DataMOCCA

Data MOdels for Call Center AnalysisProject Collaborators:

Technion: Paul Feigin, Avi Mandelbaum

Valery Trofimov, Students & RA’s, Technion Statistics Laboratory

Wharton: Larry Brown, Noah Gans, Haipeng Shen (N. Carolina),

Students, Wharton Financial Institutions Center

Companies: U.S. Bank, Israeli Cellular Phone (both Aspect-Based), …

2

Project Data MOCCA

Goal: Designing and Implementing a

(universal) data-base/data-repository and interface

for storing, retrieving, analyzing and displaying

Call-by-Call-based data/information

Enable the Study of:

- Customers

- Service Providers / Agents

- Managers / System

Wait Time, Abandonment, Retrials

Loads, Queue Lengths, Trends

Service Duration, Activity Profile

3

Project Data MOCCA

System Components:

1. Clean databases: operational histories of

individual calls, agents and sometimes customers (ID’s).

2. Friendly powerful interface: online access to (mostly) operational

and (some) administrative data (but no financial data, yet).

Currently Two Databases:

- US Bank (220/40M calls, 1000 agents, 2.5 years; 7-20GB).

- Israeli Cellular phone company (800 agents, 2.5 years, ongoing).

Primary arrivals to queue Retail August 2003

050

100150200250300350400450

07:00 09:00 11:00 13:00 15:00 17:00 19:00 21:00 23:00

Time (Resolution 5 min.)

numb

er o

f ca

ses

04.08.2003 06.08.2003

Customer service time Retail Total August 2003

0.0

0.5

1.0

1.5

2.0

2.5

0 15 30 45 60 75 90 105 120

Time (Resolution 5 sec.)

Rela

tive

fre

quen

cies

, %

04.08.2003 06.08.2003

Customers in queue(average) Retail August 2003

0

20

40

60

80

100

120

140

07:00 09:00 11:00 13:00 15:00 17:00 19:00 21:00 23:00

Time (Resolution 5 min.)

numb

er o

f ca

ses

04.08.2003 06.08.2003

4

Data MOCCA - Interface

- Daily / monthly / yearly reports & flow-charts for a complete

operational view.

- Graphs and tables, in customized resolutions (month, days, hours,

minutes, seconds) for a variety of (pre-designed) operational

measures (arrival rates, abandonment counts, service- and wait-

time distribution, utilization profiles,…).

- Graphs and tables for new user-defined measures.

- Direct access to the raw (cleaned) data.

5

Daily Report of April 13, 2004 – Regular Day

EntriesExitsAbnormal Termination

3760

3716

44

VRU

28968

2895

3

15

Announce

84645

4254

2

42103

Message

10394

1039

4

Direct

48240

603

2954247

3772

3

9130

1387

OfferedVolume

Handled

AbandonShort

AbandonCancel Disconnect

Continued

Data MOCCA - Interface

6

Daily Report of April 27, 2004 - Holiday779

22650

46849

4941

EntriesExitsAbnormal Termination

765

14

VRU22

64 7

3

Announce

2686 0

19989

Message

4941

Direct

22157

435

1072215

1858

1

2867

709

OfferedVolume

Handled

AbandonShort

AbandonCancel

Disconnect

Continued

Data MOCCA - Interface

7

Pre-designed Operational Measures: %Abandonment, TSF

Time Series

Week days

0.00

10.00

20.00

30.00

40.00

50.00

Jan-04

Feb-04

Mar-04

Apr-04

May-04

Jun-04

Jul-04

Aug-04

Sep-04

Oct-04

Nov-04

Dec-04

Jan-05

Feb-05

Mar-05

Apr-05

May-05

months

Rate

,

%

Abandons rate( Engineering)

Abandons rate( Technical)

Probability of waiting > 30( Engineering)

Probability of waiting > 30( Technical)

Data MOCCA - Interface

8

Daily Reports

Agent status February 2005

0

50

100

150

200

250

300

350

400

450

500

0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 0:00

Time (Resolution 5 min.)

