Back to Projects

Team Name:

Detour Labs


Team Members:


Evidence of Work

Flexi-Bus

Project Info

Team Name


Detour Labs


Team Members


Nikhil Chhetri and 3 other members with unpublished profiles.

Project Description


This project focuses on utilising AI to optimise the routing of buses that have a fixed final destination, such as a train station. Using data such as vehicle registrations, Bus Stop data, Travel and Activity data, and Train Service Passenger Count data to understand transport trends and patterns, an AI service will be able to provide riders with the most efficient route.


#aioptimization #publictransportation #routeoptimization #busroutes #traveldata #vistadataset #metrobusdata #estimatedarrivaltime #travelbehavioranalysis #infrastructureplanning #riderconvenience #datadrivendecisions #servicecoverage #peaktraveltimes #dynamicrouting

Data Story


Problem Statement
The prevalence of privately own vehicles coupled with the underutilization of bus services in Victoria has become critical challenge that defeats the Victorian Government’s commitments under the transport sector emissions reduction pledge. Bus transportation systems often struggle with aligning with peak office hours and convenient pick-ups for riders. This deficiency leads to lower rider satisfaction, increased dependence on private vehicles, increased travel times where riders have to walk long distances to catch buses, and overfilled car parks at train stations due to private car dependence. Additionally, infrastructure decisions, such as where to place extra bus stops, are a challenge if they are not people focused and are based on generalized data, leading to below-expectation services and poor accessibility.

Approach
Our approach to this predicament therefore involves developing an AI-driven application that adjusts bus routes in real time to pick up riders from nearby locations while ensuring that the bus still arrives at its fixed destination, such as a train station, on time. In this solution, there are two categories of buses, that is, those that will still follow the prescribed route from VicRoads, and those that will make optimised detours to pick up nearby riders. For the latter, the following proposals will apply to ensure an efficient, convenient and desirable outcome.
Dynamic routing:
A button will be introduced at various bus stops to enable riders that a travelling to a fixed destination such as a train station to request a bus. The request will be processed by the AI and the best positioned rerouted to make the pickup. If the calculations show that ETA will be affected, then that particular bus will not make any more detours till its destination.
On the other hand, if multiple riders all heading to various destinations are present at one or more bus stops, a screen will be available at the bus stop for riders to indicate which major destination they are heading to. In this way, the appropriate busses will be rerouted.
Data-driven monitoring:

Various datasets such as the VISTA and Metro Bus Routes data, Train Service Passenger counts data will enable the AI system to identify high-demand areas, calculate the best pick pickup points, and adjusts routes in real-time to balance rider demand, convenience and timely arrival at the set destination.

Infrastructure and Resources
The following infrastructure and resources will need to be available for the proper implementation of this solution. That is,
Mobile app Integration: an AI will have to be integrated into the PTV app so as to leverage the already existing schedules and frameworks on both the driver and rider ends. However, rides can only be requested at the station and this is only for updates.

Bus stop screen display and button: This will be used to request rides and specify the set destination required.
Policy: Incentives should be provided for early adoption of the system such as reduced fares and loyalty programs.
Interactive maps: Providing a maps application or integration that will show real time updates of but routes, locations and ETAs.

Outcome
The proposed solution results in a more efficient bus transportation system that enhances rider satisfaction by reducing wait times and travel times by improving access to bus services. It also aids in infrastructure planning by pinpointing areas where additional bus stops may be needed based on real-world travel data, ensuring better service coverage and accessibility. The result of this implementation is reduced traffic congestion, lower emissions and enhanced accessibility as riders move away from private transportation.


Evidence of Work

Video

Project Image

Team DataSets

Traffic Volume

Description of Use This data will be used to determine what is the best route for buses to take in case of a detour while maintaining the ETA.

Data Set

Whole Fleet Vehicle Registration Snapshot by Postcode

Description of Use This dataset will be used to monitor growing numbers of car ownership (which indicates growing demand for transportation means) so as to determine the areas where public transport facilities should be allocated.

Data Set

PTV Metro Bus Stops

Description of Use This dataset will be used to determine the current location and distribution of buses, to calculate detours without affecting ETA and to reroute buses to pick up clusters of riders to a fixed destination.

Data Set

Victorian Integrated Survey of Travel and Activity (VISTA)

Description of Use This dataset will be used to determine hotspots where many people may have travel needs but no easy access to facilities like bus stops and my cause walking long distances to bus stops. This will lead to better facility allocation (bus stops) and rerouting.

Data Set

Bus Routes Metro (PTV)

Description of Use Understanding the original route and calculating new routes while maintaining EAT. Avoiding conflicting routes between different providers once routes have been modified.

Data Set

Train Service Passenger Counts 2023-2024.

Description of Use Allows for the analysis of trends and flows of passenger movements. To determine what are the peak times and to adjust bus availability accordingly.

Data Set

Challenge Entries

Smart Mobility: Optimizing Urban Infrastructure for a Sustainable Future

How might we use data insights to promote the development of sustainable urban infrastructure and reduce dependency on private vehicles?

Go to Challenge | 26 teams have entered this challenge.