In this paper a multi-objective (MO) approach was developed to support bidders in making their bidding decisions in combinatorial auctions. To this end, the fast elitist multi-objective genetic algorithm (NSGA-II) was implemented and tested. The system was developed for a specific combinatorial auction: the clock proxy auction. In the simulated scenario, a government decides to allocate spectrum licenses among telecommunication industry companies using this auction. The MO decision system reports a Pareto frontier to each bidder with the best strategy for a given average profit or probability to win. Hence, bidders can select their final bids according to their risk attitude (risk seeking, risk neutral or risk averse).