Number of cases

01.02.2005 02.02.2005 03.02.2005 04.02.2005 05.02.2005 06.02.2005

07.02.2005 08.02.2005 09.02.2005 10.02.2005 11.02.2005 12.02.2005

13.02.2005 14.02.2005 15.02.2005 16.02.2005 17.02.2005 18.02.2005

19.02.2005 20.02.2005 21.02.2005 22.02.2005 23.02.2005 24.02.2005

25.02.2005 26.02.2005 27.02.2005 28.02.2005

Data MOCCA - Interface

Pre-designed Operational Measures: Arrival Rates, 2/2005

9

Histograms

Customer service time Private Caller Termination February 2005, Week days

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0

20

40

60

80

100

120

140

160

180

200

220

240

260

280

300

320

340

360

380

400

420

440

460

480

500

520

540

560

580

600

620

Time (Resolution 1 sec.)

Relative frequencies

,

%

Pre-designed Operational Measures: Service Times

Data MOCCA - Interface

10

New User-Defined Graphs

Step 2: Processing information

from the desired dates

Time series

0

2000

4000

6000

8000

10000

12000

14000

16000

0 5 10 15 20 25

days

numb

er o

f ca

ses

Private Bussiness

Step 3: Creating desired graphs

and tables

Data MOCCA – Customized Interface

Step 1: Entering Variable Definition

11

US Bank: Arrival Rates on Tuesdays in September

Data MOCCA – Interesting Scenarios

0

500

1000

1500

2000

2500

3000

3500

0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00

Time (Resolution 30 min.)

Arrival rate

04.09.2001 11.09.2001 18.09.2001 25.09.2001

12

Israeli Bank: Histogram of Waiting Time

Data MOCCA – Interesting Scenarios

0.0

0.2

0.4

0.6

0.8

1.0

1.2

22 42 62 82 102 122 142 162 182 202 222 242 262 282 302 322 342 362 382

Waiting time, sec

Relative frequencies, %

13

US Bank: Service Time Histograms for Telesales, 2001-3

Data MOCCA – Interesting Scenarios

0.0

0.5

1.0

1.5

2.0

2.5

0 100 200 300 400 500 600 700 800 900 1000

Time (Resolution 5 sec.)

Relative frequencies, %

May-01 May-02 May-03

14

Israeli Call Center: Technical Service – Scenario Analysis

Data MOCCA – Interesting Scenarios

0

50

100

150

200

250

6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00

Time (Resolution 30 min.)

Number of cases

Arrivals to queue Unhandled

0

5

10

15

20

25

30

35

40

6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 0:00

Time (Resolution 30 min.)

Number of cases

Total Available Incoming Call

Unhandled Calls on May 24th, 2005

Agents Status on May 24th, 2005

15

Daily Report of April 20, 2004 – Heavily Loaded Day

EntriesExitsAbnormal Termination

560955

67

42

VRU

26614

2658 6

28

Announce

157820

9382 7

63993

Message

15964

1596

4

Direct

62866

1024

432

1579

1

4974

8

11138

1980

OfferedVolume

Handled

AbandonShort

AbandonCancel

Disconnect

Continued

Data MOCCA – Interesting Scenarios

16

Data-Based Research: Must (?) & Fun

- Contrast with “EmpOM”: Industry / Company / Survey Data (Social Sciences)

- Converge to: Measure, Model, Validate, Experiment, Refine (Physics, Biology,…)

- Prerequisites: OR/OM, (Marketing) – for Design; Computer Science,

Information Systems, Statistics – for Implementation.

- Outcomes: Relevance, Credibility, Interest; Pilot (eg. Healthcare). Moreover,

Teaching: Class, Homework (Experimental Data Analysis); Cases.

Research: Test (Queueing) Theory / Laws, Stimulate New Models / Theory.

Practice: OM Tools (Scenario Analysis), Mktg (Trends, Benchmarking).