Content uploaded by Katerina Pramatari
Author content
All content in this area was uploaded by Katerina Pramatari on Jan 25, 2014
Content may be subject to copyright.
Proceedings of the Operational Research Society Simulation Workshop 2010 (SW10)
127
USING SIMULATION TO DESIGN & EVALUATE RFID IMPLEMENTATIONS IN THE
SUPPLY CHAIN
Angeliki Karagiannaki,
Katerina Pramatari
George Doukidis
ELTRUN Research Center, Dept. of Management Science & Technology,
Athens University of Economics & Business
akaragianaki@aueb.gr, k.pramatari@aueb.gr, gjd@aueb.gr
ABSTRACT:
Empowered by the possibility to automatically identify
unique product instances, the emerging Radio
Frequency Identification (RFID) technology is
expected to revolutionize the supply chain processes.
However, in view of the numerous possible ways that
RFID can be implemented within the supply chain, the
issue of supporting the design choices based on a
credible assessment between the current (as-is) and the
future (to-be) processes has become a matter of
considerable concern and debate for both practitioners
and academics alike. To design RFID implementations
in the supply chain using a robust dynamic analysis,
we resort to discrete event simulation. As a result, this
paper proposes a reference step-by-step framework
that demonstrates the value of simulation modeling as
a decision support tool to guide the design choices
regarding the RFID implementation in the supply
chain. To illustrate and assess the framework, a case
study in the warehouse context is described.
Keywords: RFID, supply chain, simulation
1. INTRODUCTION
The dynamic character of today’s competitive
environment forces organisations to an incessant
reassessment of their existing processes. Within this
context, the introduction of new information and
communication technologies (ICT) should be
perceived as a catalyst for better practices and not as a
cost of a business or as a voluntary responsibility.
Nowadays, the emerging Radio Frequency
Identification (RFID) technology is expected to be the
greatest of such technological enhancements (Chao et
al., 2007; Bendavid et al., 2008).
RFID is a generic technology concept that refers to the
use of radio waves to identify objects and, hence,
embraces a new and important sector of mainstream
ICT, the so-called ‘object-associated’ or ‘object
tracking’ or ‘item attendant’ ICT (CASAGRAS Final
Report, 2009). The identification is done by storing a
serial number, and perhaps other information, on a
microchip that is attached to an antenna. This bundle is
called an RFID tag. The antenna enables the chip to
transmit the identification information to a reader. The
reader converts the radio waves reflected back from the
RFID tag into digital information that can be passed on
to an enterprise information system (Kelepouris et al.,
2007). RFID has been extensively used for a diversity
of applications ranging from access control systems to
airport baggage handling, automated toll collection
systems, theft-prevention systems and electronic
payment systems (Agarwal, 2001; Smith & Konsynski,
2003; Kelly & Erickson, 2005; Hou & Huang, 2006).
Nevertheless, what has made this technology
extremely popular is the application of RFID for the
supply chain management. RFID can potentially
empower a great set of improvement opportunities
across the supply chain, ranging from upstream
warehouse and distribution management down to
retail-outlet operations, including shelf management,
promotions management and innovative consumer
services, as well as product traceability (Pramatari et
al., 2005; Ustundag and Tamyas, 2009).
From a supply chain perspective, it is obvious that
RFID technology is not solely regarded as an agent of
‘substituting the existing processes’ whose purpose is
self-evident. In fact, there are numerous possible ways
that the supply chain processes can be shaped in order
to incorporate the RFID technology. Such
dimensionality derives from the fact that RFID is an
infrastructure technology. This means that the
implementation depends not only on the functionality
of the RFID application itself but more on how the
technology is deployed in terms of infrastructure
requirements. Such abundance of RFID
implementations produces uncertainties and fears in
upper management who wants to decide on a
particular RFID implementation based on a credible
assessment between the current (hereafter the ‘as-is
system’) and the possible future (hereafter the ‘to-be
system’) views of the supply chain processes (Lee and
Özer, 2007). The starting point for this research is,
therefore, an effort to assist companies in evaluating
their current position, identifying their RFID design
choices and supporting their decision on moving to a
particular RFID implementation.
To support such decision making with a robust analysis
that depicts the dynamic behavior of the relationships
between the business processes and the information
system (i.e. RFID), there is a need to design
appropriate modelling tools (Paul and Serrano, 2004).
However, the majority of process modeling tools use
conventional techniques based on functional
Proceedings of the Operational Research Society Simulation Workshop 2010 (SW10)
128
decomposition or information engineering (Nidumolu
et al., 1998). The static models generated by such
approaches, while helpful in representing how the ‘as-
is’ processes work, are nevertheless limited in scope
because they cannot support dynamic analysis of the
‘to-be’ processes (Archer, 1995). To support the
dynamic structuring of the ‘to-be’ system, we can
resort to discrete event simulation. Discrete event
simulation can be an extremely valuable, timely and
cost-effective means to evaluate and design ex-ante
alternative RFID implementations without physically
building, amending or interrupting the real system
(Doukidis and Paul, 1985).
Our research objective is, therefore, to demonstrate the
value of simulation modeling and analysis as a
powerful decision support tool to guide the design
choices regarding the RFID implementation in the
supply chain. As such, we seek to contribute to the
body of simulation modeling and analysis by
proposing a reference step-by-step framework. The
approach is grounded on previous works and on
empirical data gathered from a case study. Because the
framework integrates both theoretical and practical
considerations, it represents an initial step in
developing guidelines for using simulation to:
design RFID implementations in supply chain
map the way the RFID capabilities affect the
processes
evaluate RFID implementations in terms of
process-driven savings (labor hours,
processing times, etc.) and the infrastructure
requirements
This paper is organized as follows. Section 2 offers a
justification for the relevance of the work. Section 3
goes through the key elements of the proposed
framework. Section 4 gives an illustrative example of
applying the framework to a particular case study in
order to provide a better understanding. Finally,
Section 5 provides a number of conclusions and further
research aims.
