Portfolio Bid Optimization Algorithm Reducing Runtime and Memory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face inefficiencies in processing asset-level bids within a portfolio due to increasing runtime and memory usage as the number of assets and bids grow, particularly when maximum and minimum constraints are applied, leading to sub-optimal computer performance and unprofitable bidding results.
Innovation Solution
The implementation of an optimized algorithm that reduces the iteration space by combining unconditional bids, excluding constrained bids, and reordering asset arrays to prioritize high bids within the constraints, allowing for efficient determination of high aggregate bid combinations while ensuring compliance with minimum and maximum purchase limits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the system processes all asset-level bids with maximum and minimum constraints using conventional algorithms, then comprehensive bidding results are achieved, but runtime grows to O(MN) exceeding cache size and taking significant memory
Solution Approach 1:
The patent segments the bidding problem by separating unconditional bids from constrained bids, and further divides constrained bids into those with only maximum constraints, only minimum constraints, or both. This segmentation allows the system to process different bid types using optimized algorithms tailored to their specific constraint requirements, reducing the overall iteration space from O(MN) to a more efficient complexity level while maintaining comprehensive bidding result accuracy.
Solution Approach 2:
The patent applies partial action by selectively processing constrained bids only when necessary, rather than uniformly processing all bids. The system identifies and processes unconditional bids separately from constrained bids, applying different processing strategies to each subset. This partial processing approach reduces the effective iteration space while ensuring that all necessary bidding scenarios are covered through targeted algorithmic treatment of specific bid subsets.
2Reliability
If the system iterates over all N×M bid combinations to ensure optimal bidding results, then complete bid evaluation is achieved, but memory usage becomes significant and cache size is exceeded
Solution Approach 1:
The patent extracts and separates constrained bids from the main bid processing flow, handling them through a specialized algorithm that operates on a reduced iteration space. By taking out the constrained bid subset and processing it separately with targeted constraint-checking logic, the system avoids the need to allocate memory for all possible N×M combinations simultaneously, thereby reducing peak memory usage while maintaining complete bid evaluation through the extracted constrained bid processing mechanism.
Solution Approach 2:
The patent applies preliminary action by pre-classifying bids into unconditional and constrained categories before the main evaluation process. The system pre-identifies which bids have maximum constraints, minimum constraints, or both, and organizes them in advance. This preliminary classification enables the main algorithm to process bids more efficiently by avoiding redundant constraint checking across all combinations, thereby reducing memory requirements while ensuring complete bid evaluation.
3Reliability
If the system applies both maximum and minimum constraints to bid arrays, then bidding compliance is ensured, but the iteration space increases and processing efficiency decreases
Solution Approach 1:
The patent segments constraint compliance checking by handling maximum constraints, minimum constraints, and combined constraints as separate processing stages. The algorithm first processes unconditional bids, then separately handles constrained bids with maximum constraints, minimum constraints, or both. This segmentation allows each constraint type to be validated through optimized logic, ensuring comprehensive compliance while reducing processing efficiency penalties by avoiding redundant constraint checking across all bid combinations.
Solution Approach 2:
The patent changes the parameter approach by transforming the constraint validation process from a uniform O(MN) iteration to a parameter-driven approach where the algorithm adapts its behavior based on the specific constraint parameters of each bid array. By detecting whether a bid array has maximum constraints, minimum constraints, or both, the system adjusts its processing strategy accordingly, changing the effective iteration space and validation logic to match the specific parameter requirements, thereby maintaining compliance while improving processing efficiency.
Data Source
AI summary
An Optimization Server reduces M×N iteration space using various techniques to enhance computer performance by improving computer runtime and memory usage. The Optimization Server determines an aggregate combination of bids that results in high proceeds on individual assets of a portfolio subject to minimum and maximum constraints imposed by a bidder across a plurality of individual asset bids. The Optimization Server stores in a memory a portfolio array including M number of elements. Each portfolio array element corresponds to a bid array of bids on the plurality of individual assets of the portfolio. Each bid array includes N number of elements and each bid array element corresponds to a bid amount on one of the plurality of individual assets in the portfolio. The bid array also includes a corresponding minimum constraint and a corresponding maximum constraint across all of the individual asset-level bids placed on the portfolio.