Book a live demonstration
Contact us to arrange a walkthrough of any Streamriver product and learn how we can help your organization.
Transit technology generates vast amounts of operational data 24/7/365.
Insights Analytics combines statistical analysis with machine learning to spot emerging issues early, and take appropriate action before serious problems arise.
Every moment of every day, transportation IT systems generate a massive volume of operational, passenger, and infrastructure data from sources including transit schedules, vehicle assignments, location and telemetry, ticketing and fare collection, passenger counting, CCTV, on-board and in-station connectivity, and service alerts. Processing this data and transforming it into meaningful information is highly complex, and a key part of a transit operator’s digitalization process. Artificial intelligence and machine learning are being successfully applied within transportation systems for tasks such as demand forecasting and predictive maintenance, but can be equally employed to help agencies deliver a high-quality transit experience for passengers. Streamriver Cortex has been designed from the ground up to ingest, process and make sense of transit data for precisely these needs, and deliver actionable analysis within Insights Analytics.
Understanding deviations from the baseline
Insights Analytics reviews every journey, every day. A multi-layer ensemble of statistical tests and machine-learning models, retrained daily on your own data, learns what ‘normal’ looks like for each of your services, including holidays. It cross-checks performance of on-board systems that may have interdependencies. Only strong, confirmed deviations are flagged, each with a severity and confidence rating. AI-generated summaries then highlight what matters most, so your team spends its time on the handful of journeys that need attention.
Insights Analytics Anomaly Detection for Passenger Wi-Fi & T2G Backhaul
taking action on service anomalies
Insights Analytics provides a rich visual interpretation of service performance across multiple metrics from IT resource data feeds such as back office systems and on-vehicle equipment. Insights anomaly detection combines statistical analysis with machine learning, and is built to minimize false alarms. Variations from baselines are graphically represented by level of severity and confidence, and Insights AI gives a plain-English anomaly synopsis with recommended actions.
Anomaly Drill-down for Issue Synopsis and Actions
Contact us today to discuss your solution needs and explore our products with a live demonstration.