Simulation Project - AHT

Download as pdf or txt
Download as pdf or txt
You are on page 1of 62

Six Sigma Green Belt Project

Reduction in IT ticket closure time


• XYZ company, who manages outsourced IT work
for ABC are getting complaints about delay in
closing IT tickets for last 3 months.

Problem
• SLA (Service Level Agreement) = 13 minutes.
• As a result,
• XYZ has to pay penalty of $10000 month.

Statement
• Has potential threat of losing the business
• May lead to 65% loss of revenue of the
company
• May result in job loss of 400 people

2
DEFINE
Voice of Customers

The ticket closure is The quality of work The employees are


taking more than done is not up-to the always busy when
expected time and mark, maybe because asked about the ticket,
have to do multiple they are not qualified we do not get a proper
follow-ups to get our enough or lack the response for the status
work done skills required of the ticket

1 Customer 2 Customer 3 Customer

Delayed ticket closure Lack of quality work Less no of employees


Proper Communication Lack of skills Improper infrastructure

CTQ
Critical to Quality

6
Affinity Diagram – Classification of VOC

Ticket
Admin HR
Handler

Slow work Insufficient No proper training for


infrastructure communication
improper
communication Skilled people should be
hired

Sufficient number of people


should be hired 5
Kano Model

Good number of
skilled employees

Proper
communication Proper
infrastructure

Quality work

On-time
work
Second level prioritization

Customer Expected Likelihood Expected Applicability Project Project


Importance total of success contribution to other priority order (G)
(A) project (C) to profit (D) areas (E) number (F)
cost (B)
Improve Quality 8 9 6 7 8 24192 2

Reduce ticket 9 7 9 9 7 35721 1


handling time
Improve 5 4 8 4 9 5760 3
Communication

So, the first priority to improve the customer experience is reducing the ticket handling time.
CTQ Drilldown Tree
Project Charter

Project Code ABC101 Version 1.0 Date 28/06/2020

Project Title:
Reduction of the IT ticket closure time.

Business Case
XYZ company manages outsourced IT work for a major client ABC. For last 3 months, ABC has been complaining about delay in closing IT tickets. The customers of ABC have to
wait more than agreed SLA (Service Level Agreement) of 13 minutes. Due to non adherence of the contractual agreement. XYZ company has already been paying penalty for
last 3 months i.e. $10000 for this violation per month. There are other vendors of ABC who are doing pretty well in comparison to XYZ due to which there is a potential threat
of losing the business which will lead to 65% loss of revenue of the company & around 400 people might lose their jobs.

Problem Statement
Since the last 3 months, the tickets are being serviced late. There are many employees working on it under various teams, and at different locations. The background of the
ticket handlers is also different. Also, for different types of tickets, the handling time is different. There is a need to find out the reason for this delay in ticket closure time,
otherwise Company may loose revenue and hence its employees.

Project Objective / Goal Statement


Reduction of IT ticket closure time to 13 minutes.
Project Charter (..Contd.)
Scope
Team
In Scope: Ensuring 100% productivity of the employees and infrastructure
Out of Scope: Except the scope
Sponsor: Mr. X Leader: Mr. P Members: Mr. A, Mr. B, Mr. C.

Project Schedule (Mention start and end dates) GANTT Chart

Define Phase 11/07/2020 to 20/07/2020 Improve Phase 16/08/2020 to 30/08/2020

Measure Phase 21/07/2020 to 31/07/2020 Control Phase 01/09/2020 to 15/09/2020

Analyze Phase 01/08/2020 to 15/08/2020 Case study Presentation 16/09/2020

Resource Estimation
Resource Required Value (in INR)
Consultant 1.5Lakh
Internal human resource 150 hours

Expected Benefits
Savings Description Value (in INR)
Retained customer 25Lakh
New business 10Lakh
SIPOC / COPIS

Received ticket
Ticket Call is received Ticket is
Customer
handler or ticket is Work on closed
from the
raised resolving the
company ABC
ticket Issue is
Infrastructure resolved
for receiving Getting inputs
call and from customer if Customer is
platform for required notified
raising ticket
Resolving the
issue

Suppliers Inputs Processes Outputs Customers


6
Process Map

Pick the call to


Customer calls
understand the
for an issue
issue Start working on Yes
Received the problem
request on Look for the
portal solution of the
problem Ticket Ticket No
can be Close the
re -
solved ticket
opened
No ?