2. RELATED STUDIES
To be cognizant of how this work contributes to
existing studies, this section draws upon literature
concerning the application of simulation in supporting
the design of RFID implementation within supply
chain processes.
Simulation models are regularly adopted in supply
chain management, in form of the traditional discrete-
event models or system dynamics or agent-based ones.
The prevalent use of wireless automatic and real-time
information technology in supply chain processes has
increased the need for this powerful tool. High initial
capital costs of such systems can produce uncertainties
in upper management who want to actually “see” how
changes will affect the performance of the processes
prior to making any investment. Simulation can
provide them with a platform to validate the
effectiveness or ineffectiveness of an altered system
without physically building, amending or interrupting
the real one (Senko and Suskind, 1990).
Since the technical problems associated with
implementing RFID have mostly been resolved, the
managerial issues emerge as critical (Angeles, 2005).
In this regard, the contributions dedicated solely to the
implementation of RFID within supply chain
management can be categorized into three domains.
The first one includes qualitative studies that discuss
general issues related to RFID technology. They can be
considered as conceptual papers that describe the
evolution of RFID, illustrate its benefits and pitfalls
and provide a roadmap for its implementation by
reviewing its success factors and impediments both in
general and within specific industries (Srivastava,
2004; Pramatari et al., 2005; Angeles, 2005; Jones et
al., 2005; Wu et al., 2006; Sellitto et al., 2007; Curtin
et al., 2007; Attaran, 2007; Reyes and Jaska, 2007).
The second domain includes papers that demonstrate
the value of RFID based on empirical evidence (i.e.
case studies, pilot projects). For instance, Karkkainen
(2003) discusses the potential benefits of RFID for
retailers based, on a trial conducted at Sainsbury’s,
while Hardgrave and Miller (2006) study the impact of
RFID on the ‘out-of-stock’ problem at Wal-Mart.
Loebbecke (2007) examines an RFID project involving
two leading European firms, department store chain
‘Kaufhof’ and fashion manufacturer ‘Gerry Weber’.
Other examples are the series of white papers
published by the Auto-ID Labs (e.g. Chappell et al.,
2003).
Finally, although research on the impact of RFID on
supply chains using analytical approaches has
proliferated significantly over the last few years (Ngai
et al., 2008), it is still at an early stage. Moreover, such
papers examine RFID potential impacts in a wide
range of contexts. For instance, Lee et al. (2004) used a
simulation model to quantify the indirect benefits
provided by RFID in inventory reduction and service
level improvement in a manufacturer-retailer supply
chain environment. Similarly, Fleisch and Tellkamp
(2005) examine the relationship between inventory
inaccuracy and performance by simulating a three
echelon retail supply chain with one product. Further
developments in this direction are provided by Doerr et
al. (2006) who provide an analysis of the costs and
benefits of fielding RFID technology for the
management of ordnance inventory by combining a
multi-criteria tool for the valuation of qualitative
factors with a Monte-Carlo simulation of anticipated
financial factors. Wang et al. (2008) focus on the
analysis of simulated impact of the radio frequency
identification (RFID) system on the inventory
Proceedings of the Operational Research Society Simulation Workshop 2010 (SW10)
129
replenishment of the thin film transistor liquid crystal
display (TFT-LCD) supply chain i n Taiwan.
Our review on the publications about simulation
modelling and RFID technology illustrates that there is
a growing body of literature that uses simulation
modelling as a performance evaluation tool to give a
quantitative assessment of the deployment of RFID in
supply chain processes. However, the research area
related to integrate simulation modelling as a decision
support tool to design RFID implementations has not
been addressed. Only one publication explicitly tries to
assist in RFID implementation design by proposing a
model-based reference model (Becker et al., 2009). In
addition, according to Cordella and Contini (2007), the
choice of a specific methodology to support the design
will affect the nature of the output of the design. As a
result, this paper tries to contribute to the domain of
research that is about proposing a systematic
methodology to efficiently design RFID
implementations. Integrating simulation modelling in
such a methodology can assist not only in extracting
realistic RFID implementations but also in evaluating
them at the shortest processing time and the lowest
operating cost.
3. GENERAL STRUCTURE OF THE
FRAMEWORK
The proposed framework is made up of several
interacting steps that are depicted in the Figure 1. For
each step, a detailed description follows that provides
its purpose, the techniques used and the obtained
outcomes.
Figure 1 The proposed framework
3.1 Simulation Modeling of the AS-IS processes
Purpose: The major aim of this step is to
conceptualise the way that current processes operate
(as-is model). As such, the objectives of this step are
to:
Model the as-is processes using business process
modeling notation (BPMN)
Develop the conceptual model that helps to
describe the system on a non-technical level
(inputs-outputs, aggregation of model
components)
Implement the as-is model on specialist simulation
software
Perform verification and validation tests to ensure
that the conceptual model has been transformed
into a computer model with sufficiently accuracy
(Davis, 1992) and that the model is sufficiently
accurate for the purpose at hand (Carson, 1986)
Analyse the results in order to understand the as-is
processes and identify potential improvements
Techniques: Business Process Modeling (BPM)
techniques and the key processes involved in a
simulation (such as those proposed by Law and Kelton,
2000; Banks et al., 2001; Robinson, 2004) are used to
model and simulate the as-is processes as a discrete
event model.
Outcomes: It is obvious that this step constitutes a
crucial prerequisite for identifying the performance
advantages behind investing in RFID (to-be model).
This means that the components that build the
particular model will be reused in the to-be model
incorporating, however, the RFID mindset.
Figure 2 Components of the as-is simulation model
Typically, a supply chain simulation model is guided
by the analyst’s mental models and the library of
building blocks offered by the simulation tool (van der
Zee and van der Vorst, 2005). As such, let us develop
the following proposition:
Proposition 1: “A simulation model of supply
chain processes constitutes a combination of four
Proceedings of the Operational Research Society Simulation Workshop 2010 (SW10)
130
interacting components: Configuration,
Workflow & Policies, Entities and Resources”.
Such proposition (Figure 2) is not ad-hoc as it is the
outcome of a combinatorial composition of previous
works on simulation modelling frameworks and
methodologies (such as Bagchi et al., 1998; Barnett &
Miller, 2000; Schunk & Plott, 2000; Siprelle, Parsons
and Phelps, 2001; Beamon and Chen 2001). Moreover,
it can be considered as a top-down approach, meaning
that the higher level components constitute constraints
for lower-level ones.
3.2 Experimental Design of the RFID
implementations
Purpose: There are numerous possible ways that the
supply chain processes can be shaped in order to
incorporate the RFID technology. Such dimensionality
derives from the fact that RFID is an infrastructure
technology. This means that the implementation
depends not only on the functionality of the RFID
application itself but more on how the technology is
deployed in terms of infrastructure requirements.
Adapted from Paul and Serrano (2004), the
relationship between processes, RFID applications and
the supporting RFID infrastructure design could be
seen as a three layered structure (Figure 3).
Figure 3 Processes, RFID Application & RFID
infrastructure (adapted from Paul and Serrano, 2004)
The figure illustrates that it is possible for a RFID
application to support more than one processes and run
on more than one infrastructure designs. Based on this
complex relationship and, in accordance with Davis
(1993) and Lee and Billington (1992), in order to guide
our research, the next proposition is adhered to:
Proposition 2: “To effectively design a RFID
implementation in supply chain one should focus
on the alternative infrastructure requirements that
a RFID application can be supported of”
In a simulation-based research, the process of
designing alternatives is the so-called “experimental
design” or “searching the solution space”. The solution
space is the total range of conditions under which the
simulation model might be run (Robinson, 2004). On
our occasion the total range of infrastructure designs
under which a RFID application can be run defines the
solution space. In other words, the RFID infrastructure
designs is our experimental factor that differentiates
the RFID experiments. As a result, the objectives of
this step are to provide some guidance about:
identifying the total values that a RFID
infrastructure design might assume for a particular
RFID application, called levels of the experimental
factor
identifying the total conditions under which the
‘to-be’ model can be run and, hence, designing the
RFID simulation experiments
Techniques: Based on an extensive literature review
on the implementation of RFID (described in Section
2) and on the experience gained by a case study, we
identify the variables that an RFID infrastructure
design involves. Then, using BPMN, we identify the
points in the processes (graphically represented by the
familiar diamond shape) that define alternative
behavior, work and information flow based on the
RFID infrastructure design variables.
Outcomes: In close correspondence with Sanchez
(2008), a RFID infrastructure design is a matrix where
every column corresponds to a variable, and the entries
within the column are values for this variable. Each
row represents a particular combination of variable
values, and is called a design point. As such, we
identify all the variables that distinguish one RFID
infrastructure design from another (Figure 4):
the location of RFID readers (check/reading
points)
the level of RFID tagging (pallet, case, item)
granularity (tagging only selected objects vs.
tagging all the objects, tagging products vs. assets)
Figure 4 RFID infrastructure design variables
Table 1 shows sample RFID infrastructure designs that
could be used for experimentation.
Table 1 Indicative RFID infrastructure designs
RFID
Design
Location of RF ID
readers
Level of
RFID
tagging
Granularity
1 Receiving portals Pallet All products
2Receiving &
Shipping portals Cases Only products
from supplier X
....
3.3 Mapping the RFID Effects in the Simulation
Model Components
Purpose: Based on the components of the simulation
model proposed in step 1, here we identify the
Proceedings of the Operational Research Society Simulation Workshop 2010 (SW10)
131
implications of change in each component due to the
introduction of RFID. The objectives of this step are
therefore to:
identify the nature of the effect on the responses
map the effects of RFID for a given RFID
infrastructure design
Techniques: A combinatorial composition of previous
works on the expected RFID’s potential for
improvement (Tajima, 2007; Lee, 2007; Roh et al.,
2009; Becker et al., 2009) is used in order to identify
and map the RFID effects in the as-is simulation model
components developed on step 1.
Outcomes: This step is built from the fundamental
characteristics of what RFID is or does: (i) RFID
provides automatic and massive identification; (ii) it
provides unique identification to an object. The way
RFID affects the simulation model components can be
seen mainly from the following aspects (Figure 5):
Processing time reduction
Error reduction
Change in Sequence
Resource labour reduction
Efficiency increase
Change in Distributions
Processes exclusion
New processes addition
Change in individual object characteristics
Figure 5 Mapping the RFID Effects in the Simulation
Model Components
3.4 Simulation modeling of the TO-BE processes
Purpose: The major aim of this step is to
conceptualise the way that the RFID-enabled processes
operate (to-be model).
Techniques: This step uses the same approach and
modeling techniques as those used to develop the as-is
model. In fact, to develop the to-be simulation model,
this step modifies the as-is model of step 3.1 so it
includes the RFID infrastructure design and effects
identified in the steps 3.2 and 3.3 respectively.
However, step 3.3 provides the nature of RFID effect.
To calculate the exact values is not a straightforward
task. Regarding the data availability, the ‘to-be’ data
can be regarded as ‘Category C’ data, meaning that it
is neither available nor collectable. This data
unavailability demonstrates even more the value of
simulation modeling to deal with this issue because
several experiments testing different values can be
performed with minimum cost and time effort.
Outcomes: After the completion of this step, a
simulation model that can incorporate different RFID
implementations is developed.
3.5 Experimentation, Results and Analysis
Purpose: Based on the design of experiments in 3.2
and the simulation model that can incorporate different
RFID implementations developed in 3.4, this step is
about deciding which experiments worth analyzing
using simulation.
Techniques: Selecting a design is an art, as well as a
science (Sanchez, 2008). In view of the numerous
RFID designs that are emerged in step 3.2, we identify
the realistic ones by taking into account two decision
variables. The first one lies in the aspect of technical
feasibility of the design and the second one includes a
cost assessment. As such we run many experiments in
order to guide the numerous design choices that can be
implemented.
Outcomes: After running a number of realistic RFID
experiments, we can reach informed conclusions
regarding the likely transition from the ‘as-is’ model of
the present supply chain processes to the ‘to-be’ views
foreseen for their future structure and workings.
4. CASE STUDY: APPLYING THE
FRAMEWORK IN THE WAREHOUSE
CONTEXT
To better illustrate the framework, it is applied to a
case study. The case concerns a third-party logistics
provider (3PL) that deals with paper trading and uses
its own assets and resources to provide a variety of
services on behalf of other companies. The study
considers one of its warehouses as a typical medium-
sized one. In the following sub-sections we will
illustrate the main points of this application of the
Proceedings of the Operational Research Society Simulation Workshop 2010 (SW10)
132
proposed framework (for a more detailed discussion of
the case study, the interested reader is referred to
Karagiannaki et al. (2007).
4.1 Simulation Modeling of the AS-IS processes
Business Process Modeling (BPM)
Four processes are spotted in the warehouse, namely
receiving, storage, picking and shipping. In this task,
we model these processes (Figure 6), using BPM
Notation (BPMN).
Containerarrives
Driveaforklift
tothecontainer
Pickuptheproducts
fromthecontainer
Scaneachproduct
(item/pallet/case)
Applyanewlabelto
eachproductfor
internaluse
isthereany
discrepancy?
isthereany
misread
(coveredor
damaged
barcodes)?
Movetheloadedforklift
toaprovisionalarea
insidethewarehouse
Unloadtheitems
fromtheforklift
NO
Double-check
thefindings
Scanmanuallyeach
rejectedproduct
YES
Getanotherco-worker
mixeduptocollaborate
onthemismatch
YES
NO
Initiateunloading
Checkforanydiscrepancy
betweentheBOLandthePO
Billof
Lading
Purchase
Order
Finishedaccepted
productsforstorage
Checkforanydiscrepancy
betweentheBOLandthePO
andtheactualshipmentactualshipment
scanned
Billof
Lading
Purchase
Order
actualshipment
scanned
Unloading
complete?
NO
YES
Confirmthe
mismatch
isthereany
discrepancy?
Reportthe
discrepancy
Discrepancyreport
(BOL-PO)
Discrepancyreport
(BOL-PO)
Discrepancyreport
(BOL-PO-actualshipment)
NOYES
Receiving process
Containerarrives
Driveaforklift
tothecontainer
Pickuptheproducts
fromthecontainer
Scaneachproduct
(item/pallet/case)
Applyanewlabelto
eachproductfor
internaluse
isthereany
discrepancy?
isthereany
misread
(coveredor
damaged
barcodes)?
Movetheloadedforklift
toaprovisionalarea
insidethewarehouse
Unloadtheitems
fromtheforklift
NO
Double-check
thefindings
Scanmanuallyeach
rejectedproduct
YES
Getanotherco-worker
mixeduptocollaborate
onthemismatch
YES
NO
Initiateunloading
Checkforanydiscrepancy
betweentheBOLandthePO
Billof
Lading
Purchase
Order
Finishedaccepted
productsforstorage
Checkforanydiscrepancy
betweentheBOLandthePO
andtheactualshipmentactualshipment
scanned
Billof
Lading
Purchase
Order
actualshipment
scanned
Unloading
complete?
NO
YES
Confirmthe
mismatch
isthereany
discrepancy?
Reportthe
discrepancy
Discrepancyreport
(BOL-PO)
Discrepancyreport
(BOL-PO)
Discrepancyreport
(BOL-PO-actualshipment)
NOYES
Receiving process
Receive
put-awaytasks
Finished
productsstored
Put-away
complete?
NO YES
Reportthatthe
put-awayisperformed
Pickascanned
productbyaforklift
Movetheloadedforklift
tothespecificclasszone
Put-awaytheproduct
attheappropriatebin
Initiateput-away
Identifyitsclass
(fast,mediumand
slowmovingproducts)
Arethere
otherproducts
storedwiththe
samecode?
Put-awaytheproductat
anarbitraryopenlocation
YESNO
Put-away process
Receive
put-awaytasks
Finished
productsstored
Put-away
complete?
NO YES
Reportthatthe
put-awayisperformed
Pickascanned
productbyaforklift
Movetheloadedforklift
tothespecificclasszone
Put-awaytheproduct
attheappropriatebin
Initiateput-away
Identifyitsclass
(fast,mediumand
slowmovingproducts)
Arethere
otherproducts
storedwiththe
samecode?
Put-awaytheproductat
anarbitraryopenlocation
YESNO
Put-away process
Receivecustomerorders
viaphone,fax,mail
Moveforklifttowards
variousstoragebins
Pickuptheproducts
fromthestoragebins
Scaneachproduct
(item/pallet/case)
istherea
divergence
Movetheloadedforklift
toaprovisionalarea
insidethewarehouse
Unloadtheitems
fromtheforklift
Manualcompliance
checkingof
outboundshipments
YES
NO
Initiatepicking
Finishedorders
fordelivery
Picking
complete?
NO
YES
isthereany
discrepancy?
Reportthe
discrepancy
NO
YES
Verifyagainstthe
customerorder
actualproducts
picked
Reportthatthe
pickingisperformed
Consolidate
customerorders
Preparethe
pickingtour
Sendthelistof
pickingintoWMS
Checkforanydiscrepancy
betweentheactualproductspicked
andthelistofpicking actualproducts
picked
Listofpicking
Generatethe
listofpicking
Listofpicking
Discrepancyreport
(Listofpicking-
actualproductspicked)
Picking process
Receivecustomerorders
viaphone,fax,mail
Moveforklifttowards
variousstoragebins
Pickuptheproducts
fromthestoragebins
Scaneachproduct
(item/pallet/case)
istherea
divergence
Movetheloadedforklift
toaprovisionalarea
insidethewarehouse
Unloadtheitems
fromtheforklift
Manualcompliance
checkingof
outboundshipments
YES
NO
Initiatepicking
Finishedorders
fordelivery
Picking
complete?
NO
YES
isthereany
discrepancy?
Reportthe
discrepancy
NO
YES
Verifyagainstthe
customerorder
actualproducts
picked
Reportthatthe
pickingisperformed
Consolidate
customerorders
Preparethe
pickingtour
Sendthelistof
pickingintoWMS
Checkforanydiscrepancy
betweentheactualproductspicked
andthelistofpicking actualproducts
picked
Listofpicking
Generatethe
listofpicking
Listofpicking
Discrepancyreport
(Listofpicking-
actualproductspicked)
Picking process
Receive
shippingtasks
Finishedloaded
containerfordeparture
Shipping
complete?
NO YES
Reportthatthe
shippingisperformed
Pickascanned
productbyaforklift
Movetheloadedforklift
tothecontainer
Initiateshipping
Unloadtheproducts
fromtheforklift
Generatethe
BillofLading(BOL)
GivetheBOLto
thecontainerdriver
Billof
Lading
Billof
Lading
Shipping process
Receive
shippingtasks
Finishedloaded
containerfordeparture
Shipping
complete?
NO YES
Reportthatthe
shippingisperformed
Pickascanned
productbyaforklift
Movetheloadedforklift
tothecontainer
Initiateshipping
Unloadtheproducts
fromtheforklift
Generatethe
BillofLading(BOL)
GivetheBOLto
thecontainerdriver
Billof
Lading
Billof
Lading
Shipping process
Figure 6 BPM of the as-is warehouse processes
Conceptual Modeling
This section involves a non-software specific
description of the model content and the warehouse
components of the as-is simulation model.
Table 2: Model Content
Products
Include within the Entities Component.
Flow through the warehouse that triggers the
processes of receiving and put-away.
Orders
Include within the Entities Component.
Flow through the warehouse that triggers the
processes of picking and shipping.
Operatives
Include within the Resources Component.
Resources responsible for unloading, scanning,
checking, storage, retrieval and loading of the
products. All resources need to be modeled to give
full statistics on queues and resource utilisation.
- Receiving
& Put-away
- Picking &
Shipping
Include within the Workflow & Policies
Component.
- unloading, checking, scanning and relabeling
- retrieval, scanning and checking time
Scanning
errors
Include within the Workflow& Policies Component.
Misreads because of unlabeled products and covered
or damaged barcodes result in rejected products that
must be carried out manually, with the expected
delay of the process.
Queues for:
- Unloading/
loading
- scanning-
in and -out
- checking-
in and -out
Include within the Workflow & Policies
Component.
Include as buffers in between steps in a process or
as storage areas for inventory.
Need to be modeled to give full statistics on queues
and resource utilisation
Traveling
times
Include within the Configuration Component
Based on the trucks’ speed that is 12,5 km/hr
Scale Include within the Configuration Component
Design of the layout based on a 1:200 scale
Table 3 Simulation Model Components of the as-is
warehouse processes
Configuration Component Workflow & Policies
Four Distinct Processes:
-Receiving
-Storage
-Picking
-Shipping
Entry points:
- Containers arrival
-Orders arrival
Exit points:
-Containers shipped
Storage piles
-the model is currently
designed to have up to 100
queues at a given time
Component
Scanning-in and out
Loading
Unloading
Checking-in and out
Picking
Consolidate: based on Use
Label Batching – lbl cues
weight
Entities Component
Containers
Orders
Resources Component
scanning crew
storing and picking crew
loading and unloading
crew
Model Coding
The computer modeling was implemented using the
standard version of SIMUL8 simulation software
(Figure7).
Figure 7 Print screen of the warehouse model
Model Validation
A confidence interval (Figure 8) has been employed to
obtain quantitative implications about the simulation
model’s quality.
Simulatedvsrealoutputsofwarehouseinventory
0
20
40
60
80
100
120
1 6 11 16 21 26 31 36 41 46 51 56 61
simulatedoutput realoutput
Figure 8 Simulated vs. real outputs
4.2 Experimental Design of the RFID
implementations
Based on our framework, the following table depicts
the conditions under which the as-is warehouse model
can be run when implementing the RFID technology.
Table 4 Indicative RFID designs in the warehouse
RFID
Design
Location of RF ID
readers
Level of
RFID
tagging
Granularity
1Receiving & Shipping
portals Pallet All products
and Assets
2
Receiving & Shipping
portals & Transportation
Modes (i.e. clarks)
Pallet
Only
products
from
supplier X
In view of the alternative RFID designs, we had to
decide on one in order to build the to-be simulation
model and then experiment with the remainder ones.
Proceedings of the Operational Research Society Simulation Workshop 2010 (SW10)
133
To do so, we decided based on a pilot implementation
that the company wanted to run. As such, RFID
readers were placed at the major gateways or portals
within the warehouse at the receiving and shipping
docks. In addition, multiple antennas were used to
minimize misreads and no reads, since a single reader
cannot guarantee 100% accuracy for all reads. RFID
tagging was taken place at pallet-level. Also, it was
assumed that all the products were tagged.
4.3 Mapping the RFID Effects in the Simulation
Model Components
Having decided on the infrastructure design, the
differences between the ‘as-is’ and ‘to-be’ model can
be stated by looking to the RFID effects in each
component.
Configuration Component
Neither the content nor the scale of the simulation
model is altered. However, in the ‘to-be’ model, some
activities such as the relabeling of products to continue
the storage process is no longer necessary. Thus, this
process has been eliminated in the ‘to-be’ model.
Workflow & Policies Component
Concerning the receiving process, instead of manually
scanning each inbound load and verifying it with the
purchase order and the shipment notification, in the
RFID model scenario each received load can be
identified and checked automatically as it passes
through the portal readers. As a result, the processing
time for scanning each product as well as the
processing time for checking for any discrepancy has
been reduced. Moreover, RFID decreases the incorrect
receipt of damaged or covered barcodes since RFID
readers are more reliable than traditional scanners. This
indicates that in the ‘to-be’ model, the efficiency of the
scanning process has been increased. Regarding the
shipping process, RFID allows staff to collect data
when products simply pass through the shipping doors.
This eliminates the need for compliance checks on the
shipping dock. Thereby, in the proposed ‘to-be’ model,
the processing time for loading has been significantly
reduced. Regarding the put-away and picking
processes, no change is needed as the selected RFID
infrastructure design cannot support these processes.
Entities Component
All the products are tagged. So there is no need for
changing individual variables of the objects.
Resources Component
Within this component, in the ‘to-be’ model, the
number of labour as well as the number and location of
equipment were remained the same.
4.4 Simulation modeling of the TO-BE processes
To develop the to-be simulation model, we modified
the as-is model so it includes the RFID infrastructure
design and effects identified in the previous steps.
However, the previous step provides the nature of
RFID effect. To calculate the exact values, data were
gathered from the pilot RFID implementation.
4.5 Experimentation, Results and Analysis
A summary of the responses, which determine
achievement of the objectives, is presented below.
Table 5 Summary of responses
Performance Measure Change
(%)
Labor Utilisation
Scanning labor Utilisation (%) -25.39
System Benchmarks
Daily Throughput SKUs received
and shipped/day -6.88
Cycle time Days from birth to
death -10.77
Inventory
Daily warehouse inventory SKUs on hand 14.94
Service Levels
Time waiting for unloading Queuing Time 39.73
Time waiting for storing Queuing Time 225.04
Time waiting for loading Queuing Time -56.62
Time from truck arrival to
products’ scanning (A)
Processing+
Queuing Time -56.35
Time from products’ scanning
to products’ storing (B)
Processing+
Queuing Time 81.43
Time from truck arrival to
products’ storing (A+B)
Processing+
Queuing Time 63.9
Time from o rder arrival to
product’s picking (C)
Processing+
Queuing Time -0.26
Time from products’ picking
to products’ loading (D)
Processing+
Queuing Time -46.41
Time from o rder arrival to
products’ loading (C+D)
Processing+
Queuing Time -7.36
Although the cycle time was not significantly reduced,
clear improvements appear in both receiving and
shipping processes, i.e. those that become RFID-
enabled. Concerning the receiving process, the
utilization of ‘scanning labor’ was reduced by 25%,
while, the ‘time from arrival to scanning’ was more
than halved (by 56.35%). This also arises from the fact
that the relabeling of products, to continue to the
storage process, is no longer necessary. Similarly,
regarding the shipping process, the ‘time from picking
to loading’ was almost halved (46.41% reduction). In
contrast, the ‘time from scanning to storing’ was
significantly increased (81.43%), since scanning is
now performed in a faster pace, thus eliminating the
queue preceding it and increasing the one following it
(i.e. the queue between scanning and storing). This is
also reflected by the substantial increase in the ‘time
from arrival to storing’. Nevertheless, this increase
does not counteract the overall improvement in
system’s performance. Still, our results indicate that
Proceedings of the Operational Research Society Simulation Workshop 2010 (SW10)
134
introducing RFID within the storing and shipping
processes could create additional benefits.
As a result, we chose a different RFID infrastructure
design that supports all the four processes and run this
experiment. Our results indicate that in order to reap
the benefits of the RFID technology, it is better for this
company to move on a RFID design that incorporates
all the processes and not on a gradually investment.
5. CONCLUSIONS
One of the top ICT trends for supply chain
optimisation is that of RFID technology. In fact, there
are numerous possible ways that the supply chain
processes can be shaped in order to incorporate the
RFID technology, each bringing its own benefits, as
well as requirements in cost and infrastructure. It is
unclear which implementation should be used in what
particular situation and, furthermore, a complete list of
these RFID implementations has not been reported in
literature up to now. Such dimensionality in RFID
implementations produces uncertainties in upper
management who want to actually ‘see’ how to assess
the benefits that a given design may bring to the
business processes prior to making any investment. As
with all novel technologies, terms such as ‘eye-ball the
data’ and ‘make some initial decisions…based on
intuition, experience and judgement’ are typical. There
is a credibility gap: to make robust investment
decisions, we need a methodology to support the
design of RFID implementations. Integrating
simulation modelling in such a methodology can assist
not only in extracting RFID implementations but also
in evaluating them at the shortest processing time and
the lowest operating cost.
The required elements of such a framework for the
design of RFID implementations are still insufficiently
addressed topics in the literature, especially with
regards to simulation modelling. In this paper,
drawing on theoretical constructs relevant to the RFID
implementation, we have proposed a step-by-step
framework and simulation modeling that, when
combined, allow for reaching informed conclusions
regarding the likely transition from the ‘as-is’ model of
the present supply chain processes to the ‘to-be’ views
foreseen for their future structure and workings. The
utility of the proposed framework is demonstrated
through a case-study application concerned with
designing RFID implementation within the warehouse
context.
Based on the knowledge gained through the case study,
the framework that integrates simulation modeling can
be regarded as a powerful decision support tool to
design and evaluate RFID implementations, by testing
various risk-free and virtual ‘to-be’ views. Managers
can conduct “what-if” analyses and study the
performance characteristics of RFID implementations
without physically building, amending or interrupting
the system. Moreover, this framework can be
considered as a communication tool as it facilitates
discussion and provokes a debate about the problem
situation as well as the possible solutions.
However, the work presented in this paper is a
preliminary effort to design, in a systematic way,
alternative RFID implementations and use them as
experiments in order to simulate the impact of RFID.
Further work is required. The framework should be
applied in other cases and incorporate even more
experiments before/after RFID is deployed. As such,
the potential strengths or weaknesses of the framework
can be demonstrated so that it can be improved.
Finally, in order to capture the financial aspect of the
RFID deployment, a cost-benefit analysis can be easily
integrated with the simulation model to poise the
various advantages that the RFID technology promises.
REFERENCES
Agarwal, V. (2001). “Assessing the benefits of Auto-
ID Technology in the Consumer Goods Industry”.
Cambridge University Auto-ID Centre.
Angeles, R., 2005. RFID technologies: supply-chain
applications and implementation issues,
Information Systems Management, 22(1), 51-65
Archer, R. (1995). “Using dynamic analysis: The role
of simulation in process engineering”, Enterprise
Reengineering (October/November 1995) 24
Attaran, M., 2007. RFID: an enabler of supply chain
operations, Supply Chain Management: An
International Journal, 12(4), 249-257
Bagchi, S., S. Buckley, M. Ettl, G. Lin (1998),
Experience using the IBM supply chain simulator,
in: Proceedings of the 1998 Winter Simulation
Conference.
Banks, J., Carson, J.S. Nelson, B.L. and Nicol, D.M.,
2001. Discrete-Event System Simulation, 3rd
edition, Upper Saddle River, Prentice Hall
Barnett, M.W., and C.J. Miller, (2000). “Analysis of
the Virtual Enterprise Using Distributed Supply
Chain Modeling and Simulation: An Application of
e-SCOR. In: Proceedings of the 2000 Winter
Simulation Conference
Beamon, B.M. and V.C.P. Chen. (2001). “Performance
analysis of conjoined supply chains. International
Journal of Production Research. 39 (14), 3195-
3218
Becker, J., Vilkov, L. Weiß, B. and Winkelmann, A.
(2009). “A model based approach for calculating
the process driven business value of RFID
investments”, International Journal of Production
Economics, doi:10.1016/j.ijpe.2009.08.025
Bendavid,Y., Lefebvre, E., Lefebvre, L., Fosso,
Wamba,S. (2008).”Key performance indicators for
the evaluation of RFID-enabled B-to-B e-
commerce applications. The case of a five-layer
Proceedings of the Operational Research Society Simulation Workshop 2010 (SW10)
135
supply chain. Information Systems and E Business
Management, 7(1), 1–20.
Carson, J.S. (1986). “Convincing users of model’s
validity is challenging aspect of modeler’s job”,
Industrial Engineering, 18(6), pp.74-85
CASAGRAS Final Report, October 2009, an EU
Framework 7 project, EU project Number 216803.
Chao, C.C.,Yang,J.M.,Jen,W.Y. (2007). “Determining
technology trends and forecasts of RFID by
historical review and bibliotmetric analysis from
1991 to 2005. Technovation, 27(5), 268–279.
Chappell, G., Ginsburg, L., Schmidt, P., Smith, J.,
Tobolski J., 2003. Auto-ID on the Line: The Value
of Auto-ID Technology in Manufacturing, White
Paper, Auto-ID Center, MIT
Cordella, A. and Contini, A. (2007) “Information
system and information infrastructure deployment:
the challenge of the Italian e-justice approach”.
Electronic journal of e-government, 5 (1).
Curtin, J., Kauffman, R. J., Riggins, F. J., 2007.
Making the ‘MOST’ out of RFID technology: a
research agenda for the study of the adoption, usage
and impact of RFID, Information Technology and
Management, 8, 87-110
Davis, P.K. (1992). “Generalizing concepts of
verification, validation and accreditation (VV&A)
for military simulation”, RAND
Davis, T. (1993), “Effective Supply Chain
Management”, Sloan Management Review,
Summer, 35-46
Doerr, K.H., Gates, W. R. and Mutty, J. E. 2006, A
hybrid approach to the valuation of RFID/MEMS
technology applied to ordnance inventory,
International Journal of Production Economics, 103
(2), 726-741
Doukidis, G.I. and R.J. Paul. (1985). “Research into
Expert Systems to Aid Simulation Model
Formulation”. Journal of the Operational Research
Society, 36(4), 319-325.
Fleisch, E. and Tellkamp, C. (2005). Inventory
inaccuracy and supply chain performance: a
simulation study of a retail supply chain,
International Journal of Production Economics,
Vol. 95, No. 3, pp. 373-385
Hardgrave, B. and Miller, R., 2006. The myths and
realities of RFID, International Journal of Global
Logistics & Supply Chain Management, 1(1), 1-16
Hou, J.L., and Huang, C.H. (2006). “Quantitative
performance evaluation of RFID applications in the
supply chain of the printing industry”. Industrial
Management & Data Systems, Vol.106, No.1,
pp.96-120.
Jones P., Clark-Hill C., Comfort D., Hillier D. and
Shears P. (2005). “Radio frequency identification
and food retailing in the UK”. British Food Journal,
Vol.107, pp.356-360.
Karagiannaki, A., Mourtos, I. and Pramatari, K.
(2007). “Simulating and Evaluating the Impact of
RFID on Warehousing Operations: a case study”.
In Summer Computer Simulation Conference, July
15-18, San Diego, CA
Karkkainen M., 2003. Increasing efficiency in the
supply chain for short shelf life goods using RFID
tagging, International Journal of Retail &
Distribution Management, 31(10), 529-536
Kelepouris, T., Pramatari, K. and Doukidis, G., 2007.
RFID-enabled traceability in the food supply chain,
Industrial Management & Data Systems, 107(2)
Kelly, E.P., and Erickson, G.S. (2005). “RFID tags:
commercial applications v. privacy rights”.
Industrial Management & Data Systems, Vol.105,
No.6, pp.703-713.
Law, A.M. and Kelton, W.D. (2000). Simulation
Modeling and Analysis, 2nd edition, McGraw-Hill
Lee, H.L., Billington, C. (1992), “Managing supply
chain: pitfalls and opportunities”, Sloan
Management Review, Spring, 65-73
Lee, Y., Cheng, F. and Leung, Y. (2004). “Exploring
The Impact Of RFID On Supply Chain Dynamics”,
In Proceedings of the 2004 Winter Simulation
Conference
Lee, H. (2007). “Peering through a glass darkly”, ECR
Journal : International Commerce Review, Spring
2007
Lee, H. and Özer Ö (2007). Unlocking the value of
RFID, Production and Operations Management,
Vol. 16, No.1, pp. 40-64Lekakos G. (2007).
“Exploiting RFID digital information in enterprise
collaboration”. Industrial Management & Data
Systems, Vol.107, No.8, pp.1110-1122.
Loebbecke, C. (2007). “Piloting RFID Along the
Supply Chain: A Case Analysis”, Electronic
Markets, Vol.17, No.1, pp. 29-37.
Ngai, E.W.T. Moon, K.K.L., Riggins, F. J. and Yi,
C.Y. 2008. RFID research: An academic literature
review (1995–2005) and future research directions,
International Journal of Production Economics,
112(2), 510-520
Nidumolu, S. A., Menon, N.M., and Zeigler, B.P.
(1998). “Object-oriented business process modeling
and simulation: A discrete event system
specification framework”, Simulation Practice and
Theory, 6, 533–571
Paul, R.J. and Serrano, A. (2004). “Collaborative
Information Systems and Business Process Design
Using Simulation,” Proceedings of the 37th Annual
Hawaii International Conference on System
Sciences (HICSS'04)
Pramatari, K.C., G.I., Doukidis, P., Kourouthanassis,
2005. Towards ‘Smarter’ Supply and Demand-
Chain Collaboration Practices Enabled by RFID
technology, Springer Berlin Heidelberg, Consumer
Driven Electronic Transformation, Berlin, 241-256
Reyes, P.M., Jaska, P., 2007. Is RFID right for your
organization or application?, Management
Research News, 30(8), 570-580
Robinson, S. (2004). “Simulation: The Practice of
Model Development and Use”, Wiley
Proceedings of the Operational Research Society Simulation Workshop 2010 (SW10)
136
Roh, J.J., Kunnathur, A. and Tarafdar, M. (2009).
“Classification of RFID adoption: An expected
benefits approach”, Information & Management,
46, (6), 357-363
Sanchez, S. M. (2008). Better than a petaflop: the
power of efficient experimental design, in
Proceedings of the 40th Conference on Winter
Simulation, 73-84
Schunk, D. and B. Plott, (2000). “Using Simulation to
Analyze Supply Chains. In: Proceedings of the
2000 Winter Simulation Conference
Sellitto, C., Burgess, S., Hawking, P., 2007.
Information quality attributes associated with
RFID-derived benefits in the retail supply chain,
International Journal of Retail & Distribution
Management, 35(1), 69-87
Senko, J. M. and Suskind, P. B. 1990. Proper Planning
and Simulation Play a Major Role in Proper
warehouse Design, Industrial Engineering, 22(6),
34-37
Siprelle, A.J., D. Parsons, R.A. Phelps (1999), SDI
Industry Pro simulation for enterprise wide
problem solving, in: Proceedings of the 1999
Winter Simulation Conference
Smith, H. and Konsynski, B. (2003). “Developments in
practice X: radio frequency identification (RFID) –
an internet for physical objects”. Communications
of the AIS, Vol.12, pp.301-11.
Srivastava, B., 2004. Radio Frequency ID Technology:
The Next Revolution in SCM, Business Horizons,
47(6), 60-68
Tajima, M. (2007). “Strategic value of RFID in supply
chain management”, Journal of Purchasing and
Supply Management, 13 (4), 261-273
Ustundag, A., Tamyas, M., (2009). “The impacts of
radio frequency identification technology on supply
chain costs”. Transportation Research Part E:
Logistics and Transportation Review, 45 (1), 29–38
Van der Zee, D.J. and J.G.A.J. van der Vorst. (2005).
“A Modelling Framework for Analyzing Supply
Chain Scenarios: Applications in Food Industry”.
Decision Sciences. 36 (1), 65-95
Wang, S., Liu, S. and Wang, W., 2008. The simulated
impact of RFID-enabled supply chain on pull-based
inventory replenishment in TFT-LCD industry,
International Journal of Production Economics,
112(2), 570-586.
Wu, N.C., Nystrom, M.A., Lin, T.R., Yu, HC., 2006.
Challenges to global RFID adoption, Technovation,
26(12), 1317-1323
AUTHOR BIOGRAFIES
ANGELIKI KARAGIANNAKI holds a BSc in
Management Science and Technology from Athens
University of Economics and Business (AUEB) and a
MSc in Management Science and Operational
Research from Warwick Business School. She is
currently PhD candidate at the Athens University of
Economics and Business (AUEB), Department of
Management Science and Technology. She is also
research officer in the SCORE group of the ELTRUN
Research Center at AUEB.
KATERINA PRAMATARI is Assistant Professor at
the Department of Management Science and
Technology of the Athens University of Economics
and Business (AUEB) and scientific coordinator of the
ELTRUN-SCORE research group operated at the
ELTRUN Research Center at AUEB. She holds a PhD
from Athens University of Economics & Business
(AUEB) and a Masters in Information Systems from
the same University. She has worked as a systems
analyst for Procter & Gamble European Headquarters
for two years, on the development of global Category
Management applications, and another year in the
Marketing Department of Procter & Gamble Greece. In
the last years she has had active participation, in the
fields of IT and Marketing, in the setup of new
business ventures in the area of e-business and supply
chain integration in grocery retailing. She has won
both business and academic distinctions, she has been
granted eight state and school scholarships and has
published more than twenty five journal and
conference articles.
GEORGIOS I. DOUKIDIS is a Professor in
Information Systems and Chairman of the Department
of Management Science and Technology at the Athens
University of Economics and Business (AUEB). He
holds and MSc and a PhD in OR/IS from the London
School of Economics, where he taught as a lecturer for
six years in the Information Systems Department, and
currently is a visiting Professor at Brunel University.
He has published 12 books and more than 150 papers
and has acted as guest editor for the Journal of
Operational Research Society, the European Journal of
Information Systems, the Journal of Information
Technology and the International Journal of Electronic
Commerce. He is founder and director of the eBusiness
Research Center of AUEB (ELTRUN) which is one of
the largest in European Business Schools with 35 F.T.
researchers, that specializes on m-Commerce, digital
TV, knowledge management, supply-chain
management, ebusiness models, digital marketing and
I.S. management. He has acted as Chairman in the
following International Conferences: European
Conference in Information Systems (1995),
International Electronic Commerce Conference (1998),
International Conference of the Decision Sciences
Institute (1999), International Conference on Mobile
Business (2002). His latest book: “Consumer Driven
Electronic Transformation: Applying New
Technologies to Enthuse Consumers”, was published
by Springer-Verlag in December 2004.