The client is a global brokerage provider that used a large volume of tickets in day-to-day operations. These tickets had to be assigned to appropriate resources such that quick and effective ticket resolutions could be achieved.
The client manually assigned tickets based on human intuition and knowledge, even though the process itself required simple Yes-No judgements with rigid guided diagnostics, making it prime for automation. However, the machine learning techniques employed for ticket assignment did not take domain knowledge into consideration, creating a roadblock in the automation process.
To resolve this issue, our team deployed DeepInsightsTM. The platform enables enterprises to engage with their customers through personalized experiences and explore newer business models that leverage the potential of anywhere, any time, on any device computing capabilities. We took a step by step approach as follows:
• Guided diagnostics on a probabilistic basis embedding expert knowledge while considering utility, satisfaction, cost, and time of repair in a multi-criterion decision making framework
• Extracted key features from past ticket resolution data and used past data to predict probabilities of closing tickets by different users through Bayesian networks
• Assigned tickets to resources who are most likely to resolve them based on this calculated probability
• Created online learning algorithms to assign and learn simultaneously
Our innovative solutions allowed the client to experience the following benefits:
• Rapid email data ingestion and intelligent provisioning for analysis
• Minimized the ask on domain experts for email assessment and content summarization
• Reduced time for processing and segregating emails based on content
• Achieved real-time routing of emails to right recipient
• Deployed intelligent, end-to-end, workflow platform for email processing, routing, and knowledge management