Network Design Optimization for Fiber Optic Revenue Maximization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current network design and optimization methods for fiber optic networks struggle to maximize profit by efficiently placing fiber optic cables, as they fail to effectively balance revenue generation and cost minimization in the Prize-Collecting Steiner Tree Problem in Graphs (PCSPG), a key challenge in network planning and design.

Innovation Solution

The proposed solution involves formulating the network optimization problem as a Prize-Collecting Steiner Tree Problem in Graphs (PCSPG) with generalized subtour elimination constraint (GSEC) inequalities, using a subgradient method and Lagrangian relaxation to find near-optimal or optimal solutions, and applying a Minkoff algorithm with local search to improve feasible solutions, thereby determining the optimal placement of fiber optic cables.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional network design methods are used, then implementation is simpler, but profit maximization is insufficient

Engineering Contradiction:
Improveprofit maximization accuracyVSAvoidalgorithm complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the complex PCSPG optimization problem into manageable components by formulating it as a mixed-integer linear program with specific objective functions and constraints. The network design problem is divided into discrete decision variables (whether to build each link) that can be optimized systematically, transforming an intractable complex problem into a structured optimization model that can be solved with appropriate algorithms.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the approach from traditional heuristic methods to a mathematical optimization framework with defined objective functions and constraints. By parameterizing the problem with revenue parameters (pi), cost parameters (cij), and profit parameters (di), the solution transforms qualitative design decisions into quantitative optimization that can be solved systematically to achieve profit maximization.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If comprehensive network optimization is performed, then profit is maximized, but computational resources are consumed

Engineering Contradiction:
Improvesolution qualityVSAvoidcomputational resources
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent applies preliminary action by performing Lagrangian relaxation on selected constraints before solving the full optimization problem. This preliminary step transforms the complex constrained optimization into a simpler form that can be solved more efficiently, obtaining good approximate solutions without requiring exhaustive computation of all possible network configurations.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements partial action by using heuristics that explore only a subset of the complete solution space. Rather than exhaustively evaluating all possible network designs, the optimization algorithm focuses on promising regions of the solution space, achieving satisfactory profit maximization with reduced computational resource consumption.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS7978629B2Method for network design to maximize difference of revenue and network cost
Publication Date: 2011.07.12 AT&T INTELLECTUAL PROPERTY I L P
  • US7978629B2 patent drawing
  • US7978629B2 patent drawing
  • US7978629B2 patent drawing

AI summary

A method determines an optimal or near-optimal conveyance network layout in which revenue from serviced customer locations is maximized while the cost of installing and/or maintaining the conveyance is minimized. The conveyance may, for example, be a fiber optic telecommunications cable or a power or utility distribution system. Algorithms in the method generate primal and dual bounds in a Prize-Collecting Steiner Tree Problem in Graphs (PCSPG). Those algorithms originate from a Lagrangian Non-Delayed Relax-and-Cut (NDRC) based approach and incorporate ingredients such as a new PCSPG reduction test, an effective Local Search procedure and a modification in the NDRC framework that allows additional reductions in duality gaps to be attained.