AI Ops and Automation

Trend 5 – Movement from simple runbook automation to a sentient digital workforce

As organizations begin to adopt cognitive automation, they find that a digital workforce can perform cognitive functions to help them move toward selflearning autonomous systems. A repository of bots can help accelerate this adoption because of the varied functions they can perform – from sensing an anomaly to resolving the failure and learning from the episode to improve the prediction. The key to this acceleration is the ability of bots to interact across multiple RPA and non-RPA technologies.

Consolidated automation workflows drive a successful digital workforce under a unique identity to create a digital twin of a human worker. This aggregation with the identity is also critical for auditing and tracing the actions of the digital workforce. The depth of the cognitive system capabilities, which are evolving rapidly, will define the acceptance and transition to these digital workers.

In partnership with Infosys, a leading European consumer goods manufacturer has embarked on this journey of building a sentient enterprise by leveraging three critical solutions from Infosys: Digital Brain, a technology that builds a knowledge graph to make sense of enterprise-wide data; Live Enterprise Application Platform (LEAP), software that provides a cognitive-first dashboard to detect anomalies and predict failures; and Infosys Cognitive Automation Studio, which helps build an army of cognitive bots to leverage abilities from Digital Brain and LEAP.

AI Ops and Automation

Trend 6 – Ticket triaging, solution prediction and auto resolution

A critical task in IT operations is to service tickets that either report failures or user requests. However, improper routing and miscategorization of tickets are typical challenges facing IT, resulting in delayed mean time to resolve (MTTR). Previously, deterministic automation routed the tickets to the correct assignee based on defined rules. Today, we have solutions driven by AI that learn from historical data and identify the right category of the ticket based on the problem/ request details. These solutions enrich the tickets with appropriate information to aid in faster response and resolution, greatly reducing MTTR.

Today’s systems currently rely on deterministic rules to identify the resolution path and corresponding automation, if available. While this is a step forward from the manual triggering of a relevant automation solution, the technology is still limited by defining the rule upfront for identification. New developments in AI are helping us predict solutions based on historical trends and knowledge artifacts. Once the resolution is identified, performing that action becomes just a matter of triggering the right bots. This evolved AI will help create, in essence, self-healing systems.

An Asian tax regulatory body and a client of Infosys used our in-house solutions, including an Intelligent Automation tool, to classify, enrich and route their tickets to the right support engineer more efficiently and reduce MTTR by 20%.
A support team of FINACLE, Infosys’ banking solution, automated its ticket classification and analysis to identify top solutions. This implementation leveraging Nia is a selflearning solution that improves accuracy and relevance based on user feedback.

Subscribe

To keep yourself updated on the latest technology and industry trends subscribe to the Infosys Knowledge Institute’s publications

Infosys TechCompass