Research Area: Reinforcement Learning, Bayesian Inference, Stochastic Dynamic Programming, Optimal Control. Applications in Revenue Management Pricing and Dynamic Resource Allocation.

I am senior research scientist at Amazon. My work involves building control, estimation and optimization algorithms for Revenue Management and Supply Chain Optimization. I completed my Ph.D. in Operations Research from the School of Operations Research and Information Engineering at Cornell University in 2015 where I was advised by Prof. Mark Lewis. I completed my undergraduate degree in Mechanical Engineering from the Indian Institute of Technology, Delhi, India in 2003 and Masters degree specializing in Estimation and Control Theory from the State University of New York at Buffalo under the supervision of Prof. Tarunraj Singh.  

Research

The broad area of my research is in Stochastic Control and Estimation with applications in the area of dynamic resource allocation, revenue management and pricing. My thesis work involved developing energy efficient dynamic resource allocation polices for telecommunications and computing systems. I have also worked on robust control of mechanical systems and estimation and control algorithms for image guided radiation therapy. More recently my work has focused on sequential decision making policies for Revenue Management and Pricing in Transportation industry. I am particularly interested in studying the interaction between forecasting and dynamic optimization algorithms, and developing algorithms for combined Statistical Learning and Optimal Control. 

Education

• Ph.D., Operations Research, Cornell University, Ithaca, NY, 2015

• M.S., Operations Research, Cornell University, Ithaca, NY, 2013

• M.S., Mechanical Engineering, State University of New York, Buffalo, NY, 2009

• B.Tech., Mechanical Engineering, Indian Institute of Technology, Delhi, India, 2003

Publications

• Kumar, R., Boluki, S., Isler, K., Rauch, J., Walczak, D, A Machine Learning Based Framework for Robust Price Sensitivity Estimation  with Application to Airline Pricing. 2022 Arxiv; KDD 2023 Workshop - Causal Inference and Machine Learning in Practice; Paper (Selected as a spotlight presentation INFORMS RMP 2022) 

• Kumar, R., Wang, W., Simrin, A., Arunachalam, S.K., Guntreddy, B.R., Walczak, D. Competitive revenue management models with loyal and fully flexible customers, Journal of Revenue and Pricing Management, 2021.

• Kumar, R., Li, A., and Wang, W., Learning and Optimizing through Dynamic Pricing, Journal of Revenue and Pricing Management, 2017.

• Kumar, R., Dynamic Resource Management For Systems With Controllable Service Capacity, PhD Thesis Cornell University, 2015

• Kumar, R., Lewis, M.E., Topaloglu H., Dynamic service rate control for a single‐server queue with Markov‐modulated arrivals, Naval Research Logistics(NRL) 60 (8), 661-677, 2013.

• Kumar, R., Singh, T., Design of input shapers using modal cost for multi-mode systems, Automatica, 46 (3), 598-604, 2010.

• Kumar, R., Singh, T., Singla P., Modeling and uncertainty quantification of motion of lung tumors for image guided radiation therapy, American Control Conference (ACC), 2010, 1254-1259

• Kumar, R., Tumor motion prediction for image guided radiation therapy, MS Thesis State University of New York, 2009

Working Papers

• Kumar, R., Kaufman, D.L., Lewis, M.E., Dynamic Resource Management Policies for Parallel Queues with a Shared Server Pool

• Li, A., Kumar, R., Walczak, D., A Dynamic Pricing Model with Capacity Sharing for Airline Revenue Management

Presentations/Conferences

• Machine Learning based Framework for Robust Price-Sensitivity Estimation with Application to Airline Pricing ,  Euro Pricing and Revenue Management Workshop, University of Zurich,  August 2023

• Machine Learning based Framework for Robust Price-Sensitivity Estimation with Application to Airline Pricing ,  INFORMS Revenue Management and Pricing Conference, Chicago Booth School of Business, June 2022 (Selected as a spotlight presentation)

• Machine Learning meets Causal Inference: A Hybrid Framework for Price Sensitivity Estimation, AGIFORS RM Study Group, May 2022

• Machine Learning based Hybrid Framework for Airline Dynamic Pricing, 32nd Annual POMS, May 2022

• Using Competitive Pricing Data to Enhance Revenue Management and Pricing, 22nd IFORS Conference, Hanyang University Seoul, Korea, August 2021 

• Dynamic Pricing for Ancillaries using Reinforcement Learning, AGIFORS RM Study Group Conference, May 2021 

• Dynamic Pricing for Ancillaries In Travel Industry using Reinforcement Learning, Applied Machine Learning Days Conference, Lausanne, Switzerland, January 2020 [Video]

• Dynamic Pricing for Ancillaries in Travel Industry, INFORMS Annual Meeting, Seattle, WA, October 2019

• Assortment Optimization for Airline Ancillary Revenue Management, EURO 2019 Conference, Dublin, Ireland, June 2019

• Degrees of Information Awareness in Revenue Management, INFORMS Revenue Management and Pricing Conference, Stanford GSB, CA, June 2019

• Dynamic Pricing and Learning for Airline Revenue Management, SIAM Annual Meeting 2018, Portland, OR, July 2018 

• Dynamic Programming Decomposition Approaches to the Network Pricing Optimization , POMS 29th Annual Conference, Houston, TX, May 2018

• A Dynamic Pricing Model with Capacity Sharing for Airline RM, INFORMS Annual Meeting, Houston, TX, October 2017

• Price Learning and Optimization for Airline RM, 21st IFORS triennial conference. Quebec City, Canada, July 2017 

• Overbooking under dynamic and static policies for network, 21st IFORS triennial conference. Quebec City, Canada, July 2017

• A Dynamic Pricing Model with Capacity Sharing on a Network for Airline RM, POMS 28th Annual Conference, Seattle, WA, May 2017

• Dynamic Pricing and Learning In Airline Revenue Management, INFORMS Annual Meeting, Nashville, TN, November 2016

• Dynamic Pricing and Classless Revenue Management, INFORMS RMP Conference, New York, NY, June 2016

• Overbooking Under Dynamic and Static Policies, INFORMS RMP Conference, New York, NY, June 2016

• Dynamic Resource Management for Parallel Queues with Shared Pool of Flexible Servers, INFORMS Annual Meeting, San Francisco, CA, November 2014.

• Optimal Allocation Policy for Parallel Queues with a Shared Pool of Servers, INFORMS Annual Meeting, Minneapolis, MN, October 2013.

• Dynamic Service Rate Control for a Single Server Queue with Markov Modulated Arrivals, Canadian Operations Research Society (CORS) Annual Meeting, Niagara Falls, Canada, June 2012.

• Dynamic Service Rate Control for a Single Server Queue with Markov Modulated Arrivals, INFORMS Annual Meeting, Charlotte, NC, November 2011.

• Modeling and uncertainty quantification of motion of lung tumors for image guided radiation therapy, American Control Conference, Baltimore, MD, June 2010.

Professional Activities

• Session Chair, INFORMS Annual Meeting, 2017, 2018, 2019

• Session Chair, 21st Conference of IFORS, Quebec City, Canada, July 2017

• Secretary, INFORMS Houston Chapter, 2017

• Referee, European Journal of Operations Research , 2016, 2017; OR Spectrum, 2017; Journal of Revenue and Pricing Management, 2016, 2017; IEEE American Control Conference , 2011