Ptv Flows Uses Machine Learning To ‘see’ Unexpected Congestion In Advance

PTV Group has launched a cloud-based tool, PTV Flows, which uses machine learning to predict traffic congestion up to two hours in advance. The self-learning system analyses historical and current speeds as well as specific key performance indicators to generate forecasts and automatic alerts, enabling traffic operators to anticipate potential issues and take proactive measures to mitigate congestion. PTV Flows embeds detailed maps and automatically-updated floating car data from providers such as TomTom, and either operates directly from the browser or is integrated via API into existing systems.