Hub Location Problems (HLPs), belonging to the field of location theory, have been area of much research over the past two decades. This is due, in large measure, to the applications of hub and spoke networks in practice. Among the most classical versions of HLPs are p-hub location problems (p-HLPs), p-hub location problems are one of the most well studied variants of hub location literature. The primary goal of these models is to allocate p hub facilities in a hub and spoke network so as to concentrate flows (demands) to benefit from economies of scale in cost of transportation. The application of p-hub networks extends beyond the field of telecommunication and includes air freight systems, postal delivery systems and airline industries and several transportation related systems. p-HLPs constitute a challenging class of HLPs and are known to be NP-hard. Several solution approaches have been developed from exact solutions using integer programming techniques to the development of metaheuristics. Even though metaheuristic algorithms cannot guarantee optimality, given complexity of large scale HLPs, they are being used for solving these problems. In this thesis, we focus on the multiple allocation uncapacitated p-hub location problem. Four solution algorithms will be proposed to this problem for solving the Australian Postal (AP) data instances. We start with a very simple algorithm and continue with more complicated one in order to present an efficient high quality feasible solution and to assess the impact of the quality of initial feasible solution on local improvement phase.  Computational results from the different algorithms were compared to exact solutions to track the efficiency of the proposed algorithms.