K-Shortest Path Routing With Diverse Inclusion-Constraint Ports
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing path computation methods, particularly Yen's algorithm, fail to provide diverse k-shortest paths when an inclusion constraint is applied, often resulting in all paths sharing the same vertex at the inclusion point, which is undesirable for network flexibility.
Innovation Solution
A modified Yen's algorithm that introduces a dummy destination vertex connected to all ports of the inclusion constraint, adjusting edge costs to prioritize diversity over traditional cost-based paths, ensuring that k-shortest paths include multiple vertices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If Yen's algorithm is used to compute k-shortest paths with an inclusion constraint, then the computation efficiency is maintained, but the diversity of paths at the inclusion constraint vertex is lost (all paths share the same port)
Solution Approach 1:
The patent modifies the cost parameters of edges connected to the inclusion constraint vertex dynamically. After each path is found, the cost of the port used at the inclusion constraint is increased, forcing subsequent paths to select different ports. This parameter change resolves the contradiction by maintaining computational efficiency through algorithmic modification while achieving path diversity through cost adjustment.
Solution Approach 2:
The patent creates a modified copy of the original graph where edge costs are adjusted based on previously found paths. This copied graph structure allows the algorithm to explore diverse paths while maintaining the efficiency of the original Yen's algorithm framework, thus resolving the contradiction between efficiency and diversity.
2Manufacturing precision
If traditional k-shortest path algorithms are used without modification, then the shortest paths are found based on cost, but the resulting paths lack diversity at inclusion constraint vertices
Solution Approach 1:
The patent introduces dynamic cost adjustment to the static Yen's algorithm. The edge costs at the inclusion constraint vertex are not fixed but change dynamically as paths are found. This dynamic modification ensures both path optimality (by maintaining the shortest path property) and port diversity (by forcing different ports to be selected in subsequent iterations).
Solution Approach 2:
The patent implements a feedback mechanism where the selection of ports in previously found paths influences the cost structure for subsequent path searches. This feedback loop ensures that diverse ports are selected while maintaining optimal path properties, resolving the contradiction between optimality and diversity.
Data Source
AI summary
Systems and methods for diversifying k-shortest path computation results for multiple vertices as destination or as inclusion constraint include receiving a request for k paths, in a network from a source port at a source node to a node, wherein the node includes a plurality of ports; representing the network as a graph with vertices representing ports, edges representing connections between the ports, and with weights assigned to each of the edges; assigning a dummy destination vertex to connect to each of the plurality of ports and associated edges between each of the plurality of ports and the dummy destination vertex having a same weight; determining up to the k paths between the source port and the dummy destination vertex, and, subsequent to the determining each of the k paths, assigning a corresponding edge of the associated edges in each of the k paths a predetermined value.