methodology of emission pricing and speed limit design was effectiveon reducing transportation pollution emissions and improving traffic networkperformance.