Overview
While ride-hailing market share has been growing steadily since 2015, taxis are still in high demand in certain geographies due to stricter regulations on ride-hailing. Yet, supply-demand mismatches still persist due to poor taxi positioning. On the other hand, hotspot maps provide only current demand snapshots without forecasting future passenger needs or guiding supply responses. Similarly, ride-hailing surge pricing mechanisms fail to optimise fleet movement, providing only the current state of passenger demand.
This technology leverages a neural network-based spatio-temporal prediction model to forecast taxi demand and recommend driver repositioning strategies, to enhance operational efficiency.
Our Innovation
This system employs a spatio-temporal neural network AI model that combines real-time taxi location (obtained via publicly available APIs or with data from Fleet Management Systems) with historical trip patterns to predict short-term demand across urban areas. The model processes dynamic spatial and temporal signals to generate proactive driver guidance, enhancing the chances of successful pickups. Unlike static demand visualisations, the system enables centralised coordination and real-time responsiveness.
A two-year field trial involving 500 drivers has demonstrated a 30 - 40% improvement in passenger pickup rates.
Technology Features
- Spatio-Temporal Demand Forecasting: Combines real-time taxi positioning with historical pickup patterns to predict near-future demand at specific locations.
- Dynamic Driver Repositioning: Recommends optimal repositioning to taxi drivers, maximising the probability of securing passengers in real time.
- Data Fusion Architecture: Integrates live fleet data with demand signatures to continually update and refine predictions.
Potential Applications
- Urban taxi fleet management systems
- Predictive dispatch platforms for transport operators
- Integration with ride-hailing or mobility-as-a-service platforms e.g., on-demand bus services
Logistics and point-to-point delivery optimisation e.g., parcel delivery services
If you're interested in this technology, please contact KTC.
