Arindam Dey

The Fuzzy Robust Graph Coloring Problem

Fuzzy graph model can represent a complex, imprecise and uncertain problem, where classical graph model may fail. In this paper, we propose a fuzzy graph model to represent the examination scheduling problem of a university and introduce a genetic algorithm based method to find the robust solution of the scheduling problem that remains feasible and optimal or close to optimal for all scenarios of the input data. Fuzzy graph coloring method is used to compute the minimum number of days to schedule the examination. But problem arises if after the examination schedule is published, some students choose new courses in such a way that it makes the schedule invalid. We call this problem as fuzzy robust coloring problem (FRCP). We find the expression for robustness and based on its value, robust solution of the examination schedule is obtained. The concept of fuzzy probability of fuzzy event is used in the expression of robustness, which in turn, is used for fitness evaluation in genetic algorithm. Each chromosome in the genetic algorithm, used for FRCP, represents a coloring function. The validity of the coloring function is checked keeping the number of colors fixed. Fuzzy graphs with different number of nodes are used to show the effectiveness of the proposed method.

Computing the Shortest path with Words

Computing with words is a soft computing technique to solve the decision-making problem with the information described in natural language. In this paper, we proposed a generalised Diskrtras algorithm for finding shortest paths from a source node to every other node in a fuzzy graph where the arc lengths are represented by some natural words taken from natural language. Human being generally describes the arc lengths of any shortest path problem in real life by the terms small, large, some, etc. term which do not supply any natural numbers or fuzzy numbers. We describe those terms as words. Here, we use interval type 2 fuzzy set (IT2FS) to capture the uncertainty of the words. A perceptual computer model is introduced to determine the rank of the shortest path which is a collection of words. A numerical example of transportation network is used to illustrate the effectiveness of the proposed method.

Robust and Minimum Spanning Tree in Fuzzy Environment

This paper proposes an algorithm to find the fuzzy minimum spanning tree (FMST) of an undirected weighted fuzzy graph, in which mixed fuzzy numbers, either triangular or trapezoidal, are used to represent the lengths/costs of the arcs. In the proposed algorithm, we incorporate the uncertainty in Kruskals algorithm for MST using fuzzy number as arc length. The concept of possibility programming is used to compare between the fuzzy number (i.e., costs of arcs) and addition operation of fuzzy numbers is used to find the cost of the spanning tree. We also investigate the robust version of the FMST problem. We define two measures for robustness of an FMST: absolute robustness and relative robustness. We characterize the fuzzy worst case scenarios for a given fuzzy spanning tree for both the measures. The corresponding fuzzy robust spanning trees are respectively defined as absolute robust fuzzy spanning tree (ARFST) and relative robust fuzzy spanning tree (RRFST). We extend Kruskals algorithm to compute the ARFST and RRFST in fuzzy environment. An example of fuzzy graph is used to illustrate the effectiveness of the proposed methods.