Vous êtes sur la page 1sur 62

AN INTEGRATED INVENTORY RANKING PROCEDURE USING FORECASTING

& SUPPLIER SELECTION

By
VENKATA RAVI SHANKAR.B
(Reg. No. 412207034023)

Under the esteemed Guidance of


Dr. V.V.S.KESAVA RAO
Professor in Mechanical Engineering

Department Of Mechanical Engineering


Andhra University College Of Engineering (A)
Andhra University, Visakhapatnam
Andhra Pradesh, India
2016
ABSTRACT
This project thesis addresses a vital pre-release decision that
directly affects the operational effectiveness in an organization
that follows supply chain, inventory management solutions.
In my work, firstly ABC, FSN & VDE inventory
classification methods are used for ranking the inventories taking
both qualitative & quantitative factors.
In the second phase the Croston forecasting method is applied
to forecast quantity to be procured.
In the last phase, work is followed by usage of Multi choice
goal programming techniques to select the best suppliers therein
capable to supply the desired product and the desired quantity
fulfilling the desired objectives.
INTRODUCTION
IMPORTANCE OF INVENTORY
 Inventory is vital to an organization from financial and operational
standpoint.
 Inventory control as an integrated approach is thus essential for
determining the time, item(s) and quantity to indent, and amount of
stock, so that purchasing and storing costs become minimum without
affecting production, distribution, functional effectiveness, etc.
 The management of inventory and how it can provide insight into the
firm’s performance is a topic of interest to shareholders, investors,
business owners, and the general public..
 The main purpose of the inventory management practices in all
production companies is to have the required items ready to be
processed right on the required time with incurring minimum cost.
INTRODUCTION
 Spare parts are special independent demand items deriving their
demand from the failure characteristics of the component, and with a
specialized supply situation. Adequate spare parts are held in stock for
continuous, safe and reliable maintenance activities.
 A spare part is an item of inventory used to replace failed parts. A
spare part, spare, service part, repair part, or replacement part, is an
interchangeable part that is kept in an inventory and used for the
repair or replacement of failed units.
 No plant can operate at a high level of output without a reliable
supply of functional spare parts.
 Capital spares are spare parts although acknowledged to have a long
life or a small chance of failure would cause shutdown of equipment
for a prolonged period because of the long delivery of their
replacement.
INTRODUCTION
Inventory Classification Methods

 The proposed integrated inventory ranking procedure uses three


classification methods: ABC, FSN, and VED.
 The ABC and FSN methods consider quantitative factors such
as cost and demand, and the VED focuses on qualitative factors
such as functionality, efficiency or quality.
 All the qualitative factors have cost implications. When the cost
is difficult to estimate, VED allows to subjectively assign a
class based on past experience.
INTRODUCTION
ABC Classification
The ABC classification ranks the parts based on their usage value in a
given period.
The high value parts (about 20%) are classified as A, the middle value
parts (about 30%) are classified as B, and the lower value parts are
classified as C.
The procedure used in our study to perform the classification entails the
following steps:
1. Compute quantity and dollar usage of each item from inventory data.
Dollar usage of each item = Quantity * Unit Cost
2. Arrange parts in descending order based upon the dollar usage values.
3. Compute % of dollar value for each item.
% dollar value = (dollar usage value / total dollar usage value) * 100
INTRODUCTION

4. Compute % quantity for each item.


% quantity = (Quantity / Total quantity) * 100
5. Compute the cumulative percentages.
6. Classify the top 20% of the parts as A, the next 30% of the parts
as B and the remaining parts as C.

The cutoffs for ABC in the procedure above are used as a general
rule and the management may differ in the way they classify the
inventory. So, the cutoff % for classification should be based on
user input.
INTRODUCTION

For example, in Table 1 part#1 (6.45%) is classified as A, part#2


(32%)classified as B and part#3 (64.52%) as C.

Sl. no. Rate Qty. Value % value Cum % qty. % cumqty Class
value %
1 10 200000 2000000 71.05 71.05 6.45 6.45 a
2 40 10000 400000 14.21 85.26 25.81 32.26 b
3 50 7000 350000 12.43 97.69 32.26 64.52 b
4 20 1500 30000 1.07 98.76 12.92 77.42 c
5 55 100 35000 1.24 100 22.58 100 c
INTRODUCTION
FSN Classification
With FSN, the parts having the highest demand (top 25%) are classified
as fast moving and the parts having the least demand (bottom 25%) are
classified as non-moving. The middle 50% of the parts are classified as
slow moving. The FSN procedure used in our study toper form the
classification involves the following steps:

1. Obtain the total demand for each part in a given period.


2. Arrange the parts in ascending order based on total demand.
3. Compute the first and third quartiles (Q1 and Q3) from demand data.
4. Classify the parts using the following logic:
 If total demand > Q3 then Classify as Fast Moving
 If total demand < Q1 then Classify as Non-Moving
 Otherwise, Classify as Slow Moving
INTRODUCTION

An example of a FSN classification is shown in Table below.

