Page 1
© 2016, IJARCSSE All Rights Reserved Page | 116
Volume 6, Issue 7, July 2016 ISSN: 2277 128X
International Journal of Advanced Research in Computer Science and Software Engineering Research Paper Available online at: www.ijarcsse.com
Variant Multi Product Transhipment Model Sreenuvasalu Y
1, Suresh BabuC
2, Madhu Mohan Reddy P
3, Sundara Murthy M
4
1 Asst. Professor, Dept. of Mathematics, Narayana Engineering College, Nellore, India
2, 3 Asst. Professor, Dept. of Mathematics, Siddharth Institute of Engineering and Technology, Puttur, India
4 Professors (Rtd.) Dept. of Mathematics, SV University, Tirupati, India
Abstract: ‘Transportation’ the term which plays a vital role in shaping the nation’s destiny though it’s a common
phenomenon. In which one of the important aspects is bulk transportation. The goods transported from source to
destination there be some hauling place is more natural. So, it drives me ahead to study on bulk transshipment
models. In this paper, we take some sources and destinations. Sources and destinations maintain some multiple
products. Each destination should get its own multiple requirements from unique source only. The distance/cost
between sources and destinations are to be known. All the sources and destinations need not be performed that means
some of the given sources and destinations should satisfied by the transshipment schedule. The objective is to find the
minimum cost/distance subject to the given constraints.
Keywords: Mathematical Formulation,Lexi-Search Algorithm, Pattern recognition technique, Combinatorial
structure. multipleproducts & optimal solution.
I. INTRODUCTION
There are various types of transportation models and the simplest of them was developed by[1, 3]. Several
extensions of transportation models and methods have been subsequently developed. In these models, the given cost of
transportation is for a unit quantity of item or commodity or product. But in Bulk Transportation Problem, the bulk
transportation cost (independent of quantity transported) from a source i to destination j is taken into consideration.The
problem of Bulk Transportation was first investigated by [4] who presented an algorithm. Later [12] also offered an
algorithm and [15] give an algorithms.Both the algorithms are based on Branch and Bound procedure to solve the Bulk
transportation problems.
A set of logic problems to solve multi-index transportation problems by [2] also conducted a detailed investigation
regarding the characteristics of multi-index transportation problem model. [9]used multi-index transportation problem
model to solve the shipping scheduling suggested that the employment of such transportation efficiency but also optimize
the integral system. These references are only a single objective model and its constraints are not fuzzy member. In the
case of cost-time trade- off bulk transportation problem [7] studied multi objective models and presented efficient
heuristic for this multi objective bulk transportation problem.
Furthermore [5, 10, 11, 13] have studied a variety of the problems.These investigations come under single
commodity bulk transportation problems. [8]Presenteda Multi-Product Bulk Transportation Problem to minimize the
total cost of the bulk transportation.[16]Presented three dimensional time minimization bulk transportation problem to
minimize the total time of goods transportation.A more generalized variation of the bulk TP problems studied by [14]
called A Variant Bulk Transportation Problem with Multiple Bulk Cost Constraint, where the products are supplied
simultaneously to the destinations according to their requirement from the sources. [6] Studied two stage multi objective
variant bulk transportation problem.Both of them presented Lexi Search algorithms and mentioned that the efficiency of
this algorithm over branch and bound algorithm.
II. PROBLEM DESCRIPTION
Let there be m sources with capacity S(i,1) and S(i,2) of products k=2 in ith
source . Let there be n destinations
cities with requirement of productsDR (i, 1) and DR (i, 2). Let there be r cities DT= {𝛼1, 𝛼2 ,−−−−𝛼𝑟} ⊂ 𝐷𝑅 which are
subset of destinations. Each of the destination cities in DT can supply to another rdestination cities, which is called as
transshipment. Let C(i,j) is the bulk supply cost from ith source to j
th destination for items (k=1,2) . Each destination
should get its supply of k products from one source only subject to availability. In this case of transshipment also this
condition that the product of the destinations should be from the same source only. In this transshipment model we do
not consider supply of the products from destinations to sources and sources to sourcesas they are not natural. The
objective of the problem is to find supply schedule for supplying the requirements of multiple products from the sources
to destinations subject to its availability and subject to the conditions such that the total bulk cost of supply products is
minimum.
