Company
DogWoodAI was established on November 9, 2022, to actualize the vision of AI autonomous manufacturing plants.
In manufacturing plants, product yield and quality significantly impact company revenue. DogWoodAI's flagship is a personalized AI solution meeting each client's distinct requirements and plant characteristics. The AI solution utilizes client data and domain knowledge for precise performance. Upon request, we offer process optimization, operating data review and safety analysis derived from data analysis.
The AI solution excels in manufacturing, especially in complex sectors. Crafted with a deep understanding of client production technology, it seamlessly integrates with their unique environment. It optimizes predictive accuracy and operating parameters, tailored for specific business contexts.
DogWoodAI has successfully developed an AI solution for real-time prediction of product yields in manufacturing plants, enhancing competitiveness by improving productivity and operational stability.
In a broader sense, DogWoodAI's AI solution has the potential to be a cornerstone in successfully realizing smart factories in the industrial manufacturing sector.
CEO
Chang-Ha Lee (M) is a Distinguished Professor in the Department of Chemical and Biomolecular Engineering at Yonsei University, Korea. He served as Vice-President of Planning and Management from 2018 to 2020 and has been serving a director of The Next-Generation Converged Energy Materials Research Center (CEMRC) since 2012. He also served as President of The Korean Institute of Chemical Engineers (KIChE) in 2022.
His research activities include: (i) Process intensification and simulation (ii) Multi-scale modelling using steady-state and dynamic simulation for units and processes (iii) Machine learning for chemical processes (model-based ML, physics-based ML, data-driven ML).
He has successfully supervised 47 PhD students and 108 MS students. He has authored/co-authored more than 350 publications in refereed international journals in 2023. He holds patents in the fields of adsorbents, reactors, and processes, and three technologies were transferred to companies. His works were recognized by many awards such as the PSE model-based innovation prize, award from prime minister of Korea, etc. He is a fellow of the Korean Academy of Science and Technology (KAST), a member of The National Academy of Engineering of Korea (NAEK) and a fellow of International Adsorption Society (AIS).
CTO
Min Oh (M) is a seasoned professional in the Department of Chemical & Biomolecular Engineering at Hanbat National University, Korea.
After obtaining his Ph.D. from Imperial College, UK, in 1993, he worked as a Technical Consultant at IC Consultant in UK from 1995 to 1996, and as a Chief Engineer at LG Engineering in Korea from 1996 to 1998.
Beyond academia, Professor Oh has actively contributed to shaping national policies and strategies. Serving as a Member of the National Science & Technology Council and the Presidential Committee on Balanced National Development from 2003 to 2005 reflects his dedication to broader societal advancements. His influence extended further as a Member of the Presidential Advisory Council on Science & Technology from 2005 to 2007.
His research activities include: (i) Fluid Dynamics: Crystallizer, Water Gas Shift Reactor (ii) Process Simulation and Optimization: Combined Cycle Power Plant, Gasifier (iii) Application of Machine Learning: Emission Gas Prediction, Hydrocracker AI (iv) Defense Science and Technology: Solid Propellant, Waste Explosive Incinerator
Industries
In the petrochemical industry, several licensors possess the original design technology for producing the same products worldwide. Consequently, many companies import this technology and apply it similarly. However, companies may modify their operation process based on the feed stock, market demand, and business policy.
Operation strategy, including consistency of quality, stability of the process, and efficiency of operation, significantly impacts the rate of return for oil refining and petrochemical companies. Nevertheless, rising labor costs, difficulties in supplying and deploying field engineers, and increasing competition from rival companies require higher operational competitiveness.
To address this challenge, understanding and predicting the operating conditions of the process and the real-time process situation is crucial. However, price fluctuations of raw material and product material according to market conditions changes result in significant errors in operation forecasts, causing serious losses. To minimize these errors, machine learning-based software for process prediction will be developed and provided as a solution for processes with highly volatile raw materials and operational instability.
In particular, Korean oil refineries are an example of a manufacturing industry with high process-instability due to their use of various crude oils compared to firms in other countries. Moreover, the petrochemical industry involves diverse raw material suppliers, and the unit price of raw material is highly volatile. It is not possible to predict future data with scientific formulas alone, making the demand for machine learning software high. The team plans to develop and provide tailor-made machine learning software to meet this demand.
Previous attempts to optimize feed stock through AI took into account the situation of the crude oil market and energy efficiency, but they did not consider the effect of the operating conditions and circumstances of the process on products. Due to the extreme complexity of many processes, physical formulas alone cannot predict various products, thus increasing the demand for machine learning software. DogWoodAI aims to meet the demand of each company by providing customized machine learning software.
Customers
DogWoodAI works in collaboration with customer companies to verify the short-term, mid-term, and long-term applications of AI & ML solutions in the field. Currently, the team is working with an oil refinery company and a chemical company to apply the tailor-made ML solutions for long-term applications.
It is worth noting that many oil refinery and chemical companies across the world operate using similar processes, but with different operation modes that are adapted to the circumstances of each country and market. Moreover, a single plant in these industries can produce a variety of products by operating multiple processes that are connected to each other.
Therefore, by verifying the tailor-made AI & ML solution for a specific process, DogWoodAI can build an ML platform that can be used for similar processes across the world with customized tuning. This would result in enormous market scalability, considering the related worldwide industries.