Here we see that the fast moving parts are those parts that have a
total demand over 600, the slow moving parts are those parts
having a total demand in between 400 and 600, and the non-
moving parts are the ones with a total demand below 400.
Sl. No. Qty. Class
1 50 n
2 400 s
3 500 s
4 600 s
5 1700 f
INTRODUCTION
VED Classification
VED is based on the criticality of parts judged by the management
where parts are classified as Vital, Essential, or Desirable.
In general, an item is classified as Vital in any of the following
circumstances:
 Functionality: If the non-availability of the item shuts down the
process completely and there is no standby unit as a spare.
 Efficiency: If the non-availability of the item completely reduces the
efficiency of the manufacturing process.
 Quality: If the item is unique and/or the company involved is a world-
class manufacturer of the item.
INTRODUCTION
An item is classified as Essential in any of the following circumstances:
 Functionality: If the non-availability of the item shuts down the process but a
standby unit exists.
 Efficiency: If the non-availability of the item reduces the efficiency of the
process.
 Quality: If the quality of the item manufactured in-house is slightly better
than purchased item.

An item is classified as Desirable in any of the following circumstances:


 Functionality: If non-availability of the item does not affect the operation of
the manufacturing process
 Efficiency: If non-availability of the item does not significantly affect the
efficiency of the process.
 Quality: If the quality of the item manufactured in-house is no better than
purchased item.
INTRODUCTION
SDE Classification
SDE Analysis is based upon availability of items and is very useful in
the context od scarcity of supply.
 S refers to Scarce Items: Genrally, Imported and those which are in
short supply.
 D refers to Difficult Items: Which are available indigeniously but are
difficult to produce.
 E refers to Easy to acquire: Which are easily available in local
markets
Integrating Classification
Now that, we knew these all classification, this project further will try to
have an integrated classification approach.
INTRODUCTION
FORECASTING OF SPARE PARTS DEMAND
All machines require maintenance and spare parts. Spare parts must be
available whenever needed in order to prevent the whole production line from
being stopped and thus affecting the production rates.
Different quantitative and qualitative forecasting methods are applied to
predict the future demand of spare parts in order to consistently ensure their
availability.
Time series methods are considered the most common and reliable
quantitative methods. In time series forecasting, there is a major difference
between forecasting the demand of finished goods and forecasting the demand
of spare parts. Some traditional forecasting techniques might not be applicable
for spare parts.
Spare parts forecasting pose a large challenge for all forecasters. The reason
is that the demand is stochastic and a big proportion of the demand data is zero
for several periods of time resulting on having inaccurate results.
INTRODUCTION
FORECASTING OF SPARE PARTS DEMAND
Various methods include:
 Time series

 Moving average

 Exponential Smoothening

 Croston Forecasting

Performance measures that have been suggested in literature are:


Mean error(ME), Mean squared error (MSE) and Mean absolute deviation
(MAD).
INTRODUCTION
SUPPLIER SELECTION
Supplier selection is the process by which firms identify, evaluate, and
contract with suppliers. The supplier selection process deploys a tremendous
amount of a firm’s financial resources. In return, firms expect significant
benefits from contracting with suppliers offering high value.
Business decision making has changed over time from single criterion
(profit) to multi criteria decision. Multi criteria decision making helps the
decision maker to choose appropriate suppliers for each product by trading off
between conflicting quantitative and qualitative criteria.
Supplier selection problem criteria’s are namely, total purchase cost, quality
score, and delivery lateness. Since these criteria have different dimensions, they
are transformed to satisfaction levels which are dimensionless.
Goal programming is a technique which can be used for supplier selection.
INTRODUCTION
OBJECTIVE OF THE STUDY
 To develop an integrated inventory ranking procedure that takes
into account both quantitative factors such as cost and demand,
and qualitative factors such as functionality, efficiency, and
manufacturing expertise or quality;
 Application of forecasting technique to determine quantity to be
procured
 Supplier selection for the forecasted item
METHODOLOGY
The methodology is explained in three phases as stated below:
Phase 1: Develop an integrated inventory ranking procedure that takes
into account both quantitative factors such as cost and demand, and
qualitative factors such as functionality, efficiency, and manufacturing
expertise or quality, scarcity availability etc.

Phase 2: Application of forecasting technique to determine quantity to be


procured.

Phase 3: Supplier selection for the forecasted item.


METHODOLOGY
Phase 1
In ranking the importance of parts, several factors must be considered. When
four classification methods are integrated, it is possible that the management feels that the
levels of importance of the 4 methods are different.
The integrated mechanism that introduces lets the user (i.e., management) decide
what factors are of importance and in what order.
So, if qualitative factors need to have higher priority than quantitative factors,
higher weights need to be assigned to VED compared to ABC, SDE and FSN.
Factors considered in the classification methods are not mutually independent, the four
classification methods may assign different priority scores to the same part, resulting in
three different ranked lists.
For example,
If the part with the highest demand is the most expensive part, then both FSN and ABC
will assign the part with high priority.
If the cost to make the part is relatively low but demand is high, then this part may not
receive high priority based on ABC.
A part with low demand and low cost may also be classified as “Vital” under VED if the
quality of the part manufactured in-house is significantly better than purchased from sub-
contractor.
METHODOLOGY

Distinct priority levels for all classifications


For analysis, we consider the classification having the highest priority
level as having the weights 3x(3, 2, 1), the classification having the
second highest priority level have the weights 2x(3, 2, 1), the
classification that has the third and least priority level have the weights
1x(3, 2, 1).
In this example, VED classification is given the highest priority level
and therefore the weight assignment is as follows: V → 9, E → 6, D→3
ABC classification is given the third highest priority level and therefore
the weight assignment is as follows: A → 3 B → 2, C → 1
FSN classification is given the Second priority level and therefore the
weight assignment is as follows: F → 6, S → 4, and N→ 2
SDE classification is also given the highest priority level and therefore
the weight assignment is as follows: S→ 9, D → 6, E→3
METHODOLOGY
Distinct priority levels for all three classifications

