Summary

Establish and maintain service system representations to support capacity and availability management.

Description

Service system representations provide insight into how the service system will behave given specific work volumes and varieties. These insights are used to support decision making about resource allocation, changes to the service system, service agreements, and other aspects of service management and delivery.

For many services, demand fluctuates widely. Managing services in the face of widely fluctuating demand is one of the unique challenges characteristic of services. Depending on the patterns of fluctuation, the representations can focus on small or medium time intervals (e.g., by hour of the day for work shift scheduling, day of the week, month of the year) or longer time intervals (e.g., seasons of the year, bi-annually, annually).

Estimated growth of the use of service resources is formulated using collected capacity and availability data, estimated service requirements, and service system representations.

Measurement objectives and specific properties of the service system determine the nature and extent of a service system representation. (The service agreement has a major influence on the measurement objectives.) Experience, historical data, modeling expertise, and current resource use can also influence the nature of a service system representation.

Refer to the Measurement and Analysis (MA) (CMMI-SVC) process area for more information about establishing measurement objectives and specifying analysis procedures.


Representations can be used to analyze the impact of change requests that are likely to affect availability and capacity. Representations can also be used to characterize the range of future demand that can be met and the impact of required service levels on the service system. Before representations of future behavior or service system performance can be established, descriptions of the normal use of service resources and service system performance should be established.

 

Examples of service system representations that support capacity and availability management include the following:
  • Graphical representations showing a mix of two types of health care provider resources in a hospital with specific constraints and parameters indicating what might be the best allocation of the two resources
  • Analysis of waiting lines for bank tellers
  • Vehicle scheduling programs
  • Simulation modeling of transaction arrival rates against a specific configuration of resources (e.g., bank tellers, network servers)
  • Trend analysis of the availability, reliability, and maintainability of service system components
  • Impact analysis of service system component failure
  • Load testing to generate expected demand on a service system resource and ensure that service system components can perform according to the service agreement
  • Fault tree analysis and single point of failure analysis


Service system representations can be established to provide input to support development of the service agreement and descriptions of standard services and service levels.

Refer to the Service Delivery (SD) (CMMI-SVC) process area for more information about establishing service agreements.


Refer to the Strategic Service Management (STSM) (CMMI-SVC) process area for more information about establishing standard services.


Service system representations can be established during design of the service system. However, even if great care is taken during the design and development of the service system to ensure that it can meet service requirements over a wide range of operating conditions, service management and delivery should sustain the required levels of service system performance and quality during transition and operation.

 

SSD Addition
Refer to the Service System Development (SSD) (CMMI-SVC) process area for more information about developing service systems.


Service system representations are maintained throughout the service lifecycle.

Service system representations are generally not the same as the process performance baselines and models established in Organizational Process Performance (OPP) at levels 4 and 5. Several things distinguish representations from process performance baselines and models:
  • OPP process performance models and baselines involve the use of statistical techniques to assist in developing an understanding of the performance or predicted performance of processes. Service system representations are not typically required to be developed in this way.
  • Representations established in CAM are not required to be based on data collected from using the organization’s set of standard processes.
  • The focus of OPP is on process performance baselines and models. In addition to process data, the focus of CAM’s service system representations includes non-process data, people, and other parts of the service system such as infrastructure and automated systems.
  • Service system representations are established to support capacity and availability analysis specifically. This scope is narrower than the scope of OPP practices.


Refer to the Organizational Process Performance (OPP) (CMMI-SVC)Organizational Process Performance process area for more information about establishing performance baselines and models.


Although not required for capacity and availability management, representations provide opportunities to use statistical techniques such as statistical process control. These techniques can be used to quantitatively manage service system performance and quality and to improve service system capability.

Refer to the Quantitative Work Management (QWM) (CMMI-SVC) process area for more information about quantitatively managing the work to achieve the established quality and process performance objectives for the work.


Example Work Products



  1. Representations of resource and service use
  2. Representations of service levels
  3. Data on the use of resources and services
  4. Data on current service levels delivered
  5. Thresholds that define exception conditions and breaches


Subpractices



1. Collect measurements on the use of resources and services and the current service levels delivered.



2. Establish and maintain descriptions of the normal use of service resources and service system performance.

For some services, it may be advisable to establish general systems flow charts to identify the service system and its processes before determining the service system’s current capacity, which can require determining the capacity of service system components.



3. Establish and maintain service system representations from collected measurements and analyses.

For some services, it may be advisable to estimate the capacity of the service system at peak work volumes.



4. Review and get agreement with relevant stakeholders about the descriptions of the normal use of service resources, service system performance, and service system representations.



5. Make available the descriptions of the normal use of service resources, service system performance, and service system representations.



6. Establish and maintain thresholds associated with demand, workload, use of service resources, and service system performance to define exception conditions in the service system and breaches or near breaches of service requirements.

Thresholds are typically set below the level at which an exception condition or breach of service requirement occurs to allow corrective action to prevent the breach of service requirement, over-use of resources, or poor service system performance.