When reporting a problem or requesting a service from IT, we make calls, send emails—or in most modern enterprises—go directly to a web portal to create a service ticket. Service tickets can also be auto-generated when something goes down such as an ATM machine, toll scanner, or some other Internet of Things (IoT) device.
Property fields and categorizations are part of the ticket as incident data. If not auto-generated, information must be manually entered along with choices from drop-down lists – for anything ranging from priority, department, category, or region to specifying a time and date for the ticket to be resolved.
The information that is contained in these incident service tickets within complex digital environments is a great source of actionable insight for C-level executives, data scientists, and IT operators.
“The digital transformation required to remain or become competitive means innovating and doing more for less. Enterprises need to get the right digital experience to the right users at the right time to achieve a business outcome regardless of costs.” -IT Analyst
Service tickets and IT operations
The infrastructure and apps that run the business are the responsibility of IT operations while the IT service desk handles anything from employees wanting help with their digital office equipment to customers submitting tickets for a service that is provided directly to them. Some enterprises have multiple ticketing systems and do record keeping for just about every process including the delivery of software and documentation. Getting the most out of service ticket data requires not only responding to them in a timely manner but also integrating them into a comprehensive IT operations strategy.
Improved MTTR with efficient event management
Depending on the enterprise, service tickets can be associated with a configuration management database (CMDB) that is made up of the elements or configuration items (CIs) that are related to a service provided by IT. In some cases, there might not be a CMDB but there are IT infrastructure elements that can be related to the IT service—the network elements, the servers running the apps, the apps themselves, or even the vast storage needed by most enterprises.
It is a best practice to discover the CMDB and associate elements with monitored IT services. When there’s a performance issue with any one of the elements related to the service, other related elements can be impacted and troubleshooting begins based on priority. This event in operations management can be associated with a service ticket depending on the level of integration that is configured:
Level 1—Incident created in ticketing system and linked to IT operations event
Level 2—Incident created in ticketing system for CI and linked to IT operations event
Level 3—Two Incidents created in ticketing system, one for the selected CI that is causing the outage and a second the for service that is impacted by the causal CI and both linked to the IT operations event
These levels of integration help with the more efficient resolution of IT issues. Moreover, once patterns of incidents are identified, automation can be configured to remediate issues without human intervention.
Insights from machine learning and analytics
Another way to elevate IT operations and incident management is to apply machine learning and analytics to the volume and velocity of service tickets to help achieve the digital transformation of decision making. For example, a service ticket analysis of 1,000’s of tickets per day – or even per week – could reveal that password resets have tripled in the last month and the amount of time to address them with IT staff could be greatly reduced if the resets were handled automatically instead of with human intervention.
What if the volume of incident data unexpectedly spikes? Using a machine learning and analytics approach, the condition would first be flagged as abnormal based on what has already been learned from the past. Correlating this spike with trends based on properties like geography, problem, application, and so on could help quickly identify a source. Furthermore, if needed, you could apply natural language processing to the description field and quickly understand problem details.
Many enterprises are realizing that applying these ways of integrating service ticket incident data with the core of IT operations can help advance the business by being more efficient and putting valuable insight in the hands of decision makers across the enterprise.
Artificial Intelligence for IT Ops (AIOps) is the term that Gartner analysts have introduced to describe this approach that leverages machine learning, analytics, and a big data platform, as well as remediation through automation.
BMC’s TrueSight helps with analytics and remediation on time-series and event performance data, incident tickets, business metrics, and social sentiment by leveraging the BMC portfolio of market-leading solutions.
For more information, see:
These postings are my own and do not necessarily represent BMC's position, strategies, or opinion.