Alexandre Zolotovitski, Ph.D. [email protected] Dashboard for Statistical Visualization of Client Service Events Photo by Dave Trindade.

Alexandre Zolotovitski, [email protected]

Dashboard for Statistical Visualization of Client Service Events

Photo by Dave Trindade

2

2

Dashboard for Statistical Visualization of Client Service Events

The raw TSC data:Creation – Received –First_Contact_SW – First_Contact_HW – Request – Pending – Closed

Events: Creation – Received – First_Contact_SW – First_Contact_HW – Request – Comp_Pending – Closed

Types: HW Platforms, Products, Geo, Owner, …

3

Dashboard for Statistical Visualization of Client Service Events

4

Dashboard for Statistical Visualization of Client Service EventsConsists of three key elements:

• Trend/SPC chart for longitudinal (in time) representation of data • Matrix bar chart for types of cases• Flower bed chart for time between events representation.

The flower bed and matrix bar charts can be plotted for any specific interval of time: week, month, or year.

All features of these three charts are clickable, switching is easy between charts, and drilling down can be done to detailed lists of individual cases. These interactive capabilities facilitate identification of the most business important cases.

5

5

Longitudinal Analysis: Trend Chart

TTR = Time-To-Resolve = DateTime(CaseClosed) – DateTime(CaseCreated)

6

Cross-Sectional Analysis by Types of Cases:

Traditional way To represent 2 (pivot) tables: Freq and TTR (HW Platform) x (Product)

7

Cross-Sectional Analysis by Types of Cases: Matrix Bar Chart

Width of cuboid is proportional to time and height is proportional to square root of frequency.

8

8

Cross-Sectional Analysis –Time Between Events:

The raw TSC data:Creation – Received –First_Contact_SW – First_Contact_HW – Request – Pending – Closed

Events: Creation – Received – First_Contact_SW – First_Contact_HW – Request – Comp_Pending – Closed

Types: HW Platforms, Products, Geo, Owner, …

9

9

Cross-Sectional Analysis –Time Between Events:

10

10

Cross-Sectional Analysis –Time Between Events:

11

11

Cross-Sectional Analysis –Time Between Events:

• Traditional way to represent 2 (pivot) tables: Freq and TTR (Event x PrevEvent )

Freq

0

20000

40000

60000

80000

100000

R S H Q P C W

Event

Freq

CREATION

RECEIVED

FIRST_CONTACT_SW

FIRST_CONTACT_HW

REQUEST

COMP_PEND

CLOSED

WAREHOUSE_UPDATE

TTR

0

100

200

300

400

500

600

700

800

900

R S H Q P C W

EventTT

R, d

ays

CREATION

RECEIVED

FIRST_CONTACT_SW

FIRST_CONTACT_HW

REQUEST

COMP_PEND

CLOSED

WAREHOUSE_UPDATE

12

12

Cross-Sectional Analysis –Time Between Events: Variants of possible designs

• Area of Circle (“house”) is proportional to frequency of the event.• Width of Petal (arrow, bar) is proportional to frequency of transitions between events.• Length of Petal is proportional to mean time between events.• Color of Petal is the same as color of “To” Circle (“house”)

13

13

Cross-Sectional Analysis –Time Between Events: Flower Bed Chart

Data visualization – variants of designPetal direction points to next event, with matching petal-circle color. Circle size ~ frequency. Length of petal ~ time and width of petal ~ frequency.

14

14

Cross-Sectional Analysis – Evolution of Elements

15

Dashboard for Statistical Visualization of Client Service Events

16