Yes
Escalate it
Solve it Close the ticket

Update the customer


Close the ticket Document it
about the solution
Communication Plan
Activity Audience Media Who Backup When
Project definition Sponsor, Key Email, Meetings Project leader Back up PL Kick off date
stakeholders, team
members
Regular team All team members Email Project leader Back up PL Twice a week
meetings

Daily updates All team members, Call, email Project Leader Back up PL Daily 6 pm
Manager, Team leader

Tollgate reviews Approvers Email, official reviews Project Leader Back up PL End of each phase

Project progress All stakeholders and Email Project Leader Back up PL End of each phase
team members

Technology changes Team members Email, meeting Project Leader Back up PL Whenever needed

Process changes All stakeholders and Email, meeting Project Leader Back up PL Whenever needed
Team members
RASIC
RASIC Chart
Project Team

Solely and directly responsible for the activity (Owner) -


Includes approving authority (A)
Responsible

Back up Project Leader

Other team members


Reviews and assures that the activity is being done as

Project Sponsor
per expectations

Process Owner

Project Leader
Approve

Green Belt
Black Belt
Provides the necessary help and support to the owner
Support
Is to be kept informed of the status/progress being
made
Inform
Is to be consulted for this activity for inputs

Consult
Activities
Define
Collect VOC A I C R S
Classify VOC A I I R S
Project Charter A I I R S
SIPOC/COPIS A I R S
Process Map A I I R S
Communication Plan A I I R S
RASIC A I I R S
FMEA A I I R S
Charter appoval A I I R
Information update A I I R
RASIC
RASIC Chart
Project Team

Solely and directly responsible for the activity (Owner) -


Includes approving authority (A)
Responsible

Back up Project Leader

Other team members


Reviews and assures that the activity is being done as per

Project Sponsor

Process Owner
expectations

Project Leader
Approve

Green Belt
Black Belt
Provides the necessary help and support to the owner
Support

Is to be kept informed of the status/progress being made


Inform
Is to be consulted for this activity for inputs

Consult
Activities
Measure
Potential root cause identification C A I I R S
Data collection plan A I C R S
Measurement system analysis I A C R S
Amount of data to collect C I A R S
Choosing sampling methodology A I R S
Collecting data A I R
Validating data A I I C R S
Information update A I I R
Analyze
Calculation of process capability A I I R S
Quality tools A I R S
Identifying vital Xs A I I R S
Hypothesis testing A I I C R S
Information update A I I R
RASIC
RASIC Chart
Project Team

Solely and directly responsible for the activity (Owner) -


Includes approving authority (A)
Responsible

Back up Project Leader

Other team members


Reviews and assures that the activity is being done as per

Project Sponsor

Process Owner
expectations

Project Leader
Approve

Green Belt
Black Belt
Provides the necessary help and support to the owner
Support

Is to be kept informed of the status/progress being made


Inform
Is to be consulted for this activity for inputs

Consult
Activities
Improve
Solution generation A I C R S
Screening of Ideas C A I I R S
Risk Management using FMEA I A C R S
Cost benefit analysis I A R S
Implementation C C A I R S
Pilot testing A I I R S
Information update A I I R
Control
Control Plan C A I R S
Control charts A I C R S
Documentation I A C R
SWOT Analysis I A C R
Summarizing A I I C R
Information update A I I R
MEASURE
Fishbone Diagram
Sample size estimation

The sample size can


be calculated by Stat -
> Power and sample
size -> Sample Size for
Estimation

This sample size of 174


is calculated by taking
confidence level of
95% and margin of
error is 0.15
Data Collection Plan
Measure Name IT Ticket Closure Time (in Min) Ticket Handler Gender Ticket Handler Ticket Handler Age Group Ticket Type

Measure Type (X/Y) Y X X X X

Data type Continuous Discrete - Nominal Discrete - Nominal Discrete-Categorical Discrete - Nominal

Definition Time taken to close a ticket Name of the ticket Handler Gender of the ticket Handler Age group of the ticket handler Type of issue

No. of decimal places 0 NA NA NA NA

Format 13 Jaya, Nitin, Ravi, Vandana M/F <Under 20, >30 Years, Btw 20-30 Hardware issue/ Network issue

Measure Name CSAT Score No. of follow ups User Background Shift User Qualification

Measure Type (X/Y) X X X X X

Data type Continuous Continuous Discrete-Nominal Discrete-Categorical Discrete-Categorical

Definition CSAT score of the ticket No. of follow ups required for the ticket Background of the user Shift in which ticket is handled Qualification of the user

No. of decimal places 2 0 NA NA NA

Format 50 5 Banking, Customer Care, IT, Sales Morning, Night Under Graduate, Graduate, Post Graduate