Let a set S ={ 1,2,-------m} where m is the number of sources and S(i,1) and S(i,2) are the availability of k=2
products in ith
source. Let another set DR= {1, 2-------n} where n is the number destinations with requirement of
products. DR(i,1) and DR(i,2) are the requirement of two products in ith
destination where i∈ 𝐷𝑅.We denote a total
Page 2
Sreenuvasalu et al., International Journal of Advanced Research in Computer Science and Software Engg. 6(7),
July- 2016, pp. 116-125
© 2016, IJARCSSE All Rights Reserved Page | 117
number of sources in solution as m0 and a total number ofdestinations as no .Let C (i,j) be the bulk cost for supplying
from ith
source to jth
destinationfor all products k=1,2. The objective of the problem is to find minimum supply
schedulen0(<n)requirement of both products from m0 < m sources to destinations subject to its availability and subject
to the conditions such that the total bulk cost of supply products is minimum.
III. SPECIALITY OF THE MODEL
Present this model is a variant bulk transshipment model withmultiple products. A subject of destinations DT⊂ 𝐷𝑅
can supply to r destinations the bulk cost for k products is same as C (i, j) i.e the cost of supply as each of k products
from ith to j
th destinations. In this case of transshipment i∈ 𝐷𝑇, we consider only m0<m sources and n0<n destinations for
supply sehudele. In the above situations this model is useful and the same as specialty of the model.
IV. MATHEMATICAL FORMULATION
Minimum z(x) = 𝐶 𝑖, 𝑗 𝑋(𝑖, 𝑗)𝑗 𝑖 K------------ (1)
Where i∈ 𝑆1, j∈ 𝐷𝑅&𝑆1 =S∪ 𝐷𝑇 𝐷𝑅(𝑖. 𝑘) ≤ 𝑆 𝑖, 𝑘 𝑚
𝑖=1𝑛𝑖=1 ∀K, where k ∈ 𝐾------------ (2)
𝑆 (𝑖,𝑘) ≥ 𝐷𝑅 ( j, k) where j ∉ 𝐷𝑇& k∈ 𝐾-----------(3)
𝑠 (𝑖, 𝑘) ≥ 𝐷𝑅 (j,k) where j ∈ 𝐷𝑇& k∈ ------------ (4)
X (i, j) =0 or 1 ------------ (5)
Equation (1) represents the objective of the problem. i.e., to find minimum cost supply to the n0 destinations for its
requirement of two products.Constraint(2) represents the total requirement at the destination cities for each product of k
should be less than or equal to the total availability at the sources for each product of k.Equation (3) states that we
consider a destination j∉ 𝐷𝑇 then the requirement of DR (j, k) should be less than or equal to the availability of
kthproduct in i
th source at that point of supply i.e. denoted by𝑆 (𝑖,𝑘).Equation (4) denotes, if a destination j ∈ 𝐷𝑇 then the
requirement of kth
product in the jthdesignation along its supply to other destination should be less than or equal to
it‟ssources of theproduct k which is denoted by𝑆 (𝑖, 𝑘).Equation (5) indicates there are number of order pairs connecting
to sources to destinations or destinations to destinations.
V. NUMERICAL ILLUSTRATION:
The concepts and algorithm developed will be illustrated by a suitable numerical example for which set of
number of sources S={1,2,3,4} and set of number of destinations DR ={1,2,3,4,5,6,7} . Here m=4,n=7, m0 =3, and n0
=6, k=2 & DT=(1,5,7) ⊂ DR. 𝑆1=S∪ 𝐷𝑇=(1,2,3,4) ∪ (1D, 5D, 7D)= (1,2,3,4,1D,5D,7D) then the cost matrix C(i, j)
between the sources and destinations is as follows .
Table -1
C(i,j)=
In above table- 1, suppose C (4, 6) =21 means that the cost of the supply from source city 4 to destination city 6 is 21
units for both products k=1, 2.By the transshipment consider some of the destinations among seven destinations (i.e 1,
5&7) treated as sources. So the we add 1, 5& 7 destinations to the sources. Hence the total number of sources increases
from 4 to 7.∞ indicates irrelevant between the destinations. Consider two products are available at the sources. Let them
are S(i,1) and S(i,2). Then the availability of the given products at sources S (i,α) are as follows . Where ∝𝑖 𝑖𝑠 the type
of the product.
Table-2 (source product available table)
Sources 1 2 3 4
Product 1
capacity 60 60 60 60
Product 2
capacity 70 60 60 60
By the table-2, we have S (1, 2) =70 means source 1 has capacity of product 2 is 70 units. Consider two products are
required at each and every destination .Let them are DR(j.k) where k= 1,2,3,4,5,6 & 7.Then the required of the product at
destination DR(j,α ) are as follows . Where α is the type of the product.
