My research interests are in design and control of production systems. In particular, I’m interested in developing methods for efficiently using partial information signals emitted from the production system in controlling the material release decisions.
Production systems consist of a multitude of machines and buffers connected in a network whose state information can be used for controlling the release of material into different parts of the system. Various control policies, e.g. the CONWIP and the Kanban policies, utilize predefined structures for using this information. Optimizing a control policy for every possible information signal can improve the performance of the system but it is often computationally prohibitive. As a middle ground solution, the set of information signals can be summarized into a smaller set with a manageable size. My research is partly focused on addressing the different aspects of this process, including identifying the signals, summarizing them and optimizing the policy parameters based on them, by using tools such as joint-simulation and optimization and machine learning.
Khayyati, S. and B. Tan (2020): Data-driven control of a production system by using marking-dependent threshold policy. International Journal of Production Economics, 226, 107607.
Tan, B. and S., Khayyati (2021): Supervised learning based approximation method for single-server open queuing networks with correlated interarrival and service times. International Journal of Production Research, DOI: 10.1080/00207543.2021.1887536
Khayyati, S. and B. Tan (2021): A machine learning approach for implementing marking-dependent production control policies. International Journal of Production Research. DOI: 10.1080/00207543.2021.1910872
Khayyati, S. and B. Tan (2021): A lab-scale manufacturing system environment to investigate data-driven production control approaches. Journal of Manufacturing Systems, 60, 283-297.
Khayyati, S. and B. Tan (2021): Supervised-learning-based approximation method for multi-server queueing networks under different service disciplines with correlated interarrival and service times. International Journal of Production Research. DOI: 10.1080/00207543.2021.1887536
Khayyati, S. and B. Tan (2021): A machine learning approach for implementing data-driven production control policies. 40th Congress on Operations Research/
Khayyati, S. and B. Tan (2019): Data-driven control of a production system with partially observable correlated arrival processes. SMMSO 2019, Goslar, Germany, June 2019.
18th INFORMS Applied Probability Society Conference, Istanbul, Turkey, July 2015.