Measure Name Process Complexity Trainer Manager Issue Location

Measure Type (X/Y) X X X X

Data type Discrete-Categorical Discrete-Nominal Discrete-Nominal Discrete-Categorical

Definition Complexity of the ticket Trainer for the ticket Manager for the ticket Location of the issue

No. of decimal places NA NA NA NA

Format Tier 1, Tier 2 Rajesh, Sam, Vishal Manish, Pooja, Reena, Rohit, Saurabh Mumbai, Bangalore
MSA

The data is collected directly from a reliable source ( Online


server/CRM) , and no people were involved.
Therefore, we need not conduct Measurement System
Analysis (MSA)
Normality Test

Since p-value is
greater than equal to
0.05, the data is
normal, and therefore,
mean can be used as a
measure of central
tendency.

Stat -> Basic Statistics


-> Normality test
Graphical summary

Since p-value is
greater than equal to
0.05, the data is
normal.

Stat -> Basic Statistics


-> Graphical Summary

Currently, the mean is


18 minutes and has to
be reduces to 13
minutes.
Distribution of data

The target ticket


closure time = 13
minutes

Only about 13% of the


tickets are closed
before 13 minutes as
per the data.

Graph -> Box plot ->


Simple
Randomness test

Since p-value of
clustering, trends,
mixtures, and
oscillation is greater
than equal to 0.05, the
data is random.

Stat -> Quality tools -


>Run Charts
ANALYSE
Current Sigma Level

Cpk = -0.46.
So, ZLT = -1.38 and
ZST = 0.12
So, the current sigma level is
0.12 for the process.

Stat -> Quality tools


-> Capability Analysis
-> Normal..
Hypothesis testing of potential Xs

S. No. Potential X Data Type Test to be done

1 Ticket Handler Categorical with 2+ groups One-way ANOVA / Box plot

2 Gender of ticket Handler Categorical with 2 groups 2 Sample T test/ Box plot

3 Ticket handler age group Categorical with 2+ groups One-way ANOVA / Box plot

4 Ticket type Categorical with 2 groups 2 Sample T test/ Box plot

5 CSAT Score Numerical Correlation/Scatterplot

6 Number of follow ups Numerical Correlation/Scatterplot

7 User background Categorical with 2+ groups One-way ANOVA / Box plot

8 Shift Categorical with 2 groups 2 Sample T test/ Box plot

9 User Qualification Categorical with 2+ groups One-way ANOVA / Box plot

10 Process complexity Categorical with 2 groups 2 Sample T test/ Box plot

11 Trainer Categorical with 2+ groups One-way ANOVA / Box plot

12 Manager Categorical with 2+ groups One-way ANOVA / Box plot

13 Issue Location Categorical with 2 groups 2 Sample T test/ Box plot


Ticket Handler vs. IT ticket closure time

Since p value is less than 0.05, we can say that


Ticket handler has significant impact on the
Ticket handling time and their means are not
equal. Vandana is better at handling tickets as
compared to the other ticket handlers.
Ticket Handler gender vs. IT ticket closure time

Since p value is less than 0.05, we can


say that Ticket handler gender has
significant impact on the Ticket
handling time and their means are not
equal. Females are better at handling
tickets as compared to males.
Ticket Handler Age group vs. IT ticket closure time

Since p value is more than 0.05, we can say


that there is no evidence to say that the ticket
handler age group has significant impact on
the Ticket handling time.
Ticket type vs. IT ticket closure time

Since p value is less than 0.05, we can


say that Ticket type has significant
impact on the Ticket handling time
and their means are not equal.
Network issue tickets take less time as
compared to Hardware issue tickets.
CSAT score vs. IT ticket closure time

Since p value is more than 0.05, we can say


that there is no evidence to say that the CSAT
Score has significant impact on the Ticket
handling time.
No. of follow ups vs. IT ticket closure time

Since p value is more than 0.05, we can say


that there is no evidence to say that the
Number of follow ups has significant impact
on the Ticket handling time.
User background vs. IT ticket closure time

Since p value is more than 0.05, we can say


that there is no evidence to say that the user
background has significant impact on the
Ticket handling time.
Shift vs. IT ticket closure time

Since p value is more than 0.05, we


can say that there is no evidence to
say that the Shifts have significant
impact on the Ticket handling time.
User qualification vs. IT ticket closure time

Since p value is more than 0.05, we can say


that there is no evidence to say that the user
qualification has significant impact on the
Ticket handling time.
Process complexity vs. IT ticket closure time

