The 15-Minute City: How AI Understands the Places We Live
October 3, 2025
The 15-Minute City: A Framework for Measuring Urban Quality
The 15-Minute City: A Framework for Measuring Urban Quality
A 15-minute city is an urban planning concept that aims to make all essential services and amenities accessible within a 15-minute walk from every resident’s home. The basic idea is simple: the more diverse and numerous the amenities, the better.
The concept typically generalizes the idea of a city designed for people. However, without considering individual needs, it often evaluates the city for an ‘average’ person with standard preferences and common categories of Points of Interest (POIs).
For example, a location ideal for a young professional focused on career advancement and nightlife may not suit a family with young children seeking quality schools and green spaces. Recognizing these unique preferences allows for a more personalized evaluation of locations, ensuring that the concept truly serves the diverse needs of every individual within the community.
The challenge arises when attempting to account for each person’s individuality. Manually matching various POIs (e.g., museums and galleries for artists, offices for business professionals) to every personality type is overwhelming due to the sheer variety.
At Aino, we have created a new service that evaluates a location using the concept of the 15-minute city. Our platform utilizes POI data from Open Street Map, 15-minute isochrones from MapBox API, and AI assessment to offer a comprehensive analysis of an area. This allows users to see a list of points of interest within a 15-minute walking distance from their chosen location.
A 15-minute city is an urban planning concept that aims to make all essential services and amenities accessible within a 15-minute walk from every resident’s home. The basic idea is simple: the more diverse and numerous the amenities, the better.
The concept typically generalizes the idea of a city designed for people. However, without considering individual needs, it often evaluates the city for an ‘average’ person with standard preferences and common categories of Points of Interest (POIs).
For example, a location ideal for a young professional focused on career advancement and nightlife may not suit a family with young children seeking quality schools and green spaces. Recognizing these unique preferences allows for a more personalized evaluation of locations, ensuring that the concept truly serves the diverse needs of every individual within the community.
The challenge arises when attempting to account for each person’s individuality. Manually matching various POIs (e.g., museums and galleries for artists, offices for business professionals) to every personality type is overwhelming due to the sheer variety.
At Aino, we have created a new service that evaluates a location using the concept of the 15-minute city. Our platform utilizes POI data from Open Street Map, 15-minute isochrones from MapBox API, and AI assessment to offer a comprehensive analysis of an area. This allows users to see a list of points of interest within a 15-minute walking distance from their chosen location.
A 15-minute city is an urban planning concept that aims to make all essential services and amenities accessible within a 15-minute walk from every resident’s home. The basic idea is simple: the more diverse and numerous the amenities, the better.
The concept typically generalizes the idea of a city designed for people. However, without considering individual needs, it often evaluates the city for an ‘average’ person with standard preferences and common categories of Points of Interest (POIs).
For example, a location ideal for a young professional focused on career advancement and nightlife may not suit a family with young children seeking quality schools and green spaces. Recognizing these unique preferences allows for a more personalized evaluation of locations, ensuring that the concept truly serves the diverse needs of every individual within the community.
The challenge arises when attempting to account for each person’s individuality. Manually matching various POIs (e.g., museums and galleries for artists, offices for business professionals) to every personality type is overwhelming due to the sheer variety.
At Aino, we have created a new service that evaluates a location using the concept of the 15-minute city. Our platform utilizes POI data from Open Street Map, 15-minute isochrones from MapBox API, and AI assessment to offer a comprehensive analysis of an area. This allows users to see a list of points of interest within a 15-minute walking distance from their chosen location.
Leveraging AI for Spatial Data Assessment
Leveraging AI for Spatial Data Assessment
Large Language Models (LLMs) are AI models that process text inputs and generate text outputs. Some models are task-specific, like translation, while others, like GPT, can perform a wide range of tasks based on instructions. LLMs can address the challenges of personalized spatial assessments.
After extensive testing, we developed an optimized formula using the GPT model to assess the suitability of a 15-minute city based on OSM data. We then described it as an Index along with a short text description.
Our instruction to the model looks like this:
Large Language Models (LLMs) are AI models that process text inputs and generate text outputs. Some models are task-specific, like translation, while others, like GPT, can perform a wide range of tasks based on instructions. LLMs can address the challenges of personalized spatial assessments.
After extensive testing, we developed an optimized formula using the GPT model to assess the suitability of a 15-minute city based on OSM data. We then described it as an Index along with a short text description.
Our instruction to the model looks like this:
Large Language Models (LLMs) are AI models that process text inputs and generate text outputs. Some models are task-specific, like translation, while others, like GPT, can perform a wide range of tasks based on instructions. LLMs can address the challenges of personalized spatial assessments.
After extensive testing, we developed an optimized formula using the GPT model to assess the suitability of a 15-minute city based on OSM data. We then described it as an Index along with a short text description.
