Problem Management Trend Analysis - PDF

Document Sample
Problem Management Trend Analysis - PDF Powered By Docstoc

                                  Krishna Murthy Dasari – Satyam Computer Services Ltd.


         This paper discusses the role of Six Sigma in the Problem Management area of
Information Technology Infrastructure Library (ITIL). ITIL Problem Management has
two aspects the reactive aspect and the proactive aspect. This paper discusses on the
proactive aspect of Problem Management and provides an insight on when to initiate
correction from the trend analysis report. How Six Sigma can help in identifying when to
initiate correction, is demonstrated by a simple example.

                                   SIX SIGMA IN PROACTIVE PROBLEM MANAGEMENT

   Proactive Problem Management is concerned with identifying and resolving problems
   and known errors before incidents occur, thus minimizing the adverse impact on the
   service and business related costs. One of the main activities within proactive
   Problem Management process is trend analysis. The objective of trend analysis is to
   identify the upward pattern of incidents, which can then be analyzed to identify faults
   in the IT infrastructure and then correction is initiated. Correction of faults requires
   effort and resources and therefore can be expensive. The organization has to decide
   when exactly to initiate correction so that the time and money spent on this activity is
   less than the adverse business impact created by the faults at a later stage. Trend of
   incidents of an IT infrastructure component is given below.

                                                     Trend - Number of Incidents


      Number of Incidents




                                  Jan    Feb   Mar    Apr    May    Jun     Jul    Aug   Sep   Oct   Nov

From the chart it is obvious that there is an upward trend observed from August. Now
the challenge is when to initiate correction so that the effort and resources spent are
justifiable. Do we need to initiate correction in August itself or is it still possible to
gain business/customer satisfaction even when things go wrong. Well, an
understanding of Incident Life Cycle will help making the decision.

Incident Life Cycle:

        Detection     Response              Repair Time                        Recovery Time
       Elapsed Time     Time

                 Detection   Diagnosis       Repair                  Recover           Restoration


Incident passes through several stages till the service is restored. Downtime is the
total time elapsed from the time the incident happened to the time the service is
restored. Once we know the downtime due to incidents, we can compute the cost to
business due to downtime which in turn will help us in deciding to whether to initiate
correction or not.

Cost to Business due to Downtime or Cost of Downtime:

Cost to business due to downtime is the sum of cost of unavailability of the service
and the cost to repair it. Problem Management should work closely with Availability
Management to understand the cost of unavailability of the service. The following
formula may help in determining the cost of unavailability of service:

Cost of Unavailability = (Downtime * Users affected * Average cost per user) +
                         (Downtime * Lost Business Revenue per hour) +
                          Overtime working costs + Sundry cost

Cost to repair = (Downtime * cost of the repair team)

It is also advisable to add the soft costs such as customer perception, legal loss etc. to
arrive at the Total cost to business.
                 Regression Equation for Predicting Cost to business due to Downtime:

                 Once we know the number of incidents and the cost to business due to these
                 incidents, we can build a prediction model from the historical data. The model takes
                 the number of incidents as the predictor and gives us the response which is the cost to
                 business due to these incidents. From this value we can decide whether there is a need
                 to initiate correction or is it still possible to continue without any action as there is no
                 adverse impact yet on the business operations. The regression equation and the
                 regression plot for this case are given below. From the regression equation it is
                 obvious that the predictor which is the number of incidents has significant impact on
                 the cost to business. Once the trend chart is plotted, the value of number of incidents
                 is substituted in the regression equation to arrive at the cost to business.

Cost to Business = - 35496 + 1407* Number of Incidents

Predictor        Coef SE Coef   T   P
Constant        -35496 50189 -0.71 0.497
Number of Incidents 1407.2 196.7 7.15 0.000

S = 50621.0 R-Sq = 85.0% R-Sq (adj) = 83.4%

                                           Regression Line

Cost to Business($)




                               150   200      250          300      350        400
                                              Number of Incidents

The approach discussed in this paper is easy to implement as only basic tools are used
to demonstrate the application if Six Sigma in ITIL Problem Management. Care
should be exercised in computing the cost to business, and it is recommended that this
cost needs to be arrived from a brainstorming session with participation from
different processes like Incident Management, Problem Management, Availability
Management, Financial Management etc.. Scope for further study is to understand the
influence of other parameters on the cost to business and to build a multiple
regression model.

Author Biography:
Krishna Murthy Dasari is a software quality professional with 12 years of experience
in quality. He has worked in industries like manufacturing and IT. He has specialized
experience in ISO 9000, CMMI, ITIL, Six Sigma and Information Security
Management. He is currently working with Satyam Computer Services Ltd. He can be
reached at

Description: Problem Management Trend Analysis document sample