Since p value is more than 0.05, we


can say that there is no evidence to
say that the Process complexity has
significant impact on the Ticket
handling time.
Trainer vs. IT ticket closure time

Since p value is less than 0.05, we can say that


Trainer has significant impact on the Ticket
handling time and their means are not equal.
Rajesh is better at training as compared to the
other trainers.
Manager vs. IT ticket closure time

Since p value is more than 0.05, we can say


that there is no evidence to say that the
Managers have significant impact on the
Ticket handling time.
Issue Location vs. IT ticket closure time

Since p value is more than 0.05, we


can say that there is no evidence to
say that the Issue Location has
significant impact on the Ticket
handling time.
Vital Xs

The following potential Xs are affecting the Ticket


closure time significantly.

S. No. Potential Xs Data type


1 Ticket Handler Categorical with more than two groups
2 Ticket Handler gender .
Categorical with two groups
3 Ticket type Categorical with two groups
4 Trainer Categorical with more than two groups
IMPROVE
Regression Analysis

Since, the p value is less than 0.05, we can reject the null and
conclude that these factors (Vital Xs) impact the Ticket closure
time. But, the model is not much significant. The Adjusted R sq.
is just 35.7%. It should be at least 60% for accepting the
regression equation as shown above.
Solution Generation

Potenitial Xs Observations Solutions Solution Name

Vandana is performing better than others Vandana can train others S1


Ticket Handler
New ticket handler can be appointed S2

Performance based incentive can be given S3

Females are working better than males Hiring more females S4


Ticket Handler gender
Arranging collaboration sessions S5

Network issues gets resolved earlier Building better infrastructure S6


Ticket Type
Training on hardware issue resolution S7

Rajesh is performing better Training by some area expert S8


Trainer
Training by Rajesh S9

Arranging competition between trainers S10


Effort benefit Matrix

Quick wins = Solution 1,3,5,7,10 Major Projects = Solution 6,8


Fill ins = Solution 9, Thankless tasks = Solution 2,4
Cost time Matrix

Cost Matrix
S1 S2 S3 S4 S5 S6 S7 S8 S9 S10
Cost Scale
Member 1 1 1 2 1 1 3 1 2 1 1
Member 2 1 2 1 2 1 4 2 1 1 2 0 - 50k 1
Member 3 1 1 2 1 1 5 1 1 1 1 Upto 1 lakh 2
Upto 1.5 lakh 3
Time Matrix
S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 Upto 2 lakh 4
Member 1 1 2 2 2 1 4 1 2 1 2 More than 2 lakh 5
Member 2 2 3 1 3 2 5 2 1 2 1
Member 3 1 2 3 2 2 4 1 2 1 2
Time Scale
Cost Time Matrix 0-15 days 1
S1 S2 S3 S4 S5 S6 S7 S8 S9 S10
Member 1 2 3 4 3 2 7 2 4 2 3 Upto 30 days 2
Member 2 3 5 2 5 3 9 4 2 3 3 Upto 45 days 3
Member 3 2 3 5 3 3 9 2 3 2 3
Upto 60 days 4

Total Score 7 11 11 11 8 25 8 9 7 9 More than 60 days 5

This shows that Solution 6 is very costly and time consuming, whereas solutions 1,5,7,8,9,10 are least time
consuming and expensive.
Trainer
Ticket type
Ticket Handler

Deployment Matrix
Vital Xs

Ticket Handler Gender


Rating ( 1 to 10)

6
7
9

8
1 Lowest
10 Maximum

0
9
9

5
Vandana can train others

184
0
5
5

3
Appointment of new ticket handler

104
8
8
8

232 Performance based incentive


1
6
5

Hiring more females


117
4
8
8

Arranging collaboration sessions


200
8
9
9

Building better infrastructure


264
7
8
8

Training on hardware issue resolution


234
X & Y Relationship Matrix

2
6
7

Training by external area expert


181
0
5
9

Training by Rajesh
156
9
5
7

Competition between trainers


168
0

Completeness Matrix
234
483
675

448

Green color shows high priority tasks, Yellow color shoes medium priority tasks and Red color shows low priority tasks.
FMEA

RPN
Defined "X"s Items Failure Modes Effect on EDR Severity Occurrence Detection RMS
(S*O*D)

