Information Processing Apparatus for Two-Way One-Hot Constraint Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for solving two-way one-hot constraint optimization problems, such as the capacitated vehicle routing problem, face inefficiencies in calculation time and difficulty in reaching optimal solutions due to increased redundant depots and complex weight coefficient adjustments.
Innovation Solution
An information processing apparatus that uses a processor to determine candidate sets by exchanging values of redundant variables and performs searches based on an Ising type energy function, reducing the number of iterations required to find a solution by excluding unnecessary exchanges between redundant depots.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional MCMC method is used to solve two-way one-hot constraint optimization problems, then the problem can be solved by probabilistic state transition, but the calculation time increases exponentially and the number of iterations increases due to redundant variable exchanges
Solution Approach 1:
The patent extracts and eliminates redundant variable exchanges from the search process. By identifying that exchanges between redundant depots (depots with identical characteristics) produce equivalent solutions, the method removes these unnecessary operations from the MCMC search, thereby reducing iteration count and calculation time while maintaining solution accuracy.
Solution Approach 2:
The patent performs preliminary classification of depots into redundant and non-redundant groups before the optimization search begins. This preliminary action allows the system to pre-determine which variable exchanges are meaningless, enabling the MCMC method to skip these exchanges during the search process and focus only on productive transitions.
2Reliability
If redundant depots are added to handle two-way one-hot constraints, then the constraint satisfaction is improved, but the number of iterations increases and calculation efficiency decreases
Solution Approach 1:
The patent segments the set of all depots into two distinct groups: redundant depots (which have identical characteristics and produce equivalent solutions when exchanged) and non-redundant depots (which have unique characteristics). This segmentation allows the optimization algorithm to treat different types of depots differently, applying constraint satisfaction rules only where necessary and avoiding redundant operations.
Solution Approach 2:
Instead of applying full constraint checking to all possible variable exchanges, the patent applies constraint satisfaction rules partially - only to exchanges involving non-redundant depots. This partial action approach maintains constraint satisfaction where needed while avoiding excessive computation on redundant exchanges that would not improve the solution.
3Reliability
If weight coefficient adjustment is used to handle constraints, then the constraint terms can be balanced, but the complexity of the optimization process increases
Solution Approach 1:
The patent extracts the constraint satisfaction logic from the weight coefficient adjustment mechanism and implements it as a separate preliminary classification step. By determining redundant vs. non-redundant depot relationships before optimization begins, the system removes the need for complex dynamic weight adjustments during the search process, thereby reducing optimization complexity while maintaining constraint balance.
Data Source
AI summary
An information processing apparatus of obtaining a solution to a problem in which constraint that includes two-way one-hot constraint is given, by using an objective function that has a predetermined number of variables, the information processing apparatus including: a processor configured to exclude, when determining sets of candidates that have values which are to be changed based on the constraint from expanded variables obtained by adding a plurality of redundant variables to the predetermined number of variables, one of the sets of the candidates in which values of the redundant variables are exchanged with each other; and a search unit that changes, according to the constraint, the values of the sets of the candidates, and performs a search based on the objective function, wherein the processor is further configured to determine the solution on the basis of a result of the search by the search unit.