CIS 4120 Session 8 - Georgia State University

Download Report

Transcript CIS 4120 Session 8 - Georgia State University

CIS 4120 Fa13: Define/Innovate BP’s
Session 7:
AH-7: Simulation exercise
Richard Welke
Director, CEPRIN
Professor, CIS
Robinson College of Business
Georgia State University
Atlanta, GA
© Richard Welke 2002
AH-7: Customer service scenario
Step one
Import the .bpm for INNOV-8 from Course wiki
© Richard Welke 2013
CIS4120Fa13 AH-8 Simulation on INNOV8
2
Associated scenario & process owner PTBS
After, Inc.
Provides customer service (two types)
Some they service themselves
Some they outsource to another organization (NFLEX)
Each have different costs, service rate times, etc.
Process owner’s issue
What’s the best criteria and therefore split to use
between in-house and NFLEX customer service?
Givens:
Servicing rate
Cost of providing the two customer services (per hour)
Question to answer
Optimal split between the two service organizations
Given an overall ROI formula
© Richard Welke 2013
CIS4120Fa13 AH-8 Simulation on INNOV8
3
The “Givens” and the Question
Arrival rate of customer service requests
Assume Poisson distribution with arrival rates of
Mean (negative exponential): 0.21
Collect called info
Number of servers: 1
Call handling time (fixed; non-variable): 1 min.
Split between AfterInc and NFLEX
NFLEX: 39%
Assume Service time distributions of in-house and NFLEX
that are normally distributed with mean and standard
deviations of:
In-house: Mean: 3.8 Standard deviation: 1.0
NFLEX: Mean: 4.1 Standard deviation: 1.0
Assume current split between in-house and NFLEX decision
(business rules) gateway of:
In-house: 39 %
NFLEX: 61 %
We’ll assume the automated email task take one second
(fixed)
What’s the average end-to-end customer processing time
with this scenario?
© Richard Welke 2013
CIS4120Fa13 AH-8 Simulation on INNOV8
4
At-home exercise (AH-7)
Begin exploring the “response surface” of
alternative possibilities
Note:
This is non-linear (more or less of something won’t
necessarily improve the overall outcome)
Within the “objective function” begin to explore the
“best” solution for the process owner
You have until 15 minutes before the end-class to
state your team’s best solution so far
Best demonstrable solution “wins”
© Richard Welke 2013
CIS4120Fa13 AH-8 Simulation on INNOV8
5