NTT DATA Smart Ticket Analytics Use AI/ML To Uncover Optimization Opportunities

Ticket management systems are a vital part of an organization, and their management is key. While it may be possible to manage tickets effectively in the beginning when data is limited, it becomes cumbersome and time-consuming as the data grows. It is challenging to analyze and derive meaning from historical data for vast amounts of data from millions of tickets. As a result, many organizations shift their focus to quickly closing tickets while ignoring data analysis, which provides valuable insights that uncover optimization opportunities.

For instance, companies need to understand anticipated ticket volumes in managed services engagements before deploying their support services team and uncover optimization opportunities that will provide year-over-year (YOY) productivity gains. In short, they need to go from reactive to proactive. Companies often turn to industry-related research and/or benchmarks to plan for the future, but as each client environment is unique, this may not always address their specific needs.

Traditional ticket analysis

In the traditional approach, ticket analysis focuses on the volumetric, high-level classification of tickets, priorities and configuration groupings.

When companies perform ticket analysis to uncover optimization opportunities, it is important to determine the patterns, accuracy, and speed at which it is done.

Here are some of the challenges with traditional ticket analysis:

• Categorization of tickets relies on specific keywords and doesn’t consider semantic meanings supplied by users. For example, the keyword “reset” may have different meanings or be context-dependent. In an application, this may be due to a password reset that could be handled in minutes with a self-service option. Other resets, such as OS and infrastructure, may need expert assistance and take days to resolve. As previously noted, a simple, keyword-based system may miss this.
• Text descriptions are lengthy, which requires time and effort to uncover an issue. For instance, the “real” issue may be hidden within a lengthy description. Reading, comprehending and categorizing tickets is possible, but it’s cumbersome and time-consuming, especially when volumes increase.
• Forecasting ticket similarity and time series aren’t considered in traditional ticketing systems. Most of the extrapolation is linear and doesn’t consider spikes in ticket volume, even those due to seasonality(ies).
• Traditional ticket analysis takes multiple iterations, has limited capacity/volume capabilities for ingestion, takes longer to process, and may require manual intervention.
• There isn’t enough scope to leverage for insights or to apply learnings from historical data analysis.
• There is a lack of correlation between types of tickets, patterns, predictions and insights.

NTT DATA Smart Ticket Analytics

NTT DATA Smart Ticket Analytics uses artificial intelligence (AI) and machine learning (ML) to supplant traditional ticket analysis to better categorize tickets. Resultant insights allow IT to act before they hinder user productivity. The insights are context-sensitive and consider specific towers (Applications, Infrastructure, BPO) and domains, such as healthcare, financial services, and others.

NLP-based text processing is built into NTT DATA’s Smart Ticket Analytics system, so ticket descriptions are understood, which results in proper categorization. Smart Ticket Analytics has three major characteristics that differentiate it from traditional ticketing systems:

1. Analytical/NLP model-based text processing
By leveraging Natural Language Processing (NLP), data is accurately deciphered, which provides more comprehensive and accurate ticket categorization. Meaningful data is uncovered that is difficult and time-consuming if done manually. It has the potential to reveal the often-overlooked user opinions, especially when thousands of tickets must be reviewed. AI/ML-driven ticket analytics leverages NLP capabilities to categorize and group similar tickets.

2. Insight-Driven output
Smart Ticket Analytics provides clients with better, more targeted insights, enabling them to make better decisions from a single pane.

3. Continuous learning based on historical data
An AI/ML system uses continuous learning and adoption, which improves insights. As more historical data is made available, the system is continually trained to scale, providing deep insights into specific towers and domains.

This high-level diagram illustrates how NTT DATA Smart Ticket Analytics works.

The NTT DATA Smart Ticket Analytics system uses NLP to parse and “understand” or “summarize” the ticket, then uses learnings from the previous ML training sessions for classification. The platform uses proven ML algorithms and methods for this purpose.

Here are a few of the benefits customers enjoy with NTT DATA Smart Ticket Analytics:

• Opportunities identification — Shiftleft identification helps organizations move more straightforward tasks to less expensive technical human capital, providing savings and optimizing skills efficiency. “Access Provisioning” and “Password Resets” are typical examples of shiftleft activities. In both cases, they can be addressed in the following ways:
  • The activity can be moved from a highly technical team (Level 2) to a lower team (Service Desk, for instance) or even automated.
  • Non-technical resources can handle certain simple activities following standard ITIL processes. Complex work can be addressed by more technical personnel.
  • Access approvals can be simplified by obtaining user-based approvals instead of getting individual ones. This can also be easily micro-automated by following repeatable ITIL practices. For example, access can be provided by role-based — as opposed to individual-based.
• Forecasting future ticket volumes — Smart Ticket Analytics provides insights that help take remediation from the reactive to the proactive, enabling IT to prepare for upcoming events and staff up to ensure they’re ready. For instance, a batch system failure will likely create a steep rise in tickets, so uncovering it beforehand means reducing future tickets before affecting productivity.
• Root cause identification — Clients quickly learn why an event has taken place, saving considerable time and allowing them to create and deploy steps to prevent it from happening again.
• Reduction in costs — Clients save on both hard and soft costs, requiring fewer resources or better allocating them to work on other initiatives.
• Increased efficiency — With NTT DATA Smart Ticket Analytics, client IT teams can glean key insights through predictive analytics to streamline incident management, improve workflows and better incorporate demand planning.
• Faster remediation — The analysis of tickets is completed in hours, not days. Issues and trends are brought to light faster, enabling developers and solutions architects to address tickets before they result in unsatisfied employees and customers. It’s a shift from the reactive to the proactive.
• Enhanced knowledge sharing — While ticket categorization becomes smarter with machine learning, so do personnel by staying ahead of issues before they become disastrous.
• Aging and Backlog Ticket Analytics — Analysis of aging and backlog tickets provides valuable insights that assist in optimization and the reduction of tickets.

To learn more about how NTT DATA Advance Ticket Analytics can help you proactively remediate issues before they hinder users’ productivity, reach out to our experts. They would love to hear from you.