PAPPER: Predictive Adaptive Parking Price Estimator

Project Info

Team Name

Artificially Intelligent

Team Members

5 members with unpublished profiles.

Project Description

As our cities grow, the need for getting from point A to B also grows. City centers as the hub of economic activity feel the pressure from increasing traffic every day.

This project addresses the issue of traffic congestion and planning for the future. The first step in addresses a problem is to be able to predict when it is going to happen. To this end, using opensource datasets and machine learning, we learn to predict the long term (how traffic changes day to day in a year) and short term (how traffic changes within the span of single day) patterns, allowing very accurate congestion prediction.

How to stop it from happening
We take a two pronged approach towards addressing congestion. First, we use the model predictions to feed a dynamic parking price estimator (PAPPER). The dynamic pricing looks at the demand for parking at any given moment and adjusts parking feels accordingly. This would serve and an economic incentive for not bringing personal cars into the CBD.
Second, to encourage people transportation, we provide a tool for planning agencies to know when when is high, allowing them to provide more efficient and reliable public transportation, which is better both for the environment and traffic congestion in the city.

Data Story

Training the model : The long term model was trained using data from [1] and [5] to study take into account the effect weather on traffic congestion. The main data came from [1] which contains traffic volumes measure at various intersections in Adelaide CBD.

People coming into CBD: While the first model predicts how much traffic is passing through the CBD, it does not tell us who is staying. For this we look at the bluetooth data and extract all the instances when the last bluetooth location recorded as in the CBD. This gives us the number of people coming and parking in CBD at any given time. Datasets used [3] and [4].

Public Transporation demand was built from [2] and [6].

[1] Traffic Volumes :
[2] Metrocard validations :
[3] Bluetooth detections :
[4] Bluetooth dectection site:
[5] Weather :
[6] Adelaide Bus Stops:

Evidence of Work



Team DataSets

Adelaide Public Transport Stop Data

Data Set

Climate Projections for South Australia

Data Set

Bluetooth Detection Sites

Data Set

Bluetooth Vehicle Origins and Destinations (sample only)

Data Set

Adelaide Metrocard Validations

Data Set


Data Set

Challenge Entries

Reducing CBD Traffic Congestion

How to reduce traffic congestion or parking problems in CBD?

Go to Challenge | 39 teams have entered this challenge.

Public Transport for the Future

How might we combine data with modern technologies - such as AI/ML, IoT, Analytics or Natural Language interfaces - to better our public transport services. Outcomes could take the form of new commuter experiences, reduced environmental impact, or helping plan for the future.

Go to Challenge | 45 teams have entered this challenge.

Training AI models to deliver better human outcomes

For an outcome create two AI models based on contrasting incentive systems and examine the differing impacts on the defined outcome.

Go to Challenge | 12 teams have entered this challenge.

Local Government Information Technology Association of South Australia

How might we identify opportunities for improvement or new Council services, infrastructure and facilities to benefit community outcomes in South Australia?

Go to Challenge | 15 teams have entered this challenge.