This session is facilitated by Gopalakrishna Anegundi
About this session
*Several bus routes in Bangalore are infeasible due to insufficient buses during peak hours. This increases passenger demand and environmental pollution due to more private transport.
*BMTC AC buses are facing a loss because of less passengers. I will explore ideas on increasing ridership:
- Introducing machine learning (ML) models to predict passenger numbers, by route, according to historic demand.
- Feeding BMTC data to Wikidata to provide public information about bus timings, routes and passenger numbers to help riders make better decisions.
- Merging 1 and 2 will give the best routes to accommodate more passengers based on demand and time.
*Participants will identify variables necessary to develop a feasible algorithm for the above, and ideas for global implementation.
*I can engage people from diverse backgrounds for feedback on these ideas. [ For instance, on how to deal with queries using Wikidata, adding labels about their native public transport details etc.]
Goals of this session
The objective of this session is to discuss possible solutions for increasing the feasibility of allocating buses based on the demand of passengers in a specific bus station/locality using AI. The use case will be BMTC (Bangalore Metropolitan Transport Corporation). The outcome of the session will not just be a solution for BMTC buses but also a global solution for several countries’ bus transportation systems facing similar issues.