IC Global Routing Patching for Lower DRC Violations

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current global routing solutions in IC design often fail to provide high-quality guidance to detailed routing, leading to costly and unpredictable failures due to unacceptably high design rule check violations, and existing methods lack a direct connection between global and detailed routing stages in commercial EDA tools.

Innovation Solution

A hybrid genetic metaheuristic approach that systematically patches problematic areas of global routing solutions using pieces from other solutions, generating superior global routing outcomes that reduce design rule check violations in detailed routing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional global routing methods are used, then the routing process can be completed, but the design rule check violations in detailed routing remain unacceptably high

Engineering Contradiction:
Improvedesign rule check violation rateVSAvoidrouting iteration efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent applies preliminary action by performing global routing optimization before detailed routing begins. The genetic algorithm generates optimized route guides that serve as preliminary solutions, reducing the need for subsequent detailed routing iterations and minimizing design rule check violations from the outset.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback by using design rule check results from detailed routing to evaluate and refine global routing solutions. The genetic algorithm uses this feedback to iteratively improve route guides, creating a closed-loop system that continuously reduces violations while maintaining routing efficiency.

Inventive Principle:
Principle #23Feedback

2Manufacturing precision

If global and detailed routing stages are disconnected, then each stage can be optimized independently, but the overall routing quality deteriorates due to lack of coordination

Engineering Contradiction:
Improverouting solution qualityVSAvoidintegration of routing stages
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent merges global and detailed routing stages by establishing a direct feedback loop between them. The genetic algorithm in global routing uses design rule check results from detailed routing to optimize route guides, creating an integrated system that improves overall routing quality while managing complexity through automated coordination.

Inventive Principle:
Principle #5Merging (Combining)

3Reliability

If standard routing iterations are performed without optimization, then the process remains simple, but the number of design rule check violations remains high

Engineering Contradiction:
Improvedesign rule complianceVSAvoidrouting optimization level
Core Design Contradiction:
ReliabilityVSExtent of automation

Solution Approach 1:

The patent applies self-service by implementing an automated genetic algorithm that independently optimizes global routing solutions. The system evaluates its own performance through design rule check feedback and automatically refines route guides without manual intervention, achieving high design rule compliance through self-directed optimization.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS12572729B2Integrated-circuit global routing method
Publication Date: 2026.03.10 RGT UNIV OF CALIFORNIA
  • US12572729B2 patent drawing
  • US12572729B2 patent drawing
  • US12572729B2 patent drawing

AI summary

A method for automatically creating a global routing solution for an integrated circuit. The method includes generating n original population of GR solutions. In one or more subsequent phases the method generates succeeding populations of GR solutions. The generation of each succeeding population includes determining a plurality of base GR solutions from the current population of GR solutions, determining a plurality of DRC hotspot areas within the plurality of base GR solutions, determining a plurality of patching GR solutions from which patches may be extracted, and hybridizing patching of GR solutions into base GR solutions.