A B C F S N V E D S D E RANKING
3 0 0 6 0 0 9 0 0 9 0 0 27
0 2 0 6 0 0 9 0 0 9 0 0 26
3 0 0 0 4 0 9 0 0 9 0 0 25
0 0 1 6 0 0 9 0 0 9 0 0 25
3 0 0 6 0 0 9 0 0 0 6 0 24
3 0 0 6 0 0 0 6 0 9 0 0 24
0 2 0 0 4 0 9 0 0 9 0 0 24
3 0 0 0 0 2 9 0 0 9 0 0 23
0 2 0 6 0 0 9 0 0 0 6 0 23
0 2 0 6 0 0 0 6 0 9 0 0 23
0 0 1 0 4 0 9 0 0 9 0 0 23

Contd…
METHODOLOGY
A B C F S N V E D S D E RANKING
3 0 0 0 4 0 0 6 0 9 0 0 22
0 2 0 0 0 2 9 0 0 9 0 0 22
0 0 1 6 0 0 9 0 0 0 6 0 22
0 0 1 6 0 0 0 6 0 9 0 0 22
3 0 0 6 0 0 9 0 0 0 0 3 21
3 0 0 6 0 0 0 6 0 0 6 0 21
3 0 0 6 0 0 0 0 3 9 0 0 21
0 2 0 0 4 0 9 0 0 0 6 0 21
0 2 0 0 4 0 0 6 0 9 0 0 21
0 0 1 0 0 2 9 0 0 9 0 0 21
3 0 0 0 0 2 9 0 0 0 6 0 20
3 0 0 0 0 2 0 6 0 9 0 0 20
0 2 0 6 0 0 9 0 0 0 0 3 20
0 2 0 6 0 0 0 6 0 0 6 0 20
0 2 0 6 0 0 0 0 3 9 0 0 20 Contd…
METHODOLOGY
A B C F S N V E D S D E RANKING
0 0 1 0 4 0 9 0 0 0 6 0 20
0 0 1 0 4 0 0 6 0 9 0 0 20
3 0 0 0 4 0 9 0 0 0 0 3 19
3 0 0 0 4 0 0 6 0 0 6 0 19
3 0 0 0 4 0 0 0 3 9 0 0 19
0 2 0 0 0 2 9 0 0 0 6 0 19
0 2 0 0 0 2 0 6 0 9 0 0 19
0 0 1 6 0 0 9 0 0 0 0 3 19
0 0 1 6 0 0 0 6 0 0 6 0 19
0 0 1 6 0 0 0 0 3 9 0 0 19
3 0 0 6 0 0 0 6 0 0 0 3 18
3 0 0 6 0 0 0 0 3 0 6 0 18
0 2 0 0 4 0 9 0 0 0 0 3 18
0 2 0 0 4 0 0 6 0 0 6 0 18
0 2 0 0 4 0 0 0 3 9 0 0 18
Contd…
METHODOLOGY
A B C F S N V E D S D E RANKING
0 0 1 0 0 2 9 0 0 0 6 0 18
0 0 1 0 0 2 0 6 0 9 0 0 18
3 0 0 0 0 2 9 0 0 0 0 3 17
3 0 0 0 0 2 0 6 0 0 6 0 17
3 0 0 0 0 2 0 0 3 9 0 0 17
0 2 0 6 0 0 0 6 0 0 0 3 17
0 2 0 6 0 0 0 0 3 0 6 0 17
0 0 1 0 4 0 9 0 0 0 0 3 17
0 0 1 0 4 0 0 6 0 0 6 0 17
0 0 1 0 4 0 0 0 3 9 0 0 17
3 0 0 0 4 0 0 6 0 0 0 3 16
3 0 0 0 4 0 0 0 3 0 6 0 16
0 2 0 0 0 2 9 0 0 0 0 3 16
0 2 0 0 0 2 0 6 0 0 6 0 16
0 2 0 0 0 2 0 0 3 9 0 0 16
Contd…
METHODOLOGY
A B C F S N V E D S D E RANKING
0 0 1 6 0 0 0 6 0 0 0 3 16
0 0 1 6 0 0 0 0 3 0 6 0 16
3 0 0 6 0 0 0 0 3 0 0 3 15
0 2 0 0 4 0 0 6 0 0 0 3 15
0 2 0 0 4 0 0 0 3 0 6 0 15
0 0 1 0 0 2 9 0 0 0 0 3 15
0 0 1 0 0 2 0 6 0 0 6 0 15
0 0 1 0 0 2 0 0 3 9 0 0 15
3 0 0 0 0 2 0 6 0 0 0 3 14
3 0 0 0 0 2 0 0 3 0 6 0 14
0 2 0 6 0 0 0 0 3 0 0 3 14
0 0 1 0 4 0 0 6 0 0 0 3 14
0 0 1 0 4 0 0 0 3 0 6 0 14
3 0 0 0 4 0 0 0 3 0 0 3 13
0 2 0 0 0 2 0 6 0 0 0 3 13
Contd…
METHODOLOGY
A B C F S N V E D S D E RANKING
0 2 0 0 0 2 0 0 3 0 6 0 13
0 0 1 6 0 0 0 0 3 0 0 3 13
0 2 0 0 4 0 0 0 3 0 0 3 12
0 0 1 0 0 2 0 6 0 0 0 3 12
0 0 1 0 0 2 0 0 3 0 6 0 12
3 0 0 0 0 2 0 0 3 0 0 3 11
0 0 1 0 4 0 0 0 3 0 0 3 11
0 2 0 0 0 2 0 0 3 0 0 3 10
0 0 1 0 0 2 0 0 3 0 0 3 9
METHODOLOGY
Phase 1

B. 1) All the inventory data consisting of closing stock, rate, movement details for the
last two years.
2) Perform classification.
a) VED, SDE Classification – Selection of the file and range led to the parts
information page, where information about VED Classification was entered for
each part. After entering information about the VED classification, the Perform
ABC and FSN Classification function was chosen. SDE means scarcely available,
difficult to available, easily available.
b) ABC Classification – The ABC classification was performed by taking the % of A
parts as 20% and % of B parts as 30% with the rest of the parts classified as C.
The result of this classification was as follows:
c) FSN Classification –The FSN classification was performed by taking the % of F
parts as 25% and % of S parts as 50% with the rest of the parts classified as N.
3) Assign Weights – Once all the classifications were performed, they were assigned
weights using the Assign Weights function. In this model priority weights for ABC is taken as
on1, FSN analysis 2, VED and SDE as taken as 3. ABC analysis is done and later FSN
analysis is done
METHODOLOGY
Phase 2

As the spares demand is not always uniform forecasting plays a vital role. Croston’s
forecasting method is reasonable good for slow moving, fast moving and vital, a, b class
items the method is suggested here.

The Croston’s method estimates the mean demand per period by applying exponential
smoothing separately to the intervals between nonzero demands and their sizes. The
notation used is defined as follows:

X(t): the observed demand in period t, t = 1,...,T.


I(t): the smoothed estimate of the mean interval between nonzero demands
S(t): the smoothed estimate of the mean size of a nonzero demand
z: the time interval since the last nonzero demand
α: smoothing constant between 0 and 1.
METHODOLOGY
The Croston’s method works as follows:

If X(t) = 0 then
S(t) = S(t - 1)
I(t) = I(t - 1)
z=z+1
or else
S(t) = αX(t) + (1 - α)S(t - 1)
I(t) = αz + (1 - α)I(t - 1)
z=1

Considering the demand size and intervals together, the estimate of the mean demand per
period can be calculated as

m(t) =s(t)/i(t)
METHODOLOGY
Phase 3
The various factors that are to be considered for supplier selection are:
min qty, price, lead time, price, max tonnage, defect percentage, discount
percentage, customer satisfaction ,branches, machines ,turnover
1. Analytic Hierarchy Process (AHP): The AHP is a mathematical multi criteria decision
making model to deal with the complex problems with quantitative and qualitative data
through constructing hierarchies.
The AHP method consists of three major steps as follows.
a) The first is problem description: the problem is resolved into elements, which are
grouped at different levels in a chain hierarchy, and each element is further
decomposed into sub criteria until the lowest level of the hierarchy is reached.
b) The second is comparative analysis: the relative importance of each element at a
particular level is measured by a pair wise comparison procedure. DMs must make a
series of pair wise comparisons on the basis of nine scales.
c) The third is the synthesis of priorities: the priorities of each element are synthesized to
establish the overall priorities for the decision alternatives, and the priority weights of
the elements at each level are computed using eigenvectors.
METHODOLOGY

The AHP method with pair wise comparison is used to measure the importance for every
factor, either tangible or intangible, and a matrix is generated as a result of pair wise
comparisons and criteria weights are reached as a result of these calculations.

Also, it is possible to determine the consistency ratio (CR) of decisions, and CR reveals the
random probability of values being obtained in a pair wise comparison matrix. If CR ≤ 0.1,
the degree of consistency is satisfactory; otherwise, there are serious inconsistencies and
the AHP may not yield meaningful results.

2. Multi choice Goal Programming (MCGP): GP is an analytical approach devised to


address decision-making problems where targets have been assigned to all the attributes
and where DMs are interested in minimizing the non-achievement of the corresponding
goals.

The MCGP approach, proposed by Chang allows DMs to set multi choice aspiration levels
for each goal (i.e., one goal mapping multiple aspirations
METHODOLOGY
Phase 1:
MODEL CALCULATIONS
The data used in this work is a sample data taken close to actual situation existing in
automotive industry.

(no of times)
in one years
Closing qty

Movement
Unit rate

VED
SDE
Code Description

75007 LUG, BACKING-7 5/8", 10 3/4", 13 3/8", 21" CASING


TONG 121 13.1 e de 5
75012 HOUSING, MUB-5206 BEARING-7 5/8", 10 3/4", &
169 34.4 e sc 6
14“ TONG
75014 STUD, CAM-14" TONG 898 8.35 v de 6
75015 ROLLER, CAM-14" TONG 316 3.46 v ea 5
75035 PIN, JAW HINGE-7 5/8, 10 ¾ & 14" TONG 313 7.67 e sc 6
75036 PIN, JAW ROLLER-4 1/2" & 5 1/2" - 10 3/4" TONG 8 10.6 d sc 2
75037 ROLLER, JAW-13 3/8" & 13 3/8" HT & 10 3/4" CT-10-
300 9.9 v de 8
M015
75065 ROLLER, CAM-14" TONG 908 23 v sc 8
Contd…
MODEL CALCULATIONS
75158 ROLLER, DOOR-7 5/8" TONG 14 35.1 e sc 2
75191 PIN, SLIP HINGE; BJ 175 TON OR B+V 250 160 10.8 e de 8
75195 PIN, YOKE-175, 350 & 500 TON B J TOOLS 122 25.2 d sc 8
75200 ROD, LOCK FOR BJ 350 & 175 56 65 e de 8
75206 PLUNGER, OVERLOAD FOR BJ TOOLS MISC. 572 1.37 d sc 8
75220 ROD, LOCK-500 TON BJ 35 110 e de 2
75318 STUD, CAM-36" TONG / 21" CASING TONG 238 8 v ea 7
75319 ROLLER, CAM-36" TONG / 21" CASING TONG 202 5.05 v de 9
75557 BODY, TONG DOOR INTERLOCK VALVE 258 118 v sc 6
75558 SPOOL, TONG DOOR INTERLOCK VALVE 269 14.3 e ea 6
75561 HOUSING, A-35 TONG UNLOADER VALVE MISC. 94 55.6 e ea 7
75562 POPPET, A-35 TONG UNLOADER VALVE MISC. 60 24 e de 5
75563 SEAT, A-35 TONG UNLOADER VALVE MISC. 19 4.42 e ea 2
MODEL CALCULATIONS
ABC analysis, FSN analysis, VED, SDE analysis is done. And composite ranking is done.
The results are furnished in table.