Ticket Handler Vandana can train others Vandana resigns Complete failure 9 5 1 45 Accept
Vandana has time constraints Delay in project 7 8 3 168 Accept
Vandana is unwilling to share May not be able to identify this 8 8 9 576 Avoid
Others are not willing to get trained
May not be able to identify this 8 8 9 576 Avoid
from her
Appointment of new ticket handler New person will take time to adjust Delay in project 6 9 5 270 Reduce
It may take time to hire someone Delay in project 5 7 3 105 Accept
He/she may not perform as expectedComplete failure 8 6 7 336 Reduce
People may not cooperate and can
Performance based incentive Complete failure 9 8 8 576 Avoid
mislead others
Less quality work Future issue can be caused 9 9 6 486 Reduce
Less collaboration amongst team May not be able to identify this 8 7 9 504 Avoid
Ticket handler
Hiring more females New person will take time to adjust Delay in project 6 7 7 294 Reduce
gender
It may take time to hire someone Delay in project 5 7 4 140 Accept
She may not perform as expected Complete failure 7 8 6 336 Reduce
Arranging collaboration sessions People may not learn as expected Complete failure 7 9 8 504 Avoid
People might misguide each other May not be able to identify this 9 9 9 729 Avoid
FMEA

RPN
Defined "X"s Items Failure Modes Effect on EDR Severity Occurrence Detection RMS
(S*O*D)

Ticket type Building better infrastructure It will take time to build Delay in project 6 9 4 216 Reduce
People will need further training to
Delay in project 6 7 4 168 Accept
use it
Training on hardware issue People may not be willing to attend
May not be able to identify this 7 7 9 441 Reduce
resolution it
People might take time to start
Delay in project 5 6 4 120 Accept
following the techniques

People might be unwilling to take


Trainer Training by external area expert Complete failure 6 8 8 384 Reduce
outside training

External cater may not be able to


Complete failure 8 7 9 504 Avoid
cater to organisational needs
Training by Rajesh Rajesh resigns Complete failure 9 5 2 90 Accept
Rajesh has time constraints Delay in project 7 7 4 196 Accept
Rajesh is unwilling to share May not be able to identify this 8 8 9 576 Avoid
Others are not willing to get trained
May not be able to identify this 8 8 9 576 Avoid
from him
Competition between trainers People may not learn as expected Complete failure 8 7 7 392 Reduce
People might misguide each other May not be able to identify this 9 9 9 729 Avoid
Target validation

In week 1, The p-value is less than 0.05, we accept the Alternate hypothesis and can say that mean is not equal to 13.
From the mean and confidence level we can say that the ticket handling time is always more than 13 minutes.
In week 2, the p-value is more than 0.05, so we fail to reject the null and don’t have enough evidence to say that
mean is not equal to 13.
In week 3, the p-value is less than 0.05, therefore, we can say that mean is not equal to 13. Also, from the mean and
confidence interval we can see that ticket handling time is always less than 13 minutes.
Before and After tests

When week 1 is compared to week 3, by doing paired t-test, we can see that the p-value is 0.00.
This suggests that there has been significant change in the mean of both the weeks and from the mean, we can see
that there has been a significant improvement in week 3 as compared to week 1.
Box plot before and after

From the box plot, we can see


that there is a great
improvement in IT ticket
closing time from week 1 to
week 3. In week 1,
Control charts before and after
Sigma level before and after
Normality test before and after

The p-value is greater than 0.05, therefore, we fail to reject null. Hence, the data points are normal for both
Week 1 and Week 3.
CONTROL
Control Plan
Control Plan
Process Name:
Process step Key parameter Standard Frequency of check Review by
360 degree feedback
Training Training by internal resource Weekly Training head
mechanism

Monitoring the attendance Weekly Training head

360 degree feedback


Hiring Training by external expert Weekly Training head
mechanism

Monitoring the attendance Weekly Training head

Motivating everyone to
Collaboration Collaboration within teams speak - round robin Monthly HR
technique

Proper communication plan Monthly Manager

Competition amongst team members


Competition Open voting mechanism Monthly Manager
and trainers

Building better infrastructure for the


Building Inrastructure Tracking the progress Monthly Admin
hardwork issues
Training about the new
Weekly Training head
infrastructure
Control Charts before tests

We can see that there are


various special cases in the
data.
Also, most of the data lie
above 13 minutes, which is
the desired Upper limit of
the data.
Control Charts after tests

We can see that the


number of special cases
have reduced. There is
only one special case, that
too in the desired Upper
limit of 13 minutes. So, the
data seems to be fine now.
SWOT Analysis

• Resources will be • Resource


trained limitations
• Scope of • Can be temporary
improvement achievement

Strength Weakness

Opportunities Threats

• Can improve • Customer's attitude


infrastructure can be affected by
• Can get more previous issues
orders from • Quality may be
customers reduced
THANK
YOU !

You might also like