By Staff Reporter
Manufacturers worldwide must deliver reliable goods using less energy and material despite supply disruptions. Mechanical engineer Debasish Adhikary works across boiler design, industrial maintenance, and AI-assisted manufacturing research. At Williams and Davis Boiler, he develops mechanical designs; at Sam Houston State University, he studied image-based quality monitoring for additive manufacturing. His publications examine manufacturing efficiency, thermal distortion, and resilience. We asked how they could work globally.

Q: What global manufacturing problem deserves the most urgent attention?
Adhikary: The waste we fail to see. A rejected part wastes material, electricity, machine time, labor, and delivery capacity. Across thousands of factories, the impact becomes global. My priority is identifying avoidable losses before they multiply.

Q: How does your work designing boilers shape that thinking?
Adhikary: It keeps engineering grounded. A design must meet requirements and be built, inspected, operated, and maintained safely. I work with three-dimensional models, fabrication drawings, and production teams. A brilliant calculation is not enough if a solution cannot be manufactured reliably.

Q: Why is additive manufacturing such an important part of your research?
Adhikary: Additive manufacturing can produce complex components, but repeatability is difficult. Changes in temperature, material behavior, or processing conditions may create distortion and defects. Printing one successful component differs from delivering hundreds of dependable ones. Repeatability is essential if advanced manufacturing is to become useful worldwide.

Q: What might a solution that works across countries and factories look like?
Adhikary: I envision a framework connecting process settings, thermal behavior, inspection images, and finished-part quality. It should detect rising defect risk early and prompt corrective action. But it must be validated across machines and materials; a model trained in one factory should never be assumed reliable everywhere.

Q: Your research includes physics-informed machine learning. Why combine physics and AI?
Adhikary: Patterns alone can mislead. Manufacturing follows physical rules, especially when heat and deformation are involved. Combining those rules with data may improve predictions of thermal distortion when conditions change. The goal is not AI for its own sake; it is more dependable decisions and fewer failed builds.

Q: Can AI-based quality control solve the problem of scrap?
Adhikary: It can help flag possible defects earlier. At Sam Houston State University, I supported image data for monitoring additively manufactured products. Yet inspection tools must be judged by missed defects, false alarms, and operator usability. AI should support skilled inspectors, not create blind confidence in automated judgments.

Q: What about the energy consumed even when a part passes inspection?
Adhikary: A product can meet specifications while consuming unnecessary energy. My research on thermal management and energy-aware optimization considers how production choices shape consumption and waste. We should evaluate energy per acceptable component, not just the number of parts made.

Q: How can smaller manufacturers benefit without expensive AI infrastructure?
Adhikary: Start small. Select a recurring defect, measure its cost, and test one improvement. Track scrap, rework, downtime, and energy use. A useful approach must be affordable, explainable, and maintainable by the people running the equipment. Otherwise, its global reach will remain limited.

Q: Where does supply-chain resilience fit into this picture?
Adhikary: Fragile processes create fragile supply chains. Repeated part failures can delay entire production networks. Design-for-manufacturing improvements can make production more consistent and alternative sourcing more realistic. My published work examines that connection in U.S. additive-manufacturing supply chains.

Q: You spent years maintaining power equipment. What did that experience teach you?
Adhikary: A small technical fault can quickly become a much larger operational problem. At power facilities, maintenance planning, equipment records, and timely troubleshooting were essential. The same principle applies globally: preventing downtime is often more valuable than reacting after production or essential services are interrupted.

Q: How should the world measure success in smarter manufacturing?
Adhikary: With outcomes, not headlines: fewer rejected parts, lower energy per usable product, safer equipment, shorter downtime, and more reliable deliveries. Results need independent testing and clear limitations. An algorithm matters only when workers trust it and factories can use it.

Q: What ultimately motivates your engineering work?
Adhikary: I want efficient, resilient manufacturing to be achievable beyond a handful of highly resourced plants. When engineering principles, credible data, and shop-floor expertise work together, we can conserve materials and energy while supporting dependable production. That is the kind of problem-solving that crosses borders.

Biography
Debasish Adhikary is a mechanical engineer at Williams and Davis Boiler and a former research assistant at Sam Houston State University. He holds mechanical engineering degrees from Lamar University and Khulna University of Engineering & Technology. His research spans additive manufacturing, energy efficiency, AI-assisted quality control, and resource management. Email: debasish06me@gmail.com