The Viladecans City Council has integrated artificial intelligence (AI) in the classification system of incidents on public roads. This automation improves the efficiency of incident management, reducing response time and ensuring immediate attention to citizens.
The Viladecans&Tu mobile application allows citizens to report incidents on public roads, but the large number of types of incidents (more than 1000 different ones) makes it complex to use, causing many people to abandon the process. The incorporation of AI automates the classification, simplifying the communication process and eliminating the need for municipal staff to correct errors, offering an easier and more accessible experience for citizens.
This solution is documented in theDigiCanvis Observatory as an innovative initiative of the local administration.
Com funciona?
The City Council has implemented AI to automate the process of classifying incidents reported by citizens based on photographs. With the new system, the process of reporting an incident is as follows:
- The person accesses the mobile application and selects the option to report an incident.
- Then capture a picture of the incident on the public road and upload it to the application (it is recommended to attach a detail photo and one of the general environment).
- The AI processes the image in real time, identifies the type of incident, and presents three classification options for the user to choose the most appropriate one.
- If none of the options are correct, you can sort manually.
- If the AI does not have enough information to identify the incident type, the user persona is returned to the detected level or manual classification.
- Once the incident is classified, additional comments or details can be added. The application also geolocates the incident automatically thanks to the GPS of the mobile device.
- Finally, the user sends the communication and the incident enters the centralized management system of the town hall for follow-up and resolution.
To ensure that AI classifies incidents accurately, it is essential to train it with a wide variety of photographs. representatives of each type of incident. This training is an iterative process, which includes different stages of testing and corrections to continuously improve the system.
What are the benefits?
- Improved efficiency. Automatic classification allows incidents to be handled more quickly, reducing response and resolution times.
- Improved user experience. The solution simplifies the process of reporting incidents, saving time and facilitating an agile and understandable use of the application.
- Promotion of citizen participation. The speed of the system makes it more likely that citizens will use the application to report incidents.
The service has been in operation since April 2024 with very good results: the number of incidents sent by citizens has increased by 40%.
You can find out the key aspects to take into account to implement this initiative in the file Application of AI in incident management.
