Reliability assessment for traffic data

Abstract

Given the vast amounts of data automatically collected by traffic detectors, identifying erroneous data is an important and challenging issue. In this paper, we develop a fuzzy logic approach for quantifying the reliability of data obtained from traffic detectors. Previous researchers have proposed multiple criteria for determining erroneous data; broadly speaking, these approaches either consider fundamental consistency (is the data physically plausible?), network consistency (is the data consistent with observations at nearby detectors?), and historical consistency (is the data plausible given past observations at this location?). This paper proposes a classifier incorporating all of these criteria, applying fuzzy logic to integrate these three separate assessments. An example application is given, utilizing data collected in the Dallas, TX, region.

Method to detect malfunctioning traffic count stations

“…Usually studies perform automated data checking by comparing measured data to historical data for consistency [10]. Sometimes, however, there are no historical data and only the observed database is available.…”

“…a) Represent all inequalities by matrix A and vector b: This method could be complementary to standard pre-process that analyzes historical consistency [10]. That is, observed variables must be into a real interval, in other case the observation is considered an outliers.…”

“…The methods for trying to detect errors may be classified according to the consistency criterion [10]:…”

Designing a Comprehensive Procedure for Flagging Archived Traffic Data: A Case Study

“…Among Type II checks, Turner’s method was claimed to require a manual check on the plots ( 1 ), but this test can be automated with measures like the statistical correlation. The approach proposed by Lin et al is based on a fuzzy classifier ( 18 ); however, the complex procedure and fuzzy pass/fail decision rules make the implementation less desirable.…”

Method for accuracy assessment of aggregated freeway traffic data

“…Lin et al [13] proposed a fuzzy-logic based classifier for assessing the reliability of traffic data by combining network consistency (conformity to the conservation principle) with fundamental consistency (physical plausibility) and historical consistency (agreement with past observations). In this method, a reliability index is calculated for each sensor based on the above criteria.…”