COST-OPTIMIZED AND SUSTAINABLE PROJECT SCHEDULING IN AUTOMOTIVE SUPPLY CHAINS USING A HYBRID SIMULATED ANNEALING–GENETIC ALGORITHM
DOI:
https://doi.org/10.46121/pspc.54.2.33Keywords:
Artificial intelligence; Intellectual capital; Structural equation modeling; AI use; Governance; Digital skills; Higher education. Simulated Annealing; Genetic Algorithm; Project Scheduling; Automotive Supply Chain; Sustainability; Metaheuristic Optimization; Industry 4.0; Deep Reinforcement Learning.Abstract
This paper introduces a hybrid Simulated Annealing–Genetic Algorithm (SA–GA) framework for multi-objective project scheduling in automotive supply chains. The model jointly optimizes makespan, direct cost, and sustainability impact (CO₂ and energy), while embedding resilience testing under realistic disruption scenarios. Applied to an EV battery production dataset, the framework achieves a 22% makespan reduction (41→32 days), 22.7% cost reduction (€7.9M→€6.2M), an 8.5% improvement in resilience under supply shocks, and a 33% decrease in CO₂ and energy consumption compared with a traditional baseline. Comparisons with GA, PSO, classical SA, and a Deep RL benchmark demonstrate that the hybrid approach provides balanced trade-offs, faster runtime (≈3.2s), and higher robustness, making it well-suited for real-time industrial deployment.