Closing qty.

Years (no.
Unit rate

of times)

Priority
Value

ABC

VED

scale
FSN
SDE
Code Description

LUG, BACKING-7 5/8", 10


75007 3/4", 13 3/8", 21" CASING 121 13.13223 1589 c e De 5 f 19
TONG
HOUSING, MUB-5206
75012 BEARING-7 5/8", 10 3/4", 169 34.42604 5818 a e Sc 6 f 24
& 14"TONG
75014 STUD, CAM-14" TONG 898 8.351893 7500 a v de 6 f 24
75015 ROLLER, CAM-14" TONG 316 3.458861 1093 c v ea 5 f 21
PIN, JAW HINGE-7 5/8, 10
75035 313 7.667732 2400 b e sc 6 f 23
¾ & 14" TONG
PIN, JAW ROLLER-4 1/2"
75036 8 10.625 85 c d sc 2 s 19
& 5 1/2" - 10 3/4" TONG
Contd…
MODEL CALCULATIONS
75037 ROLLER, JAW-13 3/8" & 300 9.9 2970 b v de 8 f 20
13 3/8" HT & 10 3/4" CT-
10-M015
75065 ROLLER, CAM-14" TONG 908 23 20884 a v sc 8 f 27
75158 ROLLER, DOOR-7 5/8" 14 35.0714 491 c e sc 2 s 20
TONG 3
75191 PIN, SLIP HINGE; BJ 175 160 10.775 1724 c e de 8 f 19
TON OR B+V 250 MT
75195 PIN, YOKE-175, 350 & 500 122 25.2131 3076 b d sc 8 f 20
TON B J TOOLS 1
75200 ROD, LOCK FOR BJ 350 56 65 3640 a e de 8 f 21
& 175
75206 PLUNGER, OVERLOAD 572 1.36888 783 c d sc 8 f 19
FOR BJ TOOLS MISC. 1
75220 ROD, LOCK-500 TON BJ 35 110 3850 a e de 2 s 19
75318 STUD, CAM-36" TONG / 238 8 1904 c v ea 7 f 19
21" CASING TONG
75319 ROLLER, CAM-36" 202 5.04950 1020 c v de 9 f 22
TONG / 21" CASING 5
TONG Contd…
MODEL CALCULATIONS
75557 BODY, TONG DOOR 258 118.182 30491 a v sc 6 f 27
INTERLOCK VALVE 2
75558 SPOOL, TONG DOOR 269 14.2676 3838 a e ea 6 f 18
INTERLOCK VALVE 6
75561 HOUSING, A-35 TONG 94 55.5531 5222 a e ea 7 f 18
UNLOADER VALVE 9
MISC.
75562 POPPET, A-35 TONG 60 24.0333 1442 c e de 5 f 19
UNLOADER VALVE 3
MISC.
75563 SEAT, A-35 TONG 19 4.42105 84 c e ea 2 s 14
UNLOADER VALVE 3
MISC.
MODEL CALCULATIONS
Phase 2:

As an example 75065 roller is taken for forecasting and is used in using Croston forecasting
technique in excel.
Data considered
Period Demand Period Demand
1 0 13 150
2 100 14 80
3 0 15 0
4 135 16 150
5 0 17 100
6 140 18 100
7 0 19 0
8 125 20 125
9 0 21 0
10 0 22 150
11 100 23 0
12 0 24 200
MODEL CALCULATIONS
0.1 Actual Forecast sizes Actual Forecast Forecasts Error Scaled Absolute MASE
Croston Sizes intervals intervals Error Scaled Error
0 100 0 2 50 -50 -0.6866 0.6866
135 100 2 2 50 85 1.1672 1.1672
0 103.5 0 2 51.75 -51.75 -0.7106 0.7106
140 103.5 2 2 51.75 88.25 1.2118 1.2118
0 107.15 0 2 53.575 -53.575 -0.7357 0.7357
125 107.15 0 2 53.575 71.425 0.9808 0.9808
0 108.935 0 2 54.4675 -54.468 -0.7479 0.7479
0 108.935 0 2 54.4675 -54.468 -0.7479 0.7479
100 108.935 5 2 54.4675 45.5325 0.6252 0.6252
0 108.0415 0 2.3 46.9746 -46.975 -0.645 0.645
150 108.0415 2 2.3 46.9746 103.025 1.4147 1.4147
80 112.2374 1 2.27 49.4438 30.5562 0.4196 0.4196
0 109.0136 0 2.143 50.8696 -50.87 -0.6985 0.6985
150 109.0136 0 2.143 50.8696 99.1304 1.3612 1.3612
100 113.1123 0 2.143 52.7822 47.2178 0.6484 0.6484
100 111.801 0 2.143 52.1703 47.8297 0.6568 0.6568
0 110.6209 0 2.143 51.6197 -51.62 -0.7088 0.7088
125 110.6209 6 2.143 51.6197 73.3803 1.0076 1.0076
0 112.0588 0 2.5287 44.3148 -44.315 -0.6085 0.6085
150 112.0588 0 2.5287 44.3148 105.685 1.4512 1.4512
0 115.8529 0 2.5287 45.8152 -45.815 -0.6291 0.6291
200 115.8529 0 2.5287 45.8152 154.185 2.1172 2.1172 0.91
MASE: Mean absolute Scaled Error; Smoothening Parameter (0.1);
MODEL CALCULATIONS
Croston forecasting
Forecast Size = (α)Smoothing parameter * Actual Size +(1-α)*last forecast size
Forecast interval = (α)Smoothing parameter * Actual Interval +(1-α)*last forecast interval

Forecasts = Forecast Size/Forecast Interval

Error = Actual demand – Forecasts

Scaled Error = Error / Mean Absolute Error

Absolute scaled error = Absolute value ( Scaled Error)

Mean Absolute Scaled Error = Mean ( Absolute Scaled error)


MODEL CALCULATIONS
Phase 3:
Goal programming
Min qty: Minimum order quantity to be placed against supplier
Weights are taken as geometric means
For example lead time versus price =
geometric mean of opinions of experts 1.111 = (geometric mean(1,2,1,1,1,2,1,1,1,1)
Sub criteria Weight

Min qty 1 1.0060 1.019


Price 0.943 1 0.9
Lead time 0.840 1.111 1

Min qty 1 1.006 1.019


Price 0.943 1 0.9
Lead time 0.84 1.111 1
Column sum 2.783 3.117 2.919
MODEL CALCULATIONS
Normalization of columns
Min Qty Price Lead time
Min qty 0.359324 0.322746 0.349092
Price 0.338843 0.320821 0.308325
Lead time 0.301833 0.356432 0.342583

Each cell is normalized = cell value / sum; example = 1 / 2.783 = 0.359324


Min qty Price Lead time Row sum
Min qty 0.359324 0.322746 0.349092 1.031163
Price 0.338843 0.320821 0.308325 0.967989
Lead time 0.301833 0.356432 0.342583 1.000848

Each cell row sum is divided by total row sum; ex =1.03 / 3 to obtain final weights

Min qty Price Lead time Row sum


Min qty 0.36 0.32 0.35 0.343721
Price 0.34 0.32 0.31 0.322663
Lead time 0.30 0.36 0.34 0.333616
MODEL CALCULATIONS
Calculation of consistency ratio; each cell is multiplied by weight factor
Multiplication factors 0.344 0.322 0.334
Min qty 1 1.006 1.019
Price 0.943 1 0.9
Lead time 0.84 1.111 1

Result table 1 2 3 4
Min qty 0.344 0.32 0.34
Price 0.33 0.322 0.31
Lead time 0.29 0.357742 0.3336
Calculation of row totals and multiplication with weights
1 2 3 4 row total (5) Weights (6) Multiplication
=2+3+4 (6*5)
Min qty 0.344 0.32 0.34 1.03 0.34 3.003
Price 0.33 0.322 0.31 0.96 0.322 3
Lead time 0.29 0.357742 0.3336 0.99 0.343 3
Total 9.003
Average (lambda x)= 9.003/3 = 3.001
Consistency index = Lambda x- 3 / (3-1) = 0.0005
Consistence ratio = Consistence index /0.58 = 0.0005/0.58 = 0.001 < 0.10
Consistency ratio is calculated and found to < 0.1; hence the AHP is consistent.
MODEL CALCULATIONS
Similarly for others

Flexibility 1 1.680 1.283 1.241 0.309


Service level 0.595 1 0.447 0.609 0.155
Defects return 0.779 2.236 1 0.951 0.275
Acceptable quality 0.806 1.680 1.052 1 0.261
Cr = 0.011

Number of branches 1 1.070 1.250 0.366


Number of machines 0.943 1 0.325 0.335
Max lot size 0.800 0.909 1 0.299
Cr = 0.000
MODEL CALCULATIONS

Summary of various factors:

Order details 0.311 Min qty 0.34 0.105 4


Price 0.320 0.098 6
Lead time 0.33 0.102 5
Performance 0.493 Flexibility 0.309 0.152 1
Acceptable quality 0.155 0.076 7
Defects return 0.275 0.136 2
Service level 0.261 0.129 3
Factory size 0.196 Number of branches 0.366 0.072 8
Number of machines 0.335 0.066 9
Max lot size 0.299 0.059 10
MODEL CALCULATIONS

s1 s2 s3 s4 s5 s6 s7 s8
Min qty SC1 52 180 45 129 102 55 60 47
price SC2 12 24 27 23 43 25 12 10
Lead time SC3 2 5 4 4 4 3 3 6
Max tonnage SC4 20479 23197 21821 40378 39374 31189 23025 36962
Defect percentage SC5 5 10 3 6 13 3 2 3
Discount percentage SC6 12% 14% 2% 10% 11% 11% 11% 13%
Customer satisfaction SC7 40% 60% 40% 70% 60% 30% 40% 50%
Branches SC8 4 5 5 3 3 2 4 4
Machines SC9 6 6 5 6 6 4 6 6
Turnover SC10 1524 1274 1870 2035 1631 1159 1171 1120
MODEL CALCULATIONS
MCGP Modeling
To proceed to utilize the MCGP in selecting an ideal station to start self-service, DMs
(managers of the case company) were asked to specify the minimum requirements to meet in
each goal. Note that the goal priority in GP is set by the criterion importance obtained from
AHP. The goal with the first priority is the criterion which weights the most; the second-
priority goal is the criterion having the second largest weight, and so forth.

Goal 1. The max tonnage (SC4) must exceed 20,000 tonnage; the more, the better.
Goal 2. The discount rate (SC6) must be more by 10%; the more, the better.
Goal 3. The customer satisfaction (SC7) must exceed 30%; the more, the better.
Goal 4. The Min qty (SC1) must be lesser than 45; the lesser, the better.
Goal 5. The lead time (SC3) should be less than 6 months and, of course, less is better.
Goal 6. The price (SC2) must less 10 per unit; the lesser, the better.
Goal 7. The defect percentage (SC5) should be less than 13, and, of course, less is better.
Goal 8. The number branches (SC8) must exceed 3; the larger, the better.
Goal 9. The number of machines (SC9) must exceed 4; the larger, the better.
Goal 10. The turn over must exceed lakhs (SC10) must exceed 1100; the larger, the better.
MODEL CALCULATIONS
First eliminate those suppliers that fail to meet the minimum requirements specified in
each goal.
The underlined data show that the discount rate (SC6) at S3 is below 10%, and the
number of branches (SC8) at S6 is less than three; That is, supplier S3 and S6 fail to satisfy
Goal 2 and Goal 8, respectively.
They were thus excluded from the list which the proposed MCGP method was applied to.
By doing so, it can be more efficient to find a solution from the remaining six candidates.
Then, to overcome the differences in measure units of different goals and make it
possible to compare data from different categories, we adopted the satisfaction membership
function, developed by Zadeh which maps data into satisfaction levels ranging from zero to
one.
To compare those suppliers’ locations, we identified their satisfaction (or aspiration)
levels on each goal based on DMs’ subjective preferences and evaluation consensus such as
“the more, the better.”
For example, for Goal 1 or SC4, if the max tonnage covers 35000, the satisfaction level
would be 0.8; the satisfaction level would be 1 if the max tonnage reaches41000.
However, if “less is better” is remarked in Goal 5or SC3, for instance, then higher source
data obtains a lower satisfaction level instead.
MODEL CALCULATIONS
Table lists how the original data of all sub criteria correspond to different satisfaction
levels in the order of goal priority.
Note that the lowest satisfaction levels of different goals may differ. A piecewise linear
interpolation was then applied to estimate the corresponding satisfaction level on each goal.
Table summarizes the results for each of the six suppliers under consideration, in which
their weighted satisfaction levels over the ten goals are reported at the bottom.
For example, the italic number next to SC4 (or Goal 1) in Table shows that
the supplier S1 originally owned tonnage at an average of 20479.
A linear interpolation or proportional estimate to the hypothetical source data by using
the satisfaction membership function of Goal 1 and got the satisfaction level 0.22 shown on
the upper part of Table.
The transformation is justified as follows:

(0.2 + (0.5−0.2)* (20479−20000) / (28000−20000)) = 0.22


MODEL CALCULATIONS
(SC4) Max Tonnage 20000 28000 35000 41000 up
Satisfaction Level 0.2 0.5 0.8 1
(SC6) Discount percentage (%) 6 10 12 14 up
Satisfaction Level 0.1 0.6 0.8 1
(SC7) Customer satisfaction 20 40 60 80 up
Satisfaction Level 0.1 0.4 0.7 1
(SC1) Min qty 180 150 90 40 below
Satisfaction Level 0.2 0.6 0.8 1
(SC3) Lead time 6 5 4 2 below
Satisfaction Level 0.2 0.5 0.8 1
(SC2) Price 40 30 20 10 below
Satisfaction Level 0.2 0.5 0.8 1
(SC5) Percentage of defectives 13 10 5 2 below
Satisfaction Level 0.5 0.6 0.7 1
(SC8) Number of Branches 2 4 5 6 up
Satisfaction Level 0.2 0.6 0.8 1
(SC9) Number of Machines 3 4 5 6 up
Satisfaction Level 0.6 0.8 0.9 1
(SC10) Turnover 1000 1300 1500 1600 up
Satisfaction Level 0.5 0.7 0.3 1
MODEL CALCULATIONS
MODEL CALCULATIONS
MODEL CALCULATIONS
MODEL CALCULATIONS
MODEL CALCULATIONS
RESULTS, CONCLUSIONS & FUTURE STUDY

RESULT:
The problem is solved is lingo software
satisfaction

achieved

achieved
satisfaction

values
value

value
value
Goal

achievement Positive Negative level


level
percentage deviation deviation deviations
negative
positive
G1 0.98 0 d1 0 d1 0 e1 0 e1 0.02
G2 0.6 0 d2 0 d2 0 e2 0 e2 0.4
G3 0.85 0 d3 0 d3 0 e3 0 e3 0.15
G4 1 0 d4 0 d4 0.33 e4 0 e4 0
G5 1 0 d5 0 d5 0.2 e5 0 e5 0
G6 1 0 d6 0 d6 0.29 e6 0 e6 0
G7 1 0 d7 0 d7 0.32 e7 0 e7 0
G8 0.4 0 d8 0 d8 0 e8 0 e8 0.6
G9 1 0 d9 0 d9 0 e9 0 e9 0
G10 1 0 d10 0 d10 0 e10 0 e10 0
RESULTS, CONCLUSIONS & FUTURE STUDY
CONCLUSIONS

Inventory classification based on multi categories is done. As the classification based


on individual ABC, XYZ, FSN etc. may not give true picture about the priority of spare,
stock for procurement.

Forecasting of spare demand also plays a vital role for inventory holding instead of
moving average crompston based method gives better approximation. Selection of proper
supplier with conflicting goal I s another area addressed in this project.

This study presents an AHP-MCGP method that aids decision makers s to apply
appropriate weights from AHP to the objective function of MCGP in the supplier selection
problem considering both qualitative and quantitative criteria. The proposed model can
integrate opinions about weight settings among goals from several DMs using AHP. Their
opinions can be integrated into aspiration of each goal in the MCGP method.
FUTURE SCOPE OF STUDY

The solution procedure is systematic and objective and it can be developed as a


decision support system (DSS) in future. With the tailored DSS, the proposed approach can
thus be adopted in solving not only supplier selection, but also general decision problems
like storages, transportations etc.
1.
REFERENCES
Altay Guvenir, H., &Erel, E. (1998). Multi-criteria inventory classification using a genetic algorithm.
European Journal of Operational Research, 105(1), 29-37.
2. Bhattacharya, A., Sarkar, B., & Mukherjee, S. K. (2007).Distance-based consensus method for ABC analysis.
International Journal of Production Research, 45(15), 3405-3420.
3. Braglia, M., Grassy, A., & Montana, R. (2004). Multi-attribute classification method for spare parts inventory
management.Journal of Quality in Maintenance Engineering, 10(1), 55-65.
4. Chen, Y., Li, K. W., Levy, J., Hype, K. W., & Kilgore, D. M. (2008). A rough set approach to multiple criteria
ABC analysis. In Transactions on rough sets VIII (pp. 35-52). Springer Berlin Heidelberg.
5. Edwards, W., & Barron, F. H. (1994). SMARTS and SMARTER: Improved simple methods for multi-attribute
utility measurement. Organizational Behavior and Human Decision Processes, 60(3), 306-325.
6. Gomes, L. F. A. M., & Ferreira, A. C. S. (1995). The multi-criteria ABC analysis – An application of rough set
theory.Foundations of Computing and Decision Sciences, 20(3).
7. Flores, B. E., & Whybark, D. C. (1987). Implementing multiple criteria ABC analysis.Journal of Operations
Management, 7(1), 79-85.
8. Flores, B. E., Olson, D. L., &Dorai, V. K. (1992). Management of multi-criteria inventory classification.
Mathematical and Computer Modelling, 16(12), 71-82.
9. Gopal, P. P., Ganesh, L. S., &Rajendran, C. (1994). Criticality analysis of spare parts using the analytic
hierarchy process.International Journal of Production Economics, 35(1), 293-297.
10. Hadi-Vencheh, A. (2010). An improvement to multiple criteria ABC inventory classification.European Journal
of Operational Research, 201(3), 962-965.
11. Kabir, G., Hasin, M. A. A., &Khondokar, M. A. H. (2011). Fuzzy analytical hierarchical process for multi-
criteria inventory classification. Proceedings of the International Conference on Mechanical Engineering, 18-
20 December, Dhaka, Bangladesh.
12.
REFERENCES
Liu, Q., & Huang, D. (2006). Classifying ABC inventory with multi-criteria using a data envelopment analysis
approach. In Intelligent systems design and applications, 2006. ISDA'06.Sixth International Conference on
IEEE (Vol. 1, pp. 1185-1190).
13. Ng, W. L. (2007). A simple classifier for multiple criteria ABC analysis.European Journal of Operational
Research, 177(1), 344-353.
14. Partovi, F. Y., & Anandarajan, M. (2002). Classifying inventory using an artificial neural network
approach.Computers & Industrial Engineering, 41(4), 389-404.
15. Partovi, F. Y., & Burton, J. (1993). Using the analytic hierarchy process for ABC analysis.International
Journal of Operations & Production Management, 13(9), 29-44.
16. Pawlak, Z. (1991). Rough sets: Theoretical aspects of reasoning about data. Dordrecht: Kluwer Academic
Publishing. ISBN 0-7923-1472-7.
17. Ramanathan, R. (2006). ABC inventory classification with multiple-criteria using weighted linear
optimization.Computers & Operations Research, 33(3), 695-700.
18. Saaty, T. L. (1995). Transport planning with multiple criteria: The analytic hierarchy process applications and
progress review. Journal of advanced transportation, 29(I), 81-126.
19. Vargas, L. G. (1990). An overview of the analytic hierarchy process and its applications. European Journal of
Operational Research, North-Holland.
20. Yu, M. C. (2011). Multi-criteria ABC analysis using artificial-intelligence-based classification techniques.
Expert Systems with Applications, 38(4), 3416-3421.
21. Zahedi, F. (1986).The analytic hierarchy process-a survey of the method and its applications. Interfaces, 16(4).
22. Zhou, P., & Fan, L. (2007). A note on multi-criteria ABC inventory classification using weighted linear
optimization. European Journal of Operational Research, 182(3), 1488-1491
THANK YOU

Vous aimerez peut-être aussi