On short-term traffic flow forecasting and its reliability

TL;DR

This paper introduces a novel approach to short-term traffic flow forecasting focusing on its reliability and traffic volatility, supported by computer simulations using real data.

Contribution

It presents a new methodology for traffic forecasting that emphasizes reliability and volatility analysis, diverging from traditional time series models.

Findings

Effective forecasting method demonstrated through simulations

Enhanced understanding of traffic volatility and reliability

Potential for improved traffic management strategies

Abstract

Recent advances in time series, where deterministic and stochastic modelings as well as the storage and analysis of big data are useless, permit a new approach to short-term traffic flow forecasting and to its reliability, i.e., to the traffic volatility. Several convincing computer simulations, which utilize concrete data, are presented and discussed.