1 IJCSMS International Journal of Computer Science & Management Studies, Vol. 11, Issue 01, May 2011 ISSN (Online): 2231 –5268 www.ijcsms.com Fuzzy optimization scheduling using coloring fuzzy graph Safar Mohammad noori 1 Department of Mathematic, Shahrood university of technology Shahrood, Republic Islamic of Iran Mohammad.noori@gmail.com Abstract The ( d , f ) extended coloring of a fuzzy graph was defined by the authors in S. Munez, M.TeresaOrtuna, J. Ramirez J. Yanez (2006). In this paper is defined a type of coloring that give the chromatic fuzzy sum for ( d , f ) extended coloring of a fuzzy graph and is expressed a model for this coloring using integer linear programming. In this paper one of the applications of this coloring in exam scheduling problem is studied that the exam primary days have higher graduation and the exam last days have less graduation. For this graduation express a membership function. This coloring, for this membership function present optimal answer, that sum of graduated assigned colors of fuzzy graph is maximum value over all proper the ( d , f ) extended coloring of the fuzzy graph. Keywords: chromatic fuzzy sum, graph theory, optimization, exam scheduling problem, fuzzy set. 1. Introduction Graph coloring is one of the most studied problems of combinatorial optimization .Many problems of practical interest can be modeled as coloring problems. The general form of this application involves forming a graph with nodes representing items of interest. The basic graph coloring problem is to group items in as few groups as possible, subject to the constraint that no incompatible items end up in the same group. Definition1.1. Let G (V , E ) be a graph with a vertex set V and an edge set E . A proper coloring of the vertices of the graph G is assigning different colors to adjacent vertices, a coloring function a mapping c : V N identifying c (i ) as the color of node i V , in such a way that two adjacent nodes cannot share the same color, i.e., c(i ) c( j ) if {i, j} E . The chromatic number (G ) is the smallest number of colors in a proper coloring of G . Definition1.3 A proper coloring c of a graph G is called a best coloring of whenever G c (v ) [1]. vV It is known that computing the chromatic sum of an arbitrary graph is an NP-complete problem [1]. The vertex-strength of the graph G , denoted by s (G), is the smallest integer s such that G is attained using colors {1, 2, . . . , s}. Fuzzy-set theory, introduced by Zadeh [2], is a mathematical tool to handle uncertainties like vagueness, ambiguity, and imprecision in linguistic variables, see also [3, 4]. The first definition of fuzzy graph was proposed by Kaufmann [5], from the fuzzy relations introduced by Zadeh [2]. Although Rosenfeld [6] introduced another elaborated definition, including fuzzy nodes and fuzzy edges, in this paper we deal with graphs with crisp nodes and fuzzy edges, see, for example, [7,8].See also [9] for an overview on some optimization problems in fuzzy graphs. We will use the classical definition of fuzzy set a defined on a non empty set X as the family: {( x, A~ ( x)) / x X } membership function and Where A~ ( x) : X I is A~ ( x) the reflects the ambiguity of the assertion x belongs to A . In classical fuzzy-set theory the set I is usually defined as the interval [0, 1], in such a way that belong to A , A~ ( x) = 0 indicates that A~ ( x) = 1 indicates that x x does not strictly belongs ~ to A , and any intermediate graduation represents the ~ degree in which x could belong to A . In general, the set I can be any ordered set, not necessarily numerical, for instance, I = {null, low, medium, high, total}. Definition 1.2 Let G (V , E ) be a graph with vertex set V and edge E .chromatic sum of graph the minimum sum G, G ,to be c(v) taken over all proper coloring 1. 2 the ( d , f ) extended coloring function of a fuzzy graph vV c of graph G using natural c : V N , c(v) c( w), (v, w) E . number i. e In order to model certain types of problems as coloring Problems, it is necessary to define dissimilarity or distance measure on the color set. It can be done in the following IJCSMS www.ijcsms.com 2 IJCSMS International Journal of Computer Science & Management Studies, Vol. 11, Issue 01, May 2011 ISSN (Online): 2231 –5268 www.ijcsms.com way [10]. Let S be the available color set. Let d be the dissimilarity measure defined by d : S S [0, ) With the following properties: 1 d (r , s) 0 r , s S 2 d (r , s ) 0 r s r , s S Definition 2.1Let vertex set V and fuzzy edges E , Extended Minimum sum ~ This dissimilarity measure d can take into account the incompatibility degree in the sense that the more incompatible two nodes are, the more distant their associated colors. In this way, an extended coloring function is introduced. Given I , the image of the ~ ~ membership function of the fuzzy graph G (V , E ) let f : I [0, ) Be a non-negative, none decreasing (with f ( ) f ( ) , I s.t (1) The above two concepts of dissimilarity measure and scale function allow the following definition. ~ ~ Definition1.3. Given a fuzzy graph G (V , E ) , color set S, a dissimilarity measure d defined on S ~ ~ G (V , E ) is a fuzzy graph with ~ 3 d (r , s ) d ( s, r ) r , s S respect to the order ) and real scale function, i.e. competition on resources, that is, two vertices are adjacent if the corresponding processors cannot run their jobs simultaneously. The objective is to minimize the average response time, or equivalently to minimize the sum of the job completion times. In MSC problem, conflict between vertices is crisp, and then vertices and edges are crisp. For condition that conflict between vertices is fuzzy, using definition of 1-3, extending MSC problem for fuzzy graph we will have bellow definitions: of coloring fuzzy graph G (EMSC) is the smallest sum of colors over all proper the ( d , f ) extended coloring of a fuzzy graph with natural numbers that can be dictated with E G. ~ Definition 2.2The vertex-strength of the fuzzy graph G , ~ s , such that E G is attained using colors {1,2,3,..., s} for short as ~ ~ s .Taking Graph G (V , Eij ) with vertices denote by s (G ) , is the smallest integer {v1 , v2 , v3 ,..., vn } ~ { ij Eij /{i, j} E} and ~ f ( ) , d (r , s ) That w ij ij fuzzy edges r s then modeling of EMSC coloring problem is brought in section of 2-2. and a scale ~ function f , a ( d , f ) extended coloring function of G , 2.2 . A linear programming model for EMSC problem denoted as C d , f , for short as c , is a mapping c : V S With the property: d (c (i ) c ( j )) f ( ij ) i , j V i j 2. Chromatic fuzzy sum the ( d , f ) extended coloring of fuzzy graph assigned colors of all the vertices of G is minimized. The MSC problem has a natural Application in scheduling theory. One of the applications is the problem of resource allocation with constraints imposed by conflicting resource requirements. In a common representation of the distributed resource allocation problem [11, 12], the constraints are given by a conflict graph G, in which the vertices represent processors, and the edges indicate There are not many known polynomial algorithms for computing chromatic sum for some families of graphs. Not surprisingly, there is an efficient linear algorithm for tree [1]. There is also a linear algorithm to compute chromatic sum for unicycle graphs and a cubic algorithm for planar graph [13]. For fuzzy graph haven’t expressed any algorithm yet. Expressed model in (2) hold in all of arbitrary fuzzy and weighted graph. n min c (v ) i 1 s.t i (2) ~ c (vi ) c (v j ) w ij i 1...n 1, j i 1...n; c(vi ), c(v j ) {1,..., s} IJCSMS www.ijcsms.com IJCEM International Journal of Computational Engineering & Management, Vol. 11, January 2011 ISSN (Online): ****-**** www.IJCEM.org c(vi ), c(v j ) {1,..., s} Are assigned colors to vertex, ~ f ( ) in crisp graph define following bellow and w ij ij wij t 4 3. Maximum sum of graduated assigned colors of fuzzy graph Following definition 1.1 that colors is defined with natural number, in normal case this number is crisp and has equal graduation, in cases uncertainties like vagueness, ambiguity, and imprecision use from fuzzy set theory and is graduated number. ~ Definition (SGF): let G (V , E ) be a fuzzy graph then via coloring of fuzzy graph with colors of { c(vi ) / i 1,..., n} , SGF is sum of graduated assigned colors to vertices of graph, so that color set of A is graduated via membership function(3): A That i ~ A {( c(vi ), A~ (c(vi )) n c (vi ) (3) i 1 i {1...n}} ~ A Is a fuzzy set in interval [0, 1] that dictate grade of each assigned color to vertex. In other words if fuzzy graph of then have: c(vi ) {1,2,3,...} be allocated colors to the vertices c(vi ) {1,..., s} be allocated {vi i 1,2,3,..., n} and s is vertex strength, then have: n This model is a model of integer linear programming that can be solved with soft ware. This coloring model can solve EMSC coloring problem. ~ (c (vi )) 1 (c (v ) 1) If colors to the vertices via coloring of EMSC that ~ f ( ) is defined in (1) and in And in fuzzy graph w ij ij ~ Proof: via coloring of C d , f and 1 {i , j }E 0 O .W this paper scale function of f be defined by I n l m h f 0 1 2 3 3 ~ ~ G (V , E ) is colored n ~ SGF (G ) A~ (c(vi ) n n n c(v ) c(v ) 1 c(v ) 1 c(v ) i 1 n i n i 1 i c(vi ) n i 1 i 1 i i 1 i n c (v ) i 1 i n n 1 n c(vi ) 1 n c(vi ) i 1 i 1 n n ( c(vi ) n) c(vi ) i 1 i 1 n n ( c (vi ) n ) c (vi ) i 1 i 1 n n n ( c (vi ) n) c(vi ) i 1 i 1 n n n ( c (v i ) n ) c (vi ) i 1 i 1 n 1 ( c ( v ) 1 ) c ( v ) 1 1 i 1 n 1 (c(v2 ) 1) c(v2 ) i 1 n 1 (c(v3 ) 1) c(v3 ) ... i 1 n 1 ( c ( v ) 1 ) c(vn ) n i 1 n 1 (c(v1 ) 1) c(v1 ) i 1 n 1 ( c ( v ) 1 ) c(v2 ) 2 i 1 i 1 3.1. Theorem: Let G (V , ) be a fuzzy graph, via coloring of EMSC, amount of SGF is Maximum over another proper coloring. IJCEM www.ijcem.org 4 IJCSMS International Journal of Computer Science & Management Studies, Vol. 11, Issue 01, May 2011 ISSN (Online): 2231 –5268 www.ijcsms.com 1 (c(v3 ) 1) 1 (c(vn ) 1) i{1,..., n} ~ A (c(vi )) c(v3 ) ... i 1 n c(vn ) i 1 i{1,..., n} n ~ A (c (vi )) (3) n In the coloring of EMSC, c(v ) is i 1 According to (3), i minimum, then ~ (c (vi )) is maximum i{1,..., n} A ~ (c (vi )) i{1,..., n} A over all proper the (d, f)-extended coloring of the fuzzy graph. Application of This theorem is considered in next subsection 3.2 for scheduling problem with fuzzy time that all of times don’t have equal graduation. 3.2. Application of EMSC problem in exam scheduling problem Following Bullnheimer [14], the examination scheduling problem consists of assigning a number of exams to a number of potential time periods or slots within the examination period, taking into account that students cannot take more than one exam at the same time, as well as several other constraints. A simplified version of this problem can be modeled as a classical coloring problem. Each exam is a node of the graph, and the edges link those incompatible exams in the sense that there is at least one student sharing both of them. Each color is identified with one slot and all equally colored nodes are assigned to the associated slot. If incompatibility of exams is characterized by distance between examination days and graduation of days of exam is equal so accompanying coloring of C can be amount and an Upper bound for time has lowest graduation so therefore coloring EMSC give us value maximum of i{1,..., n} ~ A (c (v i )) . Therefore via coloring of ESMC obtain that sum of times for exam that days are valued with membership function (3) is value maximum over another coloring. Bring two examples to compare the coloring of ESMC and other coloring for amount of SGF. Example 3.1There are five exams {A, B, C, D, E} and, taking into account their different characteristics (difficulty, the implied specialties or profiles, etc.), the incompatibility between them is valued following way: n if they are compatible, and l, m and h if the incompatibility degree is, respectively, low, medium and high. This problem can be modeled by means of the fuzzy graph ~ G (V , ) where V = {A, B, C, D, E} m A a B t h m t C l E D h Fig1. Fuzzy graph ~ G for Example 3.2 d, f achieved optimal time period of exam days. Now if the graduation the examination days not equal and this graduation be characterized by membership function (3), considering the colors as exam days and vertices as examinations, according to theorem (3.1) Coloring of EMSC Gives maximum amount of fuzzy time sum of the examination this means that Sum of the days that allocated for exams will be minimal And amount of SGF is maximum. In this issue membership function (3) means that the primal days have more graduation as compared to last days. Conditions imposed by this membership function (3) so that the first time has the highest graduation and the final m m t h m n t n t h l n m n l h t n n h This fuzzy graph is depicted in Fig.1. Via coloring of EMSC have: IJCSMS www.ijcsms.com IJCEM International Journal of Computational Engineering & Management, Vol. 11, January 2011 ISSN (Online): ****-**** www.IJCEM.org G 16 , s 6 (c(v )) 4.3125 i{1,..., 5} e ~ A m n l - m t m - h t h m n l c( A) 6 c(B) 4 c(C) 1 c(D) 4 c(E) 1 E 5 i Then of EMSC A~ (have: c(vi )) 4.3125 Then via coloring i{1,..., 5} c(v1 ) 3, c(v 2 ) 1 , c(v 3 ) 7 , c(v 4 ) 1 Via coloring of C d , f using exact algorithm in [10] will have: c( A) 1 c(B) 5 c(C) 6 c(D) 3 c(E) 6 6 then i{1,..., 5} ~ A (c(vi )) 4.142857 Taking the above example the result that if valuation of exam days is graduated according to (3) then the best schedule this is that exam A be in first day and exam B be in fifth day of exam and exam C,E be in third day of exam and exam D be in sixth day of exam. Coloring of EMSC schedule the five exams with these constrains and gives sum maximum of graduated assigned colors to vertex and coloring of se (G) 7 , E . Then iv G 12 i{1,..., 5} ~ A ~ A A~ (c(vi )) 3.33333 (c(vi )) 4 .142857 iv (c(vi )) 3.3076 And via coloring C d , f and getting chromatic number have: c(v1 ) 1 , c(v 2 ) 3 , c(v 3 ) 6 , c(v 4 ) 3 d , f 6 , iV c(v i ) 13 C d , f give minimum the overall exam period. For above information, minimum of slot is six and optimum schedule for both coloring is the following: ~ SGF (G ) via ~ coloring of ESMC is bigger than amount of SGF (G ) via coloring of C d , f . In this example s and amount of In order to comparing amount of SGF via coloring of EMSC with colors of c (v ) and coloring of C d , f with colors of c (v ) , That C d , f acquires via exact algorithm in[10](that give chromatic number), three different fuzzy graph types with 6,7,8 nodes and for each arbitrary three graph, have been considered. In Table 1 is depicted difference between iv In this example, chromatic number from coloring of Cd , f is equal with vertex strength s from coloring of ESMC. In below example vertex strength is bigger than chromatic number i.e. s . ~ Example3.2. let G (V , ) be a fuzzy graph where V= {A, B, C, D}, I = {n, l, m, h, t}; ~ A (c (vi )) , A~ (c(vi )) , c(vi ), iv c(vi ) that in some coloring despite s , have iv c(vi ) iv c(vi ) . And difference and and relation between , s , improvement between iv iv ~ A iv iv ~ A (c (vi )) , (c(vi )) via coloring of EMSC as compared to coloring of C d , f respectively. depicted in figure 2, figure 3, IJCEM www.ijcem.org 6 IJCSMS International Journal of Computer Science & Management Studies, Vol. 11, Issue 01, May 2011 ISSN (Online): 2231 –5268 www.ijcsms.com Fig 3 Difference of ~ SGF (G ) via EMSC and C d , f ~ Obtaining amount of SGF (G ) for coloring of ESMC is used from model of coloring (2).for all nine fuzzy graph ~ amount of SGF (G ) is bigger or equal than this amount for coloring of C d , f f, this difference between this two coloring represented in fig3 that for all nine graph ~ SGF (G ) bigger or equal than other coloring. 4. Conclusions In this paper, extending MSC problem for fuzzy graph is expressed the concept EMSC problem. And is expressed application of this coloring in exam scheduling problem that examination time is graduated with membership function (3). This means that sum of graduated allocating days for exams, is maximum value over all proper the ( d , f ) extended coloring of the fuzzy graph. ~ Fig 2 Comprising of SGF (G ) via EMSC and Cd , f References [1] E. Kubicka and A. J. Schwenk, An introduction to chromatic sums, Proc. ACM Computer Science Conference (Louisville), ACM Press, New York, 1989, pp. 39–45. [2] Zadeh LA. Similarity relations and fuzzy ordering .information Sciences 1971; 3(2):177-200. [3] Zadeh LA. The concept of a linguistic variable and its application to approximate reasoning—I. Information Sciences 1975; 8(3):199–249. Zadeh LA. The concept of a linguistic variable and its application to approximate reasoning—I. Information Sciences 1975; 8(4):301–57. Zadeh LA. The concept of a linguistic variable and its application to approximate reasoning—I. Information IJCSMS www.ijcsms.com IJCEM International Journal of Computational Engineering & Management, Vol. 11, January 2011 ISSN (Online): ****-**** www.IJCEM.org 7 Sciences 1975;9(1):43–80; [4] Cutello V, Montero J, Yanez J. Structure functions with fuzzy states. Fuzzy Sets and Systems 1996;83(2):189–202. [5] Kaufman A. Introduction a la theorie sous-ensembles flues Elements theoriques de base. Paris: Masson Des ET Cie; 1976. .[6] Rosenfeld A. Fuzzy graphs. In: Zadeh LA, Fu KS, ShimuraM, editors. Fuzzy sets and their applications to cognitive and decision processes. New York: Academic Press; 1975. p. 77– 95. [7] Koczy LT. Fuzzy graphs in the evaluation and optimization of networks. Fuzzy Sets and Systems 1992; 46(3):307–319. [8] Ramirez J. Extensions del problema de coloracion de grafos. PhD thesis. Universidad Complutense de Madrid, Madrid; 2001. [9] Delgado M, Verdegay JL, Vila MA. On valuation and optimization problems in fuzzy graphs: a general approach and some particular cases. ORSA Journal on Computing 1990; 2:74– 83. [10] S.munez,M.TeresaOrtuna, , J. Ramirez J.Yanez “Coloring fuzzy graphs. Omega 33, 211-221 (2005). [11] K. Chandy and J .Misra, the drinking philosopher's problem, ACM Transactions on Programming Languages and Systems 6 (1984), no. 4, 632–646. [12] N. A. Lynch, Upper bounds for static resource allocation in a distributed system, Journal of Computer And System Sciences 23 (1981), no. 2, 254–278. [13]E. Kubicka, Polynomial algorithm for finding chromatic sum for unicycle and outer planar Graphs, to appear in Ars Combinatoria [14] Bullnheimer B,.An examination scheduling model to maximize students time .in :burke E, Ross p, editors Practice and theory of automated timetabling II. Berlin Springer; 1998. p. 78– 91. IJCEM www.ijcem.org