Attaining a CMMI level (maturity level / capability level) is not a one time job. Without a clear focus it could not be sustained. There are cases where CMMI HM compliant organizations degraded to ML 3 or ML 2 at  the time of next assessment due to improper maintenance of practices. Let me pinpoint certain points to ensure a smooth sustenance of CMMI HM

  • Definitely first and foremost is senior management support. Then,
  • There has to be  teams  responsible for  metric and improvement initiatives related activities.
  • The team should be responsible for making periodical baselines, analysis of trends across baselines, fine tuning PPMs , target analysis  (baselines Vs targets), root cause analysis including analysis of common causes of variation, improvements identification and validation to meet the revised targets set by senior management.
  • This team shall also be responsible for ensuring the implementation of process improvements within projects and analyzing the data from process improvements on a continuous mode, not merely at the end (at the time of target analysis)
  •  Strong auditing practices needs to be ensured both at the project level as well as at the organizational lelvel. The progress of the above said team also should be audited.



High Maturity Clarifications

  • BO needs to be quantitative and traceable to sub process measures through PPO.
  • Organizational Innovation and Deployment (OID) changed to Organizational Performance Management (OPM).
  • CAR an advanced PA, could be applied across all Pas.
  • For more information refer  CMMI Dev V 1.2 Vs V 1.3 – ML5 PAs

More Effective GPs

  • Renamed GP 2.6 as ‘Control Work Products’ (earlier it was ‘Manage configurations’).
  • GP 2.9 is now ‘Objectively evaluate adherence of selected work products’. ‘Selected work products’ now provide more flexibility.
  • Simplified the GP 3.2. Earlier statement ‘Collect work products,

    Read More

  • Define project (say project X) goals based on the organization’s process performance objectives and customer requirements.
  • Ensure the validity of using organization PPM and Baselines. If the data used by organization for making baselines and models are not at all similar to the project X, then those models can not be used inside the project. In such cases if the project X is having any kind of past historical data, the project team can make their own PPMs.
  • Check the probability of achieving the project goals with the PPM.
  • The independent parameters/sub process measures influencing dependent parameter/PPOs is defined in process performance model.
  • Definitely with the current performance of sub process measures, the PPOs can not be achieved. Improvement initiatives needs to be identified and action taken as in article HM Practices in a CMMI ML 5 Organization.
  • After substituting the anticipated sub process specification limits in PPM, check the probability of achieving PPO. If probability value is acceptable (above the critical probability value defined by the organization), continue with the new process . If not, identify more improvement initiatives and repeat the steps till PPOs are achievable with an acceptable probability
  • Identify the critical sub process parameters for quantitative and statistical management
  • Do Run time monitoring, root cause analysis and PPM calibration

In CMMI V 1.3   Level 4 and 5 generic goals and practices are eliminated. This means there is no CL 4 and CL 5 . So here comes a question, how high maturity can be achieved in a continuous representation. Answer comes through equivalent staging method.

High maturity that is equivalent to staged maturity level 4 using equivalent staging is attained when you achieve capability level 3 for all process areas except for Organizational Performance Management (OPM) and Causal Analysis and Resolution (CAR).
High maturity that is equivalent to staged maturity level 5 using equivalent staging is attained when you achieve capability level 3 for all process areas.

A comparison of Capability and Maturity Levels is given below
CMMI HM in a Cont. Rp.

Assume current baselines of productivity inside the organization is 24 to 30 units. Now the organization decides to improves it to 30 (LSL) to 36 (USL) units based on some business requirements. Here, probability to achieve the target is 0 % with the current process. Organization needs to come up with improvement initiatives to change the process.

•Say for example, Usage of ‘x’ tool will enhance coding speed which in turn will results in an increased productivity based on the correlation between productivity and coding speed

First of all need to  decide the critical probability with which the improvement initiatives could be chosen after simulation analysis, say it as 75 %. The further steps in Identification and deployment of Improvement Initiatives are as given below.

Target analysis shall be done at the time of next baselinign to check whether the deployed process change has resulted in improvements ,in the expected range. During next PPB, the improvement in the identified factor for improvement proposal as well as improvement in PPO shall be analyzed as below..

CMMI Ml 5 -2


High Maturity practices should be implemented genuinely inside the CMMI compliant organization in order to ensure the proper institutionalization of the same. Optimized process should be used so that it won’t be burden upon the employees of the organization. Process optimization should be done probabilistically rather than deterministically.  Probabilistic analysis should be done at the organizational level as well as project level. It is an integral part in the realistic goal setting. The below figure illustrates the High Maturity practices in short, to be done at organizational level and project level.

Read the article” to have an integrated illustration of the same.

  • A PPM is a prediction model used to forecast  the performance of the identified PPO parameter. It is also used for identifying the critical sub process parameters for quantitative/statistical management
  • PPM is made in such a way that  probability to achieve the current performance of PPO parameters with current performance sub process parameters should be  99.97 % or at least  greater than 90%
  • Depending upon sub process parameter identified for improvement proposal (say it as x) to meet the targets, there are two cases to be analyzed with PPM.
    •   x could be  a parameter which was already identified as critical sub process parameter affecting PPO during last PPM building process itself.
    •   x is a completely new process introduced in to the organization, hence there is no data currently for x. Hence x need not be a parameter in the current PPM. Rather it could be added to PPM at the time of next baselining if found feasible.
  • Using PPM probability of achieving targets could be checked using  simulation methods .
  • While doing the probability analysis, a critical probability value should be opted in such a way that  below which process should not be acceptable. i.e. .Probability to achieve the PPO should be greater than 75% (here, the risk in losing the targets is 25 % and which is acceptable to the organization. )
  • Probability to achieve the targeted performance of PPO parameters with targeted performance of sub process should be  greater than  the critical probability value set.
  • Now if x (new sub process, helping to achieve the target)  is not a parameter in current PPM, probability analysis would be limited to prediction of current performance only and not an analysis towards meeting the targets. In that cases, it should be substantiated with enough evidence of implementing the new initiative inside the project and corresponding piloting/validation reports
  • If 75 % probability is there to achieve the target, the PM can go ahead with the project defined process. With PPM project should get this probability after simulation.
  • PPMs should be recalibrated if any process change is happening inside the projects/organization or based on any other triggers affecting the PPOs