Our instruction to the model looks like this:
LLM Instructions for Assessing 15-Minute City Suitability
“Take the list of amenities and their amounts: {json_object} in the area and assess it from the perspective of the 15-minute city concept for {persona}. Use the following rules:
Create a list of crucial amenities for the given persona type.
Compare this list with the amenities provided as input.
Rate the area from 0 to 100 (where 0 means it doesn’t match at all, and 100 is a perfect match).
Return a descriptive evaluation of the area in addition to the numeric rating.
Provide your answer as a JSON object with the schema: {“rate”: numeric, “description”: text}.”
There are additional, more specific rules related to data processing.
LLM Instructions for Assessing 15-Minute City Suitability
“Take the list of amenities and their amounts: {json_object} in the area and assess it from the perspective of the 15-minute city concept for {persona}. Use the following rules:
Create a list of crucial amenities for the given persona type.
Compare this list with the amenities provided as input.
Rate the area from 0 to 100 (where 0 means it doesn’t match at all, and 100 is a perfect match).
Return a descriptive evaluation of the area in addition to the numeric rating.
Provide your answer as a JSON object with the schema: {“rate”: numeric, “description”: text}.”
There are additional, more specific rules related to data processing.
LLM Instructions for Assessing 15-Minute City Suitability
“Take the list of amenities and their amounts: {json_object} in the area and assess it from the perspective of the 15-minute city concept for {persona}. Use the following rules:
Create a list of crucial amenities for the given persona type.
Compare this list with the amenities provided as input.
Rate the area from 0 to 100 (where 0 means it doesn’t match at all, and 100 is a perfect match).
Return a descriptive evaluation of the area in addition to the numeric rating.
Provide your answer as a JSON object with the schema: {“rate”: numeric, “description”: text}.”
There are additional, more specific rules related to data processing.
Personas
Personas
When the rules are input into the model, the data is processed and compared to the model’s internal context. This approach eliminates the need to hardcode lists of target amenities for each persona and location and allows the model to understand and meet individual needs autonomously.
At Aino, beyond evaluating the ‘average’ person, we designed diverse personas, including playful examples like Hobbits and Elon Musk, as well as practical ones like families with kids. These demonstrate how a location suitable for one person might be entirely unsuitable for another.
We consider the LLM to be a flexible and precise computational core, processing the data with text-based instructions instead of relying on traditional deterministic functions and conditions.
When the rules are input into the model, the data is processed and compared to the model’s internal context. This approach eliminates the need to hardcode lists of target amenities for each persona and location and allows the model to understand and meet individual needs autonomously.
At Aino, beyond evaluating the ‘average’ person, we designed diverse personas, including playful examples like Hobbits and Elon Musk, as well as practical ones like families with kids. These demonstrate how a location suitable for one person might be entirely unsuitable for another.
We consider the LLM to be a flexible and precise computational core, processing the data with text-based instructions instead of relying on traditional deterministic functions and conditions.
When the rules are input into the model, the data is processed and compared to the model’s internal context. This approach eliminates the need to hardcode lists of target amenities for each persona and location and allows the model to understand and meet individual needs autonomously.
At Aino, beyond evaluating the ‘average’ person, we designed diverse personas, including playful examples like Hobbits and Elon Musk, as well as practical ones like families with kids. These demonstrate how a location suitable for one person might be entirely unsuitable for another.
We consider the LLM to be a flexible and precise computational core, processing the data with text-based instructions instead of relying on traditional deterministic functions and conditions.
Conclusion
In summary, the 15-Minute City tool offers a quick, personalized location assessment, showcasing the potential of LLMs in evaluating spatial data. By allowing for more individualized urban planning, this system demonstrates how AI can enhance our understanding of the cities we live in.If you want to generate your 15-minute city analysis with your data and personas you can do it using the Aino platform. Contact us to onboard you for free.
Aino is an AI for spatial data analysis. Aino transforms data questions into interactive maps, charts, and graphs. Read more about us on our LinkedIn, Instagram and website.
In summary, the 15-Minute City tool offers a quick, personalized location assessment, showcasing the potential of LLMs in evaluating spatial data. By allowing for more individualized urban planning, this system demonstrates how AI can enhance our understanding of the cities we live in.If you want to generate your 15-minute city analysis with your data and personas you can do it using the Aino platform. Contact us to onboard you for free.
Aino is an AI for spatial data analysis. Aino transforms data questions into interactive maps, charts, and graphs. Read more about us on our LinkedIn, Instagram and website.
In summary, the 15-Minute City tool offers a quick, personalized location assessment, showcasing the potential of LLMs in evaluating spatial data. By allowing for more individualized urban planning, this system demonstrates how AI can enhance our understanding of the cities we live in.If you want to generate your 15-minute city analysis with your data and personas you can do it using the Aino platform. Contact us to onboard you for free.
Aino is an AI for spatial data analysis. Aino transforms data questions into interactive maps, charts, and graphs. Read more about us on our LinkedIn, Instagram and website.