Seminar by
Emre Nadar
Department of Industrial Engineering
Bilkent UniversityWe study the problem of project selection and resource allocation in a multi-stage new product development (NPD) process with stage-dependent resource constraints. We model the problem asan infinite-horizon Markov Decision Process, specifically under the discounted cost criterion. Each NPD project undergoes a different experiment at each stage of the NPD process; these experiments generate signals about the true nature of the project. Experimentation times are independent and exponentially distributed. After each experiment, beliefs about the outcome of each project are updated according to a Bayesian rule. A project thus becomes differentiated through its signals, and its category is determined by all its available signals. The state of the system is described by the number of projects in each category. A control policy specifies, given the system state, how to utilize the resources at each stage, i.e., the projects (i) to experiment with at each stage, and (ii)to terminate.
We characterize the optimal control policy according to a new strategy, state-dependent non-congestive promotion (SDNCP), for two different special cases of the general problem: (a) when there are multiple uninformative experiments; or (b) when there is a single informative experiment and projects are not terminated. An SDNCP policy implies that, at each stage, it is optimal to advance the project with the highest expected reward to the next stage if and only if the number of projects in each successor category is less than a state-dependent threshold. We further find that threshold values decrease in a non-strict sense as a later stage becomes more congested or as an earlier stage becomes less congested. These analytical findings uncover the role congestion plays in optimal policies. Our computational results support the outstanding performance of SDNCP as a heuristic for the general NPD problem.
Biographical Sketch
Emre Nadar is currently an Assistant Professor of Industrial Engineering at Bilkent University. He received his Ph.D. in Operations Management from the Tepper School of Business at Carnegie Mellon University in 2012. His research interests include Markov decision processes, queuing theory, supply chain management, and ustainable operations. He received several awards, including the first place award from the POMS Supply Chain College student paper competition and the Gerald L. Thompson doctoral dissertation award from Carnegie Mellon University. He was a finalist in both the INFORMS George Nicholson student paper competition and the MSOM student paper competition.
