Towards Detection of Bottlenecks in Modular Systems

TL;DR

This paper explores various heuristic methods for detecting bottlenecks in modular systems, including quality management, critical component analysis, and predictive forecasting, supported by numerical examples.

Contribution

It introduces heuristic approaches for identifying system bottlenecks in modular systems, covering multiple detection problems and providing practical numerical illustrations.

Findings

Heuristic methods effectively identify critical system components.

Multiple detection approaches are applicable to different system bottleneck scenarios.

Numerical examples demonstrate the practical utility of the proposed methods.

Abstract

The paper describes some basic approaches to detection of bottlenecks in composite (modular) systems. The following basic system bottlenecks detection problems are examined: (1) traditional quality management approaches (Pareto chart based method, multicriteria analysis as selection of Pareto-efficient points, and/or multicriteria ranking), (2) selection of critical system elements (critical components/modules, critical component interconnection), (3) selection of interconnected system components as composite system faults (via clique-based fusion), (4) critical elements (e.g., nodes) in networks, and (5) predictive detection of system bottlenecks (detection of system components based on forecasting of their parameters). Here, heuristic solving schemes are used. Numerical examples illustrate the approaches.