Episode 7: Challenges and the Future of Production Scheduling with Dr. Prasad Velaga

November 15, 2024

Duration: 38 mins 47 secs

Your Host

Zeeshan Zia

Dr. Zeeshan Zia is the CEO of Retrocausal. With a PhD in computer vision, Zeeshan is on a mission to transform manufacturing with AI Copilots.

Special Guest

Dr. Prasad Velaga

Dr. Prasad Velaga is a seasoned expert in production scheduling and manufacturing operations, with over 20 years as President of Optisol, where he developed advanced software for complex scheduling environments. Formerly a research scientist with projects for the Department of Defense and NASA, Dr. Velaga also served as a professor at leading institutions, including Washington State University and Texas A&M. Known for his thought leadership on lean manufacturing and scheduling methodologies, Dr. Velaga is a respected voice in the manufacturing and operations field.

November 15, 2024

Duration: 38 mins 47 secs

About this Episode

Join us for an in-depth discussion with Dr. Prasad Velaga, President of Optisol, as he examines production scheduling challenges, methodological limitations, and the real potential of AI in manufacturing operations. Learn why traditional approaches often fall short and what truly works on the shop floor.

Key Takeaways:

Challenges in Production Scheduling:

Drag-and-Drop Limitations: Current scheduling tools rely too heavily on manual interventions, masking weak underlying algorithms and creating shop floor inefficiencies.
Kanban and Pull System Limitations: These traditional systems struggle in complex, high-mix environments where demand is unpredictable, and product flows vary significantly.

Limitations of Traditional Project Management Methods:

Critical Path Method (CPM) Weaknesses: CPM ignores crucial material and resource constraints, focusing solely on precedence relations.
Theory of Constraints (TOC): While theoretically sound, TOC proves too complex for practical implementation in real-world job shops.

Operations Research and Practical Constraints:

Academic Limitations: Research-driven optimization algorithms often fail in practical applications. Simple, proprietary heuristics prove more effective.
Scope for Improvement: High-mix, job-shop environments need better scheduling methodologies that address practical constraints.

AI and Scheduling:

Role of AI in Manufacturing: AI shows promise for quality improvements but faces limitations in scheduling optimization due to data requirements and resource constraints.
Potential in Data-Heavy Tasks: AI could complement, but not replace, traditional scheduling algorithms in data-rich areas.

Real-World Insights and Industry Skepticism:

Adoption and ROI Challenges: Small job shops struggle with implementing scientific scheduling solutions due to budget constraints and implementation complexities.
A Call for Openness to Emerging Tech: Manufacturers should explore new technologies pragmatically, focusing on realistic process improvements.

This episode offers both technical depth and practical wisdom for manufacturing professionals seeking to improve their scheduling systems.