Table-3 (Destination product requirement table)
SN 1 2 3 4 5 6 7
1 5 12 10 9 8 19 11
2 15 18 14 20 16 3 17
3 19 1 21 13 2 22 17
4 7 20 16 23 22 21 3
1D ∞ 23 6 18 24 9 25
5D 24 7 26 10 ∞ 25 26
7D 14 12 13 8 27 4 ∞
Page 3
Sreenuvasalu et al., International Journal of Advanced Research in Computer Science and Software Engg. 6(7),
July- 2016, pp. 116-125
© 2016, IJARCSSE All Rights Reserved Page | 118
Destinations 1D 2 3 4 5D 6 7D
Product 1 Requirement 20 20 20 20 30 40 20
Product 2 requirement 20 30 10 20 20 20 30
By the table-3, we have DR (1, 2) =20 means first product requirement of second destination is 20 units. For this
numerical example the objective is the supply of the requirement of both products to n0 destinations frommosources with
minimum total cost subject to the conditions suchthat the total bulk cost of supply is minimum. Where transshipment
consider destination to destination.
VI. FEASIBLE SOLUTION
Feasible solution is a solution, which satisfies all the constraints in the problem. The constraints are discussed in the
mathematical formulation. Consider an ordered pair set {(4,1D) , (4,1D) , (1D,3) , (ID,3) , (4,7D) , (4,7D) , (2,6) , (2,6) ,
(3,5D) , (3,5D) , (3,2) , (3,2) } represents a feasible solution.In the following figure-1, squarerepresents sources with
capacity of products andcircle represents are destination cities. The value in square indicates availability of products
(product 1 numerator and product 2 denominator) number of that particular source. The value in circles indicates
requirement of products (product 1 numerator and product 2 denominator number) of that particular destination. A value
below and above the connectivity of the source city and destination city or destination city and destination city represents
cost between the particular source and destination cities.Consider a set of order pairs{ (4,1D) , (4,1D) , (1D,3) , (ID,3) ,
(4,7D) , (4,7D) , (2,6) , (2,6) , (3,5D) , (3,5D) , (3,2) , (3,2) } represents a feasible solution. FIGURE-1
1 1D 6 3
6
7
4 7
3 7D
6
3
5D 4
2
3 2 2
1
1
2 3 6
3
60
70
20
20
20
10
20
30
20
20
60
60
30
20
60
60
40
20
60
60
20
30
In above figure 1, source 4 supplies 20 units of product 1, 20 units of product 2 to the destination 1D with cost of
7 units .From there to supplies 20 units of product 1 , 10 units of product 2 to the destination 3 with the cost of 6 units.
Source 4 supplies 20 units of product 1, 30 units of product 2 to the destination 7D with cost of 3 units. source 3 supplies
20 units of product 1, 30 units of product 2 to the destination 2 with cost of 1 units .source 3 supplies 30 units of product
1 , 20 units of product 2 to the destination 5D with the cost of 2 units. Source2 supplies 40 units of product 1, 20 units
of product 2 to the destination 6 with cost of 3 units. Also 6 destinations among seven get their requirements of the
different products from given 3 sources among 4. Here we leave one source and one destination for truncation. So the
truncation conditions satisfied hence we called the above solution is feasible solution.Then the corresponding
transshipment cost of the above allocations of figure-1, as follows
The Solution is Z = 2(4,1D) X(4,1D) + 2(1D,3) X(1D,3) + 2(4,7D) X(4,7D) + 2(2,6) X(2,6) + 2 (3,5D)
X(3,5D) + 2(3,2) X(3,2) =14+12+6+6+4+2 = 44 units.
Page 4
Sreenuvasalu et al., International Journal of Advanced Research in Computer Science and Software Engg. 6(7),
July- 2016, pp. 116-125
© 2016, IJARCSSE All Rights Reserved Page | 119
VII. CONCEPTS AND DEFINITIONS
A tour is a feasible trip - schedule for the transshipment. Trip-schedule can be represented by an approximate n x n x
r indicator array X = {x (i, j,k); x (i, j ,k) = 0 or 1} in which x (i, j,k ) = 1 indicates that the vehicle visits city j from city
i in kth season
A) .Definition of an Alphabet –Ttable:
There are mxmordered pairs in thetwo-dimensional array C. For convenience these are arranged in ascending order
of their corresponding cost are indexed from 1, 2, 3…. . (SundaraMurrty -1979) .Let SN=( 1, 2, 3…. ) be the set of
indices. Let C be the corresponding array of cost.. If a, bSN and a < b then C(a) C(b).. Also let the arrays R, C& C
be the array of row, column and cost indices. For our convience we used same notationC for column and cost indices.
