The Town Council contacted Center2M with the following request:
Develop a system that helps to manage the city's municipal public service centers quickly and efficiently
Using the system, solve the problem of reducing lines and overload
Implement the system within a short time
Fit into a limited budget
The Center2M team solved the problem using a video analytics system that allows:
The camera sends an image to the video analytics system whose artificial intelligence determines the number of people in a few seconds and returns the data to the Customers Service Center
Counting the number of visitors in real time
Using predictive analytics, the system makes a forecast of the size of the line at certain times of the office business hours
Determining and forecasting peak load times for the service center
Center management can quickly change the work schedule of free employees to distribute them with maximum efficiency during business hours
Distributing employees appropriately based on the schedule of peak times
SPECIALLY TRAINED NEURAL NETWORK COUNTED THE NUMBER OF PEOPLE IN THE FRAME AND DETERMINED HOW MANY PEOPLE WERE IN THE PUBLIC SERVICE CENTER AT ANY GIVEN POINT OF TIME
The neural network determines the number of people in the frame
Given all that, the management can receive information and monitor the situation remotely: all information is reflected on a graph which shows on which days the employees encounter the most requests and how much time they spend solving the problem for each visitor.
Once it decides that the number of people in line is becoming critical, the system alerts the center management within a few seconds. Having received the alert, the management can send to the front line those employees who are less busy or even ask for reinforcement from the next office.
A model detecting silhouettes was used to count the number of people.
The graph of load of a municipal public service centers