Browse Prior Art Database

Derivation of Appropriate SLA Measurements From the Combinations of States

IP.com Disclosure Number: IPCOM000020504D
Original Publication Date: 2003-Nov-25
Included in the Prior Art Database: 2003-Nov-25
Document File: 2 page(s) / 14K

Publishing Venue

IBM

Abstract

For Service Level Management one of the key sources of data Incident and Change Request data is from Service Desk providers. Providing Service Level Agreements (SLAs) on this data is difficult because customers frequently modify Service Desk solutions. Thus, using a fixed set of measurements and states becomes difficult to generalize across customer environments. In addition, the tracking of key SLA measurements such as "Time to Fix a Problem" is difficult because the single measurement is actually composed of a total of the time spent in many different states. This invention defines a method by which Service Desk data can be transformed into an assessable SLA quantity from measurement data associated with a set of discrete states. There are two primary reasons for this further refinement of the data: 1) simple measurements, such as the wall clock time, do not allow the flexibility to avoid including certain measurement time from scheduled events (scheduled downtime, weekends, holidays, etc.); and 2) the ability to exclude time duration measurements that are beyond the service provider's control (i.e. such as waiting for parts, waiting for more input from the customer, etc.).

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 2

Derivation of Appropriate SLA Measurements From the Combinations of States

For Service Level Management one of the key sources of data Incident and Change Request data is from Service Desk providers. Providing Service Level Agreements (SLAs) on this data is difficult because customers frequently modify Service Desk solutions. Thus, using a fixed set of measurements and states becomes difficult to generalize across customer environments. In addition, the tracking of key SLA measurements such as "Time to Fix a Problem" is difficult because the single measurement is actually composed of a total of the time spent in many different states. This paper defines a method by which Service Desk data can be transformed into an assessable SLA quantity from measurement data associated with a set of discrete states. There are two primary reasons for this further refinement of the data: 1) simple measurements, such as the wall clock time, do not allow the flexibility to avoid including certain measurement time from scheduled events (scheduled downtime, weekends, holidays, etc.); and 2) the ability to exclude time duration measurements that are beyond the service provider's control (i.e. such as waiting for parts, waiting for more input from the customer, etc.).

The general flow has four key steps.

1. Define a collection of states (XML was used as the modeling language for this invention). Each state has a name and a description. The collection contains a minimum of 2 states (an initial state and a final state), and can also contain a subset of one or more states known as "transition states". The "transition states" define a progression from the initial state to a final state. For example, a change request is "opened" (initial state), then it is "reviewed" (transition state 1), "accepted" (transition state 2), and "implemented" (final state).
2. Define metric names which include any combination of states. Generally these measurements represent "Time To x" where x is often approve, fix, resolve, implement, restore, recover, or any other measurement that can be part of a workflow.
3. Export the state definitions and measurements. The Service Desk Providers will export the state definitions along with measurements for how long a resource was in each defined state.
4. Import the state definitions and time definitions. Consuming applications (Service Level Agreement Applications fo...