New ride-sharing algorithm promises better deal for riders & drivers

Ride sharing on app-based taxi services is catching up. Now researchers at the Indian Institute of Technology, Delhi, have developed an algorithm that can guide taxi drivers to take a route where they can find more commuters for sharing. They believe this algorithm can make shared rides more profitable for drivers and reduce waiting time for commuters as well.

To make cabs more economical for customers, many taxi-service companies are promoting ride-share among commuters. Here two or more ride-orders are completed by a single taxi driver who can complete these rides without deviating much in a given travel route. The taxi company benefits from such ride shares as it increases their customer base, reduces fuel and allows them to maximise profits from limited number of vehicles.

However, often a compatible ride order may not possible on a given route due to which company may incur loss as it has to notify charges to ride-share customers even before they take a cab.

“In the current scenario, the appearance of subsequent compatible customers is left to luck – a taxi moves along the shortest (or quickest) path for the existing customer and hopes to find additional compatible customers on its way,” said Dr. Amitabha Bagchi, Associate Professor at IIT-Delhi, while explaining the problem to India Science Wire.

The new algorithm is more efficient in searching compatibility between ride-orders.  Called “Share” (Sharing with History-Aware Route Recommendations), it finds routes beyond shortest paths and recommends a route that is close in length to the shortest path and offers higher chance of finding a compatible ride-order.

To verify the algorithms, researchers tested it on real taxi datasets from New York and Singapore across a range of metrics. The results showed that routes recommended by the algorithm received up to 40% reduction in orders without ride-sharing than the shortest path. Also, this reduces waiting time for customers by 20%.

“We would like to see how our algorithm works on Indian conditions by collaborating with one of the taxi operators from India. We are also extending this work for the food delivery business,” added Dr. Bagchi.


The research team included Chak Fai Yuen, Abhishek Pratap Singh, Sagar Goyal and Sayan Ranu.


The author, Dr. Aditi Jain, is a contributor at India Science Wire.


Tweet at Aditi: @AditiJain1987


Follow Indus Dictum on Facebook and Twitter for the latest updates.

The ID Staff

Daily stories by the Indus Dictum Staff

Want to share your thoughts? Leave a Comment.