Variability in Traffic Flow Quality Experienced by Drivers: Evidence from Instrumented Vehicles

Abstract

Freeway operating conditions are typically evaluated using the level of service (LOS) concept, which is defined according to the macroscopic traffic parameter, density. Although traffic density may provide general estimates of current traffic conditions, this parameter generally fails to communicate the variability in the quality of traffic flow experienced by individual drivers. This variability may be caused by factors that are not effectively being captured by density, such as travel lane, vehicle position within a platoon, characteristics of roadway geometry, etc.This paper investigates the variability of traffic flow quality as indicated by measurements of speed and acceleration noise (standard deviation of acceleration) using field data obtained from instrumented vehicles equipped with global positioning system (GPS) devices. The microscopic measures observed from individual vehicles are compared with density-based LOS, which are calculated with data from the Atlanta, GA traffic monitoring system. This comparison reveals that under the same LOS condition, drivers may experience significant differences in the quality of traffic flow. In addition, the comparison indicates that drivers under different density based LOS conditions may experience a similar level of traffic flow quality. The variability generally becomes larger as traffic conditions worsen.

Exploring the Relationship between Roadway Characteristics and Speed Variation

“…As a way to improve the explanatory power, the effects of interacting factors (e.g., facility type multiplied by grade) were also considered, but the trial failed, only introducing difficulties in interpreting the estimated parameters because of the increased number of variables and complexity. More importantly, the low explanatory power might be caused by the fact that the developed models are unlikely to remove all the variances induced by various drivers/vehicles and localized traffic conditions (e.g., interaction with other vehicles, location within a platoon, traveling lanes) (3,5). Unfortunately, such factors could not be incorporated into the models because the quantification of the factors was impractical, which may be a drawback for this type of study employing real-world instrumented vehicle data.…”

“…However, researchers have also proposed promising service quality measures using microscopic traffic parameters, such as the degree of speed variation of individual vehicles over roadway segments, assuming that the more speed variability during a trip, the greater the occupants' discomfort and the more inefficient the vehicle operation. Numerous studies have supported this assumption, finding that the smoothness of traffic flow is closely related to roadway service quality and fuel consumption (1)(2)(3).…”

Household-Level Global Positioning System Travel Data to Measure Regional Traffic Congestion

“…Because of the cost of implementation and data processing, most GPS travel studies are limited to less than a week of data collection per household. The Commute Atlanta project, in which recruited households collected data over multiple years, is probably the one notable exception; recent papers are a testament to the value of that approach for answering a wide range of research questions (15)(16)(17). The remaining sections of this paper demonstrate how a relatively small GPS travel data set can be used to define meaningful performance measures.…”

Data Management for Geographically and Temporally Rich Plug-in Hybrid Vehicle “Big Data”

“…With the development of distributed data collection technologies, many other researchers have developed techniques to collect, store and synthesize operation data from vehicle fleets [1,8,9,10,11,12,13,14]. Characterizing the operation of PHEVs is of particular interest to transportation system researchers because of the well-documented dependency of vehicle fuel consumption on individuals' driving and charging habits [1,2,3,6,8,10,11,12].…”