New method enhances AI-driven project scheduling heuristics

Researchers have developed a new method using surrogate-assisted genetic programming (GP) to improve heuristic rules for dynamic multi-mode resource-constrained project scheduling. This approach aims to reduce the computational cost of fitness evaluations in GP by using phenotypic characterization (PC) to estimate performance. The study explored three PC encoding schemes—priority-value, rank, and binary—and found that binary encoding combined with Euclidean distance was the most effective for guiding the evolutionary search. The findings suggest that controlling redundancy and preserving behavioral diversity are crucial for successful surrogate-assisted GP, in addition to identifying promising offspring. AI

IMPACT Enhances AI's capability in complex scheduling tasks, potentially improving efficiency in project management.

RANK_REASON The cluster contains a research paper detailing a novel methodology for optimizing scheduling heuristics using genetic programming. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• CatalyzeX Code Finder for Papers
• DagsHub
• dynamic multi-mode project scheduling
• Euclidean distance
• genetic programming
• Gotit.pub
• Hugging Face
• priority-value encoding
• rank encoding
• ScienceCast

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →