32st International Conference of System Dynamics Society
July 20-24, 2014
Delft, Netherlands
USING SYSTEM DYNAMICS AND FUZZY LOGIC TO ASSESS
THE IMPLEMENTATION RFID TECHNOLOGY
Milton M. Herrera R.', Mauricio Becerra F.’ , Olga R. Romero Q. 2 and
Javier A. Orjuela C?
Abstract — The Technological growth has led to the acquisition and implementation of
technologies to improve the performance of organizations that make up the supply chain.
The paper presents a RFID Technology Implementation Model on picking operation under
the approach system dynamics and fuzzy logic, to analyze behavior over time and
information and material flow in the fruit supply chain assessed by means of technology
change policies.
The results show that the model has a better performance by integrating fuzzy
inference system and system dynamics simulation, allowing make decisions through policy
implementation traceability technology in fruit supply chain on the lead-time picking
operations. Whereas, an innovation approach that combines elements complexity-
uncertainty, causality-experience, behavior-knowledge through of loop causal, dynamics
simulation model and fuzzy inference system.
Key words: System Dynamics, Fuzzy Logic, Supply Chain, Technology Management
and RFID Technology.
Introduction
The focus paper shows the impact on information and material flow in technology
management activities of organizations that make up the supply chain. In this way, the
traceability technology used in the supply chain improves the performance in control
quality and setup time (picking) associated with the supply chain. Time spent picking
operations in the storage products or raw materials represent a significant percentage the
total cost. This in turn, affects the quality and food safety for consumers and organizations
of fruit supply chain. However, the traceability technology implementation requires
appropriate policies for technology management. The model simulation shows the behavior
of policy implementation traceability technology on picking operations in fruit supply
chain.
The behavior of the model under study is analyzed by means of a simulation model
based on the principles of the system dynamics methodology and fuzzy inference system.
' Universidad Piloto de Colombia-Facultad de Ingenierta. Cra. 9 N° 45%-44 Bogotd-Colombia. milton-
‘sidad Catélica de Colombia-Facultad de Ingenierfa. Diag. 47 N° 15-50 Bogotd-Colombia.
mbecerra@ ucatolic
3 Suppla S.A. Lo}
4 Universidad Distrital Francisco José de Caldas-Facultad de Ingenierfa. Cra. 7 N° 40-53 Bogoté-Colombia.
jorjuela@ udistrital.edu.co
du.co
ica Inteligente-Gerente Técnica. Diag. 22“ N° 56*-40 Bogotd-Colombia. olga.romero@ suppla.com
Therefore, the results of the integrated model (System Dynamics-Fuzzy Inference System)
were evaluated on performance measures of the flow behavior in the supply chain. This
paper is organized as follows. Firstly, section presents the background of the model.
Second, the methodology and simulation model structure. Third, simulation model
integrated into the fuzzy logic system is presented. Finally the results and conclusions are
provided.
Background
The globalization of markets has created a dynamics between organizations framed on
competitiveness and productivity. The economic policies of globalization have enabled the
integration of markets, so it is important to perform an analysis focused on identifying the
competitive advantages in the fruit supply chain (Orjuela C., Calderon, & Buitrago, 2006).
In Colombian, one of interest is the technological aspect due to market expansion and
diversification effect on product manufacturing companies. This is evidenced by the use of
various information technologies (Figure 1). These ate highlighted using traceability
techniques, technologies CRM (C Relati ip M ‘) and DRP
(Distribution Requirements Planning) and production software a 60 per cent. However, The
technology using Colombia in manufacturing companies is related to MRP application
(Material Requirements Planning), transactions specializing in EDI/XML, storage and
distribution software.
Traceability Techniques
Software Distributio: GPS Tracking Techniques
Software Storage Digital Maps
Software EDI/XML Marking Technologies
Software production Coding Technologies
Software in supply or} CRM Technologies
demand 2
Goods Location:
Techaslagles DRP Technologies
MRP Technologies
——Yes ——=No_ "None
Figure 1. Technologies Information Used in Manufacturing (Colombia).
In this sense, the warehouse management system is an essential part of supply chain.
The storage activities (receipt, picking and y distribution) are a fundamental in your design
(Frazelle & Sojo, 2007). One of the biggest concerns of a warehouse management system is
the picking activity. Therefore picking activities represent a higher cost in storage
operations (Chen, Hwang, & Chen, 2009). The picking can be understood as the activity in
which the enlistment of products and raw materials. It intended to respond to the request of
customers in the shortest time.
32st International Conference of System Dynamics Society
July 20-24, 2014
Delft, Netherlands
Traceability is an essential process in picking activities of the warehouse. In this
sense, applications of RFID on picking operations propose evaluate the performance of the
designs and procedures performed in a warehouse (Chen, Hwang, & Chen, 2009). These
uses improve picking operations and thus the internal traceability of manufacturing
processes in the fruit supply chain.
According to, the Asociacién Nacional de Empresarios de Colombia (2012) a 77,8%
companies know the picking optimization systems, but these systems have not. Although
the percentage of companies with picking optimization systems is low (22,2%) compared to
those without, it is noteworthy that 12,7% have picking optimization systems through
development internal organization, while 6,3% has been obtained by a supplier. It is
observed in Figure 2 that 3,2% of surveyed companies have picking optimization system
through ERP systems. This accounts for the need to design appropriate picking systems that
support the manufacturing companies (Asociacién Nacional de Empresarios de Colombia-
ANDI, 2012).
12,7%
3.2%
> 15%
a] —_ 6.3% &
778% 48%
= Known but unavailable ® Available - In House Development
© Available - System ERP = Available - Supplier
= Custom Development
Figure 2. Picking optimization system.
Picking procedures are therefore related to different technologies. There are two
ways to perform these procedures (picking in situ and picking stations). Studying picking
operations from Colombia represent 80% of high-impact activities in logistics, therefore
require a high level of training in the fruit supply chain (Puentes G., 2006). The different
techniques, aspects and variables that must be considered the picking operations are
relevant to determining the quality and food safety in fruit supply chain.
System Dynamics approach in planning technology
The technology implementation is an activity technology management and this from
knowledge management. This approach from system dynamics is approached by
Wolstenholme (2003) in the assessment of technology level under simulation the
contribution is the analysis of structural behavior in the technology changes. Dharmaraj et
al. (2006) proposes a model that represents the dynamics of technology in an innovative
organization, the results focus on alternatives to obtain competitiveness of the organization.
He analyzed two effects of loops, they relate to the change in technology and the adoption
of new technologies. On the other hand, Karikoski et al. (2009) and Yin & Xia (2011)
approach the effects of technology transfer and emerging technologies; the first focuses on
the analysis under various scenarios of emerging technologies, while the other focuses on
the study of integration between the system's strategic partners for transfer technology.
The relationship between integration and technology management and logistics
system of the supply chain is approach to Kalenatic et al. (2009) using a methodology that
has flows that relate both aspects and proposes the sub-systems that support the central
system in organizations.
The implementation models of RFID technology studies to Chen (2011) and De
Marco et al. (2012) proposed the implementation from the growing market for RFID
technology and the effect on retail stock. In this way Herrera and Orjuela (2012) approach
the implementing RFID technology that is developed in this paper. Analysis of these
approaches is presented in Figure 3.
Chen, Y.
011)
Herrera ae .
Y Karikoski, Yin,Q.,
iusla etal. Xia, L
De (2012) (2009) (2011)
Marco
etal.
(2012)
Implementing RFID Knowledge Transfer and Management
Technology Emerging Technologies Technology
Figure 3. System Dynamics Approach in Impl ing RFID Technology-Technology M.
In conclusion, we have proposed systemic models and innovation policy development and
implementation of food safety technologies that address the complexity of the system and
the relationship between variables making it an appropriate methodology in dealing with
the planning and implementation of technology and incidence or food product traceability
in fruit supply chain.
Dynamics System and Fuzzy Inference System
The approach of integration between System Dynamics and the Fuzzy Inference Systems
on the analysis of supply chain models is a novel approach that has allowed a better
qualitative understanding of model (Guzman & Andrade, 2009). Applications proposed by
Ghazanfari et at. (s.f.) in which causal diagrams are developed with fuzzy relations.
However, these applications take modified through fuzzy sets that are applied in supply
chains, the first work developed by Zadeh (1965) and Zimmerman (1983) suggest the
genesis of these methods.
The evolution of operations research models is evident in the development of
comprehensive and dynamic model are essentially characterized by the combination of
mathematical approaches that contribute to make decisions. Xu & Li (2011) using fuzzy
optimization and dynamic systems, proposes a conceptual perspective and comprehensive
32st International Conference of System Dynamics Society
July 20-24, 2014
Delft, Netherlands
dynamic model which is to approach five parameters: initial (C), flow (X), level (G), fuzzy
(Z) and objective (F) of the system under study, these relate to the simulation of System
Dynamics (SD) and programming objective Function (FMOP) through the initial
parameters (C) to set comprehensive and dynamic model (SD-FMOP). Later, the fuzzy
parameter optimization with genetic algorithms is performed, and finally performs a
simulation on system dynamics.
On the other hand, Mutingi & Mbohwa (2012) made a proposal that includes four
phases. The first phase developed a simulation model with system dynamics. In the second
phase the system dynamics model used a fuzzy parameters, with which the uncertainty of
the real situation was addressed. Subsequently, the mathematical model was solved with an
optimization technique. Finally, The policies are designed through the obtained parameters.
The approach of using system dynamics and fuzzy logic is discussed since 1990.
Mula, J. (2013), Campusano-Bolarin et al. (2013) and Peidro et al. (2010) discussed studies
on the behavior of inventories in the supply chain. Furthermore, Carvalho (2000) presents
studies about fuzzy cognitive maps and qualitative relation in simulation models under
system dynamics.
The work done by Xu & Li (2011) and Ng et al. (2009) presented an integrated
approach between bio-inspired techniques (genetic algorithmic) techniques of Fuzzy
Optimization and Systems Dynamics applied in supply chain and workforce. The use of
genetic algorithms as method of integrated solution with System Dynamic is proposed by
Li & Wang (2010), Lian & Jia (2012) and Ng et al. (2009) in application of inventory
system optimization. On the other hand, the studies discussed by Xu & Li (2011) and Li,
Xu & Jiang (2013) used the Linear Multi-Objective Optimization method as integrated
methodology of System Dynamics. Liu, S., Triantis, K.P., Sarangi, S. (2011), used the
linguistic variables in a model sales and service showing the simulation results
corresponding to the probability of generating new customers and profit taking considered
fuzzy rules.
Applications of fuzzy optimization models in the areas inherent in production
management and supply chain have taken hold and have been extended in order to make
decisions taking into account the uncertainty of the real system. That is why the use of such
models has been widely reported. Dejmek & Skoglund (2007) discuss the internal
traceability introduces the term fuzzy traceability, demonstrating the difficulty in tracing
the raw material used in a process in a factory line of liquid foods. Therefore, the fuzzy
optimization and simulation approach in the integration of methods used to address
problems of tracking or fuzzy traceability.
Other approach in integrated optimization model and simulation in the supply chain
is addressed by Abo-Hamad & Arisha (2011). In this sense the Integral Dynamics Models
of planning, scheduling and control approach by Kalenatic (2001) and Orjuela et al. (2010).
The first is developed in manufacturing and the second is proposed in the service sector.
In conclusion, Fuzzy Inference Systems approach the parameter uncertainty in the
real system. This is usually modeled with statistical tools, however in some cases when the
number of variables is not significant other techniques that measure are used (Kalenatic,
2001). Therefore, in the next chapter developed the methodology of model.
Methodology
The methodology was developed two approaches. The first related to the simulation model
continues of traceability technology implementation. The second supported under a fuzzy
logic system that contributes to the simulation model under dynamic systems with the
knowledge base. The integration favors the process of decision-making, because the
complexity is approach with the System Dynamics and uncertainty in the parameters of
simulation. The relationship between the methodologies is shown in Figure 4.
In a complex system involved actors and experts. The first characterized by
complex relationships in which they are immersed and the experts who have the knowledge
base of system. The integration methodology provides the actors in the case of simulation
and causal diagram construction and the experts who contributed to the design of fuzzy
system structure. In this sense, the decision rules of simulation model are related to the
knowledge base on the designed fuzzy logic system.
Fuzzy
Inference
System
System
Dynamics
Decision |
Rules
Make
Inference Decision
Engine
Figure 4. Relationship Integral and Dynamic Fuzzy Model
In the Figure 5 presents the methodology of integration between System Dynamics
and Fuzzy Inference System. The relationship of integration in the simulation with fuzzy
parameters where the simulated model is combined in order to analyze the behavior and
decisions make.
32st International Conference of System Dynamics Society
July 20-24, 2014
Delft, Netherlands
‘System Dynamics Fuzzy Inference System Integral and Dynamics Fuzzy
Assessment Problem
and Hypothesis
Define Linguistic Tags }
Define Universe of
iscourse
and Data Collection
Effects of causal
relationship
Causal Diagram
Construction
Select Study
Variables
Generate Base
Rules
Implication and
Aggregation
Perform analysis of
fuzzy relations
Defuzzification
imulation with fuzzy
parameters
Analysis with
Cognitive Maps
Simulation Sectors
Map
‘Mathematical
Modeling and
oi 1 Anal
Simulation Model
Figure 5. Methodology Integration - Integral and Dynamic Fuzzy Model
The design of the system behavior measures is presented in the following separate
to analyze the traceability technology implementation according to flow information,
materials and investment used in fruit supply chain.
Design System Performance Measures
The system performance measures were chosen based on simulation developed sectors.
These measures are presented in Figure 6, it focused on information management and
traceability system, due to the importance that this sector represents the dynamic
hypothesis. The mathematical model of system developed is presented:
GFI (%)
Management >) DT (%)
Information | CC (Kg/batch)
an e
Traceability CSTT Ce)
GTS (%)
PT (Und Teen)
MU (Kg)
TCI (S/week
Figure 6. System Performance Measures.
Structure Model of System Dynamics
Assessment Problem
The growth organization requires a change in the procedures and processes at the impact of
food safety and quality of fruit products. The expansion of markets and technological needs
associated with product specifications requires investments that provide for the inclusion of
technology. Therefore, the question asked which is addressed with the model is:
How should make the implementing traceability technology RFID in picking
operations of the fruit supply chain?
Starting from this assessment problem is proposed the dynamics hypothesis of
Integral and Dynamics Fuzzy Model.
Hypothesis
By developing a model that integrates system dynamics and fuzzy logic may determine
management policies for implementing traceability technology in order to improve the
setup time and traceability in fruit supply chain.
Causal Diagram Model
The causal diagram model for traceability technology implementation in fruit supply chain,
proposed from the perspective of System Dynamics, addresses the conceptual elements of
industry experts and several approaches and structures. Therefore, the causal diagram in the
traceability technology implementation in fruit supply chain shown in Figure 7. It can be
seen that the causal diagram proposes five main effects: Implementing Traceability
Technology, Inventories of raw materials and quality, Flow production, storage and
demand, human resources, Production capacity (infrastructure).
32st International Conference of System Dynamics Society
July 20-24, 2014
Delft, Netherlands
Picking Learning
Ability.
Production
Orders
+ ‘
Pinel ity
fone
Capacit
Fruit G)
Processing
Demand Stock In Knowledge
ae St 5 hh Capacity
Sales
+ (= ed Infrastructure
a Matetal (storage)
+
Customer L (asf
+ An) 4.
: dex
Implementation 3
Technology B3
Cay
+ . Iny
@
contamination risk
.
Co
speciation
Bee “iceas
Figure 7. Causal Diagram - Integral and Dynamics Model.
In summary, the implementation traceability technology indicators improved
traceability and product specifications, however this behavior is controlled by the system's
capacity to invest in technology.
Simulation Sectors Map
The model sectors are designed taking into account flows: Information, Material, Capital,
Money and Human Resources (Forrester, 1958). The structure of sectors map having
different feedbacks and flows. The flows are identified by structure (See Figure 8):
a) Feedback Flow Supply Chain: includes the sectors of Raw Material Supply,
Processing, and Consumer.
b) Feedback flow Traceability and Information: includes relations sectors of
Information Management and Tracking System.
c) Feedback Flow Management Technology: provides relations sectors of Planning,
Acquisition and Incorporation of Technology, Investment, Technology and Human
Resources
Squaring the sunny circle? On balancing incentives for solar
prosumers and cost causation
Author: Merla Kubli *?*
“Institute of Sustainable Development, Zurich University of Applied Sciences, Switzerland
“Institute for Economy and the Environment, University of St. Gallen, Switzerland
The manuscript is currently in publication process and cannot be published in full length. For a personal
copy please contact the author.
Abstract
Solar prosumers are about to revolutionize the power sector. Utilities are challenged in recovering the
costs of distribution grids, as parts of their revenue basis decreases through self-consumption. Adjusting
the grid tariff sets off a reinforcing feedback loop that increases the attractiveness of solar in
but also leads to a distribution effect between solar prosumers and conventional consumers. The
question is: How to recover distribution grid costs equitable without hampering the diffusion of solar
power? Can the two criteria be fulfilled at the same time, or is do we aim for squaring a circle? To
d to understand the
address this question, I present a System Dynamics simulation model de
interactions and assess these competing goals. The occurring distribution effect under the volumetric
grid tariff appears to be rather limited. Simulation experiments reveal that grid tariff designs strongly
influence investments for solar power. A capacity tariff can reduce deviations from the cost causation
li: I to reduce
principle of solar prosumers and incentivizes in in dec / storage
peak demand. Nevertheless, also the capacity tariff causes a distribution effect.
*Author contact: Merla Kubli, Institute of Sustainable Development, Zurich University of Applied Sciences,
Technoparkstrasse 2, 8401 Winterthur, Switzerland, merla.kubli@zhaw.ch, +41 58 934 72 59.
Supply
Chain
Management
Technology ~-~~__ ; { }
Traceability
Figure 8. Structure Sector Map of Integral and Dynamics Model.
The intercept Technology Management shows the relationship this has with the
supply chain and information (traceability and information management) concept that adds
value to the structure simulation applied in this paper. Consequently, the Technology
Management as the cornerstone of the model relates to the interface and integrates
simulation, with the purpose of analyzing the behavior of the technology across different
aspects (planning, acquisition and incorporation, investment, diagnosis and human
resources).
Systematization of Simulation Model
The continuous simulation model was developed in three stages of modeling: Design
interface (simulation sectors), Design Simulation Model and Control System Design. The
first stage was analyzed in the previous section, in which the structure of model and
simulation sectors is presented.
The second stage of modeling and build is the Simulation Forrester Diagram as
shown in Figure 9. This is divided into three areas comprising the flow information,
materials, human resources, capital (technology) and money. The area that relates the flow
of material and information is called supply chain. This simulation includes four sectors
(supply of raw materials, processing, shipping and consumer).
Traceability area show in the Figure 9 relates the flow area information and material
supply chain. Also, this simulation covers two sectors (Information Management and
Tracking System). Finally, the Technology Management that includes the key areas for the
implementation and analysis technology in the system.
32st International Conference of System Dynamics Society
July 20-24, 2014
Delft, Netherlands
Traceability
Management
Technology
Supply
Chain
Figure 9. Forrester Diagram Simulation Model Fruit Supply Chain - Technol M Ti
The results of performance include implementing policies Traceability Technology,
measures system performance and behavior of the graphs of information flows, human
resources, equipment, technology and money.
Structure of Fuzzy Inference System
A Fuzzy Inference System was the ability to make a specific input value and through a
fuzzy inference process based on rules, throw a crisp output value. The purpose of
designing the Fuzzy Inference System model for implementing traceability technology is
based on two aspects: analysis of the causal diagram structure and design values to improve
policy technology management.
The type of Fuzzy Inference System is used Mandani. This system includes the
following steps: fuzzification, application of fuzzy operators, implication, aggregation and
defuzzification. Each applies to diagram causal variables and parameters associated with
picking operations and quality (traceability). Next, the structure analysis of simulation
model is shown.
Structure Analysis of Simulation Model
Design policies
The design policies of sector planning traceability technology are based on two algorithms
that determine the area (sector) intervention technology and alternative technology. The
first comprises the information flow, material flow and investment capacity. The algorithm
in Figure 10 determines the area to intervene.
j: Sector implementation of supply chain (supply, processing, shipping).
k: Subscript represents a type level k.
CFIjk: Level Information Flow capacity of sector j.
CFMjk: Level of Material Flow Capacity of sector j.
VPNijk: Net Present Value of Investment sector j.
EFI jk: Input Flow Information sector j.
SFIjk: Output Flow Information sector j.
EFMjk: Input Flow Material sector j.
SFMjk: Output Flow Material sector j.
PFIj: Policy Flow Information sector j.
PFMj: Policy Flow Information sector j.
PFInvj: Policy Investment sector j.
Material
IN (CFijkit) } and
IN (CFMIk(t) )
ee
dt
Cr =a Roe
P1: Determination of
technologically
IF PFiny| = PFI) THEN
PLeMIN(CFk(t)) ELSE IF
PFinvj=PFM] THEN PL=MIN
(CFMjk(t)) ELSEO
Max
PFIn
(VPNiK(t))
VPNikit) = VPNik(t- dt) + (FVPNjRl
ilk) * dt
Figure 10. Algorithm to determine the policy to intervene technologically area.
A second algorithm is designed to determine the technological alternative in
accordance with the levels of technology that presents the company (old technology, new
technology and technology development). The algorithm is presented in Figure 11.
i: Alternative technology (acquisition, development, adaptation).
k: Subscript represents a level type k.
TIR (i): Internal Rate of Returns type of alternative i.
CP (i): Cost of alternative type i.
Nik (t): State (level) of alternative type i.
NPVi: Net Present Value of type alternative i.
Calculate Internal Rate of Return
type i [TIR {i)]
P2: Determination of
Calculate state (level) of
type policy i
Technology
Calculate Cost of apparent policy type
iter)
Nik(t) = Nikit - dt) +
(F(i-2)ik - F(i#a)Kt) * dt
Figure 11. Algorithm to determine the policy alternative technologically.
32st International Conference of System Dynamics Society
July 20-24, 2014
Delft, Netherlands
The mathematical model supports the structure of each sector simulation, whereby a
dimensional analysis of the mathematical model presented is conducted.
Simulation Model Validation
In order to verify that the designed simulation model presents a representative of the real
system behavior are performed statistical tests for comparison of means and correlation of
the actual system demand and the demand simulated by the model. As well, was performed
Test error rate on the average percentage error and the variation proposed by Barlas (1989).
At this stage were defined the parameters of average quality (real) generated in the
pulp production (e.g. blackberry), with which the simulation is set in a period of 48 weeks
(1 year). From this, the following hypothesis is proposed for validation through a statistical
test for comparison of means between the actual and simulated variable-demand system:
Ho: Hr = Ms
Ai Ur # Ms
Ho: The average actual demand is the same simulated system demand.
H,: The average actual demand differs from the simulated system demand.
According to the results in Table 1 was decided not to reject the null hypothesis, and
it is concluded that a 0.05 significance level data can not determine that there are
differences between kilograms of real system with respect to the simulated.
Table 1. Comparison of mean demand for pulp (real vs. Simulated).
ANOVA
IDPTk
Sum of
Squares Df Mean Square EF Sig.
Between Groups 7405639 35 211,590 1,609 191
Within Groups 1578170 12 131,514
Total 8983809 47
Fopssas =2.041>1,609 Do not Reject Null hypothesis
In this way the statistical test of correlation in demand for pulp in the Table 2 to
measure the relationship between real and simulated data shows a positive correlation
(0.306) and a suitable value of significance (p-value = 0, 05> 0.034).
Table 2. Correlation of demand for pulp (real vs. Simulated).
Correlations
DemMora IDPTk4
DemMora Pearson Correlation 1 3060)
Sig. (2-tailed) 034
N 48 48
IDPTk Pearson Correlation 3060) 1
Sig. (2-tailed) 034
N 48 48
* Correlation is significant at the 0.05 level (2-tailed).
The next section presents the model of the Fuzzy Inference System applied in the
parameters representing the delay (setup time-picking and time-tracking data) in the
system.
Integration System Dynamics - Fuzzy Inference System
The integration System Dynamics and Fuzzy Inference System is made through the
variables forming the causal diagram. These have an effect on the parameter study (setup
time in the areas of supply chain fruit) with some uncertainty (positive, negative or medium
effect). In the case of Figure 12 setup time picking operations in the sector of raw material
supply is affected by the variables: products, amount of raw material required and number
supplier.
Rule Viewer: DAMP | (=3 soo
File Edit View Options
Demanda - 163, Cantidad_do_M.P.— 342 Tene 2S
| a __| | a_i _
> Ez a Ee —| = a = ]
3 | =] | =~] aal| L |
+ | | x ]
= | ail = ]
— = ms [ =|
7 fe —
° —— all aa
AP = SF ee a
os 1
[Pee neon en [rarer | [Reve ten [sare [eown] ue |]
[ Spened system Dame, 5 rues l| Help | cose | |
Figure 12. Fuzzy Inference System-Time picking supply sector
After obtaining the values of different effects (negative, positive and means) takes
the lower value of setup time (DAMP, DAPP and DAPD) and simulations with these
values are performed in order to obtain the behavior in the variables used in decision-
making. Figure 13 shows the behavior of the information flow policy of RFID technology
with the parameters obtained from the Fuzzy Inference System, which represents a better
stable performance compared with barcode technology is presented in Figure 14.
32st International Conference of System Dynamics Society
July 20-24, 2014
Delft, Netherlands
@ © crIDe 2: CFISk 3: CFITk
1004 .
F |
er
: 2
7% "0.00 3.60 6.00 0 12.
moe Weeks 003 via, 17 deena cn oat
Q3seF ? Comportamiento Flujo de Informacién-Pofiticas
Figure 13. Flow Behavior Information Technology RFID with fuzzy parameters.
@ + crink FISK 2: OFITk
4 Rue crensrnorcrsle ccoysoreureerr grr
3 50
4, ™
‘
ca : |
0 ato abs 720
Weaks O41 vie, 17 de ane du 20
Fito de
Figure 14. Flow Behavior Information Technology Barcode with fuzzy parameters.
The results of the integrated System Dynamics and Fuzzy Logic model are
presented.
Results Integrated Model-DS and Fuzzy Inference System
The results of Integral and Dynamics Fuzzy Model address two aspects: traceability
(quality) and picking operation (logistic). First sensitivity analysis taking into account the
technology management policies, later measures performance to the simulation model, and
finally the integration of the fuzzy logic methodology and System Dynamics.
Sensitivity Analysis for the lead-time parameter
Sensitivity Analysis of the simulation setup time variable used in each sector is selected, as
shown in Table 3, it presents a comparison between the RFID tracking system and
barcodes. Addressed the setup time (DAMP, DAPP and DAPD) in three states show better
efficiency in the flow capacity of RFID traceability systems with shorter enlistment in the
areas of simulation supply chain.
vs. capacity planning technology.
Table 3. Sensitivity Analysis -time pickin,
, Flow Capacity Flow Capacity
cae ereccnas by Information-CFI Material- CFM
SECIOES (Batch) (Kg/Und. T i
State DAMP DAPP. DAPD RFID/EPC Barcodes RFID/EPC Barcodes
1 05s 05 07 86,62 78,25 1&. 15
2 0,75 0,75 0,725 0 0 145 0
3 1 I 0.75 50,69 3431 14 14
In the case of Figure 15, the graph represents the behavior of stabilization system
for tracking technology employs barcode (simulation: 1, 2, 3), and RFID / EPC technology
(simulation: 4, 5, 6). After several runs are identified that lower data transmission time in
the fields of simulation, the technology has a better behavior is stable RFID / EPC
compared barcode contrast improves when the transmission time data is high.
@ cout +
S59
24.00 36.00 48.00
Page 12 Weeks 9:00 mar, 24 de dic de 2012
Neer ?
Figure 15. Sensitivity analysis - time data transmission vs. Flow Capacity Material in Supply Chain
Table 4 summarizes the changes in the simulated investment time parameter, the
highest Net Present Value (NPV) of the investment is made for the two technologies in the
shipping sector; industry that matches the policy for determining the area in which to
invest. These values reflect the phenomenon having the system stability after the third
month.
Table 4. Sensitivity analysis-time i by sector vs. technol planning
Net Present Value Net Present Value Net Present Value
Time (Delay) Investment- Investment- Investment-
Investment by sector Shipping Sector Supply Sector Manufacturing Sector
(VPNDk) (VPNSk) (VPNTk)
Sta | Tis | TID | Ter | RFID/EPC | Barcodes | RFID/EPC | Barcodes | RFID/EPC | Barcodes
1 1 1 1 2563,71 2550,98 592,99 592,35 2174,7 2174,7
2 2 2 5 2675,39 2674,83 594,69 594,27 2174,7 2174,7
3 3 3 9 2675,85 2674,89 594,84 594,48 21747 2174.7
Finally, sensitivity analysis is performed on the parameters associated with the
selection policy of technological alternatives in the supply chain. The alternative
technology of less NPV associated with traceability RFID this indicates that alternative
32st International Conference of System Dynamics Society
July 20-24, 2014
Delft, Netherlands
development is not suitable, however the alternative technology purchasing has a better
stability in the NPV, indicating that the alternative technology policy, as a result sheds
technology purchases in the state of simulation | (see Table 5).
Table 5. Sensitivity analysis-Internal Rate of Return for | alternative.
Internal Rate of Return for > ee br Altemative Buy-ANT Alternative SS
State | TIRAT TIRCT TIRDT RFID/EPC | Barcodes |RFID/EPC | Barcodes | RFID/EPC | Barcodes
1 0,09 0,008 0,08 7237,05 7033.88 | 12164603 | 122079,31 | _20074,42 20315,9
2 Ou 0,0387 0,103 5625,07 5499.75 25721,62 | 2599636 1693133 17098 62
3 0,13 0,0693 0,127 4612,73 4532.84 14575,1 1476587 1477547 14892,22
4 0,15 0,1 0,15 3920,82 3869.4 1035143 | 10492,19 13205,97 13287,13
Then the results of the performance measures of the simulated system are
presented.
Results Performance Measures-Simulated System
Performance measures were evaluated for traceability technologies RFID / EPC and
Barcode comparing implementation through performance indicators designed for each
sector of the fruit supply chain. The performance indicators traceability technology for
barcode shows an increase in the Information Flow Grade-GFI also affects the increase
Traceability Grade-GT. In this way, the indicator System Reliability Technology
Traceability CSTT presents a stable behavior over time of 74% (see Table 6).
Table 6. Sector Performance Indicators-Traceability Technology: barcode.
E a : Usin; Investment
Information Flow | paccrmance | Tcesbility Grade ene eens
Gutle-Gt Technology -DT sectors CT (Kg) ($US/week)
1,26 0,25 0,77 0,22 0,50
‘ing 1,02 0,32 071 0,29 0,22
1,60 049 0,83 02 0,22
System Reliability-CSTT 0.74
The behavior of the indicators of the RFID tracking technology shows an increase
in the Information Flow Grade-GFI also affects the increase Traceability Grade-GT,
indicating that the RFID tracking technology has a greater capacity in the information flow.
The growth delay GFI provides adaptive technology traceability therefore its growth is
slower. In this way, the indicator System Reliability Technology Traceability CSTT
presents a stable behavior over time of 89% is presented in Table 7.
Table 7. Sector Performance Indi Traceability Technology: RFID / EPC.
5 . a Usin; Investment
nee low | Performance | Traceability fe Margin Growth Rate
pede Gi ‘Technology -DT CON ACHE (Kg) ($US/week)
Supply 124 0.38 0.96 0.18 0.40
‘ing 1,04 0,46 0,85 0,12 0,18
Shipping 1.55 0.67, 0.93 0,17 0.18
System Reliability-CSTT 0.89
In conclusion, RFID tracking technology compared to barcodes presents a behavior
system reliability traceability technology (CSTT) increased, this implies a substantial
improvement in the flow of information, materials and capital, which is reflected indicators
for each sector. In the following apart are presented results of Implementing Technology
Model with a fuzzy logic system.
Results Fuzzy Logic System and System Dynamics
In this section is presented the results of the integration of System Dynamics and the Fuzzy
Inference System. The purpose of this integration is to combine the skill of the conceptual
model (causal diagram) and use fuzzy simulation parameters (knowledge base) to improve
the decision making process of a complex system with some degree of uncertainty in causal
variables. In the case of fruit supply chain objective focuses on improving times associated
enlistment picking operations (logistics) and the transmission times of the data (quality).
In this sense, the causal diagram variables (input variables) were used which have
an impact on the simulation parameters, setup time (picking), the output variable of the
Fuzzy Inference System (see Table 8). At the stage of defuzzification of the output values
taking time enrollment in picking operations (DAMP, DAPP and DAPD) expert qualitative
variable (fuzzy inference system) is associated.
Table 8. Fuzzy Logic System for simulation p 's under System Dynamics.
Input Output
Parameters Membership Parameters Value
Variable a b 1 Function a b [i i
Demand 163 | 214 | 265 05 [075 [1 0,582 Low
Raw Material 342 | 658 | 50 Triangular Triangular 075 | Mean
Supply 1 3 5 DAMP 0.75
Fruit Manufacturing 2 25 | 3 Triangular 05 [075 [| 1 0,582 Low
Production Order 13 [ 192 Gaussiana Triangular 0.75 | Mediu
Storage process 0 50_[ 100 Triangular pee 0.918 | High
Productive Capacity 0,064 | 0,12 Gaussiana
Sales 102 | 112 [| 122 Triangular 07 [0.725 [0,75 | 0.708 [Low
Customers in 20 [29 Triangular Triangular 0,725 | Mediu
Enlistment 13_[_ 128 Gaussiana DAPD 0,742 _| High
In line with the development of the integrated model output values of fuzzy
inference system is defusificar and the minimum of setup time (picking) in the system
dynamics simulation is replaced, which improved the capabilities of information flow and
material flow (see Table 9).
city (output) of the Fuzzy System and System Dynami
Table 9. Comparison of parameters (input) and cap
Seon RFID/EPC | Barcode | RFID/EPC | Barcode
(minimum)
Flow Capacity Flow Capacity
State DAMP] DAPP|DAPD| _ Information-CFI Material- CFM
(Batch) (Kg/Und. T ical)
Integral and Dynamies | 9 535 | 9.592 | 0,708 | 90.6 82 19 | 17
Fuzzy
32st International Conference of System Dynamics Society
July 20-24, 2014
Delft, Netherlands
[_ System Dynamies | 05 | 05 | 07 | 8662 | 78.25 | is [| is |
[Example Company [08 [08 [07 [ ND. [78 [ ND. | 1 |
N.D.: Data not available from the Example Company.
The real system (research company) has a setup time in the picking operation of
362.88 min / week on average (4.8 days / average * 75.6 min / day), compared with the
Integral and Dynamics Fuzzy Model with a time of enlistment of 264.6 min / week on
average (3.5 days / average * 75.6 min / day).
Conclusions
The relationship of the actors in the fruit supply chain is framed in different dynamics that
govern their behavior. In this sense, the market dynamics traceability technologies in
supply chain presents a growth that generates impacts on flows of material, information,
capital, human resources and money. Therefore, the implementation of traceability
technologies in the food supply chain requires models oriented comprehensive analysis of
relationships and flows between the actors of the chain.
An increased logistics activity requires improvements for analyzing information
using appropriate technologies to reduce the risk of contamination in the food supply chain.
In this way the model developed allows analysis of relations between actors and contribute
to the management of technology in the supply chain through logistics policies. The
proposed integration from the conceptual model (causal diagram), the system dynamics
simulation and applied knowledge base through a fuzzy inference system, allow you to
associate the expertise of unit production company and the actors studio in the fruit chain
(suppliers, processors and distributors), which allows to generate policies for implementing
traceability technologies in fruit supply chain.
The implementation model of RFID tracking technology developed in this paper
uses the methodology of System Dynamics and Fuzzy Inference Systems. The work
previously developed with these methodologies do not outline the concept of
comprehensiveness and integration from a conceptual model (causal diagram) so the
knowledge base of the expert is not addressed from the perspective of uncertainty of causal
variables related to the simulation model continuous. Whereas, the comprehensive and
dynamic model developed here combines elements of complexity-uncertainty, causality-
experience, knowledge-base-behavior and links the analysis of the causal structure by
means of a Fuzzy Inference System. In this way, it is considered that the comprehensive
and dynamic model is an innovative approach applied in the method used for the
integration of simulation methodologies and expert systems.
Delays (time) generated in logistics readiness activities (picking) and traceability
technology planning, influencing the performance of the fruit supply chain flows of
material, information, human resources, capital and investment. The inclusion of simulation
systems and fuzzy logic to analyze the uncertainty in the time associated with managing
technology and logistics activities.
The performance indicators tracking technology designed for RFID and barcode
shows an increase in the Grade Information Flow-GFI also affects the increase Traceability
Grade-GT. This indicates how similar response of the system as the flow of information
and materials. In this way, the indicator System Reliability Technology Traceability CSTT
(grade tracking and monitoring) has a stable behavior over time of 74% for the traceability
technology with barcode and 89% for RFID.
Referencias
Abo-Hamad, W., & Arisha, A. (2011). Simulation-Optimisation Methods in Supply Chain Applications: A.
review. Irish Journal of Management (116).
Amin, S., & Zhang, G. (2012). An integrated model for closed-loop supply chain configuration and supplier
selection: Multi-objective approach. Expert Systems with Applications , 8 (39), 6782-6791.
Asociacién Nacional de Empresarios de Colombia-ANDI. (2012). Encuesta Nacional Logistica: Resultados
Bechmarking Logistico . Bogota: ANDI.
Barlas. (1989). Multiple tests for validation of system dynamics type of simulation models. European
Journals of Operational Research (42), 59-87.
Campuzano-Bolarin, F., Mula, J., & Peidro, D. (2013). An extension to fuzzy estimations and system
dynamics for improving supply chains. International Journal of Production Research , 10 (51),
3156-3166.
Campuzano, F., Mula, J., & Peidro, D. (2010). Fuzzy estimations and system dynamics for improving
supply chains. Fuzzy Sets and Systems , 11 (161), 1530-1542.
Carvalho, J. P., & Tome, J. A. (2000). Rule Based Fuzzy Cognitive Maps - qualitative systems dynamics.
Annual Conference of the North American Fuzzy Information Processing Society - NAFIPS ,407-
411.
Chao-Ching, C., & Min-Ping, K. (2010). Ambidextroux effects of relational specific investments in the
OEM Transactions. The 28th International Conference of the System Dynamics Society .
Chen, K.-Y., Hwang, Y.-F., & Chen, M.-C. (2009). Applying RFID to Picking Operation in Warehouses in
Global Perspective for Competitive Enterprise . Economy and Ecology , Springer London, 531-540.
Chen, Y. (2011). Understanding technology adoption through system dynamics approach: A case study of
RFID technology. Proceedings - 2011 IFIP 9th International Conference on Embedded and
Ubiquitous Computing, EUC 2011 (art. no. 6104551), 366-371.
Dangelico, R., Garavelli, A., & Petruzzelli, A. (2010). A system dynamics model to analyze technology
districts' evolution in a knowledge-based perspective. Tech ion , 2 (30), 142-153.
De Marco, A., Cagliano, A., Nervo, M., & Rafele, C. (2012). Using System Dynamics to assess the impact
of RFID technology on retail operations. International Journal of Production Economics , I (135),
333-344.
Dharmaraj, N., Rodrigues, L., & Shrinivasa Rao, B. (2006). Technology management in innovative
organization: A system dynamics based perspective. JCMIT 2006 Proceedings - 2006 IEEE
International Conference on Management of Innovation and Technology , 2 (art. no. 4037182),
1059-1062.
Forrester, J. W. (1958). Industrial Dynamics: A Major Breakthrough for Decision Makers. Harvard
Business Review , 36 (4), 37-66.
Frazelle, E., & Sojo, R. (2007). Logistica de almacenamiento y manejo de materiales de clase mundial.
Bogota D.C.: Norma.
Ghazanfari, M., Jafari, M., & Alizadeh, S. (s.f.). An Approach to solve fuzzy system dynamic problem.
Tehran: Iran University of Science and Technology.
Giraldo, D., Betancur, M., & Arango, S. (2010). Efectos de la disponibilidad de alimentos sobre la
seguridad alimentaria. VII Congreso Latinoamericano de Dindmica de Sistemas .
Guzman, G., & Andrade, H. (2009). Una aproximaci6n al tratamiento de la cualitatividad en dindmica de
sistemas usando ldégica difusa. 7° Encuentro Latinoamericano de Dindmica de Sistemas.
32st International Conference of System Dynamics Society
July 20-24, 2014
Delft, Netherlands
Haleh, H., & Hamidi, A. (2011). A fuzzy MCDM model for allocating orders to suppliers in a supply chain
under uncertainty over a multi-period time horizon. Expert Systems with Applications , 8 (38),
9076-9083.
Herrera R., M., & Orjuela C., J. (2014). Perspective of Trazability in the Food Supply Chain: An approach
from System Dynamics. Revista Ingenierta Universidad Distrital, 19 (2).
Herrera R., M., & M., B. F. (2010). Perspectiva de gestidn de tecnologia en redes logisti
Colombiano de Dindmica de Sistemas.
Herrera, M., & Orjela, J. (2012). Evaluacién de Tecnologia de Trazabilidad en la cadena de suministro
fruticola: un enfoque bajo dindmica de sistemas. X Congreso Latinoamericano de Dindmica de
Sistemas , 1, 615-622.
Kalenatic, D. (2001). Modelo integral y dindmico para el andlisis, planeaci6n, programacion y control de
las capacidades productivas en empresas manufactureras. Bogota D.C.: Universidad Distrital
Francisco José de Caldas.
Kalenatic, D., & Gonzalez, L. (2009). El sistema de gestion tecnoldégica como parte del sistema logistico en
la era del conocimiento. . Cuadernos de administraci6n , 39 (22).
Karikoski, J., Heikkinen, M., & Haimmiinen, H. (2009). Scenario analysis and system dynamics in new
emerging technology research: Case mobile peer-to-peer content distribution. /st International
Conference on Advances in P2P Systems (art. no. 5359108), 60-65.
Kosko, B. (1986). Fuzzy Cognitive Maps. International Journal of Man-Machine Studies , 24, 65-75.
Li, F., Zhu, Y., & Wu, H. (2013). Modeling and optimization of traceability system for agriculture products
supply chain. Advanced Materials Research , 574-579.
Li, X., & Wang, X. (2010). A study of the function-based policy optimization in system dynamics model.
International Journal of Innovative Computing, Information and Control , 6 (6), 2847-2856.
Lian, Q., & Jia, S. (2012). Research on two-stage supply chain ordering strategy optimization based on
system dynamics. Advances in Information Sciences and Service Sciences , 4 (23), 16-24.
Luong, N., & Saeed, K. (1992). Food self-sufficiency in vietnam: a search for a viable solution. System
Dynamics Review .
Matos Chain, R., & Nascimento, N. (2011). Difusao da innovagao tecnolégica-andlise dindmica de um
modelo. IX Congreso Latinoamericano de Dindmica de Sistemas .
Mula, J., Campuzano-Bolarin, F., Diaz-Madrofiero, M., & Carpio, K. (2013). A system dynamics model for
the supply chain procurement transport problem: Comparing spreadsheets, fuzzy programming and
simulation approaches. International Journal of Production Research , 13 (51), 4087-4104.
Musango, J., Brent, A., Amigun, B., Pretorius, L., & Miiller, H. (2012). A system dynamics approach to
technology sustainability assessment: The case of biodiesel developments in South Africa.
Technovation , 11 (32 ), 639-651.
Mutingi, M., & Mbohw, C. (2012). Fuzzy system dynamics and optimization with application to manpower
systems. International Journal of Industrial Engineering Computations , 5 (3), 873-886.
Ng, T., Khirudeen, M., Halim, T., & Chia, S. (2009). System dynamics simulation and optimization with
fuzzy logic. IEEM 2009 - IEEE International Conference on Industrial Engineering and
Engineering Management (art. no. 5373149), 2114-2118.
Orjuela C., J. (2004). Modelo logistico Basado en Dindmica de Sistemas para la Cadena de Abastecimiento
de la Sabana de Bogota. Pereira: Universidad Tecnoldgica de Pereira.
Orjuela C., J., Calderon, E., & Buitrago, S. (2006). La cadena agroindustrial de frutas: Uchuva y tomate de
drbol. Bogota D.C.: Fondo de Publicaciones Universidad Distrital Francisco José de Caldas.
Orjuela C., J., Kalenatic, D., & Huertas, I. (2010). Modelo Integral y dindmico para la gestién de empresas
de servicios. Bogota D.C.: Universidad Catélica de Colombia.
Pankaj, S. K. (1994). A fuzzy set theoretic approach to qualitative analysis of causal loops in system
dynamics. European Journal of Operational Research , 3 (78), 380-393.
s. VII Encuentro
Peidro, D. (2010). A fuzzy linear programming based approach for tactical supply chain planning in an
uncertainty environment. European Journal of Operational Research , I (205), 65-80.
Posada, J., & Franco, C. (2010). Acercamiento desde el enfoque sistémico a la problematica de seguridad
alimentaria en la ciudad de Medellin: politicas para la superacién. VIII Congreso Latinoamericano
de Dindmica de Sistemas .
Puentes G., H. (2006). Caracterizacién de la logistica en Colombia. Bogota D.C.: Sena: Mesa sectorial de
Logistica.
Skoglund, T., & Dejmek, P. (2007). Fuzzy Traceability: A process simulation derived extension of the
traceability concept in continuos food processing. . Food and Bioproducts Processing , 85 (C4),
354-359.
Wolstenholme, E. (2003). The use of system dynamics as a tool for intermediate level technology
evaluation: Three case studies. Journal of Engineering and Technology Management - JET-M , 3
(20), 193-204.
Xu, J., & Li, X. (2011). Using system dynamics for simulation and optimization of one coal industry system
under fuzzy environment. Expert Systems with Applications , 38 (9), 11552-11559.
Yin, Q., & Xia, L. (2011). The knowledge transfer influence factors of industrial technology innovation
coalition based on system dynamics research. BMEI 2011 - Proceedings 2011 International
Conference on Business Management and Electronic Information , 2 (art. no. 5918005), 684-688.
Zadeh, L. (1965). Fuzzy Sets. Information and Control systems , 8, 338-353.
Zimmerman, H.-J. (1983). Using fuzzy sets in operational research. . European Jorunal of Operational
Research (13), 201-216.
Appendix A
GFID = DELAY (IDPTk/OCPTk,DAPD) @
GFI: Grade Information Flow (%)
of Finished Product Distribution (Kilograms).
inished Product Purchase Orders (Kilograms).
DAPD: Delay associated picking Dispatched Product (weeks).
DELAY: Specifies the delay equation of the first order.
DTMP = IMPk/TUMPk Q)
DT: Performance Technology (%)
IMPk: Raw Materials Inventory (Kilograms).
TUMPk: Traceability Technology Used in the field of Supply of Raw Material (Kilograms).
CMP = RDMPk/LMPk (3)
CMP: Strengthening Capacity of Raw Material Supply (Kgs / Batch) sector.
RDMPk: Data Record Raw Material Used (Kgs).
LMPk: Level Number of Lots Raw Material Used (Batch).
CSTT = (1-(-DTD)*(1-DTMP)*(1-DTT)) (4)
CSTT: Reliability Technology in Traceability System (%).
DTD: Performance Technology in the Distribution sector (%).
DTMP: Performance Technology for Raw Materials sector (%).
DIT: Performance Technology in the Transformation sector (%).
GTMP = OSMPk/RDMPk (5)
GTMP: Grade Raw Material Traceability (%)
OSMPK: Level order in Raw Materials Supply Process (Kilograms)
RDMPKk: Data Record Raw Material Used (Kgs).
32st International Conference of System Dynamics Society
July 20-24, 2014
Delft, Netherlands
MUTMP = TPMPk-TUMPk 6)
MUTMP: Margin Using Traceability Technology in the sector Raw Material (Kilograms).
TPMPk: Planned Traceability Technology in the Supply of Raw Material sector (Kilograms).
TUMPK: Traceability Technology Used in the Supply of Raw Material sector (Kilograms).
PNTP = (CNTPk*PNT) + (CTAK*PTA) )
PNTP: Productivity Planned New Technology (Technological Units).
CNTPk: New Capability Level (Purchased) Planned Technology (Technological Units).
PNT: Productivity New Technology, (%).
CTAk: Capability Level Old Technology (Technological Units).
PTA: Old Technology Productivity (%).
TCIS = DELAY3 ((CITSk/TUMPk)*GPMPKI, TIS) (8)
TCIS: Growth Rate of Investment in the supply sector (US Dollars | week).
CITSk: Capacity Supply Technology Investment ($).
TUMPK: Traceability Technology Used in the Supply of Raw Material sector (Kilograms).
GPMPKI: Degree of Traceability Technology Planning in the Raw Materials sector(Kg / week)
TIS: Time Delay for Investment in the Supply sector, (week).
DELAY3: Specifies the delay equation of the third order.
Appendix B
Tranaformacion 2s
FLOFAKI
FPFTIk
ROPTK
Figure 17. Forrester Diagram Sector-Supply Chain Transformation
LP
one
one!
‘Lae
rue
wm suis
) ven
one '
roar
Q on cst
er
pre rsupna ee
oar 4
wwen.
OWT ,
rsuProu es ;
veo
FoPtK
Figure 18. Forrester Diagram Traceability Sector- Traceability System
32st International Conference of System Dynamics Society
July 20-24, 2014
Delft, Netherlands
oo ee Ae
Planning
