Combinatorial Optimization Search Using Probabilistic Path Relinking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional path relinking methods for combinatorial optimization problems often generate initial solutions that are too similar to the best solution, limiting the search space and hindering the discovery of better solutions.
Innovation Solution
A data processing apparatus and method that determines higher-ranked solutions based on evaluation function values, adjusts selection probabilities to favor solutions closer to the best solution, and generates new solutions using path relinking with randomly selected diverse solutions to expand the search space.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the best solution is selected from the solution pool to generate an initial solution using path relinking, then the initial solution is similar to the best solution, but the search space is limited and diversity is reduced
Solution Approach 1:
The patent changes the selection parameter from deterministic (always selecting the best solution) to probabilistic (selecting from top-k solutions with probability proportional to their evaluation function values). This allows the system to balance between exploiting high-quality solutions and exploring diverse regions of the solution space, thereby maintaining both solution quality and diversity.
Solution Approach 2:
The patent introduces dynamic selection of the first solution by making the selection process adaptive rather than static. The selection probability dynamically adjusts based on the evaluation function values of different solutions, enabling the system to adaptively balance exploration and exploitation during the optimization process.
2Adaptability or versatility
If constraints are temporarily relaxed during the search process to ensure diversity, then solution diversity increases, but search efficiency may be reduced
Solution Approach 1:
The patent performs preliminary action by pre-selecting a set of high-quality candidate solutions (top-k solutions) from the solution pool before initiating the path relinking process. This preliminary selection ensures that the subsequent search starts from promising regions, maintaining search efficiency while still allowing diversity through probabilistic selection among the pre-selected candidates.
Data Source
AI summary
A processing unit determines, from a plurality of solutions to a combinatorial optimization problem stored in a storage unit, a plurality of higher-ranked solutions based on evaluation function values of the plurality of solutions, determines, for each of the plurality of higher-ranked solutions, a selection probability such that higher-ranked solutions having evaluation function values closer to that of the highest-ranked solution among the plurality of higher-ranked solutions are more likely to be selected, selects a first solution from the plurality of higher-ranked solutions according to the selection probabilities, selects a second solution different from the first solution, from the plurality of solutions, generates a third solution with a path relinking method using the selected first solution and second solution, and performs a solution search on the combinatorial optimization problem using the third solution as an initial solution.