An Investigation For The VKT Distributions And Emissions Estimation Based On Traffic Performance Index

Traffic congestion and vehicle pollutions are two major problems faced by the traffic industry.Quantitative indices have been put forward by the administrations and academies for evaluating both the problems.The frequent occurrence of traffic congestion results in an increase of the time fraction of unstable vehicle operating conditions such as idling,low-speed cruising,and aggressive accelerations/decelerations,which leads to a significant increase of vehicle emissions.The vehicle emissions are correlated with traffic congestions.As a quantitative index of describing the network traffic operating conditions and of evaluating the level of traffic congestion,the Traffic Performance Index(TPI)has been used for traffic management and decision-making.However,the relationship between TPI and emissions has not been clear.The motivation of this research is to understand the quantitative relationship between the TPI and the network emission quantities,and to provide theoretical foundation for a wider use of the TPI,so that an efficient evaluation of the environmental impact of traffic policies targeting relieving congestion can be realized.In this study,the Traffic Performance Index(referred as the TPI in later context),which is the evaluation index of the national standard GB29107-2012,Road Traffic Information Service.Traffic Condition Description,is selected to be investigated.The relationship between the traffic index and the network emissions is studied,and an analysis is conducted based on the features of the network traffic operations for explaining the relationship.First,the traffic index model is selected for calculating TPI based on the literature review,as well as the microscopic emission model for Beijing based on the distributions of Vehicle Specific Power(VSP).The TPIs,and the network emissions of CO2,CO,THC and NOx are calculated.The data from Remote Traffic Microwave Sensors(RTMS)and Floating Car Data(FCD)are comparatively analyzed for estimating the link-specific speeds.It is indicated that there are significant differences between the two sources of speeds.Using the RTMS speed for establishing the traffic flow models leads to an overestimation of free-flow speed and speed-at-capacity.The differences are especially considerable when the speeds are low,which causes bigger errors for the emission estimation.The link-specific speeds covering the entire network are obtained by adopting the flow-speed model for Beijing.Second,the TPIs and network emissions of the 33 work days are analyzed.It is indicated that there is a positive correlation between the TPIs and network emissions.The increment of emissions with the TPIs are the most rapid when the TPIs are low,rather rapid when TPIs are larger than 8.7,and mild when the TPIs are moderate.The coefficient of variance is used as the index for evaluating the uncertainty of the TPI-emission relationship.It is found that the uncertainty is small when TPI is larger than 3.Categorizing the data based on time period is found eligible to significantly reducing the uncertainty.The verification for the period-specific TPI-emission relationship is conducted with the sample of two days,which are excluded for the establishment of the relationship.It is found that the differences between total network emissions based on aggregated road link emissions and ones from TPI-emissions relationship are small,with relative differences of 3.46%,3.23%,3.53%,and 3.36%for CO2,CO,THC and NOx respectively,for the period of 6:00 to 22:00.Third,the VKT-TPI and emission-TPI distributions for speed bins are established for all road classes,based on the analysis above.The shape of the expressway-distribution performs one-peak for low-TPI and two-peak for high-TPI.A significant increase of the VKT for expressway is found for low-speed intervals.The distributions of the major and minor arterials each performs only one peak.The speed of the peak is found to decrease for major arterials,while the peak of the minor arterials is rather stable.A two-case comparison is conducted for two cases with close VKTs but different TPIs.It is found the emissions of expressways and major arterials have large impact on the network emissions,and the VKT fraction at speed intervals smaller than 10 km/h is considerably influential of the network emissions.Finally,the second quartiles of 2016 and 2017 are taken as examples for the case study.The TPIs are used as the inputs to the TPI-emission relationship,and an efficient evaluation of the network emissions is thus realized.This study conducts a quantitative analysis of the relationship between the TPI and network emissions.The traffic operations of the road classes are investigated for emission estimation.This study proves the theoretical feasibility of the integration of the TPI and network emissions,and provides a technical support for future efficient evaluation of network emissions based on TPI.