Towards an analysis of dynamic environments

Although the interest in nature-inspired optimization of dynamic problems has been growing constantly over the past decade, very little has been done to analyze and characterize a changing fitness landscape. However, it would be very helpful for algorithm development to have a better understanding of the nature of fitness changes in dynamic real-world problems. In this paper, we propose a number of measures that can be used to analyze and characterize the dynamism in a problem changing over time. Additionally, we introduce a new dynamic multi-dimensional knapsack problem as a close-to-real-world test problem. Categories and Subject Descriptors: I.2.8 Heuristic Algorithms General Terms: Algorithms.

• Type: Conference
• Year: 2005
• Where: GECCO
• Authors: Jürgen Branke, Erdem Salihoglu, Sima Uyar
• Keywords: Dynamic Problems, Dynamic Real-world Problems, GECCO 2005, Multi-dimensional Knapsack Problem