A Scheduling Framework for Green Supplier Selection and Dynamic Order Allocation: A Case Study in the Textile Industry

Abstract

This study introduces an innovative framework for order allocation and scheduling in the textile industry, addressing the complex challenges of multi-objective production planning across diverse processes, multiple outsourcing entities, and interconnected production departments. The proposed approach uses advanced genetic algorithms to deliver efficient and sustainable scheduling solutions, integrating a state-of-the-art production process simulator. A core innovation of this research is a novel chromosome representation, complemented by refined genetic operators and a heuristic pruning and selection mechanism. Green supplier criteria are integrated through a lexicographic scoring mechanism that governs allocation tie-breaking, ensuring that sustainability preferences are structurally embedded in the decision process without displacing primary economic objectives. The production process simulator is designed to replicate real-world industrial dynamics, significantly boosting the model’s accuracy and practical applicability. The study pioneers advanced methodologies, including a gendered genetic algorithm and a hybrid genetic algorithm, specifically developed to optimize scheduling in multi-supply chain and open shop environments. These techniques improve computational efficiency and promote sustainable practices by optimizing resources, reducing waste and minimizing production costs. Validation using real industrial data underscores the model’s capability to outperform conventional practices, demonstrating higher performance relative to the company's prior manual scheduling practice, mostly in terms of operational efficiency, cost-effectiveness and sustainability. The framework offers substantial potential for practical implementation, providing robust tools to enhance market competitiveness and support greener and resilient production strategies within the textile industry and beyond.

Article history: Received (September 16, 2025); Revised (May 7, 2026); Accepted (June 4, 2026); Published online (October 5, 2026)