Execution quality: An analysis of fulfillment errors at a retail distribution center

Date01 September 2015
DOIhttp://doi.org/10.1016/j.jom.2015.07.007
Published date01 September 2015
Journal
of
Operations
Management
38
(2015)
25–40
Contents
lists
available
at
ScienceDirect
Journal
of
Operations
Management
j
o
ur
na
l
ho
mepage:
www.elsevier.com/locate
/jom
Execution
quality:
An
analysis
of
fulfillment
errors
at
a
retail
distribution
center
Nathan
Craiga,,
Nicole
DeHoratiusb,
Yan
Jiangc,
Diego
Klabjanc
aFisher
College
of
Business,
The
Ohio
State
University,
United
States
bBooth
School
of
Business,
University
of
Chicago,
United
States
cDepartment
of
Industrial
Engineering
and
Management
Sciences,
Northwestern
University,
United
States
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Received
15
December
2014
Received
in
revised
form
21
July
2015
Available
online
31
July
2015
Accepted
by
Daniel
R.
Guide
a
b
s
t
r
a
c
t
Purchase
orders
specify
many
aspects
of
a
fulfillment
process,
including
item
quantity,
delivery
time,
carton
labeling,
bar
coding,
electronic
data
interchange,
retail
ticketing,
and
others.
These
fulfillment
terms
are
instrumental
for
highly
optimized
retail
supply
chains
employing
automation
and
techniques
such
as
pack-by-store.
When
fulfilling
a
purchase
order,
a
supplier
may
commit
a
fulfillment
error,
i.e.,
the
supplier
may
fail
to
adhere
to
the
terms
specified
by
the
retailer.
The
retailer
may
then
penalize
the
supplier
for
the
fulfillment
error
via
a
chargeback
deduction,
which
reduces
the
supplier’s
revenue.
We
present
a
study
of
the
fulfillment
errors
and
chargebacks
that
occur
in
practice
using
data
collected
from
a
major
retailer’s
distribution
center.
While
fulfillment
errors
involving
incorrect
product
quantities
and
delivery
times
have
received
the
most
attention
in
the
literature,
we
find
that
the
majority
of
fulfillment
errors
in
the
context
we
study
involve
documentation,
bar
coding,
and
retail
ticketing.
We
refer
to
these
as
correctable
fulfillment
errors,
since
they
are
amended
at
the
retailer’s
distribution
center
through
rework.
We
develop
a
model
of
inventory
management
with
correctable
fulfillment
errors
and
use
the
retailer’s
data
to
assess
the
cost
of
these
correctable
fulfillment
errors
to
the
retailer’s
inventory
system.
Our
research
provides
guidance
to
managers
in
identifying
products
and
suppliers
that
impose
large
fulfillment
error
costs
as
well
as
in
setting
appropriate
chargebacks
for
fulfillment
errors.
©
2015
Elsevier
B.V.
All
rights
reserved.
1.
Introduction
Retailers
replenish
inventory
through
supply
networks
com-
prising
suppliers,
retail
distribution
centers
(DCs),
and
retail
stores.
Suppliers
typically
ship
products
ordered
by
retailers
to
DCs,
which
break
bulk
and
distribute
items
to
retail
stores.
Inventory
manage-
ment
at
DCs
is
complicated
and
challenging
(de
Koster
and
Balk,
2008),
not
only
because
of
product
variety
and
long
lead
times,
but
also
due
to
supplier
shipments
that
do
not
conform
to
the
terms
specified
by
retailers’
purchase
orders
(POs).
These
terms
stipu-
late
item
quantities,
delivery
times,
carton
labeling,
electronic
data
interchange,
item
packaging,
bar
coding,
retail
ticketing,
and
other
aspects
of
the
fulfillment
process.
If
the
supplier
fails
to
adhere
to
these
terms,
the
supplier
commits
a
fulfillment
error.
Retail
supply
chains
are
increasingly
reliant
on
fulfillment
that
complies
with
the
terms
specified
by
POs.
Supply
chain
techniques
such
as
pack-by-store
and
cross-docking
require
prod-
uct
packaging,
retail
ticketing,
and
carton
labeling
that
conform
Corresponding
author.
to
specifications.
Automation,
such
as
automated
storage
and
retrieval
systems
and
the
robots
produced
by
Kiva
Systems,
is
unable
to
function
without
appropriate
bar
coding
and
electronic
data
interchange.
Lapses
in
execution
quality,
such
as
the
ful-
fillment
errors
identified
above,
impair
the
operation
of
highly
optimized
supply
chains
and
undermine
firms’
investments
in
technology.
Fulfillment
errors
also
negatively
affect
suppliers.
Retailers
penalize
fulfillment
errors
through
chargeback
deductions,
or
chargebacks,
which
reduce
supplier
revenue.
These
penalties
rep-
resent
a
significant
cost
for
suppliers,
reducing
overall
supplier
revenue
by
2–10%
(Zieger,
2003).
Moreover,
retailers
and
suppliers
regularly
contest
the
cost
of
fulfillment
errors
through
the
dis-
putes
and
negotiations
that
surround
chargebacks
(Chain
Store
Age,
2002).
The
first
goal
of
our
research
is
to
build
on
existing
studies
of
execution
quality
by
describing
the
fulfillment
errors
and
charge-
backs
that
occur
in
practice
within
a
retail
supply
chain.
A
recent
survey
of
42
retailers
found
that
an
average
of
13%
of
shipments
received
by
the
retailers
had
inaccurate
advance
ship
notices,
or
ASNs
(Retail,
2010).
We
collected
data
on
purchase
orders
and
http://dx.doi.org/10.1016/j.jom.2015.07.007
0272-6963/©
2015
Elsevier
B.V.
All
rights
reserved.
26
N.
Craig
et
al.
/
Journal
of
Operations
Management
38
(2015)
25–40
fulfillment
errors
from
a
major
retailer,
Omega.1These
data
include
audit
reports
that
record
the
type
of
fulfillment
error
that
occurred,
the
chargeback
levied
against
the
supplier
for
the
error,
and
the
time
required
for
Omega
to
perform
rework
on
items
affected
by
the
error,
if
applicable.
We
find
that
7%
of
Omega’s
POs
experienced
a
fulfillment
error.
Quantity
shortages,
ticket
errors,
and
ASN
errors
are
the
most
common
types,
accounting
for
52%
of
all
fulfillment
errors.
Moreover,
we
observe
that
fulfillment
errors
within
retail
sup-
ply
chains
can
be
classified
as
either
correctable
or
non-correctable,
depending
on
whether
they
are
amended
through
rework
by
the
retailer.
For
example,
quantity
shortages
and
late
shipments
are
not
correctable
by
the
retailer.
On
the
other
hand,
many
fulfillment
errors
can
be
corrected
by
the
retailer
alone
through
on-site
rework.
If
the
supplier
fails
to
document
a
shipment
properly
(e.g.,
with
a
packing
list)
or
transmit
an
ASN,
the
retailer’s
employees
can
man-
ually
inspect
the
shipment
to
identify
its
contents.
Similarly,
if
the
supplier
attaches
extraneous
packaging
or
tickets
with
an
incorrect
bar
code
to
products,
employees
of
the
retailer
amend
such
errors.
Correctable
errors
represent
a
substantial
portion—56%—of
the
fulfillment
errors
we
observed.
Industry
reports
make
similar
observations.
Three
of
the
five
most
common
fulfillment
errors
identified
by
a
logistics
provider
are
correctable:
invalid
ASNs,
incorrect
Uniform
Commercial
Code
(UCC)
128
labels,
and
incorrect
tickets
(Ma,
2013).
In
contrast
to
non-correctable
fulfillment
errors,
which
have
been
studied
as
random
yields
(Yano
and
Lee,
1995)
and
lead
times
(Bagchi
et
al.,
1986;
Eppen
and
Martin,
1988),
research
on
the
cost
of
correctable
fulfillment
errors
is
limited.
Correctable
fulfillment
errors
impose
both
direct
and
indirect
costs
on
retailers.
Consider
the
case
of
a
ticket
error.
The
retailer
faces
a
direct
labor
cost
and
a
decrease
in
labor
productivity
since
employees
remove
incorrect
tickets
and
affix
the
appropriate
tickets.
In
addition,
the
error
increases
lead
time
and
lead
time
variability,
which
can
cause
damaging
stockouts
at
the
retailer’s
stores.
While
certain
aspects
of
the
cost
of
fulfillment
errors,
such
as
labor,
are
straightforward
to
calculate,
many
retailers,
including
Omega,
do
not
know
the
over-
all
cost
of
correctable
fulfillment
errors
to
their
inventory
systems
(Retail,
2010).
The
second
goal
of
our
research
is
to
extend
prior
research
on
the
cost
of
fulfillment
errors,
which
focuses
on
non-correctable
errors,
by
proposing
a
stochastic
(Q,
R)
model
that
incorporates
correctable
fulfillment
errors
and
associated
rework.
We
study
this
model
via
numerical
experiments
using
parameters
estimated
from
Omega’s
data
on
13,500
replenishment
stock
keeping
units
(SKUs).2We
find
that
correctable
fulfillment
errors
for
the
SKUs
we
study
impose
a
substantial
cost
on
inventory
management
at
Omega,
namely,
the
cost
of
these
errors
is
between
1%
and
4%
of
the
operating
budget
of
the
DC
we
studied.
Managers
have
multiple
approaches
for
addressing
fulfillment
errors.
First,
since
fulfillment
errors
make
the
inventory
supply
process
less
efficient
by
imposing
unnecessary
costs
on
retailers
and
suppliers
alike,
managers
can
work
with
suppliers
toward
the
elimination
of
fulfillment
errors
(Wang
et
al.,
2014).
Kulp
et
al.
(2007)
describe
the
various
methods
retailers
use
to
collaborate
with
suppliers
to
reduce
fulfillment
errors.
Second,
in
the
absence
of
collaboration,
retailers
can
use
incentives,
such
as
chargebacks,
to
pass
the
cost
of
fulfillment
errors
to
suppliers.
Chargebacks
should
reflect
the
cost
of
vendor
non-compliance
to
retailers
(Aron,
1998).
1We
disguise
the
retailer’s
name
at
its
request.
2See
Somlo
et
al.
(2011),
Baron
et
al.
(2010),
van
Donselaar
et
al.
(2010),
and
Gaukler
et
al.
(2007)
for
recent
research
that
also
uses
empirical
data
to
inform
numerical
experiments.
See
Shafer
and
Smunt
(2004)
for
a
discussion
of
the
use
of
empirical
data
in
simulations
and
numerical
experiments.
Third,
retailers
can
adjust
their
inventory
policies
to
accommodate
fulfillment
errors.
These
approaches
are
not
mutually
exclusive.
Understanding
the
cost
of
fulfillment
errors
in
detail
allows
retailers
collaborating
with
suppliers
to
communicate
this
cost
and
to
prioritize
products
and
vendors
that
will
benefit
most
from
a
reduction
in
fulfillment
errors.
Further,
understanding
the
cost
of
fulfillment
errors
helps
retailers
ensure
that
their
chargebacks
are
appropriate.
We
find
that
Omega’s
chargebacks
are
too
low
in
many
cases
and
potentially
too
high
in
others,
which
suggests
misaligned
incentives
with
respect
to
execution
quality
in
Omega’s
supply
chain
(Narayanan
and
Raman,
2004).
Finally,
retailers
that
elect
to
adopt
an
adjusted
inventory
policy
often
incur
a
per-SKU
cost
when
modifying
an
existing
inventory
system.
For
example,
under
the
software-as-a-service
business
model,
inventory
man-
agement
software
vendors
like
Predictix
charge
per
SKU
to
modify
their
standard
inventory
system.
Retailers
may
thus
elect
to
mod-
ify
their
inventory
policy
for
SKUs
that
will
benefit
most
from
the
change.
In
sum,
our
research
characterizes
the
types
and
prevalence
of
fulfillment
errors
and
associated
chargebacks
that
occur
in
practice
at
a
representative
retailer,
identifies
correctable
fulfillment
errors
in
retail
supply
chains
that
are
distinct
from
traditional
random
yields
and
lead
times,
and
conducts
numerical
experiments
using
empirical
data
to
understand
the
impact
of
such
correctable
ful-
fillment
errors
on
inventory
system
cost.
The
rest
of
this
paper
is
organized
as
follows.
In
Section
2,
we
review
related
literature
and
position
our
work.
We
present
an
exploratory
study
of
fulfillment
errors
and
rework
time
in
Section
3.
In
Section
4,
we
introduce
an
inventory
management
model
that
incorporates
correctable
fulfill-
ment
errors.
We
present
an
empirical
analysis
of
fulfillment
errors
and
rework
time
in
Section
5.
In
Section
6,
we
report
the
results
of
numerical
experiments
conducted
on
the
inventory
model
using
parameters
estimated
from
Omega’s
data.
Further,
we
propose
pro-
cedures
for
identifying
SKUs
with
high
costs
of
fulfillment
errors.
In
Section
7,
we
discuss
the
findings
of
our
study
and
suggest
direc-
tions
for
future
research.
2.
Literature
review
We
survey
three
areas
of
the
literature.
First,
we
review
prior
empirical
studies
of
execution
quality
in
supply
chains.
Second,
since
fulfillment
errors
in
retail
distribution
centers
are
related
to
the
errors
that
occur
in
a
production
context,
we
discuss
research
on
production
quality,
inspection,
and
rework.
Finally,
since
retail-
ers
penalize
fulfillment
errors
using
chargebacks,
we
describe
how
our
study
relates
to
research
on
supply
chain
incentives.
Our
research
builds
on
empirical
research
on
supply
chain
execution.
In
a
related
study,
Srinivasan
et
al.
(1994)
examine
shipping
discrepancies
among
suppliers
to
an
auto
manufacturer
and
demonstrate
the
importance
of
electronic
data
interchange
in
a
just-in-time
setting.
In
contrast,
we
study
the
link
between
a
retailer
and
its
suppliers,
and
we
examine
the
types
of
fulfillment
errors
that
occur,
the
rework
necessary
to
address
errors,
and
the
costs
and
penalties
imposed
due
to
errors.
Related
issues
involv-
ing
execution
quality
in
retailing
include
misplaced
inventory
(Camdereli
and
Swaminathan,
2010),
inventory
record
inaccuracy
(Heese,
2007;
DeHoratius
et
al.,
2008),
and
backroom
inventory
(Gaukler
et
al.,
2007;
Eroglu
et
al.,
2012).
In
addition,
our
research
is
related
to
studies
of
the
impact
of
execution
quality
on
the
relationships
between
firms
and
their
suppliers
(Wang
et
al.,
2014).
The
inventory
model
we
propose
extends
prior
research
regarding
fulfillment
errors.
The
errors
we
classify
as
non-
correctable
have
been
studied
as
random
yields
(Yano
and
Lee,
1995)
and
random
lead
times
(Bagchi
et
al.,
1986;
Eppen
and

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT