SUPPLY CHAIN MODELING WITH UNCERTAINTY IN DEMAND AND SUPPLY USING FUZZY LOGIC

http://www.iaeme.com/IJARET/index.asp 333 editor@iaeme.com
International Journal of Advanced Research in Engineering and Technology (IJARET)
Volume 10, Issue 2, March- April 2019, pp. 333-340, Article ID: IJARET_10_02_032
Available online at http://www.iaeme.com/IJARET/issues.asp?JType=IJARET&VType=10&IType=02
ISSN Print: 0976-6480 and ISSN Online: 0976-6499
© IAEME Publication
SUPPLY CHAIN MODELING WITH
UNCERTAINTY IN DEMAND AND SUPPLY
USING FUZZY LOGIC
Chidurala Srinivas
Professor, Mechanical Engineering Department
Vaageswari College of Engineering
Karimnagar, Telangana State, INDIA
ABSTRACT
In this paper, we have analyzed a serial operational level Supply Chain
Performance like fill rate, Build –To - Stock etc., under an uncertainty environment
for the optimality using a stochastic customer order. It evaluates minimum cost in
supply chain, simultaneously optimize quantitative decision variables and illustrates
the significance of capacity variable. Numerical examples are presented to illustrate
the benefit of the proposed strategies and the effects of changes on the cost and
parameters are studied.
Notations used
Keywords: Supply Chain, Fill rate, Uncertainty, Fuzzy logic, Inventory
Cite this Article: Chidurala Srinivas, Supply Chain Modeling with Uncertainty in
Demand and Supply Using Fuzzy Logic, International Journal of Advanced Research
in Engineering and Technology, 10 (2), 2019, pp 333-340.
http://www.iaeme.com/IJARET/issues.asp?JType=IJARET&VType=10&IType=2
1. INTRODUCTION
Supply Chain (SC) is generally viewed as a network of facilities that perform procurement of
raw materials, processing them into end products, distribution and selling them to end
customers. A supply chain consists of all stages involved, directly or indirectly in fulfilling a
customer request which includes manufacturers, suppliers, logistic providers, warehouse,
retailers and finally the end customer. Each stage of supply chain performs different processes
and shares the information with other stages of SC. Inventories may exist in the network of
SC due to uncertainty in Business Environment. In the review on safety stock determination
pointed out that there are two kinds of uncertainties in inventory control: Demand side
uncertainty and Supply side uncertainty [1]. However most inventory models deal with the
 : Fuzzy membership function
Qc : Customer order quantity
Qw : Ware house inventory
BTS : Build – To – Stock
Pi : Product, i=1 to n
rmi : raw material inventory
Pri : Process i
epi : end-product inventory
drc : Random Customer Demand
dl : Low Customer Demand
dh : High Customer Demand
dm : Medium Customer Demand

Chidurala Srinivas
Demand side uncertainty and research on supply side uncertainty is relatively scant. In a
typical two-echelon SC the inventories have been considered with either warehouse or
supplier. Treating uncertainty is an important issue in supply chain modeling. Two inherent
sources of uncertainties in SC environment are identified as customer demand and supply of
raw material. If the warehouse has less than the fuzzy customer order quantity i.e., Qw< Qc
units in stock, first the partial order of size will be supplied from the stock, then the unfilled
customer ordered quantity (Qc-Qw) will be backordered from the warehouse. The supply of
raw material uncertainty is correlated accordingly. In this paper these two uncertainties are
represented by fuzzy logic based on fuzzy arithmetic rules and interpreted using fuzzy
modeling.
The performance of Supply Chain is evaluated and the competitive and cooperative
selections of stock level policies were determined. The customer fill rate at each stage in the
two-echelon Supply Chain with one supplier and Pi products for minimum total inventory cost
of the entire serial production Supply Chain is modeled. The performance measures can be
classified in to two categories Qualitative (customer satisfaction, product quality) and
Quantitative measures (inventory levels, resource utilization, fill rate, total cost). The fill rate
is the percentage of customer end product order quantity full filed by the ware house
immediately from the stock on hand. Customer service level is a function of performance
indices like the order fill rate, which is the fraction of customer demands that are met from the
stock at ware house. [2]
2. FUZZY LOGIC MODELING
Fuzzy set theory provides a means for representing uncertainties [3]. Past few years have
witnessed a rapid growth in the number and variety of applications of fuzzy logic. A fuzzy set
contain the elements that have varying degree of membership in the set is measure as a
function that attempts to describe vagueness and ambiguity [2]. This function maps elements
of a fuzzy set (Ã) to a real numbered value on the interval 0 to 1. If an elements in the
universe, say x is a member of fuzzy set Ã, then this mapping is given by Ã (x) ε0, 1. The
feature of fuzzy logic modeling is shown in Figure 1. Fuzzy inference is a method that
interprets the values in the input vector and, based on some set of rules, assigns values to the
output vector. Fuzzification is the process of making a crisp quantity fuzzy. They carry
considerable uncertainty. If the form of uncertainty happens to arise because of imprecision,
ambiguity, or vagueness, then the variable is probably fuzzy and can be represented by a
membership function. This is a intuitive type, which involves contextual and semantic
knowledge about an issue. The combination of fuzzy conditions is determined by the
minimum or average membership functions of the element of each proposition to the
concerned fuzzy set. When fuzzy conditions of all rules have been quantified, fuzzy
conclusions can be calculated. A key difference between crisp and fuzzy sets is that a crisp set
has a unique membership function, whereas a fuzzy set can have an infinite number of
membership functions to represent it. Fuzzy sets are always functions, which map a universe
of objects. The function that maps onto [0,1] is a fuzzy set. Defuzzification is the conversion
of a fuzzy quantity to a precise quantity, just as fuzzification is the conversion of a precise
quantity to a fuzzy quantity. The output of a fuzzy process can be the logical union of two or
more fuzzy membership functions defined on the universe of discourse of the output variable.
The defuzzification is used to convert the fuzzy set to a crisp value. Different methods are
used for defuzzification, among which the center of the gravity is the most prevalent and
physically appealing of all the defuzzification methods [1].

Supply Chain Modeling with Uncertainty in Demand and Supply Using Fuzzy Logic
Figure 1 Fuzzy Logic Modeling
Figure 2 Customer service for different uncertainties Figure 3 Raw material Inventory
A : Low degree uncertainty 68.2%
B : Medium degree uncertainty 95.45%
C : high degree uncertainty 99.73%
D : Reliable Supplier
E : Un-Reliable Supplier
F : Raw Material Inventory Range
3. CASE STUDY
In order to examine SC behavior in an uncertain environment, the authors have performed a
simulation study and the flow chart of the program is given in Figure 5. The customer demand
and external supplier reliability are specified by possibility distributions and transforms into
the corresponding probability distribution and the corresponding datum described by a
linguistic term which is based on a random number. The performance calculations have been
done at operational level of supply chain network with BTS (Build–To–Stock) system of
partial inventory at warehouse and BTO (Build-To-Order) system of partial back order[6]
Some work has been done at multi-echelon serial raw material inventory/distribution systems
to observe the performance variations of the system for different variables (Figure 4). It is
found that the most of the cases the customer uncertainty demand have found an inevitable
which causes to make some inventory levels at the ware house. The raw material requirement
(external supplier) for the random backorders (unfilled customer demand) will be placed by
the warehouse to the production facility which intern will be informed to the external supplier.
After each replenishment of the customer order, the warehouse updates the inventory status
Fuzzy Inference EngineMembership Function
A(x)0,1
Deffuzification
Fuzzification
Crisp Input
Crisp Output
Constraints

Chidurala Srinivas
every time5
. For the various iterations of analysis we have taken some of the input data (fuzzy
sets) from the work done by the Dobrila Petrovic[1,4,5].
Figure 4 Schematic view of a serial supply chain
Figure 5 Case study modeling

3.1. Iterations analysis
Six test case iterations have been conducted with the customer uncertainty
Iteration 1. High degree of customer demand uncertainty and reliable raw material supplier
Iteration 2. Medium degree of customer demand uncertainty and reliable raw material supplier
Iteration 3. Low degree of customer demand uncertainty and reliable raw material supplier
Iteration 4. High degree of customer demand uncertainty and Un-reliable raw material supplier.
Iteration 5. Medium degree of customer demand uncertainty and Un-reliable raw material supplier.
Iteration 6. Low degree of customer demand uncertainty and Un-reliable raw material supplier.
High degree customer uncertainty: The customer demand in each time can have 13
possible integer values in the range from 4 to 16 and it is modeled by a discretized time
stationary fuzzy set.
D = {0.25/4, 0.25/5, 0.25/6,0.25/7,0.5/8, 0.75/9, 1/10, 0.75/11, 0.5/12, 0.25/13, 0.25/14,
0.25/15, 0.25/16}
Medium degree of uncertainty: Customer demand in this can have 7 possible integer
values in the range from 7 to 13 and it is modeled by a discretized time stationary fuzzy set.
D = {0.25/7, 0.5/8, 0.75/9, 1/10, 0.75/11, 0.5/12, 0.25/13}
Low degree of uncertainty: Customer demand in each time can have 3 possible integer
values in the range from 9 to 11 and it is modeled by a discretized time stationary fuzzy set.
D= {0.75/9,1/10, 0.75/11}
The random customer demand uncertainty drc will be in the range of drc >dl, drc < du (dl
=4, du = 16) Supply of raw material delivery along the Supply Chain may be unreliable in the sense
that part of the quantity ordered may only be received. The reasons for that might be of different
nature, in all the means the supplier reliability can be divided using imprecise linguistic terms, such as
reliable or unreliable raw material supplier which can be determined by discretized time stationary
fuzzy set (Figure 3).
R= {0.25/8, 0.5/10, 0.75/12, 1/14}
For reliable fuzzy raw material can be determined by discretized time stationary fuzzy set,
R= {0.25/20, 0.5/18, 0.75/16, 1.0/14}
Type of uncertainty n 
High degree of uncertainty 0.48 0.1
Moderate degree of uncertainty 1.9072 0.09
Low degree of uncertainty 3.7 0.08
Number of raw material supplied = average fuzzy inventory (x) + n*
Fill Rate : Inventory at warehouse/Demand
Build-To-Stock : Inventory at Warehouse + Raw Material Inventory – Demand
Total Cost : Raw Material Cost + Manufacturing Cost + Inventory Carrying Cost

Chidurala Srinivas
Figure 6 Raw Materials used for each Production Iteration
Figure 7 Inventories at warehouse for each product after the end of each Iteration
Figure 8 Total Cost in each Iteration

Figure 9 Standard Deviation Graph for Input (Production Required) Vs Output (BTS and Fill Rate)
Figure 10 Input variables (Production Required) Vs Output variables (BTS and Fill Rate)
4. CONCLUSIONS AND FUTURE WORK
It has shown the uncertain customer demand and supply delivery along the SC have great
impact on SC behavior. The average production requirement (all iterations) is in the range of
9-10(Input variable) and the out put variables like Fill-Rate and Built-To-Stock are in the
range of 10-16(appr.) and 0.6-1.0(appr.) respectively. The output results shows that the fill
rate is sensitive to customer uncertainty demand and supplier reliability (Figure 10). Iteration
No. 2 gives less total cost (average of all the products) due to its nature of characteristics
(medium degree of customer demand uncertainty and reliable raw material supply) where as
for unreliable raw material supplier the total cost increases with decrease in uncertainty
(Figure 8) and it is found that the Standard deviation for the entire Iteration No. 2 the Build-
To-Stock & Fill Rate is very less and more over it is towards downward trend of BTS (Figure
9). With random input variables lowest end-product stock levels reached before
replenishments during the simulated time horizon in iteration 1 for cycles 1, 2, 3, 6 are -5, -3,
-2, -4 and for iteration 4 for cycles1, 4 are -5, -5 respectively. The negative stock levels denote
the existence of shortage i.e., in all these cases, the inventory levels before replenishments are

Chidurala Srinivas
Zero/Negative, indicating insufficient order-up-to levels. For other stocks it is observed that
the stock levels amplified along the chain, starting from the raw material inventory towards
the end product inventory. The raw material being used in all iteration cycle varies between
82-124 (97 avg.) Where as there is always some items existing most of the times at
warehouse for each product after the end of each iteration and varies from 0 to 19 (7 avg.)
with the customer average order range of 9-10(avg.). There is lot of scope for improving with
object library and can be added wider spectrum classes as generic reusable classes for the
backorders.
REFERENCES
[1] Dobrila Petrovic, Rajat Roy, Radivoj Petrovic (1998). “Modeling and Simulation of a
Supply
[2] Chain in an Uncertain Environment” European Journal of Operations Research, vol.109,
pp.299-309.
[3] Timothy J.Ross (1997). “Fuzzy Logic with Engineering Application” 1st
ed,. McGraw-Hill
International Editions, Singapore.
[4] Zadeh, L. A. (1973), Outline of a new approach to the analysis of complex systems and
decision processes, IEEE Trans. SMC SMC-3, 98.
[5] Dobrila Petrovic, Rajat Roy, Radivoj Petrovic, (1999). “Supply Chain modeling using
Fuzzy sets”, International Journal of Production Economics, vol.59 pp.443-453.
[6] Dobrila Petrovic, (2001).“Simulation of Supply Chain behavior and Performance in an
Uncertain environment”, International Journal of Production Economics, vol.71 pp .429-
438.
[7] Jeremy F.Shapiro, (2002). “Modeling the Supply Chain”, First Reprint, Thomson Asia
pte. Ltd., Singapore.

SUPPLY CHAIN MODELING WITH UNCERTAINTY IN DEMAND AND SUPPLY USING FUZZY LOGIC

More Related Content

What's hot (11)

Similar to SUPPLY CHAIN MODELING WITH UNCERTAINTY IN DEMAND AND SUPPLY USING FUZZY LOGIC (20)

More from IAEME Publication (20)

Recently uploaded (20)

SUPPLY CHAIN MODELING WITH UNCERTAINTY IN DEMAND AND SUPPLY USING FUZZY LOGIC