An enhanced exact procedure for the absolute robust shortest path problem Maria Elena Bruni and Francesca Guerriero Dipartimento di Elettronica, Informatica, Sistemistica, Universita ` della Calabria, Via P. Bucci 41C, 87030 Rende (Cosenza), Italy E-mail: mebruni@deis.unical.it [Bruni]; guerriero@deis.unical.it [Guerriero] Received 16 August 2008; received in revised form 19 December 2008; accepted 5 February 2009 Abstract The aim of this paper is to investigate the use of heuristic information to efficiently solve to optimality the robust shortest path problem. Starting from the exact algorithm proposed by Murty and Her, we describe how this algorithm can be enhanced by using heuristic rules and evaluation functions to guide the search. The efficiency of the proposed enhanced approach is tested over a range of random generated instances. Our computational results indicate that the use of heuristic criteria is able to speed up considerably the search and that the enhanced exact solution method outperforms the state-of-the-art algorithm proposed by Murty and Her in most of the instances. Keywords: robust optimization; scenarios; heuristic search 1. Introduction The problem of finding the shortest path, from a specified source node to a destination node, is a fundamental problem in combinatorial optimization that has widespread applicability in many contexts ranging from route selection to routing algorithms in computer networks (Pio´ ro and Medhi, 2004; Sanso´ and Soriano, 1999). Because of the importance of this problem, both from a theoretical and a practical point of view, considerable attention has been devoted to it, along with the investigation of structural properties and the development of tailored algorithms for interesting variants of the problem. These variants involve, in most of the cases, distinguishing properties as to whether the network is weighted or unweighted, directed or undirected, and the static or the dynamic nature of the problem. Scant attention has been paid to the case of uncertain data, arising from lack of information, approximate forecasts or measurements errors, which render the search of optimal solutions inappropriate and require the use of robustness analysis. The robust shortest path problem has Intl. Trans. in Op. Res. 17 (2010) 207–220 DOI: 10.1111/j.1475-3995.2009.00702.x INTERNATIONAL TRANSACTIONS IN OPERATIONAL RESEARCH r 2009 The Authors. Journal compilation r 2009 International Federation of Operational Research Societies Published by Blackwell Publishing, 9600 Garsington Road, Oxford, OX4 2DQ, UK and 350 Main St, Malden, MA 02148, USA.