Combinatorial Exchange Allocation Method for Truthful Bidding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Combinatorial exchanges, such as forward and reverse auctions, fail to encourage bidders and bid takers to reveal their true valuations, leading to economically inefficient allocations and potential economic losses due to strategic bidding practices.
Innovation Solution
A method that determines an allocation in combinatorial exchanges by inputting bids, forming bid vectors, and associating probability values to optimize exchange objectives and constraints, promoting truthful bidding through participation and truth-promotion constraints, and using optimizing routines like CPLEX or Xpress-MP to select the final allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If bidders bid strategically (not revealing true valuation), then they may achieve better individual outcomes, but the exchange fails to achieve economically efficient allocations and loses potential revenue
Solution Approach 1:
The patent transforms the bidding mechanism by introducing probability values and utility functions that change the parameters of the bidding process. Instead of direct price bidding, bidders submit valuations that are processed through mathematical transformations (probability assignments, utility calculations) to determine allocations. This parameter transformation resolves the contradiction by making truthful bidding the optimal strategy while achieving efficient allocations.
Solution Approach 2:
The patent introduces an intermediary mechanism - the probability-value-based allocation system - that mediates between bidder valuations and final allocations. The bid vector formation, probability assignment, and utility function evaluation act as intermediaries that translate true valuations into optimal allocations without requiring bidders to engage in complex strategic manipulation.
2Productivity
If the exchange uses traditional auction mechanisms, then the process is simple to operate, but bidders cannot reveal their true valuations leading to inefficient allocations
Solution Approach 1:
The patent segments the bidding process into distinct components: bid vector formation (grouping bids), probability value assignment (evaluating bid vectors), utility function calculation (determining bidder satisfaction), and allocation optimization (matching bids to items). This segmentation allows the complex mechanism to process true valuations efficiently while maintaining clarity in each individual step.
Solution Approach 2:
The patent adds dimensional complexity by introducing probability values as a new dimension to the bidding process. Instead of simple price-quantity pairs, bids are evaluated across multiple dimensions (bid vectors, probability weights, utility scores) that collectively enable efficient allocation while managing the complexity through structured mathematical relationships.
3Measurement precision
If bidders reveal their true valuations, then economically efficient allocations can be achieved, but bidders risk receiving unfavorable prices or allocations
Solution Approach 1:
The patent implements feedback mechanisms through utility functions that evaluate bidder satisfaction and probability assignments that reflect allocation likelihoods. Bidders receive feedback about their expected outcomes based on true valuations, allowing them to assess reliability before committing. The system adjusts allocations to provide feedback that truthful bidding leads to favorable or at least acceptable outcomes.
Solution Approach 2:
The patent creates equipotential conditions by designing the allocation mechanism so that truthful bidding becomes the dominant strategy - all bidders operate from the same advantageous position where revealing true valuation maximizes their expected utility. The probability-based system equalizes the playing field by objectively evaluating all bids according to the same mathematical criteria, making truthful revelation reliable for all participants.
Data Source
AI summary
In an apparatus and method of solving a combinatorial exchange, a plurality of candidate allocations of input bids is determined from input bids and a set of bid vectors is formed from the input bids. A probability value, representing how likely the bids of a bid vector will comprise the bids from which a final allocation is determined, is associated with each bid vector. A mapping is then determined between each bid vector and at least one of the candidate allocations based on a set of objectives and a set of constraints. A candidate allocation is selected as a final allocation based on a comparison of another set of bids that define an exchange event bid vector with at least one of the bid vectors.