Pricing-Allocation Bi-level Model in Combinatorial Auctions for Full Truckload Transportation Procurement

The auction-based Transportation Procurement Problem (TPP) involves solving the Bid Generation Problem (BGP) for carriers and the Winner Determination Problem (WDP) for shippers. Combinatorial auctions [1] allow a single bid of distinct lanes (i.e. a bundle), a lane is defined as a pick-up/delivery location pair. Combining distinct lanes into a single bid enables carriers to propose attractive prices, which will also lead to procurement cost reduction for shippers. In BGP, carriers place their bids (prices) for the most profitable bundles; in WDP, shippers allocate bundles to carriers based on their bids. To address this problem, we assume a market place representing shippers uses combinatorial auctions to allocate lanes to carriers. We study the Pricing-Allocation Problem (i.e. price-based BGP-WDP) under combinatorial auctions specially for full truckload transportation procurement. The BGP and WDP are two of the main auction phases in the combinatorial auctions. Even if they share strong relationship, it has not been investigated so far in the literature [3]. Our work is the first to merge BGP and WDP in a single bi-level formulation to model more accurate this interaction from a practical point of view.