PrivatBank has taught the neural network to deal with the queues in the offices
- Company news
PrivatBank has taught the neural network to deal with the queues in the offices
© Dnia
Today, the network of PrivatBank there are 2 183 branches, 7 246 12 612 ATMs and self-service terminals.
Ukrainian PrivatBank has implemented a system of analysis of service quality based on neural networks. Ukraine’s first “smart” control system of quality of service automatically monitors data from the surveillance cameras installed in each Bank branch. If any office near the cash registers a queue forms, the system analyzes the information and switches the script service mode “Turn,” and also sends a message to managers to make organizational decisions and redistribution of clientpolicy.
“In the process of training the neural network, we analyzed about half a million live images from the cameras in our offices and taught the network to correctly identify the queue outside banks, – said Deputy Chairman of PrivatBank Vadim Kovalyov. – We are working on the application of neural learning networks in practice identification of risk that will significantly accelerate and improve the reliability of risk assessment when lending to retail borrowers and entrepreneurs”.
System analysis of queues with the use of neural network will enable the Bank to optimize the performance of branch staff and improve the quality of customer service. The pilot launch of deep neural network system showed almost perfect recognition accuracy turns. While on an average day, more than 50% of all Bank branches, the neural network estimates as offices, where no large queues.
According to the developers, in the near future PrivatBank will teach a neural network to recognize the queue at the ATMs and POS terminals. These data will be used when planning the load of cash and expanding the ATM network.
Today, the network of PrivatBank there are 2 183 branches, 7 246 12 612 ATMs and self-service terminals.