Giovanni Lanfranchi

Gaining Insight from Operational Data for Service Optimization

Service optimization is key in the drive to further automate and increase the effi-ciency of the services provided to clients. A given client may be ahead, in one or more areas, versus where other clients are. Learning from these best-of-breed cli-ents benefits the other clients in optimizing their services. Each of the clients normally generates a large number of data points for all the elements comprising the services they consume. Important data points are generally contained within incident tickets and change tickets. These tickets contain not only information about the nature of the issue, but also contain resolution information.

Practical Advice for Introducing AI into Service Management

In the following chapter, we summarize best practices for the introduction of AI in service management, for the benefit of IT leaders and their teams. In our view, the following points are essential: 1. Introducing AI has to be part of a holistic strategy of transforming service management. It requires both top-down leadership support as well as community engagement, and clear communication of expected results as well as limitations. It is not just about selection of technologies and training of practitioners but has deep implications for processes and organizational culture. 2. AI has great potential to significantly change and elevate the work that performed by human experts by identifying automation opportunities. This requires different skills and organizational setups and should be use to accelerate agile transformation. We suggest to run the introduction of AI as an agile project, with disciplined iterations and clear measurements [1] 3. A successful AI strategy for IT requires establishing a data-driven approach and culture. Understanding IT operational data (especially ticket data for incidents and service requests) is a prerequisite to knowing what questions to ask and how to accelerate the training of AI. 4. When building ground truths, leverage your organizational knowledge and treat it as an enterprise asset, but also be prepared to challenge hidden assumptions. Establish an end-to-end knowledge lifecycle strategy that engages the subject matter experts.

Gaining Insight from Operational Data for Automated Responses

Facilitating autonomic behavior is largely achieved by automating routine maintenance procedures, including problem detection, determination and resolution. System monitoring provides effective means for problem detection. Coupled with automated ticket creation, it ensures that a degradation of the vital signs, defined by acceptable thresholds or known patterns, is flagged as a problem candidate. It is a known practice to define thresholds or conditions that are conservative in nature, thus erring on the side of caution. This practice leads to a large number of tickets that require no action (false positives). Elimination of false positive alerts is imperative for effective delivery of IT Services. It is also critical for the subsequent problem determination and resolution. All operational data, including events and problem records, will be used for automated resolution recommendation.

Conversational IT Service Management

Conversational AI permeates our daily lives through chatbots such as Alexa, Siri, Google etc. We can ask questions about weather, music, shop for flowers or control various home electronics devices in a conversational manner. We believe that conversational interaction will dominate the next generation of interactions even in IT Service Management world. To that effect, in this chapter we describe an AI infused conversational self-service system that enables a user to get precise answers to various domain specific questions in a natural conversational manner. These questions span the gamut of 5 W + H (What, Where, Which, Who, Why, How) queries in an increasingly refined conversational context including some that may require performing transactions on behalf of the user – such as adding memory to a virtual machine or starting/stopping a database.

AI for Solution Design

Before we explain how AI could be leveraged to dramatically transform the Solutioning phase, it is important to outline the role of this phase across the Service Delivery lifecycle as well as the major personas involved.