CD be an array of cumulative sum of the elements of C. The arrays SN, C, CD, R, and C for the numerical example are
given in the table-6. If aSN then (R(a),C(a),c(a) ) is the ordered triple and D(a) = D (R(a), C(a)) is the value of the
ordered pair and CD (a) = 𝑐(𝑖)𝑎𝑖=1 Then the alphabet of the given cost matrix (table-6) is as follows.
Table-6 (Alphabet table)
S N C CD R C
1 1 1 3 2
2 2 3 3 5D
3 3 6 2 6
4 3 9 4 7D
5 4 13 7D 6
6 5 18 1 1D
7 6 24 1D 3
8 7 31 4 1D
9 7 38 5D 2
10 8 46 1 5D
11 8 54 7D 4
12 9 63 1 4
13 9 72 1D 6
14 10 82 1 3
15 10 92 5D 4
16 11 103 1 7D
17 ∞ 114 5D 5
18 12 126 1 2
19 12 138 7D 2
20 13 151 3 4
21 13 164 7D 3
22 14 178 2 3
23 14 192 7D 1D
24 15 207 2 1D
25 ∞ 222 1D 1
26 16 238 2 5D
27 16 254 4 3
28 17 271 2 7D
29 17 288 3 7D
30 18 306 2 2
31 18 324 1D 4
32 19 343 1 6
33 19 362 3 1D
34 20 382 2 4
35 20 402 4 2
36 21 423 3 3
37 21 444 4 6
38 22 466 3 6
39 22 488 4 5D
40 23 511 4 4
41 23 534 1D 2
42 24 558 1D 5D
43 24 582 5D 1D
44 25 607 1D 7D
45 25 632 5D 6
Page 5
Sreenuvasalu et al., International Journal of Advanced Research in Computer Science and Software Engg. 6(7),
July- 2016, pp. 116-125
© 2016, IJARCSSE All Rights Reserved Page | 120
46 26 658 5D 3
47 26 684 5D 7D
48 27 711 7D 5D
49 ∞ 738 7D 7D
Let us consider 15ЄSN. It represents the ordered (R (15), C (15)) = (5D, 4). Then C (15) =C (5D, 4) = 10 and CD
(15) =92
B) Definition of a Word:
Let SN= (1,2...) be a set of indices, let c be an array of cost ,CD is an array of cumulative sum of elements in C.
Let arrays R, C and C be the row, column and cost indices of the ordered triples. Let C be the array of cost. Let Lk={a1,
a2,......,ak}, ai €SN be an ordered sequence of k indices from SN. The pattern represented by the ordered triples whose
indices are given by Lk is independent of the order of ai in the sequence. Hence for uniqueness the indices are arranged
in the increasing order such that ai ≤ai+1, i=1,2,........k-1. The set SN is defined as the “Alphabet-Table” with alphabet
order as (1,2,....,) and the ordered sequence Lk is defined as a “word” of length k. A word Lk is called a “sensible word”.
If ai ≤ai+1 , for i=1,2,....k-1 and if this condition is not met it is called a “insensible word”. A word Lkhas at least one
feasible word or equivalently the partial pattern represented by Lk has at least one feasible word or equivalently the
partial pattern represented by Lk is said to be feasible if the block of words represented by Lk has at least one feasible
word or, equivalently the partial pattern represented by Lk should not have any inconsistency.
In the partial word Lk any of the letters in SN can occupy the first place.Since the words of length greater than n-1
are necessarily infeasible,as any feasible pattern can have only n unit entries in it. LK is called a partial word if k<n-1, and
it is a full length word if k=n-1,or simply a word.A partial word Lk represents, a block of words with Lk as a leader i.e., as
it‟s first k letters. A leader is said to be feasible, if the block of word, defined by it has at least one feasible word.
C) Value of the Word:
The value of the( partial) word Lk, V(Lk) is defined recursively as V(Lk)=V(Lk-1)+D(ai) with V(L0)=0, where D(ai) is
the cost array arranged sush that D(ak)<D(ak+1).For our convince k be the value of bulk unit of D(ak). V(Lk) and V(X)
the values of the pattern X will be the same, since X is the (partial) pattern represented by Lk, (Sundara Murthy –
1979).Consider the partial word L4 = {(1(a), 1(b),2(a),2(b))}Then V (L4) =1+1+2+2=6.
D) Lower Bound of a Partial Word LB (Lk):
A lower bound LB (Lk) for the values of the block of words represented by Lk= (a1,a2, .....ak) can be defined as
follows. 𝐿𝐵(𝐿𝑘) = 𝑉(𝐿𝑘) + 𝐶𝐷(𝑎𝑘 + 𝑛 + 2 − 1 − 𝑘) − 𝐶𝐷(𝑎𝑘). Consider the partial word L4 = (1(a), 2(a), 3(a), 5(a))
and V (L4) =1+2+3+3=10. Then LB (L4) =V (L4) +CD (a4+n+2- 1-4) - CD (a4); LB (4) = 10+CD (5+7+1-4) - CD (5)
= 10+CD (9)- CD (5) =10+38-13 =35 units
VII. ALGORITHMS
The recursive algorithm for checking the feasibility of a partial word Lp is given as follows. In the algorithm first we
equate IX = 0, at the end if IX = 1 then the partial word is feasible, otherwise it is infeasible. For this algorithm we have
RA=R (ak+1) and CA=C (ak+1).
A) Arrays and Notations:
m = Number of Sources, n = Number of Destinations, k= For First Product is 1 and For Second Product is 2,
SA(i,j)= Capacity of two Products at Particular Source, DR (i,j) = Requirement of two Products at Particular
Destinations, DX (i) = Destination cities act as sources or List of Sub Sources, CDR (i,j) = Requirement of two Products
at current position.
B) Algorithm-1: (Feasible Checking) STEP0 : IX = 0 IF YES GO TO 1 & IF NO GO TO 6
STEP1 : IS (IC [CA] = 1) IF YES GO TO 6 &IF NO GO TO 2
STEP2 : IS (DX (RA) =1) IF YES GO TO 3 & IF NO GO TO 4
STEP3 : IS CDR (RA, 1) = DR (RA, 1) +DR (CA, 1)
IS CDR (RA, 2) = DR (RA, 2) +DR (CA, 2) IF YES GO TO 5
STEP4 : IS SA (RA, 1) ≥DR (CA, 1)
IS SA (RA, 2) ≥DR (CA, 2) IF YES GO TO 5 & IF NO GO TO 6
STEP5 : IX=1
STEP6 : STOP.
We start with the partial word L1 = (a1) = (1). A partial word Lk is constructed as Lk = Lk-1 * (ak). Where * indicates
chain formulation. We will calculate the values of V (Lk) and LB (Lk) simultaneously. Then two situations arises one for
branching and other for continuing the search.
1. LB (Lk) < VT. Then we check whether Lk is feasible or not. If it is feasible we proceed to consider a partial word
of under (k+1). Which represents a sub-block of the block of words represented by Lk? If Lk is not feasible then
consider the next partial word p by taking another letter which succeeds ak in the position. If all the words of
order p are exhausted then we consider the next partial word of order (k-1).
Page 6
Sreenuvasalu et al., International Journal of Advanced Research in Computer Science and Software Engg. 6(7),
July- 2016, pp. 116-125
© 2016, IJARCSSE All Rights Reserved Page | 121
2. LB (Lk) > VT. In this case we reject the partial word Lk. We reject the block of word with Lk as leader as not
having optimum feasible solution and also reject all partial words of order p that succeeds Lk.To find Optimal
Feasible word a Lexi-Search Algorithm using PRT is developed and is given in below.
C) Algorithm-II: (Lexi-search algorithm):
STEP 0 : Initialization The arrays SN, IC, R, C, SS, DRS, SW, SA, DR, DC, RA, CA, L, V, LB and values
Q, m, n, k are made available. The values I=1, J=0, VT=9999 and Max= m×n–n
STEP 1 : J = J + 1, IS [IF J > Max IF YES GOTO 15& IF NO GOTO 2
STEP2 : V [I] = V [I-1] + C [J]; V [0] = 0
LB [I] = 2[V (I) + DC (J + n – I) – DC (J) ]
IS (LB [I] ≥ VT)IF YES GO TO 14 & IF NO GO TO 3
STEP : RA = R [J], CA = C [J] IF YES GOTO 4
STEP4 : Check Feasibility using algorithm 1
IS (IX = 1) IF YES GO TO 5& IF NO GO TO 1
STEP5 : IS (I = n-1) IF YES GO TO 10& IF NO GO TO 6
STEP6 : IS DX (RA=1) IF YES GOTO 7& IF NO GO TO 8
STEP7 : DR (RA, 1) = CDR (RA, 1)
DR (RA, 2) = CDR (RA, 2) IF YES GO TO 9
STEP8 : SA (RA.1) = SA (RA, 1)-DR (CA, 1)
SA (RA.2) = SA (RA, 2)-DR (CA, 2) IF YES GO TO 9
STEP9 : IC (CA) =1, L(I)=J, I-I+1, Max=Max+1 IF YES GO TO 1
STEP10 : VT = V [I], L [I] = J, L [I] is full length word and is feasible
Record L [I] and VT IF YES GO TO 11
STEP11 : I = I – 1, J = L [I], RA = R [J], CA = C [J]
IC [CA] = 0 & IS[DX (I)] =1 IF YES GO TO 12& IF NO GO TO 13
STEP12 : DR (RA.1) =DR (RA, 1)-CDR (RA,1)
DR (RA.2) =DR (RA,2)-CDR (RA,2) IF YES GO TO 1
STEPP13 : SA (RA.1) =SA (RA,1)+DR (RA,1)
SA (RA.2) =SA (RA,2)+DR (RA,2) IF YES GO TO 1
STEP14 : IS IF YES GOTO 15& IF NO GO TO 11
STEP15 : STOP
The current value of VT at the end of the search is the value of the optimal feasible word. At the end if VT = 999 it
indicates that there is no feasible solution.
VIII. SEARCH TABLE
The working details of getting an optimal word using the above algorithm for the illustrative numerical example is
given in the following table-6. The columns named (1), (2),(3)............gives the letters in the first, second, third and so on
places respectively. The columns V, LB, R, C and C letters indicates Value of that word ,lower bound of the word , row,
column and cost indices of accepted for connectivity. The last column denotes the remarks regarding the acceptability
of the partial word. In the following table „A‟ indicates ACCEPT and „R‟ indicates REJECT.
Table-7 (Search table)
SN 1 2 3 4 5 6 7 8 9 10 11 12 V LB R C RE
1 1(a) 1 36 3 2 A
2 1(b) 2 36 3 2 A
3 2(a) 4 36 3 5D A
4 2(b) 6 36 3 5D A
5 3(a) 9 36 2 6 A
6 3(b) 12 36 2 6 A
7 4(a) 15 36 4 7D A
8 4(b) 18 36 4 7D A
9 5(a) 22 7D 6 R
10 5(b) 22 7D 6 R
11 6(a) 23 1 1D R
12 6(b) 23 1 1D R
13 7(a) 24 44 1D 3 A
14 7(b) 30 44 1D 3 A
15 8(a) 37 44 4 1D A
Page 7
Sreenuvasalu et al., International Journal of Advanced Research in Computer Science and Software Engg. 6(7),
July- 2016, pp. 116-125
© 2016, IJARCSSE All Rights Reserved Page | 122
16 8(b) 44 44 4 1D A=VT
17 8(b) 37 44 4 1D R,=VT
18 8(a) 31 45 4 1D R,>VT
19 7(b) 24 45 1D 3 R,>VT
20 5(a) 19 7D 6 R
21 5(b) 19 7D 6 R
22 6(a) 20 44 1 1D R,=VT
23 4(b) 15 39 4 7D A
24 5(a) 19 7D 6 R
25 5(b) 19 7D 6 R
26 6(a) 20 44 1 1D R,=VT
27 5(a) 16 7D 6 R
28 5(b) 16 45 7D 6 R,>VT
29 4(a) 12 39 4 7D A
30 4(b) 15 39 4 7D A
31 5(a) 19 7D 6 R
32 5(b) 19 7D 6 R
33 6(a) 20 44 1 1D R,=VT
34 5(a) 16 7D 6 R
35 5(b) 16 45 7D 6 R,>VT
36 4(b) 12 42 4 7D A
37 5(a) 16 7D 6 R
38 5(b) 16 45 7D 6 R,>VT
39 5(a) 13 46 7D 6 R,>VT
40 3(b) 9 39 2 6 A
41 4(a) 12 39 4 7D A
42 4(b) 15 39 4 7D A
43 5(a) 19 7D 6 R
44 5(b) 19 7D 6 R
45 6(a) 20 44 1 1D R,=VT
46 5(a) 16 7D 6 R
47 5(b) 16 45 7D 6 R,>VT
48 4(b) 12 42 4 7D A
49 5(a) 16 7D 6 R
50 5(b) 16 45 7D 6 R,>VT
51 5(a) 13 46 7D 6 R,>VT
52 4(a) 9 42 4 7D A
53 4(b) 12 42 4 7D A
54 5(a) 16 42 7D 6 A
55 5(b) 20 42 7D 6 A
56 6(a) 25 42 1 1D A
57 6(b) 30 42 1 1D A
58 7(a) 36 42 1D 3 A
59 7(b) 42 42 1D 3 A,=VT
60 7(b) 36 43 1D 3 R,>VT
61 7(a) 31 44 1D 3 R,>VT
62 6(b) 25 44 1 1D R,>VT
63 6(a) 21 45 1 1D R,>VT
64 5(b) 16 45 7D 6 R,>VT
65 5(a) 13 46 7D 6 R,>VT
66 4(b) 9 46 4 7D R,>VT
67 3(a) 7 40 2 6 A
68 3(b) 10 40 2 6 A
Page 8
Sreenuvasalu et al., International Journal of Advanced Research in Computer Science and Software Engg. 6(7),
July- 2016, pp. 116-125
© 2016, IJARCSSE All Rights Reserved Page | 123
69 4(a) 13 40 4 7D A
70 4(b) 16 40 4 7D A
71 5(a) 20 7D 6 R
72 5(b) 20 42 7D 6 R,=VT
73 5(a) 17 43 7D 6 R,>VT
74 4(b) 13 43 4 7D R,>VT
75 4(a) 10 43 4 7D R,>VT
76 3(b) 7 43 2 6 R,>VT
77 2(b) 4 40 3 5D A
78 3(a) 7 40 2 6 A
79 3(b) 10 40 2 6 A
80 4(a) 13 40 4 7D A
81 4(b) 16 40 4 7D A
82 5(a) 20 7D 6 R
83 5(b) 20 42 7D 6 R,=VT
84 5(a) 17 43 7D 6 R,>VT
85 4(b) 13 43 4 7D R,>VT
86 4(a) 10 43 4 7D R,>VT
87 3(b) 7 43 2 6 R,>VT
88 3(a) 5 43 2 6 R,>VT
89 2(a) 3 41 3 5D A
90 2(b) 5 41 3 5D A
91 3(a) 8 41 2 6 A
92 3(b) 11 41 2 6 A
93 4(a) 14 41 4 7D A
94 4(b) 17 41 4 7D A
95 5(a) 21 7D 6 R
96 5(b) 21 43 7D 6 R,>VT
97 5(a) 18 44 7D 6 R,>VT
98 4(b) 14 44 4 7D R,>VT
99 4(a) 11 44 4 7D R,>VT
100 3(b) 8 44 2 6 R,>VT
101 3(a) 6 45 2 6 R,>VT
102 2(b) 3 45 3 5D R,>VT
103 1(b) 1 41 3 2 A
104 2(a) 3 41 3 5D A
105 2(b) 5 41 3 5D A
106 3(a) 8 41 2 6 A
107 3(b) 11 41 2 6 A
108 4(a) 14 41 4 7D A
109 4(b) 17 41 4 7D A
110 21 7D 6 R
111 21 43 7D 6 R,>VT
112 5(a) 18 44 7D 6 R,>VT
113 4(b) 14 44 4 7D R,>VT
114 4(a) 11 44 4 7D R,>VT
115 3(b) 8 44 2 6 R,>VT
116 3(a) 6 44 2 6 R,>VT
117 2(b) 3 45 3 5 R,>VT
118 2(a) 2 46 2 6 R,>VT
At the end of the search table-7, the current value of VT is 42 and it is the value of the optimal feasible word L12
={(1(a),1(b),2(a),2(b),4(a),4(b),5(a),5(b),6(a),6(b),7(a),7(b))}.It is given in the 59th row of the search table and the
Page 9
Sreenuvasalu et al., International Journal of Advanced Research in Computer Science and Software Engg. 6(7),
July- 2016, pp. 116-125
© 2016, IJARCSSE All Rights Reserved Page | 124
corresponding order paires are(3,2),(3,2),(3,5D),(3,5D),(4,7D),(4,7D),(7D,6),(7D,6) (1,1D),(1,1D),(1D,3),(1D,3) .Then
the following figure-3 represents the optimal solution. FIGURE-3
1 5 1D 6 3
5 6
2 2 4
1
3 1 5D
2
2
4 3 7D 6
4
3 4
60
70
20
20
20
10
30
20
60
60
20
20
60
60
20
30
60
60
20
30
40
20
In above figure 3, Source 1 supplies 20 units of product 1, 20 units of product 2 to the destination 1D with cost of
5 units .From there to supplies 20 units of product 1 , 10 units of product 2 to the destination 3 with the cost of 6 units.
Source 3 supplies 20 units of product 1, 30 units of product 2 to the destination 2 with cost of 1 units .Source 3
supplies to 30 units of product 1 , 20 units of product 2 to the destination 5D with the cost of 2 units. Source 4
supplies 20 units of product 1, 30 units of product 2 to the destination 7D with cost of 3 units .From there to supplies 40
units of product 1 , 20 units of product 2 to the destination 6 with the cost of 4 units. 6 destinations among seven get
their requirement of the different products from given sources 3 among 4. Here we leave one source and one destination
for truncation. So all the conditions satisfied hence we called the above solution is optimal solution. Then the optimal
solution is
Z = 2(1,1D) X(1,1D) + 2 (1D, 3) X(1D,3)+ 2(3, 2) X(3,2) + 2 (3,5D) X(3,5D) + 2 (4,7D) X(4,7D)+ 2 (7D, 6)
X(7D,6)= 10+12+2+4+6+8 = 42 units.
IX. CONCLUSION
In thispaper,first the model is formulated as a transshipment problem. Write the corresponding mathematical
formulation and Developed Lexi-search Algorithm by using pattern recognition technique for solving this model.The
concepts and steps involved in the algorithm are explained with a suitable numerical example for understand.we find an
optimum solution with the given all constraints are satisfied.Based on this experience we can say that this algorithm is
efficient. The following References are used for making the model.
REFERENCES
[1] Dantzig G.B., and Ramser, J.H. (1959): “The Truck Dispatching Problem,” Management Science, Vol. 6, No.
1, pp.80-91.
[2] Junginger w.(1993) European journal of operational research 66,353-371.
[3] Koopmans, Tjalling, C. (1949): “Optimum utilization of the transportation system,”Econometrica 17
(Supplement) 136–146.
[4] Maio, A. D., Roveda, C. (1971): “An all zero-one algorithm for a certain class of transportation problems,”
Operations Research, 19, 1406-1418.
[5] Naganna B. Operations Research. Ph. D thesis, S. V. University, Tirupati, India. (2007)
[6] P. Guravaraju, purusotham S, Suresh Babu C and Madhu Mohan Reddy P: two stage multi objective variant
bulk TP model - International Journal of advanced research in computer science and software engineering -
ISSN 2277-128X Volume 5 (10), October (2015), pp. 115-125.
Page 10
Sreenuvasalu et al., International Journal of Advanced Research in Computer Science and Software Engg. 6(7),
July- 2016, pp. 116-125
© 2016, IJARCSSE All Rights Reserved Page | 125
[7] Prakash, S., Kumar, P., Prasad, B.V.N.S., and Gupta, A. (1986): “Pareto optimal solutions of a cost– Time
trade-off bulk transportation problem,”
[8] Purusotham, S., Sundara Murthy, M. (2011): “An Exact Algorithm for Multi- Product Bulk Transportation
Problem”, International Journal on Computer Science and Engineering, Vol. 3(9), 3222-3236.
[9] Rautman .C.A. Ried R.A and Ryder E.E(1993) operational research 41,459-469.
[10] Sangeetham Prasad, Suresh Babu C, Purusotham S and Sundara Murthy M: A Variant Constrained Bulk
Transshipment Problem - Global Journal of Pure and Applied Mathematics ISSN 0973-1768 Volume 4 (2014)
[11] SobhanBabu, K. And Sundara Murthy, M. (2010): “An Efficient Algorithm for Variant Bulk Transportation
Problem,” International Journal of Engineering Science and Technology, vol: 2(7), pp: 2595-2600.
[12] Srinivasan, V and Thompson, G.L. (1973). An algorithm for assigning uses to sources in a special class of
transportation problem. Operations Research, 21, 284-295.
[13] Sundara Murthy, M - (1976). A Bulk Transportation Problem Opsearch, 13, pp. 143-155.
[14] Suresh Babu, C. (2013): “Lexi-Search Exact Algorithm for Variant TSP, Bulk TP and Minimum Spanning
Models” Ph.D thesis, S. V. University, Tirupati, India.
[15] VanitaVerma, Puri, M.C. (1996): “A branch and bound procedure for cost minimizing bulk transportation
problem,”Opsearch, Vol. 33(3), 145-161.
[16] Vidhyullatha, A. (2012): “Pattern Recognition Lexi – Search Exact Algorithms for Variant ASP and Bulk TP
Models”, a Ph. D. thesis, Sri Venkateswara University, Tirupati, India.