Thesis (Ph.D)-Cairo University, 2025.

Bibliography: pages 132-138.

Optimization is paramount in manufacturing, serving as the cornerstone for maintaining
competitiveness and long-term viability. This thesis addresses the intricate task of
integrating three planning decisions (quantity, quality, and maintenance) within a serial
multistage production system. At its core, this research introduces a comprehensive joint
optimization model, which serves as the main building block for the solution framework
and thoroughly examines the interconnections among these three critical aspects.

The novel joint model is designed to encompass buffer allocation for efficiently managing
stochastic processing times, part quality inspection planning to minimize defects, and
preventive maintenance strategies to mitigate machine deterioration. By integrating these
functions, the manufacturing system achieves enhanced overall performance and
significant cost reductions. The research demonstrates an average cost reduction of 15%
compared to existing models in the literature.

The proposed model is formulated as a stochastic mixed-integer nonlinear programming
(MINLP) problem, characterized by its NP-hard and combinatorial nature. To effectively
address this complex problem, a comprehensive approach is employed, combining a
genetic algorithm (GA) as the primary generative method, augmented with hybrid GA
variations. Additionally, a simulation-based recursive approach serves as the evaluative
technique, both of which collectively form the solution framework.

The model and solution algorithm undergo rigorous verification and validation through
computational experiments, supported by a design of experiments to fine-tune its
parameters. Additionally, a comprehensive discussion of the research findings and their
implications is provided, offering valuable insights for practitioners and decision-makers
aiming to optimize production processes.

This research makes a significant contribution to the existing body of knowledge by
presenting a robust framework. This framework not only enhances performance but also
reduces costs, effectively addressing the interconnected challenges of quantity, quality,
and maintenance in serial multistage production systems.

تقدم الرسالة إطارًا شاملاً للتحسين المشترك في قرارات التخطيط في أنظمة التصنيع التسلسلي، من خلال دمج القرارات المتعلقة بالكمية والجودة والصيانة ضمن إطار عمل تكاملي. يعالج النموذج تخصيص المخازن، وفحص جودة الأجزاء، والصيانة الوقائية بهدف تقليل التكاليف وتحسين كفاءة الإنتاج. تم صياغة النموذج كمشكلة برمجة غير خطية مختلطة الأعداد باستخدام خوارزمية جينية مع تحسينات هجينة ونهج محاكاة تكراري للتقييم. أظهر الإطار انخفاضاً في متوسط التكلفة بنسبة ١٥% مقارنة بالنماذج الحالية. تم التحقق من فعالية الإطار من خلال تجارب حسابية دقيقة، مما يعزز من موثوقيته وقابليته للتطبيق في تحسين أنظمة التصنيع.

Issues also as CD.

Text in English and abstract in Arabic & English.