Better late than never| Planet, With People
Planet, With People
HELP = Hybrid Environment for Life and Prosperity
Summary
With the human population rising every year, people will need more places to live a normal life. Currently there are a lot of locations that are overpopulated or deserted. We are also building in areas that can potentially endanger the people living there (natural disasters). Our idea was to create a tool that will help the government officials decide where to build new infrastructures, where to renovate/upgrade current ones and where to relocate the people.
How We Addressed This Challenge
The idea is to develop an AI platform which can predict the maximum number of people living in a specific location. The maximum number represents the Urban carrying capacity (UCC) and is calculated including parameters already in use in scientific studies.
Carrying capacity is the maximum number of individuals of a given species that an area's resources can sustain indefinitely without significantly depleting or degrading those resources. Determining the carrying capacities for most organisms is fairly straightforward. For humans carrying capacity is much more complicated. The definition is expanded to include not degrading our cultural and social environments and not harming the physical environment in ways that would adversely affect future generations. (https://people.wou.edu/~courtna/ch371/lecture/popgrowth/carrying.htm)
There are studies which aim to develop a practical UCC assessment framework to guide urban development towards achieving sustainability.
(Using Urban-Carrying Capacity as a Benchmark for Sustainable Urban Development: An Empirical Study of Beijing Yigang Wei 1, Cui Huang 1,*, Patrick T. I. Lam 2, Yong Sha 3 and Yong Feng 2)
For example, the study done on Mumbai city. They used UCC parameters to remove bottlenecks to ensure future sustainability. (https://www.researchgate.net/publication/309576384_TRANSFORMING_MUMBAI_CITY_REMOVING_THE_BOTTLENECKS_TO_ACHIEVE_FUTURE_SUSTAINABILITY)
The platform shows the current state of the chosen location. For this purpose, the framework set by Jin Yeu Tsou et al (2017) is used. Their framework uses remote sensing data only. However, our aim is to add some of the urban living components that influence the parameters and overall quality of life. The problematic aspects of the living areas for example, too crowded, risk of floods, lack of infrastructure, bad air quality, etc. Moreover, it can determine the safeness of urban environments, by analysing past events and environmental state.
At first, with the combination of factors AI determines the UCC and draws you the red, orange or yellow zones. The user can then select which of the fields in question would like to address and chooses between environmental and infrastructural factors.
Seeing the results, we hope governments will easier acknowledge the problematic areas and develop projects to improve quality of life for their citizens and make them safer.
For the future, it is planned for AI to be able to allocate missing infrastructure, depending on the population’s age structure and mark areas that can potentially be under the influence of the national disasters. For example if the population is older you would need more retirement homes and medical centers in a short range. Or if the area has a lot of younger families we would need more kindergartens and schools. Maybe another way to solve the problem would be to provide better public infrastructure and populate more remote and deserted areas. However the most suitable decision will be up to the user.
How We Developed This Project
Traveling around the world made us see the problems of either overpopulated or deserted areas. Governments are acting upon problems differently sometimes, just following the money and not caring about their people. So, we wanted to present the tool which can help government officials discover problematic areas more easily and supported with smart suggested solutions. And also adapt the infrastructure according to changes in population.
- At first we familiarise ourselves with the available data set.
- Then followed the discussion about the problems humanity is facing due to the lack of specific data collection. After looking for inspiration from academic research papers, we encountered the use of UCC factorisation. Suddenly, we came to realisation, that we have all this data but no tool to generate influencing factors such as UCC. As our population is growing it will become more and more important to know how many people a piece of land with belonging infrastructure can support.
- As the model we decided to use an already validated method from study already mentioned before. With this method we calculate UCC and apply it to the standard of choice. Then our map produces the maximum population grid map (heat map).
- In ideal version, the next step AI would already try to optimise the maximum number by considering the exposure to natural disasters.
- To additional value of our AI the user would get to pick which other human factors would also be considered. Here we are talking about infrastructure and urban services (basic municipal facilities such as water, electricity, and gas supply and communication links, cultural and recreational facilities, housing conditions, healthcare facilities, and public transport)
- Over this heat map, the population estimation layer is put. This gives us the inside which areas are over or under populated, showing it in colour (red, orange, yellow, green)
- Ideally, AI would learn enough to suggest ideal changes following something like the Green city guidelines, by Michelle de Roo.
As we are just creating an idea to develop an AI platform, we mostly used Zoom for effective communication. We used Python with geographical library and Jupyter. Everything was done on our computers, none of them used specialised hardware.
Among our best achievements is that we have found the method to use on our platform to produce validated outputs. However, due to the restricted time frame, we did not produce the finished version, but just some previews. We have estimated the time needed for a challenge like this to be completed at least two years to become fully functional. To get the whole picture the AI would have to be able to create its own library for some of the infrastructure and urban services components as they cannot be remotely collected data. However, it could gain those data from official reports, like ones made from the United Nations. Similarly, it can incorporate some of the data into the calculations and predictions for climate change initiatives/solutions. For example in cities the initiative is to lessen the amount of carbon dioxide with planting more green surfaces.
How We Used Space Agency Data in This Project
At first we reviewed the data we were presented with. Then we tried to visualise how we could use as much of it as possible. Of datasets provided we used Grided Population of the World, version 4.11 and Population Estimation Service. We also planned to use HBASE and SEDAC’s visualisation tool. ESA’s resources were also used for our preview results. The plan in the future is for AI to make constant use of the presented data in collaboration with the library of non-remote sensing data.
Project Demo
Video of presentation:
https://www.youtube.com/watch?v=xTW8u_R5Yu0&feature=youtu.be
Powerpoint presentation:
Data & Resources
Michele de Roo. 2011.The Green City Guidelines. Available on: http://aiph.org/wp-content/uploads/2015/04/Green%20City%20-%20Guidelines.pdf (3. oct. 2020)
Jinyeu Tsou et al. 2017.Evaluating Urban Land Carrying Capacity Based on the Ecological Sensitivity Analysis: A Case Study in Hangzhou, China. Available on:
https://www.researchgate.net/publication/317255093_Evaluating_Urban_Land_Carrying_Capacity_Based_on_the_Ecological_Sensitivity_Analysis_A_Case_Study_in_Hangzhou_China (3. oct. 2020)
R.L.McConnell & Daniel C. Abel. Population size. https://people.wou.edu/~courtna/ch371/lecture/popgrowth/carrying.htm (3. oct. 2020)
Yigang Wei et al. 2015. Using Urban-Carrying Capacity as a Benchmark for Sustainable Urban Development: An Empirical Study of Beijing. Available at: https://www.mdpi.com/2071-1050/7/3/3244/htm (3. oct. 2020)
Amit Chatterjee et al. 2016. TRANSFORMING MUMBAI CITY: REMOVING THE BOTTLENECKS TO ACHIEVE FUTURE SUSTAINABILITY. Available at: https://www.researchgate.net/publication/309576384_TRANSFORMING_MUMBAI_CITY_REMOVING_THE_BOTTLENECKS_TO_ACHIEVE_FUTURE_SUSTAINABILITY (3. oct. 2020)
Resources that were available for this challenge.
Scrapped Google for health facilities in Ljubljana.