Machine Learning-based Ticket Triaging with 95% accuracy for a provider of IT and telecommunication services to the air transport industry

Client

One of the world’s largest provider of IT and telecommunication services to the air transport industry.

Challenges

1. Over 500 service tickets/month were being raised by users round the clock from over 250 countries
2. Teams had to service the global network of suppliers 24x7
3. the agreed SLAs had to be met during the service hours - 9.30 AM to 1.30 AM IST

Mindtree solution

Mindtree’s Cognitive Technologies Center of Excellence team was involved to evaluate if machine learning could be applied to learn from the past ticket data to achieve automated triaging

Machine learning model using Artificial Neural Net Framework was developed to analyze tickets raised during past 10 months to develop a machine learning model to predict the team that is best positioned to address the ticket

The prediction model was then applied to new ticket descriptions to predict the resolution team

Benefits delivered

• 95% ticket prediction accuracy
• 90% saving in the manual effort
• 24/7 monitoring and triaging
• Round robin ticket assignment
• Automated periodic training for machine learning