how we implemented ai analytics to improve comfort and safety

In June 2025, we launched a large-scale project in parks across the Moscow Region. The goal was to equip public green spaces with intelligent monitoring systems. 

Park management teams operate in a demanding environment. They need to maintain a high standard of service, respond quickly to potential incidents, and manage resources efficiently across vast territories.

In this case study, we explain how the L-Labs team developed and implemented an AI-powered platform that transformed the way parks are monitored, analyzed, and managed.

The Challenge

The client set three core objectives:

1. Ensure a high standard of service.
Guarantee high-quality cleaning, visitor safety, and reliable infrastructure for millions of park guests.

2. Create a unified and transparent monitoring system.
Provide continuous, real-time oversight of the condition of park areas.

3. Digitize management processes.
Replace routine inspections and manual reports with modern AI analytics, freeing up resources for strategic development.

Our Solution

Our team developed an intelligent platform that:

Automates routine processes.

The system collects and structures data on infrastructure conditions and the completion of scheduled maintenance tasks.

Serves as a single source of truth.

It consolidates information from all monitored sites into clear dashboards for analysis and effective management.

Detects problems at an early stage.

The platform identifies defects and emergency situations before they affect visitor comfort, enabling teams to respond quickly and proactively.

How the L-Labs System Works

The system receives video streams from surveillance cameras, analyzes them using computer vision algorithms, and generates reports in a format convenient for the client. The model is self-learning and was additionally trained on thousands of images and video streams from parks across the region.

System Benefits:

For rapid response: notifications.

Efficient Data Transmission:

All events are displayed on a secure web dashboard, allowing managers to monitor the current situation, analyze data, and make informed decisions.

For analytics and reporting: dashboards.

Implementation Stages:

A large team worked on the project on our side, including:

On the client’s side, domain experts helped adapt the system to real operational needs.

The process consisted of three stages:

1. Joint design.
Together with the client, we defined priority metrics and key quality indicators.

2. Pilot implementation.
We launched the system at selected sites to test and optimize its performance.

3. Iterative fine-tuning.
Based on feedback, we adjusted the algorithms to improve accuracy.

Metrics we tracked

The project was implemented in ten major parks:

Maintenance monitoring

Each park has a daily cleaning schedule. Our system automatically checked whether the required work was completed on time and to the expected standard.
The monitored metrics included:

Type: Manual cleaning of pedestrian paths

Scheduled time: 09:00–10:00

Actual completion time: 09:09–09:20

Quality: Visual evidence of completion confirmed. Work completed to standard.

Camera ID: 243913

Work completed:

* Children’s playgrounds
** e.g., benches, pavilions, trash cans

Defect detection

Our technology automatically identified the following issues across park areas:

Defect detected!

Type: Overflowing bin

Camera: 250699

Detection time: 09:07

Resolution deadline: 11:07

Incident detection

The system also monitored situations requiring immediate response:

Camera: 219980

Detection time: 12:48

Resolution deadline: 13:18

Urgent incident

Type: Unattended bag

This is how AI analytics helped improve safety and operational responsiveness in parks across the Moscow Region.

After the intelligent platform was introduced in parks across the Moscow Region, park management became more transparent and responsive. This had a positive impact on operational efficiency and the quality of visitor service.

Here are the key outcomes:

Enhanced safety. The system became a reliable assistant in emergency situations, instantly notifying security teams or emergency services and helping to protect visitors.

Higher quality standards. Clear, objective indicators made it possible to unify service quality across all regional parks included in the project.

Data-driven decision-making. Analytical tools allow management teams to plan budgets, redistribute resources, and improve operational efficiency based on objective data.

Project results

AI-powered control and analytics helped transform the management of the region’s park infrastructure and integrate it into a unified network. Instead of subjective assessments, managers now rely on objective data and predictive analytics, creating the foundation for consistently high standards of service and safety.

L-Labs acted as a reliable partner in implementing digital solutions that not only address current operational needs but also set the direction for the future development of the entire park network in the Moscow Region.