Using AI in OVANET's Service Desk system

OVANET a.s. is a provider of IT services and solutions in the field of information systems and technical asset management. It focuses on the operation of video surveillance systems, metropolitan network, call center and IT outsourcing. In addition to the Ostrava City Council and city district offices, its customers are also contributory organizations of the Statutory City of Ostrava.

Since 2018, OVANET has implemented a Service Desk from the Czech manufacturer ALVAO in cooperation with ARICOMA (formerly AUTOCONT). In 2023, this system was significantly improved by the deployment of artificial intelligence (AI).

The main tasks of AI are:

• When a new ticket is created, automatically search the knowledge database for potential matches in the database of resolved tickets and offer the right solution based on previous experience.
• When passing a ticket between different solvers, summarize the previous steps and comments into a brief summary.
• Accelerate the detection of major outages and high-impact incidents with an AI assistant that automatically identifies major incidents by analyzing signals and data from different systems.

These features have significantly reduced the time it takes to resolve routine and recurring technical issues. It has also significantly sped up the transfer of tickets between solvers, with the time it takes for a new solver to navigate a problem dropping to less than a quarter.

The implementation of AI into OVANET's Service Desk system was carried out in cooperation with ALVAO, which provided its know-how and technology. The project started with a detailed analysis of the current processes and identification of key areas that required improvement.

The first step was to set up an AI service where the AI was adapted to OVANET's specific needs. The AI was integrated into the incident classification process, which greatly accelerated the identification and assignment of the correct resolver. This enabled different types of requests to be dealt with quickly and efficiently without the need for manual intervention for each incident. The next step was the introduction of AI to oversee ticket resolution. The AI monitored the progress of incident resolution and ensured SLAs were met. The system automatically highlighted potential problems and suggested solutions.

However, the main benefit of AI was handling situations where the assignment was not completely clear. The AI was able to analyse ambiguous incident descriptions, and instead of the solvers having to think long and hard about how to interpret the assignment, the system offered relevant suggestions for next steps. This approach contributed to a significant reduction in the time taken to solve more complex and ambiguous technical problems.

One of the key technologies that played a vital role in this process was the configuration database (CMDB), which allowed AI to quickly analyze dependencies between systems and identify areas affected by failure.