Development of a fuzzy knowledge-based system for local traffic control for incident management

Multistage Fuzzy Logic Controller for Expressway Traffic Control during Incidents

“…Traffic-responsive ramp metering, on the other hand, adjusts the ramp control in response to the real-time traffic conditions on the mainline and the ramp during the metering period. The adjustment is conducted either in the reactive manner or proactive manner (Toan 2008;Zhong et al 2014). The former adjusts the ramp metering rates using real-time measurements in order to maintain a prespecified value of the expressway traffic conditions, while the latter attempts to improve the traffic conditions based on traffic variables anticipated for a certain time horizon.…”

“…With the help of fuzzy sets, the vagueness and uncertainties of the real world are handled smoothly. The key motivations behind the application of fuzzy logic for traffic control rest on the following advantages: (i) the linguistic expressions are general and easy to be perceived by the traffic operator, which is important for a decision support system; (ii) the transition from one fuzzy set to another is gradual, representing continuity in human perception; and (iii) the capability to combine several input quantities to provide a single output for the traffic operator to make a control decision (Toan 2008;Toan and Wong 2021). Support Vector Machine (SVM) is a family of machine learning algorithms.…”

“…The MS-FLC gathers real-time traffic and incident data in order to analyze and predict traffic situations, as well as to suggest alternative control methods to the traffic operator in the form of linguistic expressions. Given these functions, the decision-3 making support under these situations typically includes semi-structured decisions (Toan, 2008) that employ both structured modules for data collection, data analysis, and information processing, and non-structured component to help the operators when confronted with qualitative type of decisions. Thus, for the MS-FLC to execute in its totality, SVM is employed as a subset of MS-FLC model for short-term traffic flow prediction for anticipation of incident related traffic condition.…”

Developing a fuzzy-based decision-making procedure for traffic control in expressway congestion management

“…However, coordinated and integrated controls may provide system-wide benefits over local control because more extensive information is used and more robust control actions are coordinated and applied [7,8]. Nevertheless, for complicated situations, determining the appropriate type of control is not straightforward since this depends on network topology, background congestion level, and the availability of control facilities [9].…”

“…The former derives the ramp metering rates off-line over different periods of the day using historical data, and are applied for repeatable and steady-state traffic conditions. The latter adjusts the ramp metering rate to take account of traffic dynamics in a reactive or proactive manner [8,9].…”