Dynamic Help Desk Modeling - iceaaonline.com · •SAP software •Cost projections for 20+ year...

Dynamic Help Desk Modeling

SCEA Conference 2011- Albuquerque, NM

Presented at the 2011 ISPA/SCEA Joint Annual Conference and Training Workshop - www.iceaaonline.com

Background / Introduction

2

• Objective – Accurately estimate the long‐term annual helpdesk cost for a major DoD ERP System

• 65,000+ Users

• Financial Management, Acquisition, Contracts, Personnel Mgmt., Supply Chain Mgmt. functionality

• Rolling wave deployment strategy over 6 years

• SAP software

• Cost projections for 20+ year period


Three Tiered Help‐Desk Structure

3


Data from First of six Deployments

4

0

500

1000

1500

2000

2500

3000

3500Total Tickets Received / Wee

k

Pre‐Go Live

Go Live Sustainment

0

500

1000

1500

2000

2500

3000

3500Total Tickets Received / Wee

k

Pre‐Go Live

Go Live Sustainment

• Data from the first go‐live indicated three distinct periods of helpdesk support requirements• Pre‐Go live – low ticket volume, slowly increasing over time

• Go Live – high ticket volume, sharply decreasing after Go Live

• Sustainment – medium ticket volume, slowly decreasing over time


Data from First of six Deployments

5

y = 0.0007x ‐ 5E‐05

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Pre Go‐Live Trouble Tickets Submitted/Week/User


Linear (Pre Go‐Live Trouble Tickets Submitted/Week/User) y = 0.0007x ‐ 5E‐05

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30



Linear (Pre Go‐Live Trouble Tickets Submitted/Week/User)

y = 0.0339x‐0.118

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Sustainment Trouble Tickets Submitted/Week/User

Sustainment Trouble Tickets Submitted/Week/UserPower (Sustainment Trouble Tickets Submitted/Week/User)

y = 0.0339x‐0.118

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Sustainment Trouble Tickets Submitted/Week/User

Sustainment Trouble Tickets Submitted/Week/UserPower (Sustainment Trouble Tickets Submitted/Week/User)

• Regression equations were developed for the 3 distinct periods around Go-live• Equations predict number of tickets submitted / user / week based on deployment phase and week


Key Variables and Cost Drivers

6

Variable Impact on Help Desk Staffing Requirements

How the Model addresses this

Total Users & Active Users More users = more tickets; active users is a better predictor of volumes than total users

Regression equation normalized based on user counts; predicts per user ticket volume

User Maturity Gradual decrease in tickets over time

Regression equation for sustainment has a negative slope

System Maturity Volume of tickets / active user decreases w/ each deployment

System maturity factor is used to calibrate the model

Service Rate Improvements

Tickets are closed faster as helpdesk resources gain experience

Service rate improvement factor is used to adjust service rates over time

New functionality and system upgrades

Spike in helpdesk calls around the implementation of new functionality

Deployment dates and affected # of users can be quickly added on the import form


System / User Maturity

7

Actual Helpdesk Tickets Submitted per Quarter

Deployment 1

Deployment 2

Deployment 3

Actual Helpdesk Tickets Submitted per Quarter

Deployment 1

Deployment 2

Deployment 3

• System maturity reflected by decrease in ticket volume / user for each deployment

• User maturity demonstrated by decrease in tickets over time


Model Output

8

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

0

20

40

60

80

100

120

Total N

umber of System Users

Total N

umber of Tier 2 and Tier 3 Helpdesk FTEs

Help Desk FTEs Required by Month

Tier 2 Help Desk Tier 3 Help Desk Total Users

• Once calibrated, the model projected the # of tickets to within 5% of actuals


0

200

400

600

800

1,000

1,200

1,400

2008 2009 2010 2011 2012 2013

Total System Users / Helpdesk FTE

Dynamic Help Desk Users / FTE Step Function Users / FTE

0

200

400

600

800

1,000

1,200

1,400

2008 2009 2010 2011 2012 2013

Total System Users / Helpdesk FTE

Dynamic Help Desk Users / FTE Step Function Users / FTE

Comparison to Step Function Methodology

9

0

20

40

60

80

100

120

10/1/2008 10/1/2009 10/1/2010 10/1/2011 10/1/2012

Total H

elp Desk FTEs Required

Step Function vs. Dynamic Help Desk Model

Dynamic Help Desk Model ProjectionStep Function Projection

0

20

40

60

80

100

120

10/1/2008 10/1/2009 10/1/2010 10/1/2011 10/1/2012

Total H

elp Desk FTEs Required

Step Function vs. Dynamic Help Desk Model

Dynamic Help Desk Model ProjectionStep Function Projection


Benefits of the Model

10

• Dynamic model output significantly more accurate than step function / SME estimates

• Enabled accurate long‐term staffing forecasts

• Identified significant future funding shortfalls

• Provided DoD leadership with confidence in cost forecasts; key enabler for significant funding increases

• Rapid what‐if scenario modeling to support programmatic decisions, cost forecasts, and budgeting

• Enabled the program to recoup costs associated with added users

• Provided an objective tool to validate subjective SME inputs


Q&A

11


Dynamic Help Desk Modeling - iceaaonline.com · •SAP software •Cost projections for 20+ year...

Documents

Transcript of Dynamic Help Desk Modeling - iceaaonline.com · •SAP software •Cost projections for 20+ year...