An exact and two heuristic strategies for truthful bidding in combinatorial transport auctions
This work addresses the challenge for freight carriers to participate effectively in combinatorial auctions by reducing computational burden while maintaining high sales potential, representing an incremental improvement in auction bidding strategies.
The paper tackles the problem of enabling freight carriers to bid efficiently in combinatorial transport auctions by proposing an exact strategy and two heuristic strategies for bidding on request subsets, with the heuristics achieving 91% and 81% of available sales potential while reducing bundle bids by 64% and 96% compared to the exact method.
To support a freight carrier in a combinatorial transport auction, we proposes an exact and two heuristic strategies for bidding on subsets of requests. The exact bidding strategy is based on the concept of elementary request combinations. We show that it is sufficient and necessary for a carrier to bid on each elementary request combination in order to guarantee the same result as bidding on each element of the powerset of the set of tendered requests. Both heuristic bidding strategies identify promising request combinations. For this, pairwise synergies based on saving values as well as the capacitated p-median problem are used. The bidding strategies are evaluated by a computational study that simulates an auction. It is based on 174 benchmark instances and therefore easily extendable by other researchers. On average, the two heuristic strategies achieve 91 percent and 81 percent of the available sales potential while generating 36 and only 4 percent of the bundle bids of the exact strategy. Therefore, the proposed bidding strategies help a carrier to increase her chance to win and at the same time reduce the computational burden to participate in a combinatorial transport auction.