Combinatorial Auction Bidding Language for Winner Determination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Winner determination in combinatorial auctions is NP-hard, making current methods inefficient for auctions involving numerous goods and bids, as they struggle to maximize seller revenue effectively.
Innovation Solution
A method utilizing stochastic local search techniques with Boolean operators to allocate goods among bids, constructing neighboring allocations and comparing values to find high-quality allocations, potentially optimizing revenue through logical connectives in bids.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional winner determination methods are used in combinatorial auctions, then the allocation can be computed, but the computational time becomes excessively long for large-scale auctions
Solution Approach 1:
The patent implements a dynamic allocation process that evolves through iterative reallocation cycles. The system starts with an initial allocation and repeatedly constructs neighboring allocations by reallocating goods between bids, accepting improvements when found. This dynamic search process adapts to the specific bid structure and converges on high-quality allocations efficiently, transforming the static NP-hard optimization into an iterative improvement process that scales well.
Solution Approach 2:
The patent employs partial action by focusing the search on neighboring allocations that differ from the current allocation by small amounts (single good reallocations) rather than exhaustively searching all possible allocations. This partial exploration of the solution space is sufficient to find high-quality allocations without requiring complete enumeration, dramatically reducing computational time while maintaining solution quality.
2Measurement precision
If complex bidding requirements with multiple bids are submitted to reflect complements, then the bidding accuracy improves, but the complexity of winner determination increases
Solution Approach 1:
The patent segments the complex bidding requirements into atomic units by representing each bid as a set of goods with associated Boolean operators. This segmentation allows the system to handle complex requirements systematically by processing individual bid components and their logical relationships, making the overall complex problem manageable through structured decomposition.
Solution Approach 2:
The patent implements feedback mechanisms where the system evaluates the value of neighboring allocations and uses this information to guide the search process. By comparing the values of current and neighboring allocations and accepting improvements, the system adapts its search strategy based on the specific structure of the bids, efficiently navigating through complex bidding requirements to find optimal solutions.
Data Source
AI summary
In a combinatorial auction, a plurality of bids is received each having a plurality of sub bids and Boolean operators logically connecting each pair of sub bids. A current allocation is determined by allocating goods to at least one of the bids and a best allocation is initialized with the current allocation. A neighboring allocation is constructed by reallocating within the current allocation at least one good from at least one bid to another bid. The best allocation is updated with the neighboring allocation when the value of the neighboring allocation is greater than the current value of the best allocation.