Project Description
The Happy Parking! project aims to make the most use of the open-source data from government solving practical problems and tell stories of urban development according to historical car park data.
Part 1
Parkers aim to create a real-time web app for end-users, helping drivers to find a vacant spot efficiently.
It allows users to select the location they want to go and get them all vacant car parks in the area within 500 m and match the car park restrictions with duration they are willing to park. It has a priority rank for the matching function -- free first, within the area, outside area. If there are no vacant spots in this area, users can also use the off-street car park locations to find a car park.
At the government end, the real-time parking solution can help reduce the air pollution or traffic problem that caused by circling vehicles for vacant spaces as well as reduce the time wasted in finding a vacant spot, especially in the city area.
On the other hand, the visualization shows the land utilization in a more straightforward way, which supports the government to monitor and make a decision to improve quality of life.
Part 2
In addition, the web app also provides a visualization regarding the analysis of the parking area occupancy rate based on historical data of the past years.
It demonstrates the occupancy rate changes in each street with inground sensors according to their timestamps. Regarding these car park analysis data, it enables users to understand the urban planning in Melbourne and the changes in economic centres. Moreover, support the government’s urban planning decision in car parking, Remove car parks with low occupancy rate and build greenspace or improve public transportation network construction to reduce car park stress.
Data Story
The datasets provided by the City of Melbourne identifies the location of car park bay and vacancy in real-time told by in-ground sensor systems, which allow us to consolidate information and generate parking solutions for drivers.
For historical archived data, they give a knowledge about how well the parking systems are utilized, thus, the government could benefit from them to make a decision in respect to city resources planning.
we judged the occupancy rate on the peak hour(6:30-9: 00 am & 3:00-18: 30 pm) and whether it is the weekdays and weekends to make a comparison for each car par zone.