Location-based price estimation and synergy-based bid price optimization in simultaneous truckload transportation auctions

The widespread use of the Internet has made digital freight transport networks an effective marketplace where carriers and shippers come together. These web sites enable the cargo to be transported at a lower cost, and in a faster and safer manner. This leads to the problem of determining the bid price of the truckload carrier in the fast auction-based procurement process. In simultaneous independent transportation auctions, truckload carriers must estimate the probability distribution of competitive carriers' bid prices in order to estimate the possible revenue they will have as a result of the auction. Historical data can be used for this purpose. For this type of estimation problem, the only estimation method in the literature takes into account only the length of the auction lanes. This thesis focuses on the bid determination problem of a full truckload carrier that wants to complement the existing freight networks under a long-term contract with loads from a spot transportation market. Location-based methods that estimate lane-specific parameters that filter historical data for multiple simultaneous independent single lane auctions have been developed. In addition, synergy-based methods that do not require historical data dependence have been developed. In order to measure the effectiveness of the proposed methods, computational experiments have been carried out in the simulated spot transportation market and the results have shown that these methods can increase the profitability of the carriers' profits